In vitro formation of the endoplasmic reticulum occurs independently of microtubules by a controlled fusion reaction

Abstract

We have established an in vitro system for the formation of the endoplasmic reticulum (ER). Starting from small membrane vesicles prepared from Xenopus laevis eggs, an elaborate network of membrane tubules is formed in the presence of cytosol. In the absence of cytosol, the vesicles only fuse to form large spheres. Network formation requires a ubiquitous cytosolic protein and nucleoside triphosphates, is sensitive to N-ethylmaleimide and high cytosolic Ca(2+) concentrations, and proceeds via an intermediate stage in which vesicles appear to be clustered. Microtubules are not required for membrane tubule and network formation. Formation of the ER network shares significant similarities with formation of the nuclear envelope. Our results suggest that the ER network forms in a process in which cytosolic factors modify and regulate a basic reaction of membrane vesicle fusion.

Figure 1

Formation of membrane networks in vitro requires cytosol. (A) The light membrane fraction was analyzed in a fluorescence microscope after staining with a hydrophobic fluorescent dye. (B) Washed light membranes were incubated for 60 min with cytosol and an energy regenerating system and visualized as in A. (C) As in B but without cytosol. (D) Washed light membranes were allowed to attach to the glass surface of a microscope slide and visualized as in A. (E) Washed light membranes were incubated with cytosol for 60 min, and analyzed as in D. (F) As in E but without cytosol. Bars: (A–C) 40 μm; (D–F) 10 μm.

Figure 2

EM analysis of the membrane networks. (A) The membrane network was analyzed by EM after attachment to an EM grid and negative staining with uranyl acetate. (B) The light membrane fraction was analyzed by EM after thin sectioning. (C) The light membrane fraction was incubated for 60 min with cytosol and an energy regenerating system and analyzed as in B. Arrows indicate membrane-bound ribosomes. Bars, 500 nm.

Figure 3

The network contains ER membranes. Membrane networks were formed and attached to the glass surface of a microscope slide. (A) After fixation, the network was incubated with an antibody against the ER marker protein TRAPα, followed by a fluorescently labeled secondary antibody. (B) The anti-TRAPα antibody was preincubated with the peptide against which it was raised and used as in A. (C and D) A membrane network was labeled with both rhodamine-labeled ConA and Oregon green–labeled WGA and analyzed with rhodamine (C) and fluorescein (D) channels. (E and F) A membrane network was labeled with both the anti-TRAPα antibody and the nuclear pore antibody mAb414. The inset in F shows staining with mAb414 of the nuclear envelope of in vitro assembled nuclei. Bar, 10 μm.

Figure 4

Formation of the ER network does not require microtubules or actin filaments. ER formation was carried out with light membranes and cytosol in the presence of 200 μM nocodazole (A), 100 μM colchicine (B), tubulin-depleted cytosol (C), tubulin-depleted cytosol plus 200 μM vinblastine (D), or 200 μM latrunculin A (E). All pictures show surface-attached membrane networks. Bar, 10 μm.

Figure 5

Quantitation of ER network formation in the absence of microtubules or actin filaments. (A) ER network formation was carried out for 1 h with egg extracts in the presence (black columns) or absence (grey columns) of the indicated inhibitors. Quantitation was done by counting and averaging the number of three-way junctions of surface-attached membrane networks from 10 fields (65 × 65 μm²). The numbers below the bars indicate different experiments. (B) The time course of network formation was determined in the presence or absence of the indicated inhibitors. (C) Microtubules formed in egg extract were visualized with fluorescently labeled taxol. (D) Egg extract was pretreated with 100 μM colchicine before the addition of fluorescently labeled taxol. Bar, 20 μm.

Figure 6

ER network formation in the presence of microtubules. (A) ER formation was carried out for 1 h with egg extracts in flow chambers that were either precoated with microtubules (grey columns) or had no microtubules (black columns). When no microtubules were present, nocodazole was also added to the extract. Quantitation was done by counting and averaging the number of three-way junctions from 10 fields (31 × 31 μm²). The number of three-way junctions is given per 65 × 65 μm² as in Fig. 5. (B) ER formation was carried out and quantitated as in A, but with a mixture of light membranes and cytosol instead of an unfractionated extract.

Figure 7

A protein is required for maintenance of the membrane tubules. (A) Proteinase K was added with the staining solution to preformed membrane networks in bulk solution. The pictures shown were taken after a further 15 min incubation. (B) A control without proteinase K is shown. Bar, 20 μm.

Figure 8

ER network formation is inhibited by NEM, GTPγS, or ATPγS. (A) The light membrane fraction was pretreated with 10 mM NEM, and after removal of unreacted NEM, used in an ER formation reaction with cytosol. (B) GTPγS was added at a concentration of 1 mM to a reaction with light membranes and cytosol. (C) ATPγS was added at a concentration of 1 mM to a basic fusion reaction with washed light membranes and buffer. (D) Basic fusion reaction without ATPγS. (E) ATPγS was added at a concentration of 1 mM to an ER network formation reaction with washed membranes and cytosol. (F) ER formation reaction without ATPγS. All pictures show membranes in bulk solution. Bar, 20 μm.

Figure 9

Time course of ER network formation. An ER formation reaction with light membranes and cytosol was incubated for the indicated time periods and analyzed in solution. Note the early appearance of small aggregated membrane structures (10 min), followed by the formation of membrane tubules (20–60 min) that eventually form a dense network (90 min). Bar, 40 μm.

Figure 10

The ER network is sensitive to high cytosolic Ca²⁺ concentrations. (A) A BHK cell line that stably expresses a fusion of the ER protein Sec61β to the green fluorescent protein was treated with 10 μM of the Ca²⁺ ionophore ionomycin for 10 min at 37°C. (B) Untreated cells are shown. (C) 2 mM Ca²⁺ was added with the staining solution to membrane networks preformed in vitro. Images were taken in bulk solution after 15 min. Arrows indicate large membrane vesicles. (D) A control without the addition of Ca²⁺ is shown. Bars: (A and B) 20 μm; (C and D) 40 μm.

Figure 11

Cytosolic Ca²⁺ inhibits ER network formation. 200 μM Ca²⁺ (A), no Ca²⁺ (B), 100 μM ionomycin (C), or 100 μM ionomycin plus 10 mM EGTA (D) were added to an ER formation reaction containing light membranes and cytosol, and incubated for 1.5 h. Bar, 20 μm.

Figure 13

Effects of Ca²⁺ chelators on ER network formation. 2 mM 5,5′-dibromo BAPTA (A and B) or 2 mM BAPTA (C and D) were added to an ER formation reaction and incubated for 45 or 90 min. Bar, 40 μm.

Figure 12

Different stages of nuclear envelope formation. Xenopus sperm was added to a mixture of light membranes and cytosol and the formation of a nuclear envelope around the chromatin was followed over time after staining with a hydrophobic, fluorescent dye with a fluorescence microscope (A and B, 10 min; C, 30 min; D, 60 min). In B–D, images from several focal planes were taken, deconvolved, and a three-dimensional picture was calculated. Note the network of membrane tubules and tubules that connect small patches of membranes on the chromatin surface after short incubations (A and arrows in B). Arrows in C point to areas on the chromatin surface that are not yet covered by the nuclear envelope. D shows a completely enclosed nuclear envelope. Bar, 10 μm.

Figure 14

ER network formation is strongly reduced in mitotic extracts. Formation of the ER network in interphase extract (− cyclin B1, gray columns) and in a mitotic extract, obtained by addition of nondegradable cyclin B1 Δ90 (+ cyclin B1, black columns), was quantitated by counting the number of three-way junctions as in Fig. 5 A. Shown are results from two experiments.