The endoplasmic reticulum membrane protein complex localizes to the mitochondrial - endoplasmic reticulum interface and its subunits modulate phospholipid biosynthesis in Trypanosoma brucei

Abstract

The endoplasmic reticulum membrane complex (EMC) is a versatile complex that plays a key role in membrane protein biogenesis in the ER. Deletion of the complex has wide-ranging consequences including ER stress, disturbance in lipid transport and organelle tethering, among others. Here we report the function and organization of the evolutionarily conserved EMC (TbEMC) in the highly diverged eukaryote, Trypanosoma brucei. Using (co-) immunoprecipitation experiments in combination with mass spectrometry and whole cell proteomic analyses of parasites after depletion of select TbEMC subunits, we demonstrate that the TbEMC is composed of 9 subunits that are present in a high molecular mass complex localizing to the mitochondrial-endoplasmic reticulum interface. Knocking out or knocking down of single TbEMC subunits led to growth defects of T. brucei procyclic forms in culture. Interestingly, we found that depletion of individual TbEMC subunits lead to disruption of de novo synthesis of phosphatidylcholine (PC) or phosphatidylethanolamine (PE), the two most abundant phospholipid classes in T. brucei. Downregulation of TbEMC1 or TbEMC3 inhibited formation of PC while depletion of TbEMC8 inhibited PE synthesis, pointing to a role of the TbEMC in phospholipid synthesis. In addition, we found that in TbEMC7 knock-out parasites, TbEMC3 is released from the complex, implying that TbEMC7 is essential for the formation or the maintenance of the TbEMC.

Fig 1. Predicted domain structure of individual subunits of the TbEMC complex drawn to scale.

Predicted transmembrane domains are indicated in red and unique domains are indicated in colors. PQQ, pyrrolo-quinoline quinone repeat; TPR like, tetratricopeptide-like helical domain; DUFs, domains of unknown function; RAB, Rab5-interacting protein family; Carb bd, carbohydrate-binding like fold.

Fig 2. Growth of T. brucei procyclic forms.

A, B. Growth of TbEMC7-KO (A) and TbEMC10-KO (B) parasites (red symbols) was recorded in parallel to growth of parental parasites (black symbols). C-E. Parasites were cultured in the absence (black symbols) or the presence (red symbols) of tetracycline to maintain or ablate, respectively, expression of TbEMC1 (C), TbEMC3 (D) or TbEMC8 (E). All data points represent mean values ± standard deviations from 3 independent experiments. For some data points, the size of the symbol is larger than the standard deviation.

Fig 3. Expression of in situ-tagged TbEMCs.

Protein extracts from parasites expressing HA-tagged (A) or cMyc-tagged (B) TbEMC proteins were analyzed by SDS-PAGE and immunoblotting using anti-HA (A) or anti-cMyc (B) antibodies, in combination with the corresponding secondary antibodies. Molecular mass markers (in kDa) are indicated in the left margins.

Fig 4. Localization of the TbEMCs.

Parasites expressing in situ-tagged TbEMCs were fixed and analyzed by immunofluorescence microscopy using the corresponding first and second antibodies to visualize the TbEMC proteins (in green; middle panels; the numbers refer to the individual TbEMC proteins) and in combination with antibodies against BiP or ATOM40 (in red; left panels). DNA was stained with DAPI in the merge (in blue; right panels). Scale bar = 5 μm.

Fig 5. Interactions between individual TbEMCs.

A-G. Proteins from parasites co-expressing differently tagged TbEMC proteins (indicated below each panel) were immunoprecipitated using TbEMC5-cMyc (A-D) or TbEMC3-HA (E-G) as bait. Proteins from the input (1% of total), washed beads, and both immunoprecipitations (IP; with 3–10% of the sample being applied to detect the bait protein and 90–97% to detect the putative interaction partner) were analyzed by SDS-PAGE and immunoblotting using anti-HA or anti-cMyc antibodies as indicated, in combination with the corresponding second antibodies. Heat shock protein 70 (Hsp70) was used as negative control and visualized by specific antibodies. H. Schematic representation of co-immunoprecipitated TbEMC subunits using TbEMC5 and TbEMC3 as baits. Double headed arrows indicate reciprocal co-immunoprecipitation. I. Parasites co-expressing different pairs of in situ-tagged TbEMCs (indicated in the individual panels) were fixed and analyzed by immunofluorescence microscopy using the corresponding first and second antibodies to visualize the two TbEMC proteins (in red for cMyc; in green for HA; left two panels). DNA was stained with DAPI in the merge (in blue; right panels). Scale bar = 5 μm.

Fig 6. Analysis of TbEMC high molecular mass complexes.

A, B. Crude membrane fractions from parasites expressing in situ-tagged TbEMC proteins (indicated below the panels), or from parental cells (control) were analyzed by native PAGE and visualized after immunoblotting using anti-HA (A) or anti-cMyc (B) antibodies. Molecular mass markers (in kDa) are indicated in the left margins.

Fig 7. Analysis of TbEMC integrity.

A-D. Crude membrane fractions from control cells (A; EMC3; C: EMC8) and TbEMC7-KO parasites (A: EMC7 KO + EMC3; C: EMC7 KO + EMC8) expressing in situ-tagged TbEMC3-cMyc or TbEMC8-cMyc, respectively, were analyzed by native PAGE (A, C) and immunoblotting using anti-cMyc antibody and visualized by the corresponding second antibody. The samples were also analyzed by SDS-PAGE and immunoblotting using anti-cMyc antibody (B) or an antibody against T. brucei ADP/ATP carrier 1 (AAC1; as loading control) (D) and visualized by the corresponding second antibodies. Molecular mass markers (in kDa) are indicated in the left margins. E. Control cells (EMC3) and TbEMC7-KO parasites (EMC7 KO + EMC3) expressing in situ-tagged TbEMC3-cMyc were fixed and analyzed by immunofluorescence microscopy using anti-cMyc antibody and the corresponding second antibody. Scale bar = 5 μm.

Fig 8. TbEMCs at the mitochondrial-ER interface.

A. Overview of the purification method for mitochondrial vesicles⁴⁷ and the mitochondrial outer membrane⁴⁸. B-C. SDS-PAGE/immunoblot analyses of the four fractions whole cell (WC), cytosol and microsomal ER fraction (cyto and ER), crude (cruMi) and pure mitochondria (puMi). Abbreviations indicated in black italics in the sketch are analyzed in panel (B), fractions in white text are analyzed in panel (C). D. Protein abundance profiles of TbEMC subunits and control protein TbErp1, reproduced from Niemann et al., 2013⁴⁸. E. Parasites co-expressing TbEMC3-HA and TbPSS2-cMyc were fixed and analyzed by immunofluorescence microscopy using the corresponding first and second antibodies. DNA was stained with DAPI in the composite (in blue; right panels). Scale bar = 5 μm.

Fig 9. Volcano plots of a SILAC-based quantitative MS analysis of whole cell extracts of uninduced and induced RNAi cell lines.

EMC3-RNAi (A, B) and EMC8-RNAi (C, D) parasites grown in SILAC media were harvested after 3 (A, C) or 3.5 (B, D) days of induction. Proteins were quantified in three independent biological replicates, with the mean log₂ of normalized (norm.) ratios plotted against the −log₁₀ P value (two-sided t-test). The horizontal dashed line indicates a significance level of p = 0.05. The vertical dotted lines mark proteins with a 1.5-fold change in abundance compared to control uninduced cells. Known TbEMC subunits are represented as red dots. Non-TbEMC proteins downregulated more than 1.5-fold are labeled with accession numbers.

Fig 10. Analysis of [³H]-PC and [³H]-PE formation in TbEMC-depleted parasites.

Parasites cultured in the absence (black) or the presence (red) of tetracycline to maintain or ablate, respectively, TbEMC expression (A-F), or parasites lacking individual TbEMC proteins (G-J), were labeled with [³H]-choline or [³H]-ethanolamine, in combination with [³H]-inositol as internal standard, to measure de novo formation of phosphatidylcholine (PC) or phosphatidylethanolamine (PE), respectively. After extraction, phospholipids were separated by TLC and the amounts of radioactivity in individual peaks were quantified by radioisotope scanning. The data points are from independent experiments and are expressed relative to control values (means ± standard deviations). The asterisks indicate statistical significance (P<0.05).