EMC is required for biogenesis of Xport‐A, an essential chaperone of Rhodopsin‐1 and the TRP channel
2021; Springer Nature; Volume: 23; Issue: 1 Linguagem: Inglês
10.15252/embr.202153210
ISSN1469-3178
AutoresCatarina J Gaspar, Ligia Vieira, Cristiana C. Santos, John C. Christianson, Dávid Jakubec, Kvido Střı́šovský, Colin Adrain, Pedro Domingos,
Tópico(s)Genetics, Aging, and Longevity in Model Organisms
ResumoArticle17 December 2021free access Source DataTransparent process EMC is required for biogenesis of Xport-A, an essential chaperone of Rhodopsin-1 and the TRP channel Catarina J Gaspar Catarina J Gaspar orcid.org/0000-0002-5295-6165 Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa (ITQB-NOVA), Oeiras, Portugal Membrane Traffic Lab, Instituto Gulbenkian de Ciência (IGC), Oeiras, Portugal Search for more papers by this author Lígia C Vieira Lígia C Vieira Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa (ITQB-NOVA), Oeiras, Portugal Search for more papers by this author Cristiana C Santos Cristiana C Santos orcid.org/0000-0003-3896-3446 Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa (ITQB-NOVA), Oeiras, Portugal Search for more papers by this author John C Christianson John C Christianson orcid.org/0000-0002-0474-1207 Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, Botnar Research Centre, University of Oxford, Oxford, UK Search for more papers by this author David Jakubec David Jakubec Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences, Prague, Czech Republic Search for more papers by this author Kvido Strisovsky Kvido Strisovsky orcid.org/0000-0003-3677-0907 Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences, Prague, Czech Republic Search for more papers by this author Colin Adrain Corresponding Author Colin Adrain [email protected] orcid.org/0000-0001-7597-4393 Membrane Traffic Lab, Instituto Gulbenkian de Ciência (IGC), Oeiras, Portugal Patrick G Johnston Centre for Cancer Research, Queen's University, Belfast, UK Search for more papers by this author Pedro M Domingos Corresponding Author Pedro M Domingos [email protected] orcid.org/0000-0003-4523-5631 Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa (ITQB-NOVA), Oeiras, Portugal Search for more papers by this author Catarina J Gaspar Catarina J Gaspar orcid.org/0000-0002-5295-6165 Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa (ITQB-NOVA), Oeiras, Portugal Membrane Traffic Lab, Instituto Gulbenkian de Ciência (IGC), Oeiras, Portugal Search for more papers by this author Lígia C Vieira Lígia C Vieira Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa (ITQB-NOVA), Oeiras, Portugal Search for more papers by this author Cristiana C Santos Cristiana C Santos orcid.org/0000-0003-3896-3446 Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa (ITQB-NOVA), Oeiras, Portugal Search for more papers by this author John C Christianson John C Christianson orcid.org/0000-0002-0474-1207 Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, Botnar Research Centre, University of Oxford, Oxford, UK Search for more papers by this author David Jakubec David Jakubec Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences, Prague, Czech Republic Search for more papers by this author Kvido Strisovsky Kvido Strisovsky orcid.org/0000-0003-3677-0907 Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences, Prague, Czech Republic Search for more papers by this author Colin Adrain Corresponding Author Colin Adrain [email protected] orcid.org/0000-0001-7597-4393 Membrane Traffic Lab, Instituto Gulbenkian de Ciência (IGC), Oeiras, Portugal Patrick G Johnston Centre for Cancer Research, Queen's University, Belfast, UK Search for more papers by this author Pedro M Domingos Corresponding Author Pedro M Domingos [email protected] orcid.org/0000-0003-4523-5631 Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa (ITQB-NOVA), Oeiras, Portugal Search for more papers by this author Author Information Catarina J Gaspar1,2, Lígia C Vieira1,6, Cristiana C Santos1, John C Christianson3, David Jakubec4, Kvido Strisovsky4, Colin Adrain *,2,5,† and Pedro M Domingos *,1,† 1Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa (ITQB-NOVA), Oeiras, Portugal 2Membrane Traffic Lab, Instituto Gulbenkian de Ciência (IGC), Oeiras, Portugal 3Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, Botnar Research Centre, University of Oxford, Oxford, UK 4Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences, Prague, Czech Republic 5Patrick G Johnston Centre for Cancer Research, Queen's University, Belfast, UK 6Present address: Center for Genomics and Systems Biology, New York University Abu Dhabi, Abu Dhabi, United Arab Emirates † These authors contributed equally to this work as senior authors *Corresponding author. Tel: +44 028 9097 2700; E-mail: [email protected] *Corresponding author. Tel: +351 21 446 9322; E-mail: [email protected] EMBO Reports (2022)23:e53210https://doi.org/10.15252/embr.202153210 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions Figures & Info Abstract The ER membrane protein complex (EMC) is required for the biogenesis of a subset of tail anchored (TA) and polytopic membrane proteins, including Rhodopsin-1 (Rh1) and the TRP channel. To understand the physiological implications of EMC-dependent membrane protein biogenesis, we perform a bioinformatic identification of Drosophila TA proteins. From 254 predicted TA proteins, screening in larval eye discs identified two proteins that require EMC for their biogenesis: fan and Xport-A. Fan is required for male fertility in Drosophila and we show that EMC is also required for this process. Xport-A is essential for the biogenesis of both Rh1 and TRP, raising the possibility that disruption of Rh1 and TRP biogenesis in EMC mutants is secondary to the Xport-A defect. We show that EMC is required for Xport-A TMD membrane insertion and that EMC-independent Xport-A mutants rescue Rh1 and TRP biogenesis in EMC mutants. Finally, our work also reveals a role for Xport-A in a glycosylation-dependent triage mechanism during Rh1 biogenesis in the endoplasmic reticulum. Synopsis Xport-A, an essential chaperone for two multi-pass membrane proteins (Rhodopsin-1 and TRP), is a new client of the ER membrane protein complex (EMC). The data suggest that the EMC shapes the membrane proteome not only directly via its client proteins, but also indirectly by mediating the insertion of essential co-factors of certain multi-pass membrane proteins. The ER membrane protein complex (EMC) is required for biogenesis and membrane insertion of Xport-A. EMC independent Xport-A can rescue Rhodopsin-1 (Rh1) and TRP biogenesis defects in EMC3 mutant clones. ER resident, hyperglycosylated Rh1 TMD1-5 accumulates in Xport-A and EMC6 mutants. Introduction Membrane proteins comprise ~30% of the eukaryotic proteome and confer many essential functions to biological membranes (Wallin & Von Heijne, 1998; Fagerberg et al, 2010). These proteins contain hydrophobic transmembrane domains (TMDs) that must be inserted into the membrane of the endoplasmic reticulum (ER) through evolutionarily conserved molecular machineries. The vast majority of membrane proteins are inserted into the ER membrane through the co-translational pathway, in which proteins containing signal peptides or TMDs are recognised by the signal recognition particle as they emerge from the ribosomes. Eventually, the ribosome-nascent chain complex is delivered to the Sec61 translocon, where membrane insertion takes place (Guna & Hegde, 2018). The co-translational insertion pathway is the predominant mechanism for membrane protein insertion, but it is unable to deal with the biogenesis of a specific class of membrane proteins called tail anchored (TA) proteins, which lack a signal peptide and contain a single TMD at their C-terminus (Kutay et al, 1993). Consequently, their TMD remains shielded inside the ribosomal tunnel until the termination codon is reached, and TMD recognition can only occur post-translationally (Hegde & Keenan, 2011). Accordingly, TA proteins utilise a dedicated conserved TMD recognition complex (TRC) pathway, which facilitates the targeted release of these proteins into the ER in eukaryotic cells, in a post-translational manner (Stefanovic & Hegde, 2007). The crucial component of this pathway is the ATPase TRC40 (or in yeast Get3), which captures and shields TMDs of TA proteins (Stefanovic & Hegde, 2007; Schuldiner et al, 2008; Mateja et al, 2015), until they are released to a receptor complex composed of WRB-CAML (yeast Get1-Get2), which mediates TMD insertion into the ER membrane (Mariappan et al, 2011; Stefer et al, 2011). Structural and biochemical studies have shown that this pathway displays a preference for TMDs of high hydrophobicity (Wang et al, 2010; Rao et al, 2016; Guna et al, 2018). Recently, the ER membrane protein complex (EMC) was identified as an insertase for TA proteins with TMDs of moderate to low hydrophobicity (Guna et al, 2018; Tian et al, 2019; Volkmar et al, 2019). The EMC is a highly conserved, multi-subunit protein complex, with nine subunits in mammals (Wideman, 2015). The EMC was initially described in yeast in a high throughput genetic screen for genes required for protein folding (Jonikas et al, 2009) and its mammalian counterpart was later identified as part of the interaction network of the ER-associated protein degradation machinery (Christianson et al, 2011). The EMC was subsequently shown to serve as an insertase for specific polytopic membrane proteins that contain a signal anchor sequence (SAS), including a subset of G-protein coupled receptors (GPCR; Chitwood et al, 2018). A variety of experiments suggested that the EMC coordinates the insertion of the first TMD of the GPCRs, after which subsequent TMD insertions are EMC-independent and require the Sec61 translocon, via a "handing-off" mechanism that remains to be fully elucidated (Chitwood et al, 2018). The structure of both the human and yeast EMC have recently been determined using cryo-electron microscopy (cryo-EM) (Bai et al, 2020; Miller-Vedam et al, 2020; O'Donnell et al, 2020; Pleiner et al, 2020) and a model has been proposed for EMC-mediated co- and post-translational insertion of client proteins (Pleiner et al, 2020). According to this model, a captured client protein is directed towards the membrane by the flexible cytosolic loop of EMC3, the insertase subunit of EMC; the EMC then presumably reduces the energetic cost of insertion by inducing a local thinning of the membrane and by arranging polar and positively charged residues within the bilayer. The client would then dissociate from EMC3, encountering EMC1's β propellers, which could act as a scaffold for co-factor recruitment (Bai et al, 2020; Pleiner et al, 2020). The requirement for the EMC in the biogenesis of some TA proteins and the first TMD of some polytopic membrane proteins containing a SAS has been dissected in vitro, leading to a rationalisation of how EMC can act mechanistically. However, the EMC has also been shown to be required for the biogenesis of multi-pass membrane proteins enriched with "challenging" TMDs (Shurtleff et al, 2018). Indeed, the majority of proteins found to date to be affected by loss of EMC are neither TA nor SAS-containing membrane proteins, suggesting that much remains to be understood about EMC specificity for its client proteins (Shurtleff et al, 2018; Tian et al, 2019) These include the ABC transporter Yor1 in yeast (Louie et al, 2012; Lakshminarayan et al, 2020), acetylcholine receptor in C. elegans (Richard et al, 2013), rhodopsins (Taylor et al, 2005; Satoh et al, 2015; Hiramatsu et al, 2019; Xiong et al, 2020), the transient receptor potential (TRP) channel in Drosophila (Satoh et al, 2015), ABCA1 in mice (Tang et al, 2017) and mutant connexin32 (Coelho et al, 2019). Furthermore, EMC loss of function has also been associated with defects in phospholipid trafficking (Lahiri et al, 2014; Janer et al, 2016), cholesterol homeostasis (Volkmar et al, 2019), autophagosome formation (Li et al, 2013; Shen et al, 2016), viral pathogenesis (Bagchi et al, 2016, 2020; Savidis et al, 2016; Barrows et al, 2019; Lin et al, 2019), neurological defects (Harel et al, 2016) and male fertility (Zhou et al, 2018). Although this diversity in phenotypes is still an area of investigation, several of these examples impact unrelated membrane proteins with multiple TMDs (Chitwood & Hegde, 2019; Volkmar & Christianson, 2020) and many of these candidate EMC clients are neither TA nor SAS-containing proteins. An alternative possibility to EMC acting as an insertase for a broader range of TMD protein topologies is that some of these proteins may be "indirect" or "secondary clients" of the EMC, for example, proteins whose biogenesis or stability depends on a direct EMC client protein. In this study, we interrogated the Drosophila proteome in an effort to bioinformatically identify TA proteins that could have a dependency on the EMC for their biogenesis. Using a Drosophila larval eye imaginal disc assay, we identified two EMC clients: fan (farinelli), which controls sperm individualisation (Ma et al, 2010) and Xport-A (exit protein of rhodopsin and TRP-A) (Rosenbaum et al, 2011; Chen et al, 2015b), an essential chaperone for the biogenesis of both Rhodopsin-1 (Rh1) and the TRP (Transient Receptor Potential) channel, and their targeting to the rhabdomere, the light sensitive compartment of the photoreceptors. Interestingly, the biogenesis of Rh1 and TRP has been shown to be deficient in EMC mutant clones (Satoh et al, 2015; Hiramatsu et al, 2019; Xiong et al, 2020). We generated a mutant of Xport-A (Xport-A4L) whose biogenesis proceeds independently of the EMC. Xport-A4L is able to rescue the expression of Rh1 and TRP in EMC mutant tissue, suggesting that the latter proteins are not direct clients of the EMC but rather, depend on EMC indirectly via Xport-A. Overall, our results suggest that EMC is required for sperm and photoreceptor differentiation in Drosophila, due to EMC's role in the biogenesis of fan and Xport-A, respectively. Crucially, our data establish that EMC impacts the biogenesis of some multi-pass membrane proteins indirectly, by governing insertion of the cofactors they require for assembly and deployment. This paradigm expands the potential governance of the EMC to a greater portion of the membrane proteome. Results EMC is required for the biogenesis of a subset of tail-anchored (TA) proteins The EMC has been shown to function as an insertase for TA proteins with TMDs of moderate to low hydrophobicity (Guna et al, 2018), but the identities of its clients have not been fully determined. To ascertain the range of EMC-dependent TA proteins, we began by screening the Drosophila melanogaster proteome for all possible TA proteins. TA proteins are defined by their cytosolic N-terminal domain that is anchored to the lipid bilayer by a single hydrophobic TMD proximal to the C-terminus (Borgese et al, 2003). We bioinformatically interrogated the proteome using prediction algorithms for: signal peptide, TMD and topology (Käll et al, 2005) as well as subcellular localisation (Almagro Armenteros et al, 2017) (Fig 1A). This analysis yielded a total of 254 candidate TA proteins in the Drosophila proteome, with predicted membrane localisations (Figs 1A and EV1B, Dataset EV1). We subsequently characterised the distribution of different biophysical features such as TMD length, TMD hydrophobicity, tail length and charge (Fig EV1C–F). These analyses show that within the list of predicted TA proteins, those that are ER-localised tend to have higher TMD hydrophobicity and length than their mitochondrial and peroxisomal counterparts (Fig EV1C–F), which is consistent with published predictions for both human (Costello et al, 2017) and Arabidopsis (Kriechbaumer et al, 2009) proteomes. Figure 1. EMC is required for the biogenesis of a subset of TA proteins A. The Drosophila proteome was screened to identify TA proteins using the PolyPhobius (Käll et al, 2005) algorithm within the TOPCONS (Tsirigos et al, 2015) web server. Proteins with a single TMD within 35 AAs or less upstream of the C terminus and lacking a signal peptide or mitochondrial targeting peptide were selected for further interrogation. The DeepLoc (Almagro Armenteros et al, 2017) predictor of protein sub-cellular localisation was used to eliminate soluble proteins from the list. These procedures yielded 254 proteins, which were scored for their tail charge and hydrophobicity using the "transmembrane tendency" score of Zhao & London (Zhao & London, 2006). B. A subset of the predicted TA proteins was screened by overexpression in Drosophila larval eye imaginal discs containing clones of a previously isolated EMC3 mutant allele, EMC3Δ4 (Satoh et al, 2015). TA protein predicted membrane localisation is shown to the left: G (Golgi apparatus), ER (endoplasmic reticulum), C (cytoplasmic membrane), N (nucleus), M (mitochondria) and P (peroxisome). The ratio of fluorescence intensity for the tested TA proteins in EMC3 homozygous mutant cells over WT cells (EMC3Δ4/WT) was measured and plotted. Proteins were classified into four groups according to the ratio measured: strongly reduced (ratio < 0.5; bars in red), reduced (ratio 0.5–0.8; bar in pink), normal (ratio 0.8–1.2; bars in blue) and increased (ratio > 1.2; bars in dark blue). For quantification, at least three mutant patches were quantified per eye imaginal disc, and at least three eye imaginal discs derived from distinct flies were used (N = 3). Error bars correspond to standard deviation (SD). C–J. All panels show Drosophila third instar larval eye imaginal discs with eyeless-Flippase-induced clones of cells homozygous for EMC3Δ4 (Satoh et al, 2015), labelled by the absence of ubiGFP (green). The red channel shows ELAV, which labels the nuclei of photoreceptor cells. All UAS constructs were expressed under the control of GMR-GAL4. (C) Expression of UAS-Rh1 (4C5, in blue) and (D) Na+K+ATPase (A5-C, in blue) is strongly reduced in EMC3Δ4 homozygous mutant cells. (E) Expression of UAS-Xport-A-HA (anti-HA, in blue) and (F) UAS-fan-HA (anti-HA, in blue) is strongly reduced in EMC3Δ4 homozygous mutant cells. (G) Expression of UAS-rtv-HA (anti-HA, in blue) and (H) hid (anti-hid, in blue) is increased in EMC3Δ4 homozygous mutant cells. (I and J) Normal expression of two TA proteins in EMC3Δ4 homozygous mutant cells. (I) UAS-CG8814-HA (anti-HA, in blue) is a TA protein containing a TMD of low hydrophobicity (J) UAS-Syx13-HA is a TA protein with high hydrophobicity TMD. Scale bars represent 10 μm. Source data are available online for this figure. Source Data for Figure 1C–J [embr202153210-sup-0004-SDataFig1CJ.zip] Download figure Download PowerPoint Click here to expand this figure. Figure EV1. Characterisation of the predicted TA proteins A. Pie chart showing the distribution of subcellular localisation in tested TA proteins. TA candidates with all predicted sub-cellular membrane localisations were tested. The tested proteins had the following sub-cellular localisations: ER membrane (~30%), mitochondrion membrane (~30%), cell membrane (~13%), nuclear membrane (~8.7%), Golgi membrane (~8.7%) and peroxisome membrane (~4.3%). B. Pie chart showing the distribution of the cell localisation in predicted TA proteins. The predicted TA proteins were predominately localised to the ER membrane (35%), mitochondrial membrane (~20%) and cell membrane (~20%). Other subcellular localisations predicted were Golgi membrane (~13%), nuclear membrane (9%) and peroxisome membrane (~3%). C–F. Violin plots showing the distribution of biophysical feature such as (C) TMD length, (D) average TMD hydrophobicity (TMD hydrophobicity in Zhao & London score, averaged to the number of AAs in each TMD), (E) tail length and (F) tail charge in predicted TA proteins, categorised by different cellular localisation. The mid-line in the violin plots represents the median of the population and the extremities below and above the median correspond to the minimum and maximum values. Significance was determined by Kolmogorov–Smirnov test: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001. Download figure Download PowerPoint The TMDs of most predicted TA proteins (~67%) exhibited low hydrophobicity (Guna et al, 2018) (hydrophobicity value below 22 in the Zhao & London hydrophobicity scale (Zhao & London, 2006)). This proportion was maintained in most sub-cellular localisations including the ER, but was altered in the Golgi apparatus (75% proteins with high hydrophobicity), the peroxisome and mitochondria (100% of proteins with low hydrophobicity). Next, we selected 23 candidate TA proteins to test for EMC dependency by monitoring their expression in EMC3 mutant clones in Drosophila larval eye imaginal discs (Figs 1B, E–J, and EV1A). This model enables candidate TA protein expression levels in WT and adjacent EMC3 mutant territories to be compared side-by-side, within the same tissue. Of the 23 proteins tested, ~87% had low predicted TMD hydrophobicity, were from within all predicted sub-cellular membrane localisations and had similar distribution of localisation to the overall pool of predicted TA proteins (Fig EV1A and B, Appendix Table S1). Candidate TA proteins were overexpressed under the control of GMR-GAL4 in larval eye imaginal discs containing EMC3 mutant (EMC3Δ4 (Satoh et al, 2015)) clones. EMC3 is considered a "core" subunit of the complex, with mutations in EMC3 and other core subunits (EMC1, 2, 5 and 6) causing significant impairment of EMC functionality (Guna et al, 2018; Volkmar et al, 2019; Volkmar & Christianson, 2020). Candidate TA protein expression was assayed by immunofluorescence using antibodies raised against the respective proteins or, when no suitable antibodies existed, against an appended HA (Hemagglutinin) tag. Rh1 and the Na+K+ATPaseα subunit were also evaluated (Fig 1C and D), as their expression has been reported previously to be EMC dependent (Satoh et al, 2015; Hiramatsu et al, 2019; Xiong et al, 2020). We evaluated TA candidate expression by determining the ratio of fluorescence intensity between the signal associated with the candidate TA protein in EMC3 mutant clones versus non-mutant (WT) clones (EMC3Δ4/WT) (Fig 1B). TA candidate expression was classified into four groups according to the EMC3Δ4/WT fluorescence intensity ratio: strongly reduced (ratio < 0.5, bars in red), reduced (ratio 0.5–0.8, bar in pink), normal (ratio 0.8–1.2, bars in blue), and increased (ratio > 1.2, bars in dark blue). Rh1 (Fig 1C) and the Na+K+ATPaseα subunit (Fig 1D) exhibited strongly reduced expression in EMC3Δ4 mutant clones, consistent with previously published results in the adult/pupal eye (Satoh et al, 2015; Hiramatsu et al, 2019; Xiong et al, 2020). Expression of Dpck was reduced in EMC3Δ4 mutant clones while rtv (Fig 1G), hid (Fig 1H) and PIG-X showed increased expression. Notably, two TA candidates, Xport-A (Fig 1E) and fan (Fig 1F), showed strongly reduced expression in EMC3Δ4 mutant clones compared to WT. Both proteins are predicted to localise to the ER membrane and contain TMDs of low hydrophobicity. Of the 23 TA proteins tested, 17 exhibited normal ratios of expression, demonstrating that the GMR-GAL4 driven transcription of the candidate proteins is not affected by the loss of EMC function. Examples of TA proteins with TMDs of low hydrophobicity (CG8814 – Fig 1I) and high hydrophobicity (Syx13 – Fig 1J) showing normal EMC3Δ4/WT fluorescence intensity ratios could also be found. EMC is required for sperm differentiation in Drosophila Our screen identified the predicted TA protein fan, whose expression was defective in EMC3Δ4 homozygous mutant clones (Fig 1F). As fan reportedly plays an important role in sperm individualisation and male fertility (Ma et al, 2010) (Fig EV2A), we asked whether the EMC was also required for male fertility in Drosophila. We selected RNAi lines targeting different EMC subunits (EMC1, EMC3, EMC4, EMC5, EMC6, EMC7, EMC8/9) and mated adult males in which individual EMC subunits were knocked down, with wild-type virgin females (Fig EV2B and C). We counted the mean number of progeny obtained from these crosses and observed a statistically significant reduction for all RNAi lines of EMC5 and EMC6, and for one RNAi line of EMC1 (Fig EV2B). We analysed sperm vesicle size and observed that, in accordance with the reduced male fertility, all EMC5 and EMC6 RNAi lines tested exhibited a statistically significant reduction in sperm vesicles (Fig EV2D), which correlates with a reduction in sperm production (Ma et al, 2010; Chen et al, 2015a). Furthermore, the seminal vesicles of flies expressing EMC5_2 RNAi line appeared devoid of mature sperm, as the needle shaped nuclei of mature sperm were absent (Fig EV2E). Finally, we tried to address the cause of male fertility in an EMC5 RNAi line, observing that while staining with an antibody against active Caspase-3 was present in multiple cystic bulges (cb) in wild-type testis, this staining was deficient in EMC5 RNAi testis (Fig EV2F), demonstrating sperm differentiation defects, as previously shown (Arama et al, 2003). Click here to expand this figure. Figure EV2. EMC is required for sperm differentiation in Drosophila A. Box-and-whisker plot of progeny per female per day for fan RNAi lines. Thirty individual males of the represented RNAi lines were mated to wild-type virgin females in separate vials and the mean number of progeny for crosses of the different genotypes was counted. The mid-line in the box plots represents median progeny (per female/day), the box represents the 25th–75th percentiles and whiskers below and above the box indicate the sample range. One of the tested fan RNAi lines (fan_1) shows a highly significant reduction of progeny produced. Significance was determined by Welch's t-test: *P ≤ 0.05, **P ≤ 0.01, ****P ≤ 0.0001. B, C. Box-and-whisker plots of progeny per female per day for EMC RNAi lines. Thirty individual adult males of the represented RNAi lines were mated to wild-type virgin females in separate vials and the mean number of progeny for crosses of the different genotypes was counted. The mid-line in the box plots represents median progeny (per female/day), the box represents the 25th–75th percentiles and whiskers below and above the box indicate the sample range. All tested EMC5 and EMC6 RNAi lines, and one EMC1 RNAi line (EMC1_1) have a significant reduction of progeny produced. Significance was determined by Welch's t-test: ****P ≤ 0.0001. D. Box-and-whisker plot of seminal vesicle size measured for 15 males per RNAi line. The mid-line in the box plots represents the median and the box represents the 25th–75th percentiles. Whiskers below and above the box indicate the minimum and maximum seminal vesicle size. All tested EMC5 and EMC6 RNAi samples show a significant reduction in sperm production compared to controls. Significance was determined by Welch's t-test: ****P ≤ 0.0001. E. Detection of sperm by nuclear staining with DAPI in seminal vesicles. Boxed areas are magnified below. The characteristic needle-like nuclei of mature sperm can be seen in control samples (Control_2), but not when UAS-EMC5_2 RNAi is expressed (EMC5_2). Scale bars represent 50 μm. F. Immunostaining of testis with an antibody against active caspase-3 (red), showing presence of multiple cystic bulges (indicated by arrows →) in Control_2 (left panel). When UAS-EMC5_2 RNAi is expressed (right panel), cystic bulges are absent. Nuclear staining was performed with DAPI (blue). Scale bars represent 20 μm. Source data are available online for this figure. Download figure Download PowerPoint EMC3 is required for the expression of Xport-A, but not Xport-A4L Although the specificity of EMC for its clients remains to be fully delineated, the EMC exhibits a preference for TA proteins with low TMD hydrophobicity and/or containing polar/charged amino acid residues (Guna et al, 2018; Tian et al, 2019; Volkmar et al, 2019). As the Xport-A TMD has a reduced hydrophobicity due to the presence of one polar and multiple charged residues, an Xport-A TMD with increased hydrophobicity would be expected to relieve its EMC dependency. To that end, we performed site-directed mutagenesis on the Xport-A TMD to substitute the four most hydrophilic amino acids with leucine, creating the mutant Xport-A4L (Fig 2A). As hypothesised, whereas WT HA-Xport-A was dependent on the EMC (Fig 2B and G), expression of HA-Xport-A4L (Fig 2C and G) or Xport-A4L-HA (Fig 2D and G) no longer required EMC3 in larval eye imaginal discs. Importantly, the levels of the Na+K+ATPaseα control protein remained defective in EMC3 mutant clones (Fig 2D and G), confirming the presence of "bona fide" EMC3 mutant clones in the eye disc and demonstrating the specificity of the rescue effect of the Xport-A4L mutant for Xport-A targets, but not for EMC clients in general. A similar impact of the Xport-A4L mutant was observed in experiments carried out in adult retinas (Figs 2E–G and EV3A). Altogether, these results indicate that increasing the hydrophobicity of the Xport-A TMD renders its expression independent of the EMC and that tagging Xport-A and Xport-A4L at either the N- or C-terminus yields identical results. Figure 2. EMC3 is required for the expression of Xport-A, but not Xport-A4L A. The TMD of Xport-A was identified using TOPCONS (Tsirigos et al, 2015) and is in bold/underlined, together with the immediate flanking residues. Red re
Referência(s)