Genome-Wide Identification of Alternative Splicing Events That

SHORT REPORT Genome-wide identification of alternative splicing events that regulate protein transport across the secretory pathway Alexander Neumann1, Magdalena Schindler1, Didrik Olofsson1, Ilka Wilhelmi2, Annette Schürmann2 and Florian Heyd1,*

ABSTRACT Nevertheless, the regulation of protein transport and the adaptation to Alternative splicing (AS) strongly increases proteome diversity and different cellular requirements remain incompletely understood. functionality in eukaryotic cells. Protein secretion is a tightly controlled Alternative splicing (AS), which strongly increases proteome process, especially when it occurs in a tissue-specific and diversity and often leads to severely different interaction surfaces (Pan differentiation-dependent manner. While previous work has focussed et al., 2008; Yang et al., 2016), has been widely overlooked as a on transcriptional and post-translational regulatory mechanisms, the regulator of membrane trafficking. AS has been shown to react to impact of AS on the secretory pathway remains largely unexplored. various external stimuli in a highly dynamic manner (Heyd and Here, we integrate results from a published screen for modulators of Lynch, 2011; Preußner et al., 2017), and would thus be perfectly protein transport and RNA-Seq analyses to identify over 200 AS events suited to control protein secretion in response to changing cellular as secretion regulators. We confirm that splicing events along all environments. Although such a connection has been suggested, this in silico stages of the secretory pathway regulate the efficiency of membrane was mainly based on predictions (Blue et al., 2018). We have SEC16A trafficking using morpholino and CRISPR/Cas9 experiments. We previously described how AS of exon 29 increases the furthermore show that these events are highly tissue-specific and efficiency of the early secretory pathway (Wilhelmi et al., 2016) but mediate an adaptation of the secretory pathway during T-cell activation this remains one isolated example. Here, we identify over 200 and adipocyte differentiation. Our data substantially advance the candidate AS events that could act to modulate the secretory pathway understanding of AS functionality, add a new regulatory layer to a in a tissue-specific and activation- and differentiation-dependent fundamental cell biological process and provide a resource of manner. Our genome-wide approach, together with selected alternative isoforms that control the secretory pathway. experimental validations provides evidence for a global control of protein secretion by AS. KEY WORDS: Alternative splicing, Secretory pathway, Protein secretion, Tissue-specific adaptation RESULTS AND DISCUSSION Potential transport-regulating AS events identified in a INTRODUCTION genome-wide fashion After biosynthesis in the endoplasmic reticulum (ER), proteins are To identify regulators of the secretory pathway, several RNAi screens transported to their destination via the secretory pathway. Transport have been performed (Farhan, 2015). One of these screens used the processes are highly flexible and adapt to new requirements, for VSVG reporter to monitor protein transport upon knockdown (KD) instance during differentiation or after activation (Farhan and of 22,000 human genes (Simpson et al., 2012). In this screen, four Rabouille, 2011; McCaughey and Stephens, 2018). This adaptation RNA-binding proteins (RBPs) that regulate AS and whose KD led to has been studied in regard to differential gene expression (Dunne a strong inhibition of secretion were identified (Table S1): et al., 2002; Coutinho et al., 2004; Schotman et al., 2009), changes in HNRNPA1, PTBP1, RBM27 and SRSF1 (Fig. 1A). As these membrane morphology and dynamics (Forster et al., 2006; Guo and RBPs have no known direct function in protein secretion, we Linstedt, 2006; Farhan et al., 2008), and altered activity of kinases and considered that they could exert an indirect effect, through controlling phosphatases (Farhan et al., 2010). Furthermore, expression of AS of components of the secretory pathway, which then changes different paralogs of the COPII coat has been suggested to be cargo flux (Fig. 1B). We therefore used published KD RNA-Seq involved in conferring tissue-specific functionality. For example, the datasets for these RBPs (see Table S1 for accession numbers) to COPII inner coat SEC24A, SEC24B, SEC24C and SEC24D proteins globally determine their effect on AS. We propose that the impact of have been reported to show differential interaction patterns (Adolf these RBPs on the secretory pathway is mediated by a network of et al., 2016) and knockout of SEC24C leads to tissue-restricted shared AS events, and thus generated an overlap of alternative neurological phenotypes in mice (Wang et al., 2018), implying a role isoforms that showed differential splicing in the single KDs compared for paralog expression in controlling tissue-specific functionality. to control datasets. We propose that splice variants that are regulated by at least three of the four RBPs are potential secretion modulators, 1Freie Universität Berlin, Institute of Chemistry and Biochemistry, Laboratory of RNA and henceforth call their formation a secretion-related AS event Biochemistry, Takustrasse 6, 14195 Berlin, Germany. 2Department of Experimental (Fig. 1C). Meta-analysis of these events, which mostly involve Diabetology, German Institute of Human Nutrition Potsdam-Rehbrücke (DIfE), cassette exons, indicates that they have a modulatory effect on protein 14558 Nuthetal, Germany. function instead of regulating total gene expression, as they lie almost *Author for correspondence ([email protected]) exclusively within the coding sequence and consist of fewer nonsense-mediated decay (NMD)-inducing exons than is seen for A.N., 0000-0002-8512-6643; M.S., 0000-0003-3363-1864; I.W., 0000-0002- 0287-8402; F.H., 0000-0001-9377-9882 the the skipped exons that are regulated by any of these RBPs (i.e. the RBP background, defined as all AS events that show differential

Received 22 January 2019; Accepted 9 March 2019 splicing upon any of the KDs; Fig. S1). When performing gene Journal of Cell Science

1 SHORT REPORT Journal of Cell Science (2019) 132, jcs230201. doi:10.1242/jcs.230201

Fig. 1. Genome-wide identification of secretion-modulating AS events. (A) Venn diagram of proteins implicated in secretion phenotypes upon KD and alternative splicing (AS)-regulatory RBPs. (B) Model of how RBPs indirectly regulate cargo flux by modulating AS. (C) Venn diagram of RBP KD RNA-Seq AS analyses. We define exons that are differentially spliced in at least three KDs as candidate AS events that might impact on protein secretion (in green, termed secretion-related AS events). Number of respective AS events outlined. (D) Biological process GO terms for secretion-related AS event genes (all significantly enriched terms shown), with all genes that change splicing in any of the KDs defined as the background (RBP background). (E) Network of the secretion-related AS events describing their function according to RefSeq gene annotations. Proteins of the secretory pathway are color-coded based on their localization. ontology (GO) analysis, only three terms for biological processes pathways. Second, mitochondrial proteins could impact on were found to be significantly enriched, all of which fit the hypothesis mitochondria-related traffic. Finally, there are several chaperones, a that the secretion-related AS events play a role in membrane group of proteins which have been shown to influence protein trafficking (Fig. 1D). We then grouped the AS-controlled proteins by secretion as well (Roth et al., 2012). Closer inspection of the proteins function (Fig. 1E). As expected, a large cluster is directly connected known to be a direct part of the trafficking machinery revealed that to the secretory pathway or to cytoskeletal organizers that can provide they are localized in all compartments along the secretory pathway scaffolding functions for vesicle transport. Additionally, we find (Fig. 1E). While we focussed our analysis on AS events controlled phosphoinositide homeostasis and ARF signalling, which are known by the four RBPs, it is interesting to note that all four RBPs have to regulate vesicular trafficking (Ooms et al., 2009) and proteins been reported to shuttle between the nucleus and the cytoplasm involved in kinase and phosphatase signalling, which can post- (Kamath et al., 2001; Iervolino et al., 2002; Kavanagh et al., 2005; translationally control trafficking proteins (Farhan et al., 2010). Apart Twyffels et al., 2011). A further impact of one or several of these from these targets, there are several groups that may have an indirect RBPs in controlling protein secretion, for example, by controlling effect on secretion. First, transcriptional, post-transcriptional and stability or translation of mRNAs encoding for secretion-associated translational modulators can change gene and protein expression of proteins, is thus conceivable but remains to be experimentally various targets, which can be further controlled by degradation addressed. Journal of Cell Science

2 SHORT REPORT Journal of Cell Science (2019) 132, jcs230201. doi:10.1242/jcs.230201

Protein transport is impaired upon KD of specific assay was subsequently performed by staining against GFP 1 h splicing regulators after biotin addition. Although we did not observe a significant To validate the functionality of the secretion-related AS events, difference between control and either the MBNL3, SRSF6 or we employed the RUSH system (Boncompain et al., 2012) in HNRNPA1 KDs, cells treated with siRNA against PTBP1, which a GFP–GPI reporter is transported from the ER to the RBM27 and SRSF1 (denoted siPTBP1, siRBM27 and siSRSF1, plasma membrane (PM) after addition of biotin. Upon arrival of respectively) showed a substantial and highly significant decrease the reporter at the PM, GFP will be located on the extracellular in GFP surface staining (Fig. 2D,E). The lack of an influence for side of the cell where its amount can be quantified by staining the HNRNPA1 KD (Simpson et al., 2012) may be due to the use against GFP without permeabilizing the cell (Fig. 2A). A time of a different reporter in our study, which could point to a role in course experiment in HEK293T cells and the quantification cargo selectivity. However, we also notice that HNRNPA1 has of antibody-stained (PM-localized) GFP per cell is shown in the lowest number of shared splicing targets among the four Fig. 2B. To verify that expression levels of HNRNPA1, PTBP1, RBPs, as the largest overlap in our analysis are the targets RBM27 and SRSF1 had an effect on protein transport in our controlled by PTBP1, RBM27 and SRSF1 (Fig. 1C). For further assay, we performed KDs of these proteins and of two further validations, we therefore focussed on events that were controlled RBPs (MBNL3 and SRSF6), as a control (Fig. 2C). The RUSH by these three RBPs.

Fig. 2. Validation of a secretion phenotype upon RBP KD. (A) The RUSH system principle with extracellular (cell surface) GFP staining as read-out for transport efficiency. (B) Time course of a RUSH assay experiment with images and quantification. 25 (0 min)/ 21 (15 min)/ 53 (30 min)/ 62 (60 min) cells were quantified; P-values are not indicated. (C) Validation of the RBP KD. The percentage expression of each targeted gene, normalized to GAPDH and control siRNA is shown as the mean±s.d. of three (siMBNL3, siSRSF6 and siPTBP1) or four (siHNRNPA1, siRBM27, siSRSF1) independent replicates. (D) Quantification of the RUSH assay in control and RBP KD cells. A total of 70 (siCtrl), 130 (siMBNL3), 77 (siSRSF6), 280 (siHNRNPA1), 187 (siPTBP1), 125 (siRBM27) and 185 (siSRSF1) cells were quantified. (E) Representative images corresponding to experiments as in D. Blue DAPI stain, green GFP, magenta anti-GFP staining (surface GFP), yellow overlap of green and magenta. Scale bars: 10 µm. ns, not significant, P>0.05; **P<0.01, ***P<0.001. For B and D, the box represents the 25–75th percentiles, and the median is indicated. The whiskers show the 10–90th percentiles, and outliers are indicated. Journal of Cell Science

3 SHORT REPORT Journal of Cell Science (2019) 132, jcs230201. doi:10.1242/jcs.230201

Manipulation of all tested splicing events leads to impaired performed RUSH assays on these cells and indeed observed a highly protein transport significant decrease in GFP reporter transported to the PM for all To validate the influence of individual splicing events on secretion, candidates (Fig. 3B,C). we selected four targets that are directly involved in different parts of To independently validate the observed effect, we used CRISPR/ the secretory pathway (Fig. 1E): for the early secretory pathway, Cas9 to generate isoform-specific knockouts (KOs). We selected the we selected SEC31A, which is part of the outer coat of COPII vesicles alternative microexon 19 in OCRL, as it stands out both because the (Gürkan et al., 2006) and SEC22C, which is a homolog of MO-induced splicing change from ∼25% inclusion to complete the SNARE protein Sec22 in yeast (Tang et al., 1998; Yamamoto exclusion led to a drastic transport defect and as this is achieved by et al., 2017). As examples for splicing events affecting Golgi exclusion of only eight amino acids. We generated four independent and post-Golgi components of the secretory pathway, we selected OCRL exon 19 KO cell lines and validated them both on the OCRL, which acts at the trans-Golgi and later compartments as a genomic and transcriptomic level (Figs S2A, S3D). Cell lines phosphoinositide phosphatase (De Matteis et al., 2017) and EXOC7 showed expression of only the exclusion isoform with the overall (also known as EXO70), which is part of the exocyst complex expression level remaining largely unchanged (Fig. S2B). We then involved in targeting vesicles to the PM (He and Guo, 2009). We performed the RUSH assay and again observed a highly significant used splice-site blocking morpholinos (MOs) to manipulate the AS reduction in the amount of surface GFP for all clones and that was in of these targets without altering the endogenous gene expression the same range as for the OCRL exon 19 MO-treated cells, further level (Fig. 3A), thereby recapitulating changes in exon exclusion validating the MO approach (Fig. 3E). These data together validate observed upon knockdown of the RBPs (Table S1). We then our bioinformatics approach, as all tested splicing events indeed

Fig. 3. Validation of a secretion phenotype upon manipulation of secretion-related AS events. (A) Representative radioactive RT-PCR and quantifications of the percentage of AS variants seen for cells treated for 48 h with MOs against SEC31A exon 24b, SEC22C exon 7x, OCRL exon 19 and EXOC7 exon 7, respectively. In addition to the investigated exon 7, there is an alternative 3′ splice site in EXOC7 exon 8, leading to a longer (8L) and shorter (8S) isoform. Shown are mean±s.d. of three (SEC31A, SEC22C, and OCRL) or four (EXOC7) independent replicates. (B) Representative microscopy images of cells treated with control MOs and MOs against the targeted exons from A and subjected to a RUSH assay. Blue, DAPI stain; green, GFP; magenta, anti-GFP cell surface staining. Yellow represents the overlap of green and magenta. Scale bars: 10 µm. (C) Quantification of the results of RUSH assay performed with cells pre-treated with MOs for 36 h before transfection of the RUSH plasmid. A total of 29 (ctrl), 88 (SEC31A), 84 (SEC22C), 114 (OCRL) and 199 (EXOC7) cells were quantified. (D) Validation of CRISPR/Cas9-mediated knockout of OCRL exon 19 for four independent cell lines on transcriptomic level by performing radioactive RT-PCR. (E) Quantification of the RUSH assay performed with wild-type and OCRL exon 19 knockout cells (line #1–#4). A total of 64 (wt, wild-type), 76 (#1), 25 (#2), 46 (#3) and 37 (#4) cells were quantified. ns, not significant, P>0.05; **P<0.01, ***P<0.001. For C and E, the box represents the

25–75th percentiles, and the median is indicated. The whiskers show the 10–90th percentiles, and outliers are indicated. Journal of Cell Science

4 SHORT REPORT Journal of Cell Science (2019) 132, jcs230201. doi:10.1242/jcs.230201 controlled protein transport efficiency. This strongly increases the task in mature adipocytes is the rapid shuttling of the glucose confidence that our group of secretion-related AS events is genuine, transporter GLUT4 (also known as SLC2A4) from post-Golgi vesicles thereby adding a new regulatory layer to the secretory pathway and to the PM in response to insulin and the recycling of the receptor substantially increasing the number of alternative isoforms with (Stöckli et al., 2011). A major and more dynamic adaptation is known cellular functionality. therefore required for the post-Golgi trafficking machinery, which is achieved by AS. These data together strongly argue for a global role of Transport-regulating AS events are regulated in a tissue- and AS in controlling the efficiency of the secretory pathway in a tissue- differentiation-specific manner specific manner, as well as in dynamic settings such as during To address the physiological significance of the connection differentiation and activation (Fig. S3E). between AS and protein transport, we investigated whether the secretion-related AS events act in a tissue-, differentiation- or Conclusions activation-dependent manner. To this end, we initially used In summary, we combine data from a genome-wide screen for RNA-Seq data from various human organs to calculate inclusion modulators of protein secretion with knockdown RNA-Seq datasets to levels for the secretion and the RBP background events. We indeed discover over 200 AS events that, based on our validations using the observed a tissue-specific usage for a larger proportion of the RUSH system, are high-confidence secretion regulators. Our analysis secretion-related AS events in comparison to what was seen for the shows that AS is involved in regulating all stages of the secretory RBP background (Fig. 4A; Fig. S3A). This visual impression is pathway (Fig. 4H). We furthermore show that secretion-regulating AS supported by two measures of variability. First, in comparison to the events are used in diverse biological contexts to adapt the secretory RBP background, the secretion-related AS events displayed more pathway to tissue-specific or differentiation- and activation-dependent variable splicing in all two-tissue comparisons (i.e. a higher percentage requirements (Fig. 4H; Fig. S3E). Although individual AS events of differentially spliced events, Fig. 4B). Second, the strongest have been reported to play a role in membrane trafficking (Wilhelmi percentage spliced in difference (dPSI) was on average also larger for et al., 2016; Valladolid-Acebes et al., 2015; Blue et al., 2018), we the secretion-related AS events (Fig. 4C). This strongly points to a show, in a system-wide manner, that the secretory pathway is global role of AS in adapting the secretory pathway to tissue-specific regulated by AS at all stages. Our findings thus add an additional layer requirements. Next, we turned to two cellular systems where cells with of complexity to the regulation of the secretory pathway. In addition, basal secretory requirements differentiate into a cell type with higher the dynamic nature of these splicing events in various biological secretory load (Fig. 4D): T-cell activation and differentiation of pre- contexts provides an important step towards understanding adipocytes into adipocytes. We used RNA-Seq datasets from primary differentiation-specific control of protein secretion. AS might help human CD4+ T-cells and human SGBS adipocytes at pre- and post- the cell to adapt the secretory pathway in an intermediate timeframe, differentiation to analyse gene expression and alternative splicing. supplementing the very fast modulations of kinases and phosphatases When analysing differential AS, we found significant overlaps with and the slower effects of transcription. Despite clear variations in the secretion-related AS events in both sets of data (Fig. 4E,F), from cargo load and specificity in different cells and tissues, the molecular which we validated three targets each (Fig. S3B,C), implying that AS basis for these distinct adaptations remains largelyenigmatic. Our data can adapt protein secretion upon activation and differentiation. Of provide evidence that cell-type-specific adaption of protein secretion note, the vast majority of secretion-related AS events act is, at least partially, controlled by a network of splicing changes. Our independently of transcription, as ∼80% of the corresponding genes finding that several RBPs are involved in controlling this process is show an expression fold change smaller than two in both model consistent with splicing decisions being under combinatorial control systems (Table S2). Altered AS may be controlled by HNRNPA1, of several or many cis-andtrans-acting factors, and explains how PTBP1, RBM27 and SRSF1, which all show slightly increased different cell types can adapt their splicing patterns to the respective mRNA expression upon T-cell activation and a modest reduction requirements. The expression and activity of these RBPs can be during adipocyte differentiation (Table S2). However, the activity individually controlled to result in a variety of activities that is tailored of RBPs is extensively regulated at post-transcriptional and post- to the secretory requirement of the respective cell type. translational levels, which likely plays a determining role. We observed that while the T-cell overlap events lie within proteins MATERIALS AND METHODS found throughout the secretory pathway, the adipocyte overlap events Accession numbers, RNA-Seq analysis and post-analysis lie within proteins that mainly locate in post-Golgi compartments, Accession numbers and all results from analyses are listed in Tables S1 and which is also reflected in a GO term analysis (Fig. S3D). Additionally, S2. RNA-Seq analyses were essentially performed as previously described (Herdt et al., 2017). In short, reads were mapped to the hg38 human genome we found that components of the COPII machinery are upregulated using STAR version 2.5.3a (Dobin et al., 2013). For AS analyses, the during adipocyte differentiation but not during T-cell activation ‘mixture of isoforms’ (MISO) (Katz et al., 2010) version 0.5.3 with a (Fig. 4G; Table S2). This suggests that adipocytes use transcriptional custom-made annotation was used. Differential events are defined by having regulation in the early steps of protein secretion and AS in post-Golgi a minimum ΔPSI of 10% and a Bayes factor of greater than 5. Replicates compartments to adapt their secretory pathway, whereas T-cells rely were merged for analyses after mapping. Gene expression analysis was on AS to regulate secretion capacity during activation. This difference performed using DESeq2 (Anders et al., 2012). Further analyses were in the control of membrane trafficking may be explained by specific performed using custom Python scripts (available from the corresponding requirements of these cells after differentiation. Activated T-cells author upon request). GO term analyses were performed using the produce cytokines and cytotoxins at the ER and then transport them PANTHER classification system version 13.1 (https://pantherdb.org) out of the cell, using the whole secretory pathway, which happens in a (Mi et al., 2013). Network maps were generated using Cytoscape version 3.6.1 (Montojo et al., 2010). highly dynamic and temporally controlled manner (Huang et al., 2013). Adipocytes produce and export adipokines and also need to Cell culture, transfections and genome editing adapt their secretory machinery in this regard (Kuryszko et al., 2016), HEK293T cells were cultivated in high-glucose Dulbecco’s modified but this happens during a longer differentiation process that is more Eagle’s medium (DMEM; Biowest, Nuaillé, France) containing 10% fetal suitable for stable transcriptional changes. However, a fundamental calf serum (FCS; Biochrom, Berlin, Germany) and 1% penicillin/ Journal of Cell Science

5 SHORT REPORT Journal of Cell Science (2019) 132, jcs230201. doi:10.1242/jcs.230201

Fig. 4. Tissue- and differentiation-specific usage of secretion-related AS events to adapt the secretory pathway. (A) Tissue heatmap of secretion-related AS events based on RNA-Seq data from the Illumina bodymap. PSI, percentage spliced in. (B) For every possible two-way tissue comparison, the percentage of differential splicing events that show a significant difference was calculated. Shown are the secretion-related AS events and the RBP background. (C) Maximal PSI difference (ΔPSI) between any tissues per event for the secretion-related AS events and the RBP background. (D) Model of a cell with low secretory potential that differentiates into a cell with high secretory potential after a stimulus. (E,F) Overlap of T-cell activation-specific (E) and adipocyte differentiation-specific (F) AS events with the secretion-related AS events. P values calculated via hypergeometric distribution with all expressed alternative exons as background. (G) Gene expression analysis of the COPII component fold change (FC) from the RNA-Seq data for T cells (orange) and adipocytes (violet) pre- and post-differentiation. For B and C, the box represents the 25–75th percentiles, and the median is indicated. The whiskers show minimum to maximum values. ***P<0.001. (H) Proposed model of splicing-regulated secretion modulators. Proteins encoded by genes showing secretion-related AS events that are linked to the secretory pathwayareshown, with their position indicating their function. Validated events are marked in red and differentiation-specific events with an orange T and violet A for T-cell activation- and adipocyte differentiation-specific, respectively. For B, n=120 comparisons; for C, n= 7367 for RBP background, 228 for secretion events; for G, n=11.

streptomycin (Biowest) at 37°C and 5% CO2. These cells have been Roth, Karlsruhe, Germany) following the manufacturers’ instructions. MOs extensively used in our laboratory (Herdt et al., 2017; Preußner et al., 2017; were obtained from Gene Tools. They were transfected at a final Goldammer et al., 2018) and test negative for mycoplasma contamination on concentration of 3 µM using Endo-Porter following manufacturers’ a monthly basis. Plasmid transfections were performed using RotiFect (Carl instructions. For KDs, a pool of four siRNAs (Dharmacon, Lafayette, Journal of Cell Science

6 SHORT REPORT Journal of Cell Science (2019) 132, jcs230201. doi:10.1242/jcs.230201

Colorado, US) was used at final concentration of 10 nM. They were percentage of GFP surface staining, single cells were initially defined by transfected using HiPerFect (Qiagen, Venlo, Netherlands) according to the user input. Pixels were defined as GFP-positive (green value) or GFP manual. Genome-editing using CRISPR/Cas9 was performed as previously staining-positive (red value) if the respective intensity value was above a described (Wilhelmi et al., 2016). All sequences for MOs, siRNAs, guide certain threshold (75 for green, 65 for red fluorescence). The GFP surface sequences and genotyping primers are listed in the Table S3. staining was calculated as the number of double-positive pixels divided by Human SGBS cells were cultivated and differentiated as described the number of green-positive pixels multiplied by 100. previously (Wabitsch et al., 2001; Fischer-Posovszky et al., 2008). In brief, cells were seeded in DMEM/F12 containing 10% FCS at a density of 4×104 Statistics cells per well in six-well plates for 3 days to reach confluency. Differentiation If not mentioned otherwise, either one-sample or unpaired t-tests were used was started by application of a specific cocktail (2 μmol/l rosiglitazone, to calculate P values. P values are indicated by asterisks and explained in the 25 nmol/l dexamethasone, 0.5 mmol/l methyliso-buthylxantine, 0.1 μmol/l legend of each figure; n numbers in the legends represent biological cortisol, 0.01 mg/ml transferrin, 0.2 nmol/l triiodotyronin and 20 nmol/l replicates. P-value calculations were only performed if at least three human insulin) in DMEM/F12 without serum and albumin. Medium was biological replicate values were obtained. changed every 4 days (DMEM/F12 0.1 μmol/l cortisol, 0.01 mg/ml transferrin, 0.2 nmol/l triiodotyronin, and 20 nmol/l human insulin). Cells Acknowledgements were harvested in Qiazol 3 days after seeding (pre-adipocytes) and 14 days The authors thank the HPC Service of ZEDAT, Freie Universität Berlin, for post differentiation (adipocytes) for RNA extraction using the RNeasy computing time. We thank Franck Perez for generously providing the RUSH plasmid and Rainer Pepperkok for sharing unpublished information on the genome-wide MinElute Kit (Qiagen) according to manufacturers’ instructions. screen performed in his laboratory. We also thank members of the Heyd laboratory for discussions and comments on the manuscript. Constructs The RUSH plasmid consists of a KDEL ER hook and an EGFP-GPI Competing interests reporter, a gift from Franck Perez (available from Addgene, #65294). The authors declare no competing or financial interests.

RNA extraction, RT-PCR, qRT-PCR and genomic PCR Author contributions RNAs from primary human T-cells were prepared as described (Michel Conceptualization: A.N., F.H.; Methodology: A.N.; Software: A.N., D.O.; et al., 2014). RNA extraction, radioactive RT-PCR and Phosphor imager Validation: A.N., M.S.; Formal analysis: A.N., D.O.; Investigation: A.N., M.S., I.W., A.S.; Resources: F.H.; Data curation: A.N.; Writing - original draft: A.N., F.H.; quantification were performed as previously described (Wilhelmi et al., Writing - review & editing: A.N., M.S., I.W., A.S., F.H.; Visualization: A.N.; 2016). RNA was extracted using RNATri (Bio&Sell, Feucht, Germany). For Supervision: F.H.; Project administration: F.H.; Funding acquisition: F.H. reverse transcription, 1 µg RNA was used with a gene-specific reverse primer. Low-cycle number PCRs were performed with a radioactively Funding labelled forward primer, and products were separated by denaturing PAGE. This work was supported by the Deutsche Forschungsgemeinschaft (SFB958/A21 Imaging was performed using a Phosphoimager and gels quantified using to F.H.). ImageQuantTL version 8.1 (GE, Boston, Massachusetts, US). Quantitative reverse transcription PCR (qRT-PCR) was performed in technical Data availability duplicates in a 96-well format using an Absolute qPCR SYBR Green Mix All accession numbers for RNA-Seq experiments are available in Tables S1 and S2. (Thermo Fisher, Waltham, MA) on a Stratagene (San Diego, CA) Mx3000P Supplementary information machine. Mean values of the technical duplicates were used and expression Supplementary information available online at normalized to GAPDH. Non-radioactive genomic PCR products were http://jcs.biologists.org/lookup/doi/10.1242/jcs.230201.supplemental separated on an agarose gel. See figure legends for number of independent biological replicates. Primer sequences are provided in Table S3. References Adolf, F., Rhiel, M., Reckmann, I. and Nakano, A. (2016). Sec24C/D-isoform- RUSH assay and staining specific sorting of the preassembled ER-Golgi Q-SNARE complex. Mol. Biol. Cell The day before transfection, cells were seeded on precision coverslips 27, 2697-2707. doi:10.1091/mbc.e16-04-0229 Anders, S., Reyes, A. and Huber, W. (2012). Detecting differential usage of exons (0.17 mm thickness, Sigma) that had previously been coated with poly-L- 5 from RNA-seq data. Genome Res. 22, 2008-2017. doi:10.1101/gr.133744.111 lysine (Sigma) for 1 h. For the standard RUSH assay, 1.0×10 cells were Blue, R. E., Curry, E. G., Engels, N. M., Lee, E. Y. and Giudice, J. (2018). How seeded. At 24 h post seeding, cells were transfected with the RUSH plasmid, alternative splicing affects membrane-trafficking dynamics. J. Cell Sci. 131, and at 16 h post transfection, protein synthesis was stopped by addition of jcs216465. doi:10.1242/jcs.216465 cycloheximide (10 µg/ml final concentration, Carl Roth) and D-biotin (40 µM Boncompain, G., Divoux, S., Gareil, N., De Forges, H., Lescure, A., Latreche, L., final concentration, Sigma) was added to release the reporter. The assay was Mercanti, V., Jollivet, F., Raposo, G. and Perez, F. (2012). Synchronization of stopped, usually after 1 h unless indicated otherwise, by washing with ice- secretory protein traffic in populations of cells. Nat. Methods 9, 493-498. doi:10. cold PBS (Biowest) and fixation of cells using 4% formaldehyde (Carl Roth) 1038/nmeth.1928 Coutinho, P., Parsons, M. J., Thomas, K. A., Hirst, E. M. A., Saúde, L., Campos, solution in PBS for 10 min. After 1 h of blocking using 5% goat serum I., Williams, P. H. and Stemple, D. L. (2004). Differential requirements for COPI (Sigma) in PBS, primary rabbit anti-GFP antibody (Jamieson et al., 2015; transport during vertebrate early development. Dev. Cell 7, 547-558. doi:10.1016/ Invitrogen, Carlsbad, CA; cat. #A11122, batch #1891900, 1:200) was applied j.devcel.2004.07.020 for 2 h at room temperature. After washing, the secondary donkey Alexa De Matteis, M. A., Staiano, L., Emma, F. and Devuyst, O. (2017). The 5- Fluor 594-conjugated anti-rabbit-IgG antibody (Ram et al., 2017, Invitrogen, phosphatase OCRL in Lowe syndrome and Dent disease 2. Nat. Rev. Nephrol. 13, cat# A21207, batch# 1107500, 1:200) was applied for 1 h. Coverslips were 455-470. doi:10.1038/nrneph.2017.83 mounted on microscopy slides (Roth) with ProLong Gold antifade mountant Dobin, A., Davis, C. A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P., Chaisson, M. and Gingeras, T. R. (2013). STAR: ultrafast universal RNA-seq with DAPI (Thermo Fisher). Slides were stored at room temperature for 24 h aligner. Bioinformatics 29, 15-21. doi:10.1093/bioinformatics/bts635 and either directly imaged or stored at 4°C until imaging. When the RUSH Dunne, J. C., Kondylis, V. and Rabouille, C. (2002). Ecdysone triggers the 5 assay was combined with either MO or siRNA treatment, only 0.5×10 cells expression of Golgi genes in Drosophila imaginal discs via Broad-complex. Dev. were seeded to account for the additional day of treatment. MOs or siRNAs Biol. 245, 172-186. doi:10.1006/dbio.2002.0632 were transfected 32 h before the RUSH plasmid. Farhan, H. (2015). Systems biology of the secretory pathway: what have we learned so far? Biol. Cell 107, 205-217. doi:10.1111/boc.201400065 Farhan, H. and Rabouille, C. (2011). Signalling to and from the secretory pathway. Microscopy and analysis J. Cell Sci. 124, 669-669. doi:10.1242/jcs.086991 Fluorescence microscopy was performed on a Leica SP8 confocal Farhan, H., Weiss, M., Tani, K., Kaufman, R. J. and Hauri, H. P. (2008). Adaptation microscope. Image analysis was performed using custom Python scripts of endoplasmic reticulum exit sites to acute and chronic increases in cargo load.

(available from the corresponding author upon request). To calculate the EMBO J. 27, 2043-2054. doi:10.1038/emboj.2008.136 Journal of Cell Science

7 SHORT REPORT Journal of Cell Science (2019) 132, jcs230201. doi:10.1242/jcs.230201

Farhan, H., Wendeler, M. W., Mitrovic, S., Fava, E., Silberberg, Y., Sharan, R., chemokine expression in human T cells. J. Biol. Chem. 289, 13651-13660. doi:10. Zerial, M. and Hauri, H.-P. (2010). MAPK signaling to the early secretory pathway 1074/jbc.M113.526269 revealed by kinase/phosphatase functional screening. J. Cell Biol. 189, 997-1011. Montojo, J., Zuberi, K., Rodriguez, H., Kazi, F., Wright, G., Donaldson, S. L., doi:10.1083/jcb.200912082 Morris, Q. and Bader, G. D. (2010). GeneMANIA cytoscape plugin: Fast gene Fischer-Posovszky, P., Newell, F. S., Wabitsch, M. and Tornqvist, H. E. (2008). function predictions on the desktop. Bioinformatics 26, 2927-2928. doi:10.1093/ Human SGBS cells-A unique tool for studies of human fat cell biology. Obesity bioinformatics/btq562 Facts 1, 184-189. doi:10.1159/000145784 Ooms, L. M., Horan, K. A., Rahman, P., Seaton, G., Gurung, R., Kethesparan, Forster, R., Weiss, M., Zimmermann, T., Reynaud, E. G., Verissimo, F., D. S. and Mitchell, C. A. (2009). The role of the inositol polyphosphate 5- Stephens, D. J. and Pepperkok, R. (2006). Secretory cargo regulates the phosphatases in cellular function and human disease. Biochem. J. 419, 29-49. turnover of COPII subunits at single ER exit sites. Curr. Biol. 16, 173-179. doi:10. doi:10.1042/BJ20081673 1016/j.cub.2005.11.076 Pan, Q., Shai, O., Lee, L. J., Frey, B. J. and Blencowe, B. J. (2008). Deep Goldammer, G., Neumann, A., Strauch, M., Müller-Mcnicoll, M., Heyd, F. and surveying of alternative splicing complexity in the human transcriptome by high- Preußner, M. (2018). Characterization of cis -acting elements that control throughput sequencing. Nat. Genet. 40, 1413-1415. doi:10.1038/ng.259 oscillating alternative splicing. RNA Biol. 15, 1081-1092. doi:10.1080/15476286. Preußner, M., Goldammer, G., Neumann, A., Haltenhof, T., Rautenstrauch, P., 2018.1502587 Müller-Mcnicoll, M. and Heyd, F. (2017). Body temperature cycles control Guo, Y. and Linstedt, A. D. (2006). COPII-Golgi protein interactions regulate COPII rhythmic alternative splicing in mammals. Mol. Cell 67, 433-446.e4. doi:10.1016/j. coat assembly and Golgi size. J. Cell Biol. 174, 53-63. doi:10.1083/jcb.200604058 molcel.2017.06.006 Gürkan, C., Stagg, S. M., Lapointe, P. and Balch, W. E. (2006). The COPII cage: Ram, R., Brasch, H. D., Dunne, J. C., Davis, P. F., Tan, S. T. and Itinteang, T. Unifying principles of vesicle coat assembly. Nat. Rev. Mol. Cell Biol. 7, 727-738. (2017). The identification of three cancer stem cell subpopulations within doi:10.1038/nrm2025 moderately differentiated lip squamous cell carcinoma. Front. Surg. 4, 12. He, B. and Guo, W. (2009). The exocyst complex in polarized exocytosis. Curr. doi:10.3389/fsurg.2017.00012 Opin. Cell Biol. 21, 537-542. doi:10.1016/j.ceb.2009.04.007 Roth, S. D., Schüttrumpf, J., Milanov, P., Abriss, D., Ungerer, C., Quade-Lyssy, Herdt, O., Neumann, A., Timmermann, B. and Heyd, F. (2017). The cancer- P., Simpson, J. C., Pepperkok, R., Seifried, E. and Tonn, T. (2012). Chemical associated U2AF35 470A>G (Q157R) mutation creates an in-frame alternative 5′ chaperones improve protein secretion and rescue mutant factor VIII in mice with splice site that impacts splicing regulation in Q157R patients. RNA J. 23, hemophilia a. PLoS ONE 7, e44505. doi:10.1371/journal.pone.0044505 1796-1806. doi:10.1261/rna.061432.117 Schotman, H., Karhinen, L. and Rabouille, C. (2009). Integrins mediate their Heyd, F. and Lynch, K. W. (2011). DEGRADE, MOVE, REGROUP: signaling unconventional, mechanical-stress-induced secretion via RhoA and PINCH in control of splicing proteins. Trends Biochem. Sci. 36, 397-404. doi:10.1016/j.tibs. Drosophila. J. Cell Sci. 122, 2662-2672. doi:10.1242/jcs.039347 2011.04.003 Simpson, J. C., Joggerst, B., Laketa, V., Verissimo, F., Cetin, C., Erfle, H., Huang, J., Brameshuber, M., Zeng, X., Xie, J., Li, Q.-j., Chien, Y.-h., Valitutti, S. Bexiga, M. G., Singan, V. R., Hériché, J.-K., Neumann, B. et al. (2012). and Davis, M. M. (2013). A Single peptide-major histocompatibility complex Genome-wide RNAi screening identifies human proteins with a regulatory ligand triggers digital cytokine secretion in CD4+T Cells. Immunity 39, 846-857. function in the early secretory pathway. Nat. Cell Biol. 14, 764-774. doi:10.1038/ doi:10.1016/j.immuni.2013.08.036 ncb2510 Iervolino, A., Santilli, G., Trotta, R., Guerzoni, C., Cesi, V., Bergamaschi, A., Stöckli, J., Fazakerley, D. J. and James, D. E. (2011). GLUT4 exocytosis. J. Cell Gambacorti-Passerini, C., Calabretta, B. and Perrotti, D. (2002). hnRNP A1 Sci. 124, 4147-4159. doi:10.1242/jcs.097063 nucleocytoplasmic shuttling activity is required for normal myelopoiesis and BCR/ Tang, B. L., Low, D. Y. and Hong, W. (1998). Hsec22c: a homolog of yeast Sec22p ABL leukemogenesis. Mol. Cell. Biol. 22, 2255-2266. doi:10.1128/MCB.22.7. and mammalian rsec22a and msec22b/ERS-24. Biochem. Biophys. Res. 2255-2266.2002 Commun. 243, 885-891. doi:10.1006/bbrc.1998.8194 Jamieson, C., Lui, C., Brocardo, M. G., Martino-Echarri, E. and Henderson, Twyffels, L., Gueydan, C. and Kruys, V. (2011). Shuttling SR proteins: more than B. R. (2015). Rac1 augments Wnt signaling by stimulating -catenin-lymphoid splicing factors. FEBS J. 278, 3246-3255. doi:10.1111/j.1742-4658.2011.08274.x enhancer factor-1 complex assembly independent of -catenin nuclear import. Valladolid-Acebes, I., Daraio, T., Brismar, K., Harkany, T., Ögren, S. O., Hökfelt, J. Cell Sci. 128, 3933-3946. doi:10.1242/jcs.167742 T. G. M. and Bark, C. (2015). Replacing SNAP-25b with SNAP-25a expression Kamath, R. V., Leary, D. J. and Huang, S. (2001). Nucleocytoplasmic shuttling of results in metabolic disease. Proc. Natl Acad. Sci. USA 112, E4326-E4335. polypyrimidine tract-binding protein is uncoupled from RNA export. Mol. Biol. Cell doi:10.1073/pnas.1511951112 12, 3808-3820. doi:10.1091/mbc.12.12.3808 Wabitsch, M., Brenner, R. E., Melzner, I., Braun, M., Möller, P., Heinze, E., Katz, Y., Wang, E. T., Airoldi, E. M. and Burge, C. B. (2010). Analysis and design of Debatin, K.-M. and Hauner, H. (2001). Characterization of a human preadipocyte RNA sequencing experiments for identifying isoform regulation. Nat. Methods 7, cell strain with high capacity for adipose differentiation. Int. J. Obes. 25, 8-15. 1009-1015. doi:10.1038/nmeth.1528 doi:10.1038/sj.ijo.0801520 Kavanagh, S. J., Schulz, T. C., Davey, P., Claudianos, C., Russell, C. and Wang, B., Joo, J. H., Mount, R., Teubner, B. J. W., Krenzer, A., Ward, A. L., Rathjen, P. D. (2005). A family of RS domain proteins with novel subcellular Ichhaporia, V. P., Adams, E. J., Khoriaty, R., Peters, S. T. et al. (2018). The localization and trafficking. Nucleic Acids Res. 33, 1309-1322. doi:10.1093/nar/ COPII cargo adapter SEC24C is essential for neuronal homeostasis. J. Clin. gki269 Investig. 128, 3319-3332. doi:10.1172/JCI98194 Kuryszko, J., Sławuta, P. and Sapikowski, G. (2016). Secretory function of Wilhelmi, I., Kanski, R., Neumann, A., Herdt, O., Hoff, F., Jacob, R., Preuaÿner,̃ adipose tissue. Pol. J. Vet. Sci. 19, 441-446. doi:10.1515/pjvs-2016-0056 M. and Heyd, F. (2016). Sec16 alternative splicing dynamically controls COPII McCaughey, J. and Stephens, D. J. (2018). COPII-dependent ER export in animal transport efficiency. Nat. Commun. 7, 12347. doi:10.1038/ncomms12347 cells: adaptation and control for diverse cargo. Histochem. Cell Biol. 150, 119-131. Yamamoto, Y., Yurugi, C. and Sakisaka, T. (2017). The number of the C-terminal doi:10.1007/s00418-018-1689-2 transmembrane domains has the potency to specify subcellular localization of Mi, H., Muruganujan, A. and Thomas, P. D. (2013). PANTHER in 2013: Modeling Sec22c. Biochem. Biophys. Res. Commun. 487, 388-395. doi:10.1016/j.bbrc. the evolution of gene function, and other gene attributes, in the context of 2017.04.071 phylogenetic trees. Nucleic Acids Res. 41, D377-D386. doi:10.1093/nar/gks1118 Yang, X., Coulombe-Huntington, J., Kang, S., Sheynkman, G. M., Hao, T., Michel, M., Wilhelmi, I., Schultz, A.-S., Preussner, M. and Heyd, F. (2014). Richardson, A., Sun, S., Yang, F., Shen, Y. A., Murray, R. R. et al. (2016). Activation-induced tumor necrosis factor receptorassociated factor 3 (Traf3) Widespread expansion of protein interaction capabilities by alternative splicing. alternative splicing controls the noncanonical nuclear factor κB pathway and Cell 164, 805-817. doi:10.1016/j.cell.2016.01.029 Journal of Cell Science