Lipid binding protein SYTL5 localises to mitochondria and endo-lysosomes

A. Live confocal microscopy imaging of U2OS stably expressing SYTL5-EGFP co-stained with 50 nM MitoTracker red. SYTL5-EGFP expression was induced for 24 h using 100 ng/ml doxycycline. MitoTracker red was added 30 min before imaging. Scale bars: 10 μm, 2 μm (insets).

B. Time-lapse video frames (Supplementary video 1) tracking SYTL5 vesicle movement along filaments positive for MitoTracker red (arrows). SYTL5-EGFP expression was induced for 24 h using 100 ng/ml doxycycline. Scale bars: 3 μm.

C. Live confocal microscopy imaging of U2OS stably expressing SYTL5-EGFP co-stained with 50 nM LysoTracker red. SYTL5-EGFP expression was induced for 24 h using 100 ng/ml doxycycline. LysoTracker red was added 30 min before imaging. Scale bars: 10 μm, 2 μm (insets).

D. Confocal imaging of U2OS stably expressing SYTL5-EGFP stained for endogenous LAMP1. Nuclei were stained using Hoechst. Scale bars: 10 μm, 2 μm (insets).

E. Overview of domain structure of SYTL5 full length, ΔSHD and ΔC2AB mutants.

F. 3xFLAG fusion proteins expressed in U2OS cells (SYTL5-EGFP-3xFLAG, SYTL5 (ΔSHD)-EGFP-3xFLAG and SYTL5 (ΔC2AB)-EGFP-3xFLAG) were immunoprecipitated and added to membranes spotted with lipids found in cell membranes: LPA (lipoprotein A); LPC (lysophosphatidylcholine); PI (phosphatidylinositol); PI(3)P (phosphatidylinositol 3-phosphate); PI(4)P (phosphatidylinositol 4-phosphate); PI(5)P (Phosphatidylinositol 5-phosphate); PE (phosphatidylethanolamine); PC (phosphatidylcholine); S1P (sphingosine-1-phosphate); PI(3,4)P2 (phosphatidylinositol 3,4-bisphosphate); PI(3,5)P2 (phosphatidylinositol 3,5-bisphosphate); PI(4,5)P2 (phosphatidylinositol 4,5-bisphosphate); PI(3,4,5)P3 (phosphatidylinositol 3,4,5 –triphosphate); PA (phosphatidic acid); PS (Phosphatidylserine); TAG (triglyceride); DAG (diglyceride); PG (phosphatidylglycerol); CL (cardiolipin); Cholesterol; SM (sphingomyelin); Sulfatide.

G. Live confocal microscopy imaging of U2OS stably expressing SYTL5 (ΔSHD)-EGFP-3xFLAG co-stained with 50 nM MitoTracker red added 30 min before imaging. Scale bars: 10 μm, 2 μm (insets).

H. Live confocal microscopy imaging of U2OS stably expressing SYTL5 (ΔC2AB)-EGFP-3xFLAG co-stained with 50 nM MitoTracker red added 30 min before imaging. Scale bars: 10 μm, 2 μm (insets).

Mitochondrial localisation of SYTL5 requires RAB27A GTPase activity

A. SYTL5-EGFP expression was induced for 24 h with 100 ng/ml doxycycline in U2OS cells with stable inducible expression of SYTL5-EGFP and constitutive expression of mScarlet-RAB27A. Cells were co-stained with MitoTracker DR for 30 min before live imaging. Arrows indicate mitochondrion filaments. Scale bars: 10 μm, 2 μm (insets).

B. CLEM analysis of the cells described in A. Before imaging, cells growing in monolayer were fixed in warm (≈37°C) 3.7% paraformaldehyde in 0.2 M HEPES (pH 7). After fixation, cells were imaged using a confocal microscope to acquire Z-stack of optical sections and DIC images to locate the cells of interest. The cells were finally fixed using 2% glutaraldehyde in 0.2 M HEPES (pH 7.4) for 120 min before sample preparation for TEM. Arrows in panel 1 indicate mitochondrion filaments. Scale bars: 15 μm (left), 5 μm (middle) and 2 μm (right).

C. U2OS dKO cells were rescued with mScarlet-RAB27A and co-stained with 50 nM MitoTracker green for 30 min before imaging. Scale bars: 10 μm, 2 μm (insets).

D. U2OS dKO cells were rescued with SYTL5-EGFP and co-stained with 50 nM MitoTracker red for 30 min before imaging. Scale bars: 10 μm, 2 μm (insets).

E. Live confocal microscopy imaging of U2OS dKO cells rescued with SYTL5-EGFP and mScarlet-RAB27A (upper panel); SYTL5-EGFP and mScarlet-RAB27A-T23N (middle panel) or SYTL5-EGFP and mScarlet-RAB27A-Q78L (lower panel). All cells were co-stained with 50 nM MitoTracker DR for 30 min before imaging. Scale bars: 10 μm, 2 μm (insets).

F. Lysates from U2OS cells stably expressing EGFP (control) and U2OS dKO cells expressing SYTL5-EGFP and/or mScarlet-RAB27A (wild-type, T23N or Q78L mutants) were immunoprecipitated using GFP-Trap beads and analysed by western blot using RAB27A and EGFP antibodies.

SYTL5 interacts with proteins involved in vesicle-mediated transport and cellular response to stress

A. SYTL5 interactome. SYTL5-EGFP expression in U2OS cells was induced for 24 h using 100 ng/ml doxycycline, followed by immunoprecipitation of SYTL5-EGFP using GFP-Trap beads and

identification of co-purified proteins by MS analysis. The list of co-purified proteins was compared with that obtained from cells expressing EGFP and only significant (p<0.05) SYTL5-EGFP specific protein hits are highlighted in colour. Protein hits falling into at least one of the biological processes; autophagy, regulation of mitochondrion organisation, protein insertion into mitochondrial membrane or cellular response to oxidative stress, are colour coded according to plot legend. All other significant hits are indicated in blue. Non-significant (p>0.05) are indicated in grey. Data are from 3 biological replicates.

B. GO cellular compartment term enrichment of significant protein hits co-purified with SYTL5-EGFP. Corresponding enrichment false discovery rate (FDR) value is represented inside each bar and bars are ordered from smallest to largest FDR (q-value) from top to bottom.

C. GO biological processes term enrichment of significant protein hits co-purified with SYTL5-EGFP.Corresponding enrichment FDR value is represented inside each bar and bars are ordered from smallest to largest FDR (q-value) from top to bottom. Proteins corresponding to each biological process category are listed in the table to the right.

SYTL5 interactome

Proteins co-immunoprecipitated with SYTL5-EGFP vs EGFP expressing control

Mitochondrial localisation of SYTL5 and RAB27A is not perturbed by autophagy inhibition or mitophagy stimulation

A. SYTL5/RAB27A dKO U2OS cells rescued with SYTL5-EGFP and mScarlet-RAB27A were untreated (upper panel) or treated with 1 mM DFP (2nd panel), hypoxia (1% O2) (3rd panel) or 1 mM DMOG (4th panel) for 24 h. All cells were stained with 50 nM MitoTracker DR 30 min prior to live confocal microscopy imaging. Scale bars: 10 μm, 2 μm (insets).

B. SYTL5/RAB27A dKO U2OS cells rescued with SYTL5-EGFP and mScarlet-RAB27A were treated or not with 1 mM DFP for 24 h. Cells were stained with 50 nM MitoTracker DR prior to fixation. Fixed cells were stained with a LAMP1 antibody and imaged by confocal microscopy. Scale bars: 10 μm, 2 μm (insets). Arrowheads point to SYTL5/RAB27A-positive structures.

C. U2OS cells expressing pSu9-Halo-mGFP were transfected with 20 nM control siRNA, SYTL5 siRNA #1 or RAB27A siRNA #1 for 48 h before 20 min incubation with the TMR-conjugated Halo ligand followed by 24 h incubation with 1 mM DFP or 1 mM DMOG. Cell lysates were analysed by western blot for the Halo tag. Actin was used as a loading control.

D. Quantification of data in C. Error bars represent the mean with standard deviation between replicates (n=3). Significance was determined by two-way ANOVA followed by Tukeýs multiple comparison test. * = p> 0.05, ** =p> 0,01, *** = p>0.001.

SYTL5 and RAB27A regulate mitochondrial bioenergetics.

A. Mitochondrial oxygen consumption rate was analysed in U2OS control cells, SYTL5 KO, RAB27A KO and dKO cells using the Seahorse XF analyser. OCR was measured after sequential addition of oligomycin, CCCP and a combination of rotenone and antimycin-A into the assay medium. Error bars represent the mean with standard deviation between replicates (n=3).

B. The four basal OCR measurements per well (before oligomycin addition) were averaged to obtain the basal OCR value and the non-mitochondrial respiration was subtracted to determine the basal respiration associated to each condition. The ATP production was calculated by subtracting the proton leak from the maximal respiratory capacity. Error bars represent the mean with standard deviation between replicates (n=3). Significance was determined by two-way ANOVA followed by Bonferroni multiple comparison test, ** = p>0.01, * = p> 0.05.

C. The extracellular acidification rate (ECAR) was analysed in U2OS control cells, SYTL5 KO, RAB27A KO and dKO cells using the Seahorse XF analyser. ECAR was measured upon addition of glucose, oligomycin and 2-DG using control and SYTL5 depleted cells. Error bars represent the mean with standard deviation between replicates (n=3).

D. The glycolysis rate of each condition was calculated by subtracting the ECAR value after 2-DG treatment to the ECAR after addition of glucose (Glycolysis=ECAR after addition of glucose − ECAR after 2-DG treatment). Error bars represent the mean with standard deviation between replicates (n=3). Significance was determined by one-way ANOVA followed by Tukey multiple comparison test, * = p> 0.05.

E. Cell lysates from U2OS control, RAB27A KO, SYTL5 KO and dKO cells were analysed by western blot for the indicated electron transport chain complex-II, -III, -V and -V proteins. GAPDH was used as a loading control.

F. Quantification of data in E. Band intensities were quantified relative to GAPDH and normalised to control. Error bars represent the mean with standard deviation between replicates (n=4). Significance was determined by two-way ANOVA followed by Bonferroni multiple comparison test, ** = p>0.01.

Low SYTL5 expression is related to reduced survival for adrenocortical carcinoma patients.

A. SYTL5 expression levels in normal/healthy tissues, using data from the GTEx project (v8). Shown are the 10 tissue sites for which SYTL5 is most highly expressed (i.e. based on median expression levels). TPM = transcripts per million. Median is represented by a red solid line and quartiles by red dashed lines.

B. Comparison of SYTL5 gene expression levels in normal adrenal gland samples (GTEx, n = 128) and adrenocortical carcinoma samples (The Cancer Genome Atlas (TCGA) ACC cohort, n = 77). Median expression is represented by a red solid line and quartiles by red dashed lines. Statistical analysis was performed using the unpaired t-test with Welch’s correction. TPM= transcripts per million.

C. Kaplan-Meier plot for ACC patient survival related to SYTL5 expression levels. The events related to high SYTL5 expression (expression level above median) are indicated in red and the dotted lines represent 95% confidence interval. Events related to low SYTL5 expression (expression level below median) are indicated in blue and the dotted lines represent 95% confidence interval. Statistical analysis was performed using Logrank (Mantel-Cox) test.

D. Comparison of RAB27A gene expression levels in normal adrenal gland samples (GTEx, n = 128) and adrenocortical carcinoma samples (The Cancer Genome Atlas (TCGA) ACC cohort, n = 77). Median expression is represented by a red solid line and quartiles by red dashed lines. Statistical analysis was performed using the unpaired t-test with Welch’s correction.

E. Kaplan-Meier plot for ACC patient survival related to RAB27A expression levels. The events related to high RAB27A expression (expression level above median) are indicated in red and the dotted lines represent 95% confidence interval. Events related to low RAB27A expression (expression level below median) are indicated in blue and the dotted lines represent 95% confidence interval. Statistical analysis was performed using Logrank (Mantel-Cox) test.

A-F. SYTL5-EGFP expression was induced for 24 h with 100 ng/ml doxycycline in U2OS cells with stable inducible expression of SYTL5-EGFP and constitutive expression of mScarlet-RAB4 (A), mScarlet-RAB11 (B), mScarlet-RAB5 (C), mScarlet-RAB7 (D), mScarlet-RAB9 (E) or mScarlet-RAB6 (F) followed by live confocal microscopy imaging analysis. Scale bars: 10 μm, 2 μm (insets).

A. Live confocal microscopy imaging of U2OS constitutively expressing mScarlet-RAB27A co-stained with MitoTracker green added 30 min before imaging. Scale bars: 10 μm, 2 μm (insets). B. Total cell lysate (TCL), cytosol, and a crude mitochondria fraction from U2OS control cells and cells with stable expression of mScarlet-RAB27A or SYTL5-EGFP were analysed by western blotting for EGFP, RAB27A, tubulin, TIM23 and COXIV. C. Live confocal microscopy imaging of HeLa constitutively expressing SYTL5-EGFP co-stained with MitoTracker Red added 30 min before imaging. Scale bars: 10 μm, 2 μm (insets). D. Live confocal microscopy imaging of HeLa constitutively expressing mScarlet-RAB27A co-stained with MitoTracker green added 30 min before imaging. Scale bars: 10 μm, 2 μm (insets).

A. RAB27A KO validation by genotyping. The WT RAB27A gDNA sequence was aligned with the RAB27A KO clone 21(c21) gDNA sequence. CRISPR/Cas9 editing produced an indel (single nucleotide deletion indicated by *) in a genomic location shared between all RAB27A isoforms. gRNA targeting region is underlined in the WT RAB27A gDNA sequence. B. RAB27A KO validation by western blot. Cell lysates from U2OS control cells and RAB27A KO cells (c21) were compared to cell lysates where RAB27A was depleted for 72 h using 2 independent siRNAs and probed with RAB27 and actin antibodies. C. SYTL5 CRISPR/Cas9 exon deletion was performed using two gRNA simultaneously. These gRNA target the 5’UTR or intronic regions of SYTL5 (transcript NM_138780) that are shared by all SYTL5 isoforms. Genomic deletion validation and individual clone genotyping was performed using a primer pair that flanked the region to be edited (about 1307 bp length between these primer binding sites). D. SYTL5 KO validation was performed with conventional PCR reaction. A PCR product of 1307 bp is expected for wild-type U2OS gDNA, while a PCR product of about 618 bp is expected if a clone has a deletion in the targeted region. This method was used to validate the SYTL5 KO clone2 and for dKO RAB27A/SYTL5 clone57. The deleted sequence includes a coding exon from SYTL5, resulting in a non-functional SYTL5 protein.

A. SYTL5/RAB27A dKO U2OS cells rescued with SYTL5-EGFP and mScarlet-RAB27A were treated with 1 mM DFP for 24 h. Fixed cells were stained with a LAMP1 antibody in addition to either p62 (upper panel) or LC3 (lower panels) antibodies and imaged by confocal microscopy. Arrowheads point to SYTL5/RAB27A-positive structures. Scale bars: 10 μm, 2 μm (insets). B. SYTL5/RAB27A dKO U2OS cells rescued with SYTL5-EGFP and mScarlet-RAB27A were treated with 1 μM MRT68921 for 2 h (upper panel) or 5 μM Vps34-IN1 for 2 h (lower panel). All cells were stained with 50 nM MitoTracker DR 30 min prior to live imaging confocal microscopy. Scale bars: 10 μm, 2 μm (insets).

A. U2OS cells were transfected with 20 nM control siRNA or SYTL5 siRNA oligos for 72 h. Transfection efficiency was analysed by qPCR where the relative mRNA expression of SYTL5 was normalised to the expression levels of TBP and siControl. Error bars represent the standard deviation between replicates (n=3). Significance was determined by one-way ANOVA followed by Dunnett multiple comparison test, *** = p>0.001. B. U2OS cells expressing the mitochondrial matrix reporter NIPSNAP11–53-EGFP-mCherry30 (referred to as IMLS cells) were transfected with 20 nM control siRNA or SYTL5 siRNA oligos for 48 h before 24 h incubation with 1 mM DFP in the absence or presence of 100 nM BafA1 for the last 2 h. After fixation, cell nuclei were stained with Hoechst and widefield images were obtained using a high-content imaging microscope. The area of red-only puncta per cell (>1000 cells analysed) was quantified using a Cell Profiler pipeline. The obtained results were normalised to the control siRNA and error bars represent the mean with standard deviation between replicates (n=3). Significance was determined by two-way ANOVA followed by Bonferroni multiple comparison test. C. U2OS IMLS cells expressing PARKIN were transfected with 20 nM control siRNA or SYTL5 siRNA oligos for 48 h before a 16 h incubation with 20 μM CCCP to induce mitophagy, with treatment with 100 nM BafA1 or not for the last 2 h. The area of red-only puncta per cell (>1000 cells analysed) was quantified using Cell Profiler from widefield images obtained using a high-content imaging microscope. The obtained results were normalised to the control siRNA and error bars represent the mean with standard deviation between replicates (n=3). Significance was determined by two-way ANOVA followed by Tukeýs multiple comparison test. D. U2OS IMLS cells were transfected with 20 nM control siRNA or RAB27A siRNA oligos for 48 h before 24 h incubation with 1 mM DFP in the absence or presence of 100 nM BafA1 for the last 2 h. After fixation, cell nuclei were stained with Hoechst and widefield images were obtained using a high-content imaging microscope. The area of red-only puncta per cell (>1000 cells analysed) was quantified using a Cell Profiler pipeline. The obtained results were normalised to the siControl, and error bars represent the mean with standard deviation between replicates (n=3). Significance was determined by two-way ANOVA followed by Bonferroni multiple comparison test. E. WT (control) or SYTL5 KO U2OS cells were starved in EBSS for 4 h or incubated with complete medium in the presence or absence of 100 nM BafA1 the last 2 h. Cell lysates were harvested and analysed by western blot for LC3B and actin proteins. LC3B-II band intensity was quantified relative to actin and normalised to control. Error bars represent the mean with standard deviation between replicates (n=3). Significance was determined by one-way ANOVA followed by Bonferroni multiple comparison test.

NCl-H295R cells were transfected with 20 nM control siRNA, SYTL5 siRNA or RAB27A siRNA oligos for 72 h. Cortisol secretion was measured and normalised to siControl. Error bars represent the mean with standard deviation between replicates (n=4).