Fig 1.
STED imaging reveals lumenal nanodomain periodicity in peripheral ER tubules.
(A) Representative 2D STED images of live and fixed HT-1080 or COS-7 cells expressing ERmoxGFP are shown. Magnified confocal and STED images of the boxed region highlight the improved resolution obtained by 2D STED. Scale bar, 5 μm; zooms, 2 μm. (B) Line scans of isolated peripheral ERmoxGFP-labeled tubules (dashed) from STED images of live and fixed cells show matching maxima and minima for raw (red, dashed) and deconvolved (decon, green, solid) images. Scale bar, 0.5 um. (C) Length of ERmoxGFP maxima/blob was measured in 2D STED images of peripheral ER tubules in live and fixed cells. Values plotted are mean ± SEM from three independent experiments (10–20 line scans/each repeat). Significance assessed by Student t test. Numerical values that underlie the graph are shown in S1 Data. ER, endoplasmic reticulum; ERmoxGFP, ER monomeric oxidizing environment-optimized green fluorescent protein; ns, not significant; STED, stimulated emission depletion; 2D, two-dimensional.
Fig 2.
RTN4 regulates lumenal ER nanodomain periodicity.
(A) Western blots of siCTL–, siRTN4–, and siCLIMP-63–transfected HT-1080 cells were probed with anti-RTN4, anti-CLIMP-63, and anti-β-actin as a loading control. (B) Representative images of ER tubules in HT-1080 cells transfected with ERmoxGFP and siCTL, siRTN4, or siCLIMP-63. Arrowheads indicate the tubules with increased blob length. Scale bar, 2 μm. (C) The quantification of maxima length, variation of maxima length (SD), and maxima-minima intensity differentials of ER tubules in HT-1080 cells transfected with siRTN4, siCLIMP-63, or siCTL. Bar graphs show mean ± SEM and scatter dot plots median with interquartile range. Significance assessed by Student t test from three independent experiments (20–40 line scans/each repeat). **P < 0.01; ***P < 0.001. Numerical values that underlie the graphs and plots are shown in S1 Data. CLIMP-63, cytoskeleton-linking membrane protein 63; ER, endoplasmic reticulum; ERmoxGFP, ER monomeric oxidizing environment-optimized green fluorescent protein; ns, not significant; RTN4, reticulon4; SD, standard deviation; siCLIMP-63, siRNA to CLIMP-63; siCTL, siControl; siRTN4, siRNA to RTN4.
Fig 3.
RTN4a and CLIMP-63 overexpression differentially impacts ER nanodomain periodicity.
(A) STED images of ERmoxGFP in HT-1080 cells transfected with ERmoxGFP or cotransfected with mCherry-CLIMP-63 (CLIMP-63), mCherry-RTN4a (RTN4a), or mCherry-ATL1 (ATL1). Peripheral ER regions (white boxes) are shown as zooms; line scans of selected tubules in these regions (yellow boxes) are shown with ERmoxGFP in green and ER-shaping proteins in red. Scale bar, 5 μm; zooms, 2 μm. (B) Peripheral ER tubule maxima length, variation of maxima length (SD), and maxima-to-minima intensity differential are shown for cells transfected with ERmoxGFP alone (CTL) or cotransfected with mCherry-CLIMP-63 (CLIMP-63), mCherry-RTN4a (RTN4a), or mCherry-ATL1 (ATL1). Significance assessed by one-way ANOVA from three independent experiments (40 line scans/each repeat). Bar graphs show mean ± SEM and scatter dot plots median with interquartile range. *P < 0.05; **P < 0.01; ***P < 0.001. Numerical values that underlie the graphs and plots are shown in S1 Data. (C) Based on line scan analysis of peripheral ER tubules of HT-1080 cells cotransfected with mCherry-CLIMP-63 (CLIMP-63), mCherry-RTN4a (RTN4a), or mCherry-ATL1 (ATL1), percent localization of CLIMP-63, RTN4a, and ATL1 puncta to minima or maxima of lumenal ERmoxGFP-labeled tubules was quantified. Significance assessed by one-way ANOVA from four independent experiments (40 line scans/each repeat). Bar graphs show mean ± SEM. *P < 0.05; ***P < 0.001. Numerical values that underlie the graphs are shown in S1 Data. ATL, atlastin; CLIMP-63, cytoskeleton-linking membrane protein 63; CTL, control ER, endoplasmic reticulum; ERmoxGFP, ER monomeric oxidizing environment-optimized green fluorescent protein; ns, not significant; RTN4a, reticulon4a; SD, standard deviation; STED, stimulated emission depletion.
Fig 4.
Periodic distribution of other ER markers and ER-resident proteins along peripheral ER tubules.
(A) Representative confocal and STED images of live HT-1080 and COS-7 cells overexpressing Sec61βGFP or live knock-in U2OS cells expressing GFP-calreticulin at endogenous levels. Scale bar, 2 μm. (B) STED images of fixed HT-1080 cells expressing lumenal ERmoxGFP (green) and membrane Sec61βmRFP (red) show distinct periodicity of these two ER reporters along peripheral ER tubules. Scale bar, 2 μm; zoom, 0.5 μm. (C) Untransfected HT-1080 cells labeled for derlin-1 or calnexin imaged by confocal and STED. Scale bar, 2 μm. (D) Association of ER-resident proteins derlin-1 and calnexin with ERmoxGFP-labeled peripheral ER tubules in HT-1080 cells by 2D STED. Scale bar, 2 μm. (E) Association of ER-resident proteins derlin-1 and calnexin with ERmoxGFP-labeled peripheral ER tubules in HT-1080 cells by 3D STED. Scale bar, 1 μm. ER, endoplasmic reticulum; ERmoxGFP, ER monomeric oxidizing environment-optimized green fluorescent protein; KI, knock-in; STED, stimulated emission depletion; 2D, two-dimensional; 3D, three-dimensional.
Fig 5.
ER-resident proteins calnexin and derlin-1 are enriched in nanodomains depleted of lumenal ERmoxGFP.
(A) Representative merged images of single peripheral ER tubules expressing ERmoxGFP or Sec61βGFP labeled for calnexin or derlin-1. The dashed line indicates the site of line scan analysis along tubule. Fluorescence intensities of ER reporter (green) and protein (red) from line scans are presented as graphs. Scale bar, 0.5 μm. (B) Based on line scan analysis of peripheral ER tubules, percent localization of calnexin and derlin-1 puncta to ERmoxGFP or Sec61βGFP maxima and minima was quantified. Values plotted are mean ± SEM from three independent experiments (40 tubules per repeat) with one-way ANOVA for significance. ***P < 0.001. Numerical values that underlie the graphs are shown in S1 Data. (C) Based on line scan analysis of peripheral ER tubules, percent localization of calnexin and derlin-1 puncta to ERmoxGFP maxima and minima was quantified in cells transfected with siCTL, siCLIMP-63, or siRTN4. Significance was assessed by χ2 test from three independent experiments (20–40 tubules per repeat). ***P < 0.001. Numerical values that underlie the graphs are shown in S1 Data. (D) Based on line scan analysis of peripheral ER tubules, percent localization of calnexin puncta to ERmoxGFP maxima and minima was quantified in HT-1080 cells cotransfected with mCherry-CLIMP-63, mCherry-RTN4a, or mCherry-ATL1 compared with CTL. Significance assessed by χ2 test from three independent experiments (40 tubules per repeat). ***P < 0.001. Numerical values that underlie the graphs are shown in S1 Data. ATL, atlastin; CLIMP-63, cytoskeleton-linking membrane protein 63; CTL, control; ER, endoplasmic reticulum; ERmoxGFP, ER monomeric oxidizing environment-optimized green fluorescent protein; ns, not significant; RTN4a, reticulon4a; siCLIMP-63, siRNA to CLIMP-63; siCTL, siControl; siRTN4, siRNA to RTN4.
Fig 6.
RTN4a and CLIMP-63 regulate dynamics of lumenal nanodomain compartmentalization along peripheral ER tubules.
STED live cell imaging (40 ms/frame over 4 seconds) of isolated ROIs of peripheral ERmoxGFP-labeled tubules (a) was performed for COS-7 cells cotransfected with mCherry-CLIMP-63 or mCherry-RTN4a (A) or transfected with siCLIMP-63, siRTN4, or siCTL (B). Kymograms show the distribution of ERmoxGFP at specific sites along ER tubules over time (b). From plots of normalized average intensity over time (c), we determined the CoV along the tubule length as a measure of localized distribution of ERmoxGFP to distinct domains along peripheral ER tubules (d). Scatter dot plots show median with interquartile range from three independent experiments (20–50 tubules per condition) with one-way ANOVA for significance. *P < 0.05; ***P < 0.001. Numerical values that underlie the plots are shown in S1 Data. CLIMP-63, cytoskeleton-linking membrane protein 63; CoV, coefficient of variation; CTL, control; ER, endoplasmic reticulum; ERmoxGFP, ER monomeric oxidizing environment-optimized green fluorescent protein; ns, not significant; ROI, region of interest; RTN4a, reticulon4a; siCLIMP-63, siRNA to CLIMP-63; siCTL, siControl; siRNA, small interfering RNA; siRTN4, siRNA to RTN4; STED, stimulated emission depletion.