Figure 1.
Halotag can be used to follow the aggregation/condensation of a protein in the ER.
HeLa cells expressing Halo-µΔCH1 were labelled for 24 hours with the red TMR ligand (5 µM) in complete medium. After extensive washings, cells were incubated with the green ligand R110 (5 µM) for the indicated times. To control saturation, cells were treated with cycloheximide (0.5 mM) for the last 2 hours of the first TMR incubation, and then during the incubation with R110 for 4 hours. Cells were then fixed in paraformaldheyde and analyzed by confocal microscopy. Images are shown as confocal slices. With the progression of the chase, the green signal co-localizes with the pre-existing red signal and newly formed aggregates (only green) can be visible. Pre-existing red aggregates grow by apposition of newly synthetized proteins which can be labeled in green. (Bar: 5 µm). Single channel signals in gray scale are shown for the thumbnails (Bar: 1 µm). Inset: HeLa cells transiently expressing Halo-µΔCH1 or Halo-µs were lysed, and aliquots from soluble (s) and insoluble (i) fractions resolved under non-reducing conditions on a 3–8% polyacrylamide gradient gel. Nitrocellulose membranes were decorated with anti-µ. Bands indicated with 1 represent monomeric proteins, while bands indicated with 2 represent homodimers.
Figure 2.
A Quantification of cluster growth by pulse-chase. Images were acquired as described above and analyzed with custom-written Matlab routines to quantify the size of the clusters in the two channels. When the two ligands are added in sequence (with a 24 hours chase) an average 16%+/−3% (SEM) increase in the cluster size is observed, while no difference in size can be measured when the two ligands are added simultaneously. For the pulse and chase experiment 9 cells and 99 clusters were examined. For the simultaneous addition of the two ligands 9 cells and 101 clusters were analyzed. Bar: 5 µm. B Distribution of old and young Halo-µΔCH1 molecules. HeLa cells transiently expressing Halo-µΔCH1 were subjected to fluorescent pulse-chase assays: 24 hours with the TMR ligand and 24 h with the non-labelled ligand (TMR+black) or 24 hours with non-labelled ligand and 24 hours with the TMR ligand (black+TMR). Halo-ligands were used at 5 µM. Aliquots of soluble (s) and insoluble (i) material were then resolved under reducing conditions on a 10% polyacrylamide gel. After transferring to nitrocellulose, the signal of the TMR ligand was collected using fluorescence technology by FLA900 Starion. Densitometric quantifications are shown on the right. Average of 3 independent experiments +/− standard deviation.
Figure 3.
Live-imaging of ER protein aggregation.
A HeLa cells expressing Halo-µΔCH1 were labeled for 24 h with 5 µM TMR. Living cells were then imaged by confocal microscopy at a frame rate of 2 images/s. B Aggregates display a heterogeneous dynamic behavior with some clusters moving directionally (orange arrowhead) and others showing confined diffusion (blue arrowhead). C We tracked individual clusters in 8 cells and we selected tracks that were longer than 150 frames (115 tracks) and we plotted individual MSD curves (orange thin lines) and their average (thick line). Treatment of cells with 0.3 µg/ml Nocodazole for 15 hours (blue lines) decreased of the mobility of the aggregates. D The diffusion coefficient D and the anomalous exponent α were calculated from the MSD curves obtained from the untreated samples and plotted against the estimated diameter of the clusters. On average we measured α = 0.74±0.02 and average diffusion coefficient D = 0.0045±0.0005 µm2/s. A value of α<1 is representative of anomalous subdiffusion, probably due to obstacles constraining the mobility of the clusters. While the anomalous exponent did not seem to depend on the cluster size, the diffusion coefficient was found to inversely correlate with the particle diameter. E We compared the distribution of measured particle sizes, diffusion coefficients and anomalous exponents for the untreated and the Nocodazole treated samples: the diffusion coefficient was the only parameter significantly affected by the disruption of the cytoskeleton (error bars: Mean ± SEM).
Figure 4.
Exploting Halotag to study ERAD.
A Halo-tagging does not alter the localization of unassembled Ig-µs. HeLa cells were transiently transfected with µs or Halo-µs as indicated, fixed with PFA and stained with antibodies specific for PDI or Ig-µ. In this experiment, transfection efficiency was about 25%. Bar: 5 µm. B The Halotag with or without bound ligand does not interfere with µs degradation. 48 hours after transfection, HeLa cells expressing µs or Halo-µs cells were pulsed for 10 minutes with 35S aminoacids and chased in the presence of 5 mM DTT for the indicated time points, to accelerate degradation [35]. One sample (Halo-µs+ligand) was pre-treated with TMR (2.5 µM) before and during the radioactive pulse. Aliquots from the lysates corresponding to 106 cells for each time point were precipitated with anti-µ, resolved under reducing conditions and transferred to nitrocellulose membranes which were first developed for the 35S signal and then for TMR fluorescence (532 nm). Densitometric quantification was performed by ImageJ. The graph shown in the bottom panel represents the percentage of the total µ present at time 0 remaining at individual times.
Figure 5.
Halotag does not influence the function or localization of ERp44.
A Halo-tagging does not alter the localization of ERp44. HepG2 cells were transiently transfected with Halo-ERp44 and pre-treated with TMR ligand (10 nM O/N) before fixation and staining with antibodies specific for ERGIC-53 or GM130 as markers of ERGIC or Golgi). Images were acquired with a 60X objective on an Olympus inverted fluorescence microscope and subjected to deconvolution. Single channel images and the merge are shown. Bar: 5 µm. B Halo-ERp44 retains Ero1α efficiently. Hela cells were transiently transfected with Halo-ERp44 and HA-ERp44 (alone or in combinations with Ero1α as indicated) 48 hours after transfection cells were washed and incubated for 4 hours in OPTIMEM. One sample was treated with TMR before and during the secretion. The spent media and cells were collected. The lysates corresponding to 105 cells (lanes 1–4) and the TCA precipitated supernatants of 5 times as many cells (5×105, lanes 5–8) were resolved under reducing conditions. Membranes were decorated with anti-Halo (blue signal), anti-HA (white signal), anti-Ero1α (green signal) and TMR fluorescence (red). Halo-ERp44 can be expressed and visualized by TMR staining. No significant differences were detected between Halo-ERp44 and HA-ERp44 in terms of function (Ero1α retention). C Secretion of ERp44ΔRDEL is not impaired by appending a Halotag. HeLa transfectants expressing Halo-ERp44 or Halo-ERp44ΔRDEL were incubated with OPTIMEM. After 4 hours, aliquots of lysates and supernatants (corresponding to 105 cells) were resolved under reducing conditions and the blots decorated with anti-ERp44. As HA-ERp44 ΔRDEL [6], Halo-ERp44 ΔRDEL is massively secreted implying that the tag does not interfere with folding/secretion. D Hela cells transiently expressing Halo-ERp44ΔRDEL were incubated overnight with the R110 ligand (35 nM), washed and incubated in OPTIMEM (without the Halo-ligand) for the indicated time points. Supernatants corresponding to 5*106 cells were concentrated by TCA precipitation, resolved under reducing conditions and the TMR signal collected using a fluorescent analyzer. The nitrocellulose filters were then decorated with anti-Halo antibodies. The intensities of the R110 fluorescent ligand and anti-Halo signals were quantified by densitometry using the ImageJ software and normalized to the value after 15 minutes of secretion. The graph shows the average of 2 independent experiments +/− standard deviation.