Characterization of heterochromatin and telomere motion. (A) Colocalization of telomeres using green fluorescent protein (GFP)-TRF1 with heterochromatin stained by CFP-HP1α shows that most telomeres do not colocalize with the heterochromatin foci in UMUC3 cells. (B) Track4D analysis of heterochromatin motion in three-dimensional (3-D) space shows that the UMUC3 cell heterochromatin moves with much less motion than UMUC3 cell telomeres. (C) Histogram of D values for heterochromatin motion, N = around 400 heterochromatin spots. Heterochromatin hardly moves, and D values for heterochromatin fall in the range reported (mode around 1 × 10^-4 μm²/second). (D) 3-D trajectory image of tracked telomeres in a UMUC3 nucleus. The enlarged view shows the motion path of a representative jittering telomere and its non-directional movement and telomeric motion conforms to a 'constrained non-directional walk' description. (E) A scatter plot of diffusion coefficients of telomeres against distance from the periphery shows that there is no significant association of telomere movement with the nuclear periphery, and no correlation between distance to the periphery and motion characteristics.

Heterogeneous motions of telomeres. Telomeres tracked using green fluorescent protein (GFP)-TRF1 are shown as representatives. GFP-TRF2 telomeric tracks are shown in A in Additional file 10. (A) Tracking of telomere motion by the software program Track4D, shown as end-to-end (E2E) distance in three-dimensional (3-D) space. An individual telomere has moved in space away from its 0 time-point position. Hence this displays motion of each telomere as a function of time T (seconds), and its distance (μm) from its starting point in the first timeframe. The effective diffusion coefficient, D, was calculated from these E2E measurements. (B) TrackIt4D tracking of telomere motion as demonstrated by plotting of the cumulative path distance traveled between time 0 and time T (seconds). (C) Tracking from kymographs (3-D tracks projected on to a xy plane) of a live UMUC3 cell showing one representative of a calm telomere and a jittering telomere. (D) Telomere motion statistics calculated from average E2E distance over sub-intervals. All distances traveled by the same telomere over time intervals of the same length were averaged together and plotted. Telomeres tracked for 200 seconds were plotted for intervals up to 100 seconds. (E) Telomere motion quantified as a histogram of D values, N = around 400 telomeres.

Telomere movement can be altered experimentally. (A) Azide treatment of UMUC3 cells shows that jittering telomere movements require energy. Washing away azide partially restores the movement of jittering telomeres. (B) Southern blotting (insert in right panel) shows that UMUC3 telomeres were extended by expressing an extra telomerase RNA (WT-hTER) in cells, which dampens the telomere movement (right panel). The empty vector does not change the telomere motion (middle panel). (C) 47A MT-hTer, which induces DNA damage foci colocalizing with telomeres (see E in Additional file 14), increases telomere motility in UMUC3 cells (middle panel), whereas the general DNA damaging drug methyl methanesulfonate does not (right panel). (D) A summary of histogram analysis of parental telomeres, longer telomeres under WT-hTER treatment, uncapped telomeres under 47A MT-hTer treatment, or heterochromatin (N = around 400 telomeres or heterochromatin spots). The Additional file 18 insert shows that the Komolgorov-Smirnov non-parametric comparison between histograms is statistically significant. (E) A model for the novel form of the dynamic telomere behavior reported here: short telomeres and uncapped telomeres have higher motility than longer, functionally capped telomeres or other internal regions of chromosome arms. The motility of each mobile chromosomal entity is indicated by the length/strength of the arrow.

Visualization of telomeres in live cultured cancer cells. (A) A schematic of lentivector expressing green fluorescent protein (GFP)-tagged TRF1 or TRF2. The vector map is referenced in the Methods section. (B) Western blotting analysis of UMUC3 cells stably expressing GFP-TRF1 or GFP-TRF2 using antibodies against TRF1 or TRF2. Clonal lines with the lowest average expression (indicated by red numbers) were chosen for comparison with the endogenous level of TRF1 or TRF2 for telomere dynamic analysis. (C) Indirect immunofluorescence of clonal UMUC3 cell lines expressing either GFP-TRF1 (upper row) or GFP-TRF2 (lower row). Complete colocalization was observed from GFP-TRF1-expressed telomeres stained with anti-TRF2 antibody in fixed cells in the upper row. Lower row: immunofluorescence images of GFP-TRF2-expressed telomeres stained with anti-hRap1 antibody also showed telomere colocalization. (D) A schematic of the data collection regime for a representative UMUC3 nucleus. One complete three-dimensional (3-D) stack of 8 μm deep (an entire UMUC3 nucleus) with 0.5 μm focal spacing (17 sections) was acquired every second. (E) Representative examples of projection images from one complete 3-D stack of GFP-TRF1-expressed UMUC3 nucleus that was acquired every second for 200 consecutive seconds (T, time in seconds). The photobleaching curve is shown in D in Additional file 1.