Supplemental Figure 1 Demonstration of the method for 3D distance determination
Image stacks were acquired as described in the material and methods section. (A) The YFP images are displayed in the red channel while the CFP are in the green channel. zm is the z-stack mean projection. Due to the limited resolution and exposure time, the images are very noisy. Nevertheless the nuclei appear as diffuse bright areas in all z-slices and in both channels. Each telomere, on the other hand, appears as a single bright spot in one slice of its corresponding channel.
The SpotDistance plug-in proceeds in three steps: 1) segmentation of the field of nuclei, 2) detection of the telomeres and, finally, 3) distance measurement. To detect the nucleus, the program computes the average of the two channels along the z-axis (Supp. Figure 1A). The resulting image is then smoothed and segmented using the watershed algorithm (Vincent and Soille 1991), this process separates the bright areas (nuclei) with a "dam", even when they are in contact. The images are pre-filtered with a Gaussian smoothening to avoid over-segmentation.
The detection of the telomeres is based on the maximum of the response of the LoG filter (Laplacian of Gaussian or Mexican Hat) inside a segmented nucleus. The LoG is the optimal filter under the assumption that the noise is fractal and that the telomeres have a Gaussian shape (Sage et al. 2005).
It is particularly helpful in presence of a high level of noise (Supp. Figure 1C).
To achieve sub-pixel accuracy, the two telomere positions are refined by taking the center of gravity of a small surrounding neighborhood; the distance is computed in 3D. In our present implementation, the program is able to report all distances in less than 10 seconds. This is about an order of magnitude faster than a manual method and it is also reproducible. (B) Enlargement of image #12 shows the high level of noise. (C) Results of the LoG filter (Laplacian of Gaussian =1.5) and result of the detection. In this case, the (green) telomere was detected in image #11 and the (red) one was found in image #13. The telomere-to-telomere distance is 1020.6 nm.
Supplemental Figure 2 Data examples from the SpotDistance plug-in
(A) Output of the watershed algorithm which is able to separate the nuclei. (B) Result of the detection step overlaid on a slice of the z-stack. (C) Table of distances provided by the SpotDistance plug-in. All detected spots were visually verified.
Supplemental Figure 3 Telomeric organization in S. cerevisiae
All the telomeric repeats, telomere associated sequences (TAS), and subtelomeric genes of the yeast strain S288c are shown, although Y’ elements, which are either 4.5 or 6.5 kb in length, are not drawn to scale. Telomeric regions are aligned at the centromeric side of the X elements. The size and placement of subtelomeric ORFs are indicated in bp at the top of the figure. The different elements in the map are as follows: Black rectangles: telomeric repeats; blue interrupted rectangles: Y' elements;
red unfilled rectangles: STR elements; orange unfilled rectangles: DNA sequences homologous to mitochondrial DNA; green filled rectangles: X core elements; white and colored rectangles containing arrows: subtelomeric ORFs. The arrows inside the rectangles indicate the direction of transcription.
Continuous lines indicate noncoding DNA sequences while discontinuous lines cover coding and non-coding DNA sequences.
The gray shadowed regions link DNA regions longer than 1 kb that share > 90% sequence identity with at least one fragment of the same group. Conserved genes are coordinately colored. Among COS family members 80% to 90% sequence identity with other family members was shaded. TAS generally share more than 90% sequence identity with at least one element of the same kind, with the
exception of the X core element at Tel6R that has only 70% identity with other X core elements. ORFs that do not share > 90% of identity with other ORFs are in white.
Supplemental Figure 4 The Tel6R pairing element works only in cis
Distance distributions of native telomere-pair 6L-6R in diploid cells represented as box plots. The left box plot shows the result obtained when the tagged Tel6R and Tel6L are integrated on the same chromosome (GA-3563; n=130; in cis) and right box plot presents distances separating Tel6R and Tel6L placed on one of each of the two Chr6 homologues (GA-3564; n=207; in trans).
Supplemental Figure 5 Detailed description of the chromosome swap technique
(A) Explicit scheme of the arm swap and (B) the telomere swap. (C) The chromosome swap technology described in detail for the telomere swap. We make use of an engineered URA3 allele that contains the actin intron and a loxP-kanMX-loxP cassette within the intron (Cheng et al. 2000). To do this, overlapping 5’ and 3’ portions are incorporated sequentially on the two chromosomes of interest.
In the first step, we integrated the 5’ portion (consisting of the URA3 promoter, a small portion of coding sequence, the intron, and loxP-kanMX-loxP cassette) within ORF YER186C of Tel6R by PCR-mediated gene replacement and selection for G418 resistance (Wach et al. 1994). Following integration, the loxP-kanMX-loxP cassette was reduced to a single loxP site by induced Cre recombination (Cheng et al. 2000). Next, an overlapping 3’ portion of the URA3 allele (consisting of the remainder of the URA3 ORF, as well as the intron and the loxP-kanMX-loxP cassette) was integrated within ORF YFR054C of Tel5R. This loxP-kanMX-loxP cassette was also converted to single loxP site a second brief induction of Cre recombinase. A third induction of the Cre recombinase was used to catalyze exchange between the two loxP sites on different chromosomes. Positive clones were selected by reconstitution of a functional URA3 allele, which permits growth on plates that lack uracil. The system used here to target chromosomal rearrangements is applicable to any system, in which homologous recombination is possible. The only prerequisite for the targeting of homologous recombination is the availability of two selectable markers and their arrangement in a nested fashion.
Supplemental Figure 6 Perinuclear anchoring does not correlate with clustering
(A) Image stacks of living cells that express GFP-NUP49 and a GFP-tagged telomere seen as a bright spot are taken with 200nm z-step size. To analyze the positioning of the tagged telomere we determine its distance from the NE in the image plane where the spot is the brightest, which is normalized to the nuclear diameter in the plane of focus. Binning into 3 zones allow us to determine nonrandom association with the nuclear periphery as being >33% in zone 1. (B) Positioning of telomeres in the nucleus before and after exchange of telomeric sequence. Telomere 5Rpreswap in strain GA-3143 and 5R6Rsubtelo in strain GA-3658 were scored for their presence in three zones of equal surface and the % of foci in the outermost zone 1 is shown. Tel5Rpreswap contains the
integrated LoxP-URA3-5’ but no exchange was induced. (C) Positioning of Tel6Rpreswap in strain GA-3146 and 6R5Rsubtelo in strain GA-1607 before and after the exchange are also shown. Results for G1 and S phase cells showed no significant differences and are pooled. (D) Positioning of the internal ARS514 monitored in GA-3690. The tagged locus is localized randomly in G1 and S-Phase when no lexA or lexA alone is expressed (32.6%; n=138 and 26.42%; n=193 in zone1 respectively).
Expression of lexA-Sir4-PAD and lexA-Yku80-4 confer perinuclear positioning 70.5%; n=166 and 51.6%; n=157 in zone1, respectively.
Supplemeantal Figure 7 Annotations sites for transcriptions factors in the subtelomeric regions of telomeres 6R, 5R, 6L and 6R10kb
Binding sites for transcriptions factors on the indicated subtelomeric regions according to MotEvo (Pachkov et al. 2007). Chromosome coordinates are displayed as a ruler on top of each panel together with the ORF names at the corresponding position. Only binding sites with a probability range of 0.5-1 are displayed.
Supplemental movies
(1) A full chip Rap1-GFP 4D acquisition performed as described in Materials and Methods is displayed as a Z-projection. Total number of frames: 271 or 90 min of life imaging with a time frame of 20 s. The image stacks were deconvolved using the Huygens© software. (2) SIR3-GFP time series displayed as a full rotation of a kymograph. Total time is 60 min. (3, 4) Rap1-YFP (in red) and telomere 14L (in green) time-lapse; 30 sec intervals. The image stacks were deconvolved using the Huygens© software. Changing gray levels due to bleaching were adjusted using Imaris (Bitplane).
Supplemental References
Cheng, T.H., C.R. Chang, P. Joy, S. Yablok, and M.R. Gartenberg. 2000. Controlling gene expression in yeast by inducible site-specific recombination. Nucleic Acids Res28: E108.
Pachkov, M., I. Erb, N. Molina, and E. van Nimwegen. 2007. SwissRegulon: a database of genome-wide annotations of regulatory sites. Nucleic Acids Res35: D127-131.
Sage, D., F.R. Neumann, F. Hediger, S.M. Gasser, and M. Unser. 2005. Automatic tracking of individual fluorescence particles: application to the study of chromosome dynamics. IEEE Trans Image Process14: 1372-1383.
Vincent, L. and P. Soille. 1991. Watersheds in digital spaces: an efficient algorithm based on immersion simulations. IEEE Transaction on Pattern Analysis and Machine Intelligence, IEEE 13:
583-598.
Wach, A., A. Brachat, R. Pohlmann, and P. Philippsen. 1994. New heterologous modules for classical or PCR-based gene disruptions in Saccharomyces cerevisiae. Yeast10: 1793-1808.
Schober et al., Supplemental Figure 1
Schober et al., Supplemental Figure2
DSF1 HXT13
COS1 YNL335w SNO2 SNZ2 6L
0 Kb AAD14 14L
YFL057c AAD6
MPH2 SOR2 HXT15
COS7 THI13 4L
MPH3 SOR1 HXT16 YJR157w THI11
10R
COS6 YGR294w YGR293c MAL12 7R2R
COS2 DAN3YBR300c MAL32
AAD3 RDS1 ADH7 15L
YOL166c AAD15 BDS1 YOL157c HXT11 YOL155c
14.3 Kb
IMD2 PHO12 YHR214w-A YHR214w YHR213w YHR212c
10 Kb
1L
PAU8 SEO1 YAL066w YAL065c YAL064w-B YAL064c-A
PHO11 1R
IMD1 YAR070c YAR069c YAR068w YAR066w YAR064w YAR062w YAR061wYAR060c YAR075w
Ty1-1
10L
PAU1 VTH2 FSP2 YJL220w HXT9
PAU14 YIL174w VTH1 YIL172c HXT12 HXT12 YIL169c 9L
Ty5-1
PAU21 ERR1 YOR392wHSP33 YOR390w YOR389w
16L
PAU22 ERR2 HSP32 YPL279c YPL278c YPL277c
13R
YGL260w YPS5 0 Kb VEL1
5R
YER188w YER187w YER186c
YFR057w ARR3 ARR2 ARR1YFR055w SGE1 16R6R
0 0 0
4 5000 6000 7000 2000 3000
1000 8000 9000 1000011000120001300014000 1500016000
TELOMERIC REPEATS AND TAS
YFR056c YFR054c
Schober et al., Supplemental Figure 3
X core elements
URA3 5’
4) Cre mediated recombination 1) Oligo A fwd, A rev PCR
2) Integrate
3) KanMX excision by Cre expression
Tel 6R
3) KanMX excision by Cre expression
Schober et al., Supplemental Figure 5
A
1
0005000100510002
2
1: 6L - 6R in cis in diploid (GA-3563; n=130)
2: 6L - 6R in trans in diploid (GA-3564; n=207)
Schober et al., Supplemental Figure 4
Separating distances in nm
frequency in zone 1
Schober et al. , Supplemental Figure 6
B
Manuscript in preparation for submission
71
reproduced results
72