Mouse TSCs were obtained from Dr. Janet Rossant, Hospital for Sick Children, (Toronto, Canada) and maintained in DMEM/F12 with 15 mM Hepes, 20% FBS, 2 mM glutamine, 100 ug/ml streptomycin, 1 mM sodium pyruvate, 100 uM BME, supplemented with Activin, FGF4, and Heparin following (Erlebacher et al, Developmental Biology 2004). Trophoblast stem cells were differentiated into trophoblast giant cells by replacing the FGF, Activin and Heparin in the media with retinoic acid (Yan et al., 2001; Erlebacher et al., 2004). Mature TGCs are seen after 4-6 days of differentiation (Carney et al., 1993). Cells were trypsinized for 4 minutes before the extract protocol.
Extracted molecule
genomic DNA
Extraction protocol
Replication timing data were obtained by hybridizing early and late replication intermediates to NimbleGen oligonucleotide arrays, as described in Hiratani et al [PLoS Biology (2008) 6: e245]. Briefly, replication intermediates are prepared from cells that are first pulse-labeled with BrdU and then sorted into early (1st half of S) and late (2nd half of S) stages of S-phase by flow cytometry, followed by anti-BrdU immunoprecipitation of the BrdU-substituted (nascent) replication intermediates that were synthesized either early or late during S-phase. Samples were labeled after unbiased amplification of recovered DNA by whole-genome amplification (WGA; Sigma, GenomePlex). Further details can be found in Hiratani et al [PLoS Biology (2008) 6: e245].
Label
Cy3
Label protocol
Standard NimbleGen protocol (www.nimblegen.com/products/lit/index.html). Briefly, randomly-amplified BrdU-immunoprecipitated DNA samples (1 ug each) were Cy3- or Cy5-labeled by Klenow reaction using end-labeled Cy3- or Cy5-random 9-mers.
Mouse TSCs were obtained from Dr. Janet Rossant, Hospital for Sick Children, (Toronto, Canada) and maintained in DMEM/F12 with 15 mM Hepes, 20% FBS, 2 mM glutamine, 100 ug/ml streptomycin, 1 mM sodium pyruvate, 100 uM BME, supplemented with Activin, FGF4, and Heparin following (Erlebacher et al, Developmental Biology 2004). Trophoblast stem cells were differentiated into trophoblast giant cells by replacing the FGF, Activin and Heparin in the media with retinoic acid (Yan et al., 2001; Erlebacher et al., 2004). Mature TGCs are seen after 4-6 days of differentiation (Carney et al., 1993). Cells were trypsinized for 4 minutes before the extract protocol.
Extracted molecule
genomic DNA
Extraction protocol
Replication timing data were obtained by hybridizing early and late replication intermediates to NimbleGen oligonucleotide arrays, as described in Hiratani et al [PLoS Biology (2008) 6: e245]. Briefly, replication intermediates are prepared from cells that are first pulse-labeled with BrdU and then sorted into early (1st half of S) and late (2nd half of S) stages of S-phase by flow cytometry, followed by anti-BrdU immunoprecipitation of the BrdU-substituted (nascent) replication intermediates that were synthesized either early or late during S-phase. Samples were labeled after unbiased amplification of recovered DNA by whole-genome amplification (WGA; Sigma, GenomePlex). Further details can be found in Hiratani et al [PLoS Biology (2008) 6: e245].
Label
Cy5
Label protocol
Standard NimbleGen protocol (www.nimblegen.com/products/lit/index.html). Briefly, randomly-amplified BrdU-immunoprecipitated DNA samples (1 ug each) were Cy3- or Cy5-labeled by Klenow reaction using end-labeled Cy3- or Cy5-random 9-mers.
Hybridization protocol
Standard NimbleGen protocol (www.nimblegen.com/products/lit/index.html). Briefly, Cy3 and Cy5 labeled DNA samples (31 ug each) were co-hybridized to Nimblegen CGH arrays containing one oligonucleotide probe every 3.5 kb across the entire mouse genome.
Scan protocol
GenePix 4000B scanner (Molecular Devices) and GenePix software were used per NimbleGen's standard protocol (www.nimblegen.com/products/lit/index.html).
Data processing
NimbleScan software was used to obtain .pair raw data per manufacturer's instructions. Raw early/late data (i.e. from .pair files) from three independent biological replicates in which early and late replicating DNA were labeled were loess-normalized to remove signal intensity-dependent bias, scaled to a reference data set to have the same median absolute deviation and averaged (limma package, R/Bioconductor).The mean early/late ratios were used to generate a smoothed profile (i.e. processed data) using local polynomial smoothing (loess, 300-kb span) for each chromosome using basic functions in the statistical language R. Processed data sets can be graphically displayed (and are also downloadable) on our web site (http://www.replicationdomain.org). Further details can be found in Hiratani et al [PLoS Biology (2008) 6: e245].
