|
|
|
|
|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | ||||
| ||||||||||||||||||||
Files in this Data Supplement:
Fig. S1. The cell death phenotype of xDMO morphants can be alleviated by co-injection with either xMBD4, xMLH1 or xp53 morpholinos. (A-F) The same pictures as in Fig. 1A-F but at a higher magnification so that the phenotypes of the embryos are more clearly seen. Notice the loss of pigmentation (due to death) in the xDMO morphants (B) compared with controls (A). This phenotype is clearly reduced in (xDMO and xM4MO) morphants (C), (xDMO and xMLH1MO) morphants (D) and (xDMO and xp53MO) morphants (E). Co-injection of xKMO with xDMO (F) enhances this phenotype. A, wild type stage 12-23; B, xDMO depletion, stage 13; C, xDMO and xM4MO, stage 13; D, xDMO and xMLH1MO, stage 13; E, xDMO and xp53MO, stage 13; F, xDMO1 and xKMO, stage 13.
Fig. S2. Phenotype of embryos injected with a low dose (200 ng) of mouse MBD4 mRNA. (A,B) Control embryos (A) and embryos injected with 200 pg of murine MBD4 (mMBD4) mRNA (B). Notice the bent tails in B. (C) Model for activation of apoptosis via MBD4/MLH1 in xDMO embryos. Depletion of xDNMT1 activates the MLH1/MBD4 complex to signal to the components of the DNA damage pathway, ATM and ATR, which activate p53 leading to programmed cell death.
Fig. S3. Injection of β-galactosidase mRNA does not affect Xenopus laevis development. (Left) Control uninjected embryos. (Right) Embryos injected with 750 ng β-galactosidase mRNA. The injected embryos developed normally, as evidenced by formation of the neural tube. This process is inhibited in Mbd4 mRNA-injected embryos.
Fig. S4. MBD4 interacts with DNMT1 from rat, mouse and Xenopus. (A) The indicated Myc-tagged rat DNMT1 constructs were co-expressed in HeLa cells with T7-tagged xMBD4 (T7xMBD4). Cell extracts were prepared and immunoprecipitated with an anti-Myc mAb and subsequently western blotted with an anti-T7 mMb. Lane 1, full length (1-1622 amino acid) rat DNM1 and T7xMBD4 input; lane 2, partial (1-1414 amino acid) rat DNMT1 and T7xMBD4 input; lane 3, partial (1-378 amino acid) rat DNMT1 and T7xMBD4 input; lane 4, T7xMBD4 input only; lanes 5-6, extracts were immunoprecipitated with an anti-Myc antibody and western blotted with a T7 epitope antibody. The T7xMBD4 band is indicated. Notice all three rat DNMT1 proteins interact with xMBD4. (B) Schematic of mouse DNMT1 (top) indicating functional domains present, including the PCNA-binding domain (PCNA), the Zn finger (CXXC) domain and the catalytic domains. The GST-fusion proteins used in pull-down assays are shown below. ‘300’ indicates 1-300 amino acid fusion; ‘1000’ indicates 300-1000 amino acid fusion; ‘1620’ indicates 1000-1622 amino acid fusion. A plasmid encoding Xenopus xMBD4 was transcribed and translated in vitro and incubated with GST only or with the indicated GST-fusion proteins. After extensive washing, S35-labelled bound proteins were resolved on SDS-PAGE gels, dried and exposed to indicate functional interactions. Notice the 1000-1622 amino acid fusion of mouse DNMT1 preferentially interacts with xMBD4. (C) To narrow the interaction region, we made three N-terminal xDNMT1-GST fusions corresponding to regions 1-130 amino acids (Gst1), 124-380 amino acids (Gst2) and 505-600 amino acids (Gst3). Al three were tested in pull-down experiments with S35-labelled xMBD4. Notice that the 124-380 amino acid region (Gst2) preferentially interacts with xMBD4. (D) The minimum MBD4 interaction region in DNMT1 was determined by comparing which of the different DNMT1 constructs from rat, mouse and Xenopus interacted with xMBD4. The overlapping sequences and their homology are shown. Identical amino acids are indicated by asterisks. The MECP2 interaction domain of rat DNMT1 is shown underneath, which shows a partial (underlined) overlap with the MBD4 interaction region.
Fig. S5. Models for the recruitment of DNMT1/MBD4/MLH1 to DNA-damage sites. (A) In this model, recruitment of DNMT1/MBD4/MLH1 is via PCNA, which interacts directly with DNMT1. After UV-induced damage (asterisk), PCNA scans the genome, detects the lesion and subsequently interacts with DNMT1, which can recruit MBD4/MLH1. (B) In this model, recruitment of DNMT1/MBD4/MLH1 is via MBD4/MLH1. After UV-induced damage (asterisk), MBD4/MLH1 present at the lesion subsequently interacts with DNMT1, perhaps in concert with PCNA.
Fig. S6. Inhibition of in vitro translation of either xDNMT1, xMBD4 or xMLH1 by their respective morpholinos. (A) In vitro transcription and translation experiments show efficient knockdown of xDNMT1 translation in the presence of xDMO morpholino (compare lanes 2 and 3). Lane 1 shows translation of a luciferase mRNA control. (B) In vitro transcription and translation experiments show efficient knockdown of xMLH1 translation in the presence of xMLH1MO morpholino (compare lanes 2 and 3). Lane 1 shows translation of a luciferase mRNA control. (C) In vitro transcription and translation experiments show efficient knockdown of xMBD4 translation in the presence of xM4MO morpholino (compare lanes 1 and 2). Lane 3 shows translation of mouse mMBD4 control, which is not inhibited by the Xenopus-specific xMBD4 morpholino (lane 4).
Movie 1. RFP-PCNA and GFP-MLH1 are recruited to sites of DNA damage. The left panel shows accumulation of RFP-PCNA at one microirradiated foci, while GFP-MLH1 is shown on the right. The first images after microirradiation can be identified by the bleach effect of the 405 nm laser. The time interval for live image capture is 300 seconds.
Movie 2. RFP-PCNA and GFP-MBD4 are recruited to sites of DNA damage. The left panel shows accumulation of RFP-PCNA at two microirradiated foci, while GFP-MBD4 is shown on the right. In the middle is a merge of the two recordings, which have been pseudo-coloured. Notice that accumulation of PCNA is sustained throughout the experiment, whereas MBD4 is not stably associated at the lesion and its signal disappears towards the end of the recording. The first images after microirradiation can be identified by the bleach effect of the 405 nm laser. The time interval for live image capture is 300 seconds.
Movie 3. Ch-MLH1and GFP-MBD4 are recruited to sites of DNA damage. The left panel shows accumulation of Cherry-MLH1 at one microirradiated foci, while GFP-MBD4 is shown on the right. In the middle is a merge of the two recordings, which have been pseudo-coloured. Notice that accumulation of both MLH1 and MBD4 is sustained throughout the experiment. The first images after microirradiation can be identified by the bleach effect of the 405 nm laser. The time interval for live image capture is 300 seconds.
Movie 4. RFP-DNMT1and GFP-MBD4 are recruited to sites of DNA damage. The left panel shows accumulation of RFP-DNMT1 at two microirradiated foci, while GFP-MBD4 is shown on the right. In the middle is a merge of the two recordings, which have been pseudo-coloured. Notice that accumulation of both DNMT1 and MBD4 is sustained throughout the experiment. The first images after microirradiation can be identified by the bleach effect of the 405 nm laser. The time interval for live image capture is 300 seconds. See the following references for details (Mortusewicz et al., 2005; Mortusewicz et al., 2007).
| ||||||||||||||||||||
