This page has been archived and is no longer updated
Evidence that homologous X-chromosome pairing requires transcription and Ctcf protein
Author: Na Xu,Mary E. Donohoe,Susana S. Silva,Jeannie T. Lee
Keywords
Keywords for this Article
Add keywords to your Content
Save
|
Cancel
Share
|
Cancel
Revoke
|
Cancel
Rate & Certify
Rate Me...
Rate Me
!
Comment
Save
|
Cancel
Flag Inappropriate
The Content is
Objectionable
Explicit
Offensive
Inaccurate
Comment
Flag Content
|
Cancel
Delete Content
Reason
Delete
|
Cancel
Close
Full Screen
"Evidence that homologous X-chromosome pairing requires transcription and Ctcf protein Na Xu 1,3 , Mary E Donohoe 1,3 , Susana S Silva 1,2 & Jeannie T Lee 1 X-chromosome inactivation (XCI) ensures the equality of X-chromosome dosages in male and female mammals by silencing one X in the female 1 . To achieve the mutually exclusive designation of active X (Xa) and inactive X (Xi), the process necessitates that two Xs communicate in trans through homologous pairing 2,3 . Pairing depends on a 15-kb region within the genes Tsix and Xite 2 . Here, we dissect molecular requirements and find that pairing can be recapitulated by 1- to 2-kb subfragments of Tsix or Xite with little sequence similarity. However, a common denominator among them is the presence of the protein Ctcf, a chromatin insulator 4?7 that we find to be essential for pairing. By contrast, the Ctcf-interacting partner, Yy1 (ref. 8), is not required. Pairing also depends on transcription. Transcriptional inhibition prevents new pair formation but does not perturb existing pairs. The kinetics suggest a pairing half- life of o1 h. We propose that pairing requires Ctcf binding and co-transcriptional activity of Tsix and Xite. In the mouse, the X-inactivation center (Xic) orchestrates XCI through three noncoding genes: Xite 9 , Tsix 10?12 and Xist 13?15 . Xite and Tsix designate the future Xa and are expressed on both Xs before XCI, whereas Xist initiates silencing and is expressed only from the Xi. The Xic works both in trans and in cis.In trans, Xite and Tsix coordinate the ?counting? of X chromosomes 16,17 and the mutually exclusive designation of Xa and Xi 18 . Once committed to XCI, Xite and Tsix become cis acting 9,19?21 : on the future Xa, their persistent expression blocks the upregulation of Xist 10,11 , whereas on the future Xi, their repression induces Xist expression, which in turn initiates silencing of the linked X chromosome 13?15 . Although pairing is a prerequisite for the X-chromosome binary switch, the molecular underpinnings of this process have yet to be elucidated. Previous work has shown that a 3.7-kb fragment of Tsix (p3.7, Fig. 1a)anda 5.6-kb fragment of Xite (pXite) can recapitulate pairing in a transgene- based assay 2 . In mouse embryonic stem cells, the autosomal trans- genes induce de novo pairing between the X and the autosome (X-A pairing) and prevent the formation of true X-X pairs 2 , which in turn causes a failure of XCI 16 . To pi np oi nt sp ec ifi c cis-andtrans-factor requirements, we created six small transgenes carrying specific motifs within Tsix or Xite (Fig. 1a and Supplementary Table 1 online): pNS41 spans a 2.4-kb conserved region of Tsix 10 and several Ctcf-Yy1 elements 8 ; pNS25 is a 1.6-kb fragment that contains DXPas34 17,22 and multiple Ctcf-Yy1 b Day 5 Day 8 a pNS132 pNS130 pNS135pNS82 pNS25 pNS41 Tsix Dxpas34 Xite p3.7 pXite Day 5 Day 8 0 20 40 60 80 100 Day 0 Day 4 Day 8 c WT 11-6 82-7 41-7 25-2 132-1 135-5 130-1 P ercentage cell death 1 kb 11-6 82-7 41-7 25-2 132-1 135-5 130-1 J1 25-1 41-1 132-1 130-1 135-1 Clones Control Controls Figure 1 Growth and differentiation characteristics of new transgenic cell lines. (a)MapofTsix and Xite, the previously characterized p3.7 and pXite transgenes and the six new transgenes. (b) Morphology and growth of embryoid bodies on day 5 and day 8 of differentiation. (c) Magnitude of cell death on day 0, day 4 and day 8 of differentiation. The averages of three independent experiments are shown. Error bars, s.d. Received 26 June; accepted 15 August; published online 21 October 2007; doi:10.1038/ng.2007.5 1 Howard Hughes Medical Institute Department of Molecular Biology, Massachusetts General Hospital Department of Genetics, Harvard Medical School Boston, Massachusetts 02114, USA. 2 Gulbenkian PhD Programme in Biomedicine, Rua da Quinta Grande, 6, 2780-156, Oeiras, Portugal. 3 These authors contributed equally to this work. Correspondence should be addressed to J.T.L. (lee@molbio.mgh.harvard.edu). 1390 VOLUME 39 [ NUMBER 11 [ NOVEMBER 2007 NATURE GENETICS LETTERS sites 8 ; pNS82 contains the 0.22-kb minimal Tsix promoter 23 and a truncated Ctcf-Yy1 site; pNS132 spans a 2.5-kb Xite fragment with minor promoters and DNase I hypersensitive sites 9 ;pNS135 consists of the 1.2-kb enhancer required for persistent Tsix ex- pression on the future Xa 9,23 ; pNS130 encompasses a 1.8-kb region upstream of the enhancer; and pNS11 (not shown) is a control containing only the vector. For each construct, we examined multiple clones. In all experiments that follow, three independent tests for each clone yielded similar results regardless of trans- gene copy number, so a representative clone for each transgene is shown. All wild-type and transgenic female cell lines grew normally in the undifferentiated state (day 0, data not shown). However, upon differentiation into embryoid bodies to induce XCI, cell lines carrying pNS41 (for example, clone 41-7), pNS25 (for example, 25-2), pNS132 (for example, 132-1), pNS135 (for example, 135-5) and pNS130 (for b Cum ulativ e frequency Cum ulativ e frequency 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.92 82-7 ND 41-7 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P < 0.001 ND 135-5 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P < 0.001 ND 132-1 0.0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.03 ND 25-2 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.01 ND Cum ulativ e frequency Cum ulativ e frequency 130-1 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.003 ND Day 0 Day 4 c de 130-1 0 190 0.0 2 137 1.4 16.7 Day 4 Day 8 54 199 27.1 63 136 46.3 135-5 0 181 0 31 148 20.9 41-7 1 198 0.5 5 109 4.6 Day 4 Day 8 82-7 20 51 110 126 18.2 40.5 Day 4 Day 8 25-2 1 162 0.6 0 189 0 Day 4 Day 8 132-1 2 0 210 201 0.9 0 Day 4 Day 8 Day 4 Day 8 Day 4 Day 8 Number nuclei with Xist cloud n, Total nuclei Percent 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.01 ND 82-7 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.23 ND 41-7 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.42 ND 25-2 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.536 ND 135-5 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.36 ND Day 0 Day 4 130-1 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.34 ND 132-1 X-X interactions, TgX-A interactions, Tg a 0 0.05 0.10 0.15 0.20 0.1 0.3 0.5 0.7 0.9 0 0.05 0.10 0.15 0.20 0.25 0.1 0.3 0.5 0.7 0.9 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 Day 0 Day 4 ND ND ND F requency Cum ulativ e frequency P < 0.001Day 0 Day 4 WT WT WT 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.22 41-1 ND 25-1 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.57 ND 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.002 135-1 ND 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P < 0.001 132-1 ND Cum ulativ e frequency Cum ulativ e frequency X-A interactions, Tg 0 0.1 0.2 0.3 0.4 0.5 0 0.05 0.1 0.15 0.2 P < 0.001 130-1 ND Day 0 Day 4 Figure 2 Induction of ectopic X-A pairing, disruption of endogenous X-X pairing and compromised Xist upregulation. (a) Distribution of X-X distances in wild- type female embryonic stem cells on day 0 (left) and day 4 (middle). Cumulative frequencies at a normalized distance (ND) of 0.0?0.2 are shown at right. n � 171?173 nuclei counted. Normalized distance � Xic-Xic distance/d, where d � 2 C2 (nuclear area/p) 0.5 . Normalized distance ranges from 0 to 1. The significance of the difference between day 0 and day 4 (P) was determined by the Kolmogorov-Smirnov (KS) test, a nonparametric test to examine the null hypothesis that two independent datasets have the same distribution (SPSS 12.0 software). (b) Cumulative frequency curves for X-A distances in day 0 and day 4 transgenic female lines. n � 151?189. (c) Cumulative frequency curves for X-X distances for the cell populations in b. n � 163?193. (d) Frequency of Xist upregulation in day 4 and day 8 control and transgenic embryoid bodies, as determined by RNA FISH. (e) Cumulative frequency curves for X-A interactions in male transgenic lines. n � 151?189. NATURE GENETICS VOLUME 39 [ NUMBER 11 [ NOVEMBER 2007 1391 LETTERS example, 130-1) all grew poorly (Fig. 1b), with elevated cell death between day 4 and day 8 (Fig. 1c). By contrast, wild-type embryoid bodies and vector-only embryoid bodies (for example, 11-6) differentiated normally. Embryoid bodies carrying the minimal Tsix promoter (for example, 82-7) also behaved normally. The effects were sex specific, as growth retardation and high cell death were seen in female but not male embryonic stem cells (Fig. 1c and data not shown). This is consistent with the sex-specific manifestation of Tsix and Xite defects on embryonic stem cell differentiation observed previously 9,11,16 . To determine whether the anomalies could be related to pairing defects, we followed chromosomal movements over time using FISH. As shown previously 2,3 , the two Xs were randomly distributed relative to each other in the pre-XCI state (day 0) but became closely associated or ?paired? between day 2 and day 4 (that is, a norma- lized distance of o0.1), as shown by a left shift in X-X distance RNAi-Ctcf Merge Fibrillarin Ctcf RNAi-vector WT Merge Fibrillarin Ctcf 3.7-11 RNAi-Yy1 RNAi-vector c d e f Merge Fibrillarin Yy1 Merge Fibrillarin Yy1 WT 3.7-11 Cum ulativ e frequency Cum ulativ e frequency Cum ulativ e frequency Cum ulativ e frequency ND 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.03 Vector Ctcf 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P < 0.001 ND Vector Ctcf Vector Yy1 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.98 ND 0 0.1 0.2 0.3 0.4 0 0.05 0.1 0.15 0.2 P = 0.44 Vector Yy1 ND ba RNAi knockdown 0 10 20 30 Control Yy1 Ctcf Yy1 Ctcf ns82 ns25 ns41 Tsix Dxpas34 1 kb AE D CFCtcf sites 2.5 2 1.5 1 0.5 0 Yy1 Ctcf Control g h RNAi knockdown ?-Yy1 ?-Ctcf siRNA: yy1 Control Control siRNA: Ctcf ?-Actin ?-Actin ?-Actin ?-Yy1 ?-Actin ?-Ctcf WT 3.7-11 3.7-11 Tsix A Tsix A Tsix C Mock WT H 2 O Ctcf siRNA Control siRNA No Ab Input? -Ctcf No Ab Input? -Ctcf No Ab Input? -Ctcf i Protein b lots P ercentage n uclei with Xist cloud Relativ e Oct4 le v els Figure 3 Ctcf protein is essential for pairing at Tsix.(a) Paired Ctcf-Yy1 elements at the 5� end of Tsix.(b) Protein-blot analysis of Yy1 and Ctcf protein in the knockdown cells. Actin is used as loading control. a, antibody. (c,d) RNAi knockdown of Ctcf protein (red) in wild-type (WT) female embryonic stem cells (c) and transgenic 3.7-11 female cells (d). DAPI staining outlines the nuclear morphology (blue); immunostaining of fibrillarin (green) suggests unperturbed overall nuclear architecture. Cumulative frequency curves are shown for control (vector) and Ctcf knockdowns. We repeated all experiments at least three times. n � 153?181 nuclei counted per experiment. (e,f) RNAi knockdown of Yy1 protein (red). Experiments are as outlined in c and d.(g) ChIP using anti- Ctcf at Tsix in wild-type and transgenic 3.7-11 day 4 knockdown embryoid bodies. (h) Quantification of Oct4 mRNA by real-time RT-PCR in the Yy1, Ctcf and control knockdowns shown in c and e. Each is normalized to b-actin. Primers and PCR as described 30 .(i) Frequency of Xist upregulation in Ctcf, Yy1 and control knockdown wild-type female cells. Averages of three experiments are shown. Error bars, s.d. n 4 100 nuclei per experiment. 1392 VOLUME 39 [ NUMBER 11 [ NOVEMBER 2007 NATURE GENETICS LETTERS distributions relative to day 0 (Fig. 2a). In transgenic cells, the Xs and Xic-bearing autosomes were likewise randomly distributed on day 0, but they became closely associated on day 4 as the number of nuclei showing normalized X-autosome (X-A) distances of o0.1 increased (Fig. 2b). The only exception was 82-7, the cell line with normal growth and cell death (Fig. 1b). To determine whether ectopic X-A interactions disrupted X-X interactions, we tracked X-X distances in the same cells (Fig. 2c). Indeed, the frequency of X-X pairing was markedly reduced in all transgenic lines except 82-7. The reduced pairing frequency correlated with decreased initiation of XCI, as measured by reduced formation of Xist RNA ?clouds? (Fig. 2d). Thus, de novo X-A association blocked formation of X-X pairs and inhibited XCI in female cells. Among male lines, only the Xite subfrag- ments, pNS132, pNS135 and pNS130, induced de novo X-A pairing (Fig. 2e), con- sistent with the previous observation that pXite, but not p3.7, could nucleate X-A pairing in males 2 . The data show that the Xite sequence is more effective at nucleating pairing than Tsix, as even a 1.2-kb Xite transgene induces pairing, whereas 5- to 15-kb Tsix transgenes do not 2 . However, in contrast to that in females, X-A pairing had no apparent physiologic consequence in males (Fig. 1b), further underscoring differ- ences in the XX-versus-XY response to changes in Tsix and Xite dosage 2,16 in a manner consistent with a two-factor model for counting and choice 16,11,18 . These data indicated that pairing can be recapitulated by fragments as small as 1.2 kb with little to no obvious sequence similarity. However, we noted that several fragments contained the recently described Ctcf-Yy1 motifs 8 (Fig. 3a). Both Ctcf and Yy1 have multiple zinc fingers that can be used com- binatorially to achieve functional diversity in their roles as transcription factor and chromatin insulator. Ctcf has also been implicated in long-range chromosomal inter- actions 24,25 . To test for a role of Ctcf-Yy1 in pairing, we depleted Ctcf or Yy1 by RNAi knockdown in day 2 female embryoid bodies and harvested on day 4 for analysis. Protein blotting (Fig. 3b)and immunofluorescence (Fig. 3c?f) confirmed protein depletion. Fibrillarin (nucleolar) staining and 4,6-diamidino-2-phenylindole (DAPI) staining suggested no gross disturbance to nuclear architecture (Fig. 3c?f). By contrast, DNA FISH indicated a profound effect on pairing. In wild-type embryoid bodies, Ctcf knockdown reduced the frequency of X-X pairing to background levels (Compare Fig. 3c to day 0 data in pNS132 pNS130 pNS135 Xite pXite 1 kb Ctcf 12 3 4 Xite 1 TTGGTCACTAGGTGGCAGTA Xite 2ATAGCACATGGTAGCCCGC Xite 3 AGGCCTTGATGGTGGCAGGG Xite 4 TATTGGGGGTGGGGGGAGTT Tsix A AGGGCAGCCAGAAGGCAGCC Tsix C TCCTCCGTTATGCGGCAGGC Tsix D CCAGGTCCCCGGTGGCAGGC Tsix E ATCCAAACATGGCGGCAGAA Old consens. CCGCNNGGNGGCAG 123456789101121314151617181920 C A T G A T C A T C G A C G T A T G C G A T C T C G A A T G C C G T A C A T A T G T A G G C A T T A G A T G A T G C T C A A T C G C A T G G T A C b c d a Nucleotide position Ctcf IVT IVT-Ctcf WT comp NE ?-Ctcf WT probe Xite 1 Xite 2 Xite 3 * ++++ + ++ + ++ + + + ? ?? ? ??? ???? ?? ? ?? ? ??? ? ?? ++++ + ++ + ++ + + + ? ?? ? ??? ???? ?? ? ?? ? ??? ? ?? ++++ + ++ + ++ + + + ? ?? ? ??? ???? ?? ? ?? ? ??? ? ?? * * Mock Day 0 ES Xite 1 Day 4 ES Xite 2 Xite 3 Chic1 Mock Xite 4 C Deduced Tsix/Xite Ctcf consensus InputInputInputInput No AbNo AbNo AbNo Ab ? -Ctcf ? -Ctcf ? -Ctcf ? -Ctcf H 2 O H 2 O Figure 4 New Ctcf sites in Xite.(a)Map depicting the Xite locus, associated transgenes and new Ctcf sites. (b) Ctcf elements within Tsix and Xite as compared to the old Ctcf consensus, 5�-CCGCNNGGNGGCAG-3�.Anewlydeduced Xic Ctcf 20-bp consensus is shown below; the relative size of each nucleotide indicates the base?s probability of occurrence at each position. (c) By EMSA, in vitro?translated Ctcf (IVT-Ctcf) or day 0 embryonic stem cell nuclear extract (NE) shifts labeled Xite 1, Xite 2orXite 3 oligos. Unprogrammed lysate produces a faint shift, possibly reflecting endogenous binding activity. Comp, cold competitors at 100C2 molar excess; WT, wild-type probe; arrow, Ctcf-DNA shift; asterisks, anti-Ctcf supershift; a, antibody. (d) ChIP analysis using anti-Ctcf in day 0 and day 4 wild-type female embryonic stem cells is shown for four Xite elements and the control Chic1 locus. Mock, control immunoprecipitation with no chromatin added. P < 0.001 0.5 0.4 0.3 0.2 0.1 0 0 0.05 0.1 0.15 0.2 ND siRNA: Ctcf Control Mock Wild-type d4 ES Xite 1 Ctcf siRNA Control siRNA bc de f a siRNA: Ctcf Control Mock Xite 1 Ctcf siRNA Control siRNA Xite Tg d4 ES Xite 2 Xite 3 Wild-type d4 ES Wild-type d4 ES Xite Tg d4 ES Control Ctcf 0 0.1 0.2 0.3 0.4 0.5 0 0.05 0.1 0.15 0.2 P < 0.001 ND Xite Tg d4 ES Control Ctcf Xite 3 ?-Ctcf ?-Actin ?-Ctcf ?-Actin InputInputInput No AbNo AbNo Ab ? -Ctcf ? -Ctcf ? -Ctcf H 2 O InputInputInput No AbNo AbNo Ab ? -Ctcf ? -Ctcf ? -Ctcf H 2 O Cumulative frequency Cumulative frequency Figure 5 Ctcf protein is also required for pairing at Xite.(a) Protein-blot analysis of Ctcf and actin in knockdown and control female embryoid bodies (wild type) on day 4 of differentiation. (b) ChIP analysis of Ctcf binding at Xite and Tsix in knockdown and control cells corresponding to a.(c) Cumulative pairing frequency curves for Ctcf knockdown and control in wild-type female embryoid bodies. n � 255?277. (d) Protein-blot analysis of Ctcf and actin in knockdown and control Xite-11 transgenic embryoid bodies on day 4 of differentiation. (e) ChIP analysis of Ctcf binding in knockdown and control cells corresponding to d. (f) Cumulative pairing frequency curves for Ctcf knockdown and control in Xite-11 transgenic embryoid bodies. n � 122?135. NATURE GENETICS VOLUME 39 [ NUMBER 11 [ NOVEMBER 2007 1393 LETTERS Fig. 2a). In 3.7-11 transgenic embryoid bodies, Ctcf knockdown likewise disrupted interchromosomal pairing (Fig. 3d). Chromatin immunoprecipitation (ChIP) using antibody to Ctcf (anti-Ctcf) indicated that Ctcf binding at Tsix was indeed reduced (Fig. 3g). Notably, depleting the interacting partner, Yy1, did not affect pairing in either case (Fig. 3e,f). These results suggested that the effect of Ctcf is specific and is not a consequence of altered cell differentiation rate, as downregulation of the Oct4 pluripotency factor occurred normally (Fig. 3h); nor is it the consequence of decreased cell viability, as the number of heteropyknotic cells determined by Hoechst 33258 stain- ing 26 was {5% in both Ctcf-knockdown and control-knockdown cells (Supplementary Fig. 1 online). Measurements were taken only in the nonapoptotic fraction. The Ctcf knockdown concurrently dis- rupted formation of Xist clouds (Fig. 3i), consistent with the idea that the initiation of XCI depends on X-X pairing 2,3 .WeconcludethatCtcf is a trans-acting mediator of X-X pairing at the Tsix locus. Although Ctcf sites have been mapped within Tsix, they had not been described in Xite. Given that small Xite transgenes could also mediate pairing (Fig. 2), we next investigated whether Ctcf also binds Xite. Bioinformatic analysis using established Ctcf consensus motifs 4,5?7,27 revealed four potential sites within the tested fragments (Fig. 4a,b; note that pNS130 may contain sites that elude detection by the Ctcf algorithm, which identifies only a fraction of in vivo sites 27 ). Electrophoretic mobility shift assays (EMSA) showed that in vitro synthesized Ctcf protein gel-shifted 32 P-labeled Xite sites, which were competed away by unlabeled Xite oligonucleotides (Fig. 4c and data not shown). Unprogrammed lysates produced only a faint shift (endogenous activity). Incubation with nuclear extract from the embryonic stem cells resulted in a DNA-protein complex that super- shifted when exposed to anti-Ctcf (Fig. 4c). Furthermore, ChIP showed that Ctcf binds all four Xite elements in pre-XCI female embryonic stem cells (day 0). On day 4, Ctcf binds all but Xite 4 (Fig. 4d). We conclude that, like Tsix, Xite complexes with Ctcf both in vitro and in vivo. We deduced a 20-bp Ctcf consensus for the Xic (Fig. 4b) using established methods 27,28 . To determine whether Ctcf also mediates pairing at Xite,we knocked down Ctcf in wild-type female embryoid bodies on day 2 and harvested cells for analysis on day 4. Protein analysis indicated that Ctcf protein was considerably depleted (Fig. 5a); ChIP analysis confirmed that Ctcf binding at Xite was consequently disrupted (Fig. 5b). Examination of interchromosomal distances revealed that Ctcf depletion did markedly reduce X-X pairing (Fig. 5c). We observed a similar effect on X-A pairing in the Xite-11 (ref. 16) transgenic line when Ctcf was knocked down and Ctcf binding was at least partially disrupted at Xite (Fig. 5d?f; note that residual binding may reflect residual Ctcf activity, as shown by protein-blot analysis). Thus, we propose that Ctcf mediates interchromosomal pairing at both Tsix and Xite, either by directly bridging two chromosomes or indirectly through other factors. The occurrence of Ctcf-binding sites is therefore one notable common denominator among disparate pairing-competent transgenes. Another common thread among the transgenes is their potential to drive transcription, as cryptic promoters have been shown to exist outside of the primary Tsix and Xite initiation sites?all of which are F requency 0.5 h 0 0.05 0.10 0.15 0.20 0.25 0.1 0.3 0.5 0.7 0.9 P = 0.05 ND 1.0 h 0 0.05 0.1 0.15 0.2 0.1 0.3 0.5 0.7 0.9 P = 0.001 ND 2.0 h 0 0.05 0.1 0.15 0.2 0.25 0.1 0.3 0.5 0.7 0.9 P = 0.05 ND 0 0.05 0.10 0.15 0.20 0.25 0.1 0.3 0.5 0.7 0.9 0 h ND 0 0.1 0.2 0.3 0.4 0 0.1 0.2 0 h 0.5 h 1.0 h 2.0 h ND de ca 0.0 0.2 0.4 0.6 0.8 1.0 012345 Hours b 0 0.1 0.2 0.3 0.4 0 0.1 0.2 0 h 4 h ND P = 0.05 0 0.1 0.2 0.3 0.4 0 0.1 0.2 0 h 8 h ND P = 0.04 f XiteDXPas34 Tsix 1 kb X Tsix/Xite transcription Other pairing factors Ctcf protein Xic Major promoter Minor promoter Cis pairing elements X Cum ulativ e frequency Cum ulativ e frequency Cum ulativ e frequency Relativ e Tsix le v els Figure 6 Analysis of transcription requirements. (a) Cumulative X-X pairing frequencies in day 2 wild-type female embryoid body cells treated with 5 mg/ml actinomycin D for 0 or 4 h. n � 159?208. (b) Cumulative X-X pairing frequencies in day 2 wild-type female embryoid body cells treated with 10 mg/ml a-amanitin for 8 h. n � 105?222. (c) Quantitative real-time, strand-specific RT-PCR of Tsix RNA confirmed that actinomycin D (shown) and a-amanitin (not shown) treatment disrupted Pol II activity. RNA levels are normalized to that of the 18S rRNA. Averages from three experiments. Error bars, s.d. (d) X-X distance distribution profiles in day 2 wild-type female embryoid body cells after 0, 0.5, 1.0 or 2.0 h of actinomycin D treatment. Kolmogorov- Smirnov test (P) compares each time point with the 0-h untreated sample. n � 160?171. (e) Cumulative frequency curves of the distributions shown in d. (f)Summaryofcis and trans factors for pairing at Tsix and Xite. Left, the critical pairing region and functionally redundant pairing elements therein (light green triangles). Ctcf sites are found throughout Tsix and Xite. Ctcf cannot be working alone, as pNS130 appears to lack binding sites but still nucleates pairing. Other pairing proteins (hexagons) are likely to exist. Transcriptional activity is also required (note that multiple minor and cryptic promoter elements can be found throughout the region). Right, a working model in which X-X pairing is held together by Ctcf proteins and transcriptional activity or RNA at Tsix and Xite. 1394 VOLUME 39 [ NUMBER 11 [ NOVEMBER 2007 NATURE GENETICS LETTERS DNA polymerase II (Pol II) directed 9,12,21?23 . To determine whether transcriptional activity is required for pairing, we treated wild-type embryoid bodies (day 2) with either actinomycin D for 4 h (Fig. 6a) or a-amanitin for 8 h (Fig. 6b) and confirmed Pol II inhibition by measuring a drop in Tsix transcription (Fig. 6c). Notably, Pol II inhibition resulted in a consistent disruption of X-X pairing. To establish the kinetics of pairing loss, we treated day 2 wild-type female embryoid bodies with actinomycin D for 0, 0.5, 1.0 and 2.0 h (Fig. 6d). At 2 h and beyond, the X-X distribution profiles differed significantly from those at the 0-h time point (P � 0.05) and resembled the random distribution typical of pre-XCI cells (compare Fig. 6d, 2-h time point, and Fig. 2a, day 0), indicating that pairing persists for less than 2 h in the absence of new transcription. At the 1-h time point, the distribution was intermediate, with fewer pairs in the 0.0?0.2 normalized-distance range than at the 0-h time point (P � 0.001, Fig. 6d), indicating that pairs begin to disintegrate after B1 h. Notably, at 0.5 h, the profile was bimodal: whereas the 0.2?1.0 normalized-distance profile suggested a random distribution, the number of pairs separated by a normalized distance o0.1 remained unchanged (Fig. 6d,e). Three independent experiments yielded similar results at all time points. Thus, inhibition of transcription immediately abolished the formation of new X-X pairs but did not disrupt pre- existing pairs. These results suggested that nucleation of X-X pairing requires new transcription or its associated RNA, but the maintenance of existing pairs does not. Conveniently, the sensitivity of new pair formation to transcrip- tional inhibitors enabled us to estimate the half-life (t 1/2 ) of X-X pairs. Because more than 50% of the X-X pairs with a separation of normalized distance o0.1 were lost after 1 h of actinomycin D treatment (Fig. 6d), the half-life must be o1.0 h. This range is consistent with the observations that X-X pairing occurs very tran- siently 2,3 and that an X-X separation of normalized distanceo0.1 can be observed in fewer than 15?20% of any day 2?4 embryoid body population (Fig. 2a). In conclusion, our analysis has revealed new cis and trans require- ments for homologous chromosome pairing. Within the 15-kb critical domain for counting, choice and pairing 16 , fragments as small as 1?2 kb?each with overtly different primary sequences?can promote pair formation, at least when multimerized on a transgene array. We have identified Ctcf sites as a common motif among them and shown that the Ctcf protein mediates interchromosomal pairing. Although we cannot exclude indirect and general effects of Ctcf on pairing, the abundance of Ctcf motifs in Tsix and Xite indicates that at least some effects must be locus specific. Ctcf could directly mediate pairing (Fig. 6f) or collaborate with additional factors. Given Ctcf?s ubiqui- tous nature, it seems likely that additional factors must be involved in regulation. Indeed, pairing may occur co-transcriptionally with Tsix and Xite, as new transcription is required to create new pairs. Together with Ctcf protein, Tsix and Xite RNAs may form an ?RNA-protein bridge? between two Xs (Fig. 6f). Once formed, such an RNA-protein bridge might have transient stability, accounting for the apparent insensitivity of existing pairs to transcriptional inhibitors during the first half-hour of treatment. With other epigenetic phenomena being subject to intra- and interchromosomal regulation 24,25,29 , the proper- ties described here for homologous chromosome pairing are likely to be relevant for many other mammalian loci. METHODS Embryonic stem cell lines, transgenesis and culture. Wild-type male J1 (40XY) and female 16.7 (40XX) embryonic stem cell lines and transgenic lines 3.7-11 and Xite-11 have all been previously described 11,16 . We created new transgenic lines by linearizing 30?40 mgofplasmids(Supplementary Table 1; plasmids as described in ref. 23), electroporating the DNA into 10 7 16.7 or J1 cells at 240 V and 500 mF using the Bio-Rad Genepulser, and selecting for clones in 300 mg/ml G418 for 7?9 days. We analyzed at least three independent clones for each construct (each with similar results). All clones were maintained under G418 selection. Differentiation was induced by EB suspension culture and withdrawal of LIF. On day 4, we attached embryoid bodies to gelatinized plates to promote outgrowth of differentiated cells. We carried out analysis of cell death using the trypan blue method as described 16 . FISH. Embryonic stem cells were trypsinized into single cells, cytospun on glass slides and fixed in 4% paraformaldehyde/1C2 PBS. We then carried out FISH as described 2,10 . We labeled probes with fluorescein-12-dUTP, Cy3-dUTP or AlexaFluor-555-dUTP by nick translation (Roche). We used probes pSxN and pSx9 for general detection of the Xic,aneo probe for specific detection of the transgene locus 2 and a single-stranded riboprobe cocktail for detection of the Xist RNA 23 . For analysis of pairing, we took digital images with the Zeiss AxioPlan2 and processed them using OpenLab software (Improvision; see details in ref. 2). In brief, z sections were captured at 0.2 m intervals and three- dimensional images were projected on a single two-dimensional plane. We measured Xic-Xic distances (x) and the nuclear areas (A) using the OpenLab software. Only nuclei with two resolvable X signals were scored (single dots were excluded). ?Normalized distance? (ND) is defined as x/d,whered is the nuclear diameter, defined as 2(A/p) 0.5 . Transgene copy number analysis. We estimated transgene copy number by quantitative FISH. For each clone, we carried out FISH using a probe corresponding to the Xic transgene fragment and determined the total intensity in arbitrary units (S) using Openlab software, which integrates the intensities of all the pixels in the signal area. The transgene signal (S tg ) was distinguished from the endogenous Xic signal (S x ) by colocalization to a neo probe. We determined the transgene copy number as S tg /S x and classified it as low (2?5 copies), medium (6?10 copies) or high (410 copies). RNAi knockdown. We carried out siRNA-mediated knockdowns of Ctcf and Yy1 proteins using mouse Ctcf SMARTpool siRNA and yy1 BS/U6 Pol III vectors, respectively, as described 8 . We used vector-only or gfp siRNAs in control knockdowns. In brief, small hairpin RNA (shRNA; 4 mg) or short interfering RNAs (siRNAs; 0.8 mM) were delivered into day 2 wild-type or transgenic female embryoid bodies using the Amaxa Biosystems Nucleofector/ Mouse ES Nucleofector Kit. Day 4 embryoid bodies were harvested 46?48 h later. Ctcf and Yy1 knockdowns were confirmed by protein-blot analysis using rabbit anti-Ctcf (Upstate), mouse anti-Yy1 (Santa Cruz H10) and mouse anti? b-actin (Chemicon), with the appropriate rabbit or mouse secondary anti- bodies conjugated to horseradish peroxidase for detection (Western Lightening Chemiluminescence Kit; Perkin Elmer Life Sciences). Immunostaining was carried out using anti-fibrillarin (Abcam 38F3), anti-Yy1 and anti-Ctcf, with detection using secondary Alexa555-goat anti-rabbit or Alexa488 anti-mouse (Molecular Probes). EMSA. We carried out EMSAs using 20 fmol of oligonucleotides with 1?2 mg undifferentiated embryonic stem cell nuclear extract or 1 ml in vitro?translated Ctcf (TNT-coupled reticulocyte system, Promega) 4 . Cold competitors were added at 100C2 molar excess. See Supplementary Table 2 online for oligo sequences. Supershifts were carried out with 2 ml of anti-Ctcf (Upstate). ChIP. ChIP, antibodies, real-time PCR, and Tsix and Chic1 PCR primers have been described 8 . Xite primer sequences are shown in Supplementary Table 2. For ChIP on knockdown samples, we harvested 3 C2 10 6 ?10 C2 10 6 cells nucleofected cells on day 4 to perform ChIP. Transcription inhibition. Three independent biological replicates yielded similar results. Day 2 wild-type female embryoid bodies were treated with 5 mg/ml actinomycin D or 10 mg/ml a-amanitin as described 21,22 . Embryoid bodies were then harvested at 0, 0.5, 1.0, 2.0, 4.0 and 6.0 h and subjected to FISH or quantitative RT-PCR analysis. For real-time RT-PCR, we isolated RNAs using Trizol, treated 1 mg of total RNA with RNase-free DNase, reverse transcribed the RNA using random primers, and amplified 200 ng of cDNA NATURE GENETICS VOLUME 39 [ NUMBER 11 [ NOVEMBER 2007 1395 LETTERS using the Bio-Rad iCycler iQ real-time PCR detection system in SYBRGreen. All samples were normalized to the 18S rRNA internal control in Pol II inhibition experiments 22 . Tsix RT-PCR was used to confirm Pol II inhibition using primers as described 8 . Note: Supplementary information is available on the Nature Genetics website. ACKNOWLEDGMENTS We thank all members of the Lee laboratory for stimulating discussion and feedback throughout this work. S.S.S. is partially funded by a doctoral fellowship from the Gulbenkian Institute?Portugal (Fundacao para a Ciencia e a Tecnologia SFRH/BD/9614/2002). This work was supported by an US National Institutes of Health grant (RO1-GM58839) to J.T.L., who is an investigator of the Howard Hughes Medical Institute. AUTHOR CONTRIBUTIONS N.X., M.E.D. and J.T.L. designed this study; N.X. created and analyzed transgenes and performed pairing, RNA FISH, immunofluorescence and transcription analyses; M.E.D. performed knockdowns, western blot, ChIP, EMSA, bioinformatics and qRT-PCR analyses; S.S. contributed to pairing analyses; J.T.L. wrote the paper and supervised the research. Published online at http://www.nature.com/naturegenetics Reprints and permissions information is available online at http://npg.nature.com/ reprintsandpermissions 1. Lyon, M.F. Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature 190, 372?373 (1961). 2. Xu, N., Tsai, C.L. & Lee, J.T. Transient homologous chromosome pairing marks the onset of X inactivation. Science 311, 1149?1152 (2006). 3. Bacher, C.P. et al. Transient colocalization of X-inactivation centres accompanies the initiation of X inactivation. Nat. Cell Biol. 8, 293?299 (2006). 4. Chao, W., Huynh, K.D., Spencer, R.J., Davidow, L.S. & Lee, J.T. CTCF, a candidate trans-acting factor for X-inactivation choice. Science 295, 345?347 (2002). 5. Bell, A. & Felsenfeld, G. Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene. Nature 405, 482?485 (2000). 6. Hark, A.T. et al. CTCF mediates methylation-sensitive enhancer-blocking activity at the H19/Igf2 locus. Nature 405, 486?489 (2000). 7. Kanduri, C. et al. Functional association of CTCF with the insulator upstream of the H19 gene is parent of origin-specific and methylation-sensitive. Curr. Biol. 10, 853?856 (2000). 8. Donohoe, M.E., Zhang, L., Xu, N., Shi, Y. & Lee, J.T. Identification of a CTCF co-factor, YY1, for the X-chromosome binary switch. Mol. Cell 25, 43?56 (2007). 9. Ogawa, Y. & Lee, J.T. Xite, X-inactivation intergenic transcription elements that regulate the probability of choice. Mol. Cell 11, 731?743 (2003). 10. Lee, J.T., Davidow, L.S. & Warshawsky, D. Tsix, a gene antisense to Xist at the X-inactivation centre. Nat. Genet. 21, 400?404 (1999). 11. Lee, J.T. & Lu, N. Targeted mutagenesis of Tsix leads to nonrandom X inactivation. Cell 99, 47?57 (1999). 12. Sado, T., Wang, Z., Sasaki, H. & Li, E. Regulation of imprinted X-chromosome inactivation in mice by Tsix. Development 128, 1275?1286 (2001). 13. Brockdorff, N. et al. The product of the mouse Xist gene is a 15 kb inactive X-specific transcript containing no conserved ORF and located in the nucleus. Cell 71, 515?526 (1992). 14. Brown, C.J. et al. The human XIST gene: analysis of a 17 kb inactive X-specific RNA that contains conserved repeats and is highly localized within the nucleus. Cell 71, 527?542 (1992). 15. Wutz, A., Rasmussen, T.P. & Jaenisch, R. Chromosomal silencing and localization are mediated by different domains of Xist RNA. Nat. Genet. 30, 167?174 (2002). 16. Lee, J.T. Regulation of X-chromosome counting by Tsix and Xite sequences. Science 309, 768?771 (2005). 17. Vigneau, S., Augui, S., Navarro, P., Avner, P. & Clerc, P. An essential role for the DXPas34 tandem repeat and Tsix transcription in the counting process of X chromosome inactivation. Proc. Natl. Acad. Sci. USA 103, 7390?7395 (2006). 18. Lee, J.T. Homozygous Tsix mutant mice reveal a sex-ratio distortion and revert to random X-inactivation. Nat. Genet. 32, 195?200 (2002). 19. Navarro, P., Pichard, S., Ciaudo, C., Avner, P. & Rougeulle, C. Tsix transcription across the Xist gene alters chromatin conformation without affecting Xist trans- cription: implications for X-chromosome inactivation. Genes Dev. 19, 1474?1484 (2005). 20. Sado, T., Hoki, Y. & Sasaki, H. Tsix silences Xist through modification of chromatin structure. Dev. Cell 9, 159?165 (2005). 21. Sun, B.K., Deaton, A.M. & Lee, J.T. A transient heterochromatic state in Xist preempts X inactivation choice without RNA stabilization. Mol. Cell 21, 617?628 (2006). 22. Cohen, D.E. et al. The Dxpas34 repeat regulates random and imprinted X-inactivation. Dev. Cell 12, 57?71 (2007). 23. Stavropoulos, N., Rowntree, R.K. & Lee, J.T. Identification of developmentally specific enhancers for Tsix in the regulation of X chromosome inactivation. Mol. Cell. Biol. 25, 2757?2769 (2005). 24. Kurukuti, S. et al. CTCF binding at the H19 imprinting control region mediates maternally inherited higher-order chromatin conformation to restrict enhancer access to Igf2. Proc. Natl. Acad. Sci. USA 103, 10684?10689 (2006). 25. Ling, J.Q. et al. CTCF mediates interchromosomal colocalization between Igf2/H19 and Wsb1/Nf1. Science 312, 269?272 (2006). 26. Syken, J., De-Medina, T. & Munger, K. TID1, a human homolog of the Drosophila tumor suppressor l(2)tid, encodes two mitochondrial modulators of apoptosis with opposing functions. Proc. Natl. Acad. Sci. USA 96, 8499?8504 (1999). 27.Kim,T.H.et al. Analysis of the vertebrate insulator protein CTCF-binding sites in the human genome. Cell 128, 1231?1245 (2007). 28. Workman, C.T. et al. enoLOGOS: a versatile web tool for energy normalized sequence logos. Nucleic Acids Res. 33, W389?92 (2005). 29. Spilianakis, C.G., Lalioti, M.D., Town, T., Lee, G.R. & Flavell, R.A. Interchromo- somal associations between alternatively expressed loci. Nature 435,637?645 (2005). 30. Nichols, J. et al. Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 95, 379?391 (1998). 1396 VOLUME 39 [ NUMBER 11 [ NOVEMBER 2007 NATURE GENETICS LETTERS "
Add Content to Group
|
Bookmark
|
Keywords
|
Flag Inappropriate
share
Close
Digg
Facebook
MySpace
Google+
Comments
Close
Please Post Your Comment
*
The Comment you have entered exceeds the maximum length.
Submit
|
Cancel
*
Required
Comments
Please Post Your Comment
No comments yet.
Save Note
Note
View
Public
Private
Friends & Groups
Friends
Groups
Save
|
Cancel
|
Delete
Please provide your notes.
Next
|
Prev
|
Close
|
Edit
|
Delete
Genetics
Gene Inheritance and Transmission
Gene Expression and Regulation
Nucleic Acid Structure and Function
Chromosomes and Cytogenetics
Evolutionary Genetics
Population and Quantitative Genetics
Genomics
Genes and Disease
Genetics and Society
Cell Biology
Cell Origins and Metabolism
Proteins and Gene Expression
Subcellular Compartments
Cell Communication
Cell Cycle and Cell Division
Scientific Communication
Career Planning
Loading ...
Scitable Chat
Register
|
Sign In
Visual Browse
Close
Comments
CloseComments
Please Post Your Comment