A role for nucleoprotein Zap3 in the reduction of telomerase activity during embryonic stem cell differentiation

Mechanisms of Development 121 (2004) 1509–1522 www.elsevier.com/locate/modo

A role for nucleoprotein Zap3 in the reduction of telomerase activity during embryonic stem cell differentiation Lyle Armstronga,1, Majlinda Lakob,1, Ine van Herpea, Jerry Evansb, Gabriele Saretzkic, Nicholas Holea,* a

b

School of Biological and Biomedical Sciences, University of Durham, South Road, Durham DH1 3LE, UK Institute of Human Genetics, International Centre for Life, University of Newcastle, Newcastle upon Tyne NE1 3BZ, UK c Institute for Ageing and Health, University of Newcastle, Newcastle upon Tyne NE4 6BE, UK Received 2 February 2004; received in revised form 14 July 2004; accepted 14 July 2004 Available online 24 August 2004

Abstract Telomerase, the enzyme which maintains the ends of linear chromosomes in eukaryotic cells is found in murine embryonic stem cells; however, its activity is downregulated during in vitro differentiation. Previous work has indicated that this is due to the transcriptional downregulation of murine reverse transcriptase unit (mTert) of telomerase. To investigate the factors that cause the transcriptional repression of mTert we defined a 300 bp region which is essential for its transcription and performed site directed mutagenesis and electrophoretic mobility shift assays. This analysis indicated that Sp1, Sp3 and c-Myc bind to the GC-boxes and E-boxes, respectively, within the promoter and help activate the transcription of mTert gene. We also identified a novel binding sequence, found repeated within the mTert core region, which when mutated caused increased mTert expression. Yeast one hybrid screening combined with electrophoretic mobility shift assays indicated that the nuclear protein Zap3 binds to this site and its overexpression leads to the downregulation of mTert during differentiation. This suggests that regulation of mTert transcription is a complex process which depends on a quantitative balance between transcription factors that cause activation or repression of this gene. Overexpression of Zap3 in murine embryonic stem cells results in reduction in telomerase activity and telomere length as well as reduced proliferative capacity and limited ability to contribute to the development of haematopoietic cells upon differentiation. q 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Murine telomerase reverse transcriptase; Haematopoiesis; Embryonic stem cells; TRAP assay; GFP; Zap3; Hematopoietic commitment; Cell cycle; Cell proliferation

1. Introduction Telomeres are thought to protect the ends of all linear chromosomes, allow their complete replication and help maintain genetic stability (Blackburn, 1991; Greider, 1994). In the absence of a mechanism to maintain telomeres, progressive shortening occurs with each cell division due to

Abbreviations: GFP, green fluorescent protein; HSC, hematopoietic stem cell; mTert, murine telomerase reverse transcriptase; TRAP, telomerase repeat amplification protocol. * Corresponding author. Tel.: C44-191-334-1244; fax: C44-191-3741201 E-mail address: [email protected] (N. Hole). 1 Both authors contributed equally to this work. 0925-4773/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mod.2004.07.005

the end replication problem, eventually leading to cessation of cell proliferation (Watson, 1972). Replication of telomeric DNA is carried out by telomerase, a specialised ribonucleoprotein enzyme complex that stabilises telomere length by adding hexameric (TTAGGG) repeats to the ends of chromosomes (Morin, 1989; Moyzis et al., 1988). Telomerase is found in embryonic cells and adult male germline cells (Wright et al., 1996) the proliferative cells of renewing tissues, e.g. haematopoietic stem cells and activated lymphocytes (Hiyama et al., 1995; Broccoli et al., 1995; Chiu et al., 1996), basal cells of the epidermis (Harle-Bachor and Boukamp, 1996) and intestinal crypt cells (Hiyama et al., 1996). In most normal somatic cells that lack telomerase activity for most of their cell cycle, progressive telomere shortening is observed, eventually

1510

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

leading to a limited replicative capacity (Allsopp et al., 1992; Masutomi et al., 2003). This causal relationship between telomere degradation and replicative senescence has led to the hypothesis that telomere length may act as a ‘mitotic clock’ associated with cellular aging in vivo (Harley, 1991). In contrast to most normal cell types which show at best diminished levels of telomerase activity, telomerase is upregulated in almost 90% of all human cancers (Shay, 1997). In addition to maintaining telomere length, telomerase may also prevent the loss of G-rich single stranded overhangs that participate, together with specific telomere binding proteins, in forming the T loop structure (Stewart et al., 2003; Masutomi et al., 2003). Erosion of the telomeric overhang seen in senescent cells is likely to cause collapse of the T loop structure and uncapping of the telomere end that eventually signals DNA damage and activates the senescence pathway (Masutomi et al., 2003). Consistent with this, a reduction in telomerase activity can cause erosion of the overhangs in normal human cells, whilst increased activity of telomerase maintains the overhangs at a similar length to that found in early passage cells (Masutomi et al., 2003). The telomerase complex comprises two main components: the catalytic subunit (TERT) that catalyses the addition of new repeats (Martin-Riviera et al., 1998; Harrington et al., 1997) and a structural RNA component (TR) containing the template region that binds the telomere repeats (Blasco et al., 1995). Several lines of evidence suggest that TERT plays a major role in regulating telomerase activity, although these have primarily been restricted to human cell types (Meyerson et al., 1997; Bodnar et al., 1998; Wen et al., 1998). In a recent publication (Armstrong et al., 2000) we reported the construction of a reporter cassette in which the expression of green fluorescent protein (GFP) was under the control of the promoter of the murine mTert gene and showed that the changes in the level of GFP fluorescence closely followed the expression level of mTert when the cassette was stably transfected into murine embryonic stem cells. These cells were then induced to differentiate and the levels of telomerase activity were observed to decrease as differentiation progressed. A strong correlation was established between the GFP fluorescent intensity and telomerase activity thereby establishing a clear link between mTert expression and telomerase activity (Armstrong et al., 2000). These data are consistent with observations from human cell lines where levels of hTERT expression have been correlated with telomerase activity (Meyerson et al., 1997). In addition these data indicate that transcriptional control of the Tert gene plays a key role in the activation of telomerase thus a thorough understanding of such transcriptional control is essential to understand the regulation of telomerase. At present, our understanding of this transcriptional control is incomplete. A number of transcriptional regulatory proteins have been implicated in the regulation of human or murine Tert, including c-myc, Sp1, WT-1

and NF-kb (Oh et al., 1999a,b, 2000; Kyo et al., 2000), as has histone acetylation (Takakura et al., 2001; Hou et al., 2002) and methylation (Nomoto et al., 2002; Bechter et al., 2002). However, almost all previous work on Tert regulation has used neoplastic and/or immortalised cell lines. In order to better understand telomerase regulation in primary stem cells, we have investigated Tert transcriptional regulation in murine embryonal stem cells (ES cells). We have already demonstrated their utility in the correlation of mTert expression with telomerase activity (Armstrong et al., 2000), and these cell types can be differentiated in vitro into somatic cell types (Hole et al., 1996), offering us an opportunity to examine telomerase regulation during differentiation. We report here the definition of a 300 bp region which is essential for the transcription of mTert gene and confirm the presence of a number of transcription factors such as c-myc, Sp1 and Sp3 which have also been shown to regulate the transcription of human TERT. More interestingly, one protein whose binding increased as Tert expression decreased was identified and evidence from site directed mutagenesis indicated that this protein normally functions as a repressor of mTert. Yeast one hybrid screening revealed that the protein which binds to the core promoter of mTert and controls its downregulation during the in vitro differentiation of murine embryonic stem cells is a novel transcription factor, Zap3. This has a profound influence on the functional characteristics of the cells such as telomerase activity and telomere length and their proliferative ability which may be demonstrated by their reduced rate of growth and their limited ability to contribute to the development of haematopoietic cells upon differentiation.

2. Results and discussion 2.1. Identification of the core promoter region for the murine telomerase reverse transcriptase gene The full length mouse Tert cDNA clone has been previously described (Martin-Riviera et al., 1998; Greenberg et al., 1998). For mTert transcripts isolated from fibroblasts or testis, the transcription start site has been reported as being 29 bp upstream of the translation initiation site (see Fig. 1a,TS1). However, it was necessary to identify the transcription start site in mouse embryonic stem cells. Computer searching of several databases showed the presence of two Riken EST clones (Genbank accession number BB651920 and BB618671) for mTert isolated from mouse ES cells and 8 day mouse embryo, respectively. These libraries are highly enriched in full length rare transcripts (especially at the 5 0 end) since they were created by using a subtraction procedure that eliminates abundant transcripts and cap-trapping to generate full length single stranded cDNAs (Carninci et al., 2000). Comparison of the sequence of these 5 0 ESTs to the genomic sequence for mTert located

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

1511

Fig. 1.. Definition of the core promoter region for mTert. (a) The sequence of the core region needed for the mTert transcription. The sequences of the E-boxes, GC-boxes and UNK-boxes are underlined. The initiating ATG codon is shown in boldface. The nucleotides that were changed by site directed mutagenesis are shown in italics above the consensus motifs for the binding sites. The positions of the bases are indicated relative to the ES cell transcriptional start site TS2. The transcription start site reported for other tissues by Greenberg et al. (1998) and Martin-Riviera et al. (1998) is represented by TS1. (b) Deletion analysis of mTert promoter and 5 0 untranslated region. Murine embryonic stem cells (CGR8) were transiently transfected with a series of GFP constructs containing the 5 0 flanking DNA of the mTert gene. The GFP intensity produced by the promoterless construct pEGFP-1 was normalised to a value of 1; GFP intensity of other constructs are shown relative to this control. Results are expressed as the mean of at least three independent experiments, each performed in triplicates. The standard deviations of the means are indicated by the error bars. The positions of the bases are indicated relative to the transcription start site (C1). (c) Mutation analysis of the core region of mTert. Murine embryonic stem cells (CGR8) were transiently transfected with a series of GFP constructs containing the substitution mutation for each of the factor binding site in the mTert core promoter. The GFP intensity of the promoterless construct pEGFP-1 was normalised to 1 and the relative GFP intensity of the other constructs is shown. Results are expressed as the mean of at least three independent experiments, each performed in triplicates. The standard deviations of the means are indicated by the error bars. P-values were calculated using student T-test. (d) Effects of differentiation on the mTert-GFP constructs. The core region construct of mTert (K195/105-GFP) and the other constructs containing the substitution mutation for each of the factor binding sites in the mTert core region, as well as one construct with mutations in both UNK sites (UNK1/UNK-2 mut) were stably transfected into murine embryonic stem cells. The isolated clones were allowed to differentiate for 6 days. The GFP intensity of the promoterless construct pEGFP-1 was normalised to 1 and the relative GFP intensity of the other constructs is shown before and after differentiation. Results are expressed as the mean of at least three independent experiments, each performed in triplicates. The standard deviations of the means are indicated by the error bars. The GFP intensity at D6 compared with undifferentiated cells is given by a percentage figure above each data set.

the transcription start site 114 bp upstream of the translation initiation site (see Fig. 1a, TS2). Since the transcription start site 2 was defined in ES cells we decided to use it as point of reference (C1) for our deletion constructs and definition of the minimal region needed for mTert transcription. We have previously demonstrated that telomerase activity decreases upon differentiation of murine ES cells into the various cell types which occur in the embryoid bodies (Armstrong et al., 2000) and correlated this to the corresponding decrease in mTert expression. In this work

we have examined the transcriptional control of mTert expression during differentiation of ES cells by investigating the binding of proteins in the nuclear extracts of embryoid bodies at different stages of differentiation. For this purpose, an analysis of the mTert promoter and definition of a core region was indispensable. To be able to do this we isolated various lengths DNA fragments from the promoter and 5 0 untranslated region of mouse mTert gene (accession number AF 121949) by PCR of mouse genomic DNA and inserted them into the multiple cloning

1512

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

site of the promoterless GFP vector, pEGFP1, (BD Biosciences). These reporter constructs were transiently transfected into murine ES cells (CGR8) and the mean fluorescence intensity was measured using flow cytometry. As can be seen from the Fig. 1b the longest construct (K4372/105) shows good transcriptional activity compared to the promoterless construct, pEGFP-1. Truncations of this construct from K4372 to K1371 shows some decrease in transcriptional activity suggesting the presence of one or more repressor binding sites within the segment K1371 to K371. The highest transcriptional activity was shown by the K195/105 construct and further truncations led to a significant reduction in transcriptional activity (see Fig. 1b). From this analysis we concluded that the region responsible for maximal promoter activity extends from K195 to C105 bp from the transcriptional start site. The definition of a core region permitted bioinformatic analysis (SignalScan software) of the sequence of this DNA fragment to identify potential binding sites for known transcription factors. Three GC boxes (GC-box 1 to 3) known to bind Sp-1 family proteins and two E-boxes (E-box 1 and E-box 2) known to bind bHLH proteins were identified (see Fig. 1a). Interestingly, two additional sites (here named UNK-1 and UNK-2) were identified. To investigate these potential binding sites further, mutations were introduced using site directed mutagenesis into the core promoter as shown in Fig. 1a. The mutated constructs were transiently transfected into murine ES cells and the mean GFP fluorescence intensity was measured after 24 h. As can be seen from Fig. 1c mutation of the first GC-box (GC-box1) resulted in 15% decrease in promoter activity, while mutations in GC-box 2 and GC-box 3 resulted in 32.5 and 37.8% decrease in promoter activity, respectively. Mutation of E-box 1 caused only 11.2% decrease in promoter activity which provides an interesting contrast to the results reported by Nozawa et al. (2001) wherein a 50% decrease in promoter activity was observed following mutation of the same E-box from a luciferase reporter construct transiently expressed in C2C12 myoblasts. Similarly the mutation of E-box 2 caused 15.2% decrease in promoter activity (see Fig. 1c). Mutations which may be of greater significance in this system and are statistically significant are those concerning the UNK-2 binding site and the UNK-1 and UNK-2 sites together (see Fig. 1a) which show a considerable increase in promoter activity (63.7 and 90%, respectively; see Fig. 1c). Mutations in the UNK-1 site which shows sequence similarity to UNK-2 also caused an increase in promoter activity (30.4%) albeit at a lower level (see Fig. 1c). Although these data suggest that the UNK sites may play a role in controlling the absolute levels of mTert transcription, we were interested as to whether they were involved in regulating the dramatic changes in mTert expression during ES cell differentiation. We thus generated ES cell stable transfected lines, which were then differentiated for 6 days, a process that we had previously shown to reduce normal

mTert expression (Armstrong et al., 2000). Fig. 1d shows that expression driven by the K195/105 core promoter in day 6 differentiated cells was 53% of that in undifferentiated cells, a result consistent with our previous studies with the full length promoter (Armstrong et al., 2000), suggesting that most, if not all, of the transcriptional regulation sites important in differential expression were contained within this core promoter region. Mutations in the GC boxes maintained this differential expression, although the relative change was larger, where day 6 differentiated cells had between 30 and 40% of expression in undifferentiated cells. Mutations in the two E-boxes also maintained differential expression. In marked contrast, mutations in either UNK-1 or UNK-2 produced expression in day 6 differentiated ES cells only moderately lower than expression in undifferentiated ES cells. Mutation in both UNK1 and UNK-2 sites resulted in expression at day 6 that was 92% of that of undifferentiated cells. A possible interpretation of these results is that members of the Sp1 and bHLH family which bind to the GC and E boxes, respectively cooperate together to cause activation of the mTert transcription, while another factor(s) binding at the UNK-1 and UNK-2 causes repression of mTert activity. 2.2. Identification of transcription factors that bind to the core region of mTert using electrophoretic mobility shift assays (EMSA) We have shown that differentiation of murine embryonic stem cells is accompanied by a marked downregulation of mTert transcriptional activity and a consequent decrease in telomerase activity (Armstrong et al., 2000). We were therefore interested to identify transcription factors that bind to the core promoter and cause this downregulation. For this we performed EMSA with 5 mg of nuclear extract from undifferentiated cells (CGR8) and embryoid bodies at days 2, 4 and 6 of differentiation as described in Experimental procedures. When oligonucletides containing GC-box 1, GC-box 2 and GC-box3 were used as probes two DNAprotein complexes were observed (see Supplementary information). This binding pattern did not change when nuclear extracts from differentiating EBs were used in the EMSA. The specificity of binding was confirmed by addition of wild type competitor which abolished binding and mutated competitor which did not alter binding at all. The intensity of the lower band was reduced by addition of anti-Sp3 antibody while the upper band was reduced by addition of either antiSp1 or anti-Sp3 (data not shown). Based on these results we concluded that the upper band was derived from the Sp1-DNA complex or Sp3-DNA complex while the lower one from the Sp3-DNA complex. Since mutations in the three GC-boxes cause a reduction in promoter activity in ES cells we concluded that Sp1 and Sp3 function as transactivators of mTert in murine ES cells. We observed a three-band pattern when we used E-box 1 and E-box 2 oligonucleotides as probes (data not shown). The slower migrating band was not

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

competed either by the addition of excess wild type competitor or a mutated probe in which the sequence of the E-box1 or E-box 2 binding site was altered. From this we concluded that this band was likely to represent non-specific binding. The middle band was competed by the wild type competitor and the c-Myc consensus sequence but not by the mutated probe suggesting that this band is caused by c-myc– DNA complex. The faster migrating band was competed by both the wild type and mutated competitor indicating that another transcription factor(s) other than c-Myc binds to this promoter region, however, its binding site is likely to be different to that one defined for the E-box. This is very similar to the findings reported by Tzukerman et al. (2000). They identified a novel transcription factor binding to the promoter of human TERT in a site overlapping but distinct to the proximal E-box and comparison of the sequences of human and mouse Tert promoters suggested that the site is conserved. Since mutations in E-boxes also caused a reduction in promoter activity we concluded that c-myc functions as a transactivator of Tert in murine ES cells. Nozawa et al. (2001) showed that Sp1, Sp3 and c-Myc function as potent transactivators of the mouse Tert gene in C2C12 myoblasts and suggested that downregulation of DNA binding activities of Sp1, Sp3 and c-Myc during differentiation of those cells is a determining factor for the downregulation of mTert gene transcription during muscle cell differentiation. Our EMSA assays did not show any detectable change in DNA binding activities of these three factors during differentiation of ES cells and this could be due to differences in the cell system used by us (embryonic stem cells) and Nozawa’s group (C2C12 myoblasts) and the shorter differentiation interval we have used (6 days) compared to the 10 days used for myoblast differentiation. Although EMSA is sensitive in detecting DNA-protein binding even from crude extract it cannot be used to quantify small changes in the amounts of proteins bound to the DNA. In addition other posttranslational modifications such as phosphorylation and binding of additional factors to the same sites which can result in different complexes cannot be properly resolved from the original complexes using EMSA. Lastly, whilst no changes are observed in DNA binding activities of Sp1, Sp3 and c-Myc it is possible that binding by these transcription factors is very different in the context of the entire promoter rather than the core promoter analysed in this paper. Given this and our results using site directed mutagenesis we would like to suggest that Sp1, Sp3 and c-Myc act as transcriptional activators of Tert gene in ES cells, however, their involvement in Tert downregulation is presently unclear. The most interesting binding patterns were observed when oligonucleotides containing the UNK-1 and UNK-2 binding sites were used as probes in EMSA (see Fig. 2). A comparison of the sequences of the two sites indicated that a conserved binding sequence was CAGCNNCNNC, therefore we started by changing the 3rd, 4th and 7th nucleotide to A and the last nucleotide to T (see Fig. 1a).

1513

Fig. 2. Transcription factor binding to the UNK-2 site. Nuclear extracts from murine embryonic stem cells (lanes 2–4), d2 differentiating embryoid bodies (lanes 5–7), d4 differentiating embryoid bodies (lanes 8–10) and d6 differentiating embryoid bodies (lanes 11–13) were incubated with 32P labelled probe containing the UNK-2 binding site. Lane 1 is 32P labelled probe alone. The specificity of the complex was examined by adding 100fold excess of unlabelled oligonucleotides (lanes 3, 6, 9 and 12) or unlabelled mutated oligonucleotides (lanes 4, 7, 10 and 13) as cold competitors. The two arrows at the top of the figure show the DNA-protein complexes, while the single arrow on the bottom shows the free labelled probe.

Two DNA–protein complexes were observed and they were both abolished by the wild type competitor (Fig. 2; lane 3, 6, 9 and 12) but not by the mutated competitor (Fig. 2; lane 4, 7, 10 and 13) suggesting that this binding is specific. When we used nuclear extracts from differentiating EBs in EMSA we observed a slight increase in the intensity of the faster migrating band from undifferentiated cells to day 2 and 4 of differentiation and a marked increase when nuclear extracts from day 6 of differentiation was used (see Fig. 2: lane 2, 5, 8 and 11). Site directed mutagenesis of these two potential binding sites showed that the factors that bind therein might have a repression activity. The increase in the intensity of binding to DNA as levels of mTert decrease during differentiation fits well with this suggested role. In order to define exactly the binding site for this novel transcription factor(s) we synthesised six oligonucleotides covering the UNK-2 site with mutations spread throughout the sequence and used them as unlabelled competitors in EMSA (see Fig. 3a). Only the m3 oligonucleotide (data not shown) abolished binding and acted as wild type competitor suggesting that the two mutated nucleotides therein are not important for binding of proteins to the UNK-1 and UNK-2 sites. From this analysis and comparison of the sequence of UNK-1 and UNK-2 sites we concluded that the consensus binding sequence of this repressor(s) is CAGCNNCNNC. Comparison of the mouse and human Tert core promoter sequences revealed that only the UNK-2 but not the UNK-1 consensus binding site is conserved

1514

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

Fig. 3. Identification of a consensus sequence for the UNK-1 and UNK-2 binding sites. (a) Sequences of mutated oligonucleotides used as cold competitors in EMSA for mapping the UNK-2 transcription factor binding site; (b) A comparison of the mouse and human Tert core promoters. The mTert sequence is shown at the top and the hTERT on the bottom. The UNK-1 and UNK-2 are shown in bold typeface and their binding sites are underlined.

between the two (see Fig. 3b) suggesting that this site is potentially significant in the regulation of hTERT promoter activity. 2.3. Zap3 controls transcriptional downregulation of Tert during differentiation of murine ES cells In order to identify the protein that bind to UNK sites we performed yeast-one hybrid screening as outlined in Experimental procedures. We identified 3 positive clones, namely Zap3, Nhp2 like protein and transforming protein RhoC which showed good growth in both primary and secondary plates. Protein extracts from these clones were used in electrophoretic mobility shift assays in order to confirm specific binding of these proteins to the UNK sites. Only the protein extract from Zap3 showed specific binding to the UNK-2 (see Fig. 4a) and UNK-1 site (data not shown). To confirm these results in ES cells we also subcloned Zap3, Nhp2 and RhoC cDNA fragments isolated during yeast one hybrid screening from the pACT2 vector

Fig. 4. Zap3 binds to the Unk-2 site of mTert core promoter and downregulates its expression. (a) Nuclear extracts from d6 differentiating embryoid bodies (lanes 1–2), Nhp2 yeast one hybrid clone protein extract (lane 3), RhoC yeast one hybrid clone protein extract (lane 4) and Zap3 yeast one hybrid clone protein extract (lanes 5–6) were incubated with 32P labelled probe containing the UNK-2 binding site. The specificity of the complex was examined by adding 100-fold excess of unlabelled oligonucleotides (lanes 2 and 6) as cold competitor. The two arrows at the top of the figure show the DNA-protein complexes, while the single arrow on the bottom shows the free labelled probe; (b) semi-quantitative RT-PCR analysis of mTert in cells that had been mock transfected (mock), cells transfected with pDsRed2-C1 (DsRed), cells transfected with Zap3-DsRed (Zap3), cells transfected with Nhp2-DsRed (Nhp2) and cells transfected with RhoC-DsRed (RhoC), cDNA extracted from 7.5 days murine embryo (Cve); (c) semi-quantitative RTPCR analysis for Zap3 during in vitro differentiation of murine ES cells (day1–day6 of differentiation). b-actin has been used as normalising control in both (b) and (c).

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

into pDsRed2-C1 (Invitrogen) and transfected them into murine ES cells (CGR8) which were stably transfected with the mTert core promoter-GFP beforehand. We confirmed their overexpression into ES cells using semi-quantitative RT-PCR for each of the three candidates, flow cytometry for the fusion protein and confocal microscopy (data not shown). Using semi-quantitative RT-PCR we again observed reduction of mTert levels only in cells where Zap3 had been overexpressed (see Fig. 4b). We examined the expression pattern of Zap3 during the in vitro differentiation of ES cells and very interestingly observed that it was expressed only between days 4 and 6 of differentiation (see Fig. 4c) which comprises of the window where most effective differentiation and downregulation of mTert occurs (Hole et al., 1996; Armstrong et al., 2000). By increasing the number of cycles in the semi-quantitative RT-PCR we noticed that low levels of Zap3 could also be detected in the ES cells and d1–d3 of differentiation, however, this expression was much lower than in the days 4–6 of differentiation (data not shown). This result fits well with the EMSA results which show fainter DNA binding complexes in nuclear extracts prepared from ES cells and day 2 embryoid bodies (see Fig. 2) compared to extracts prepared from day 4 and day 6 embryoid bodies. Zap3 was first identified by Misawa et al. (2000) and subsequently by Sutherland et al. (2001) as a novel protein with a nuclear speckle expression which fits well with the proposed role of regulating mTert promoter activity. Isolation of numerous ESTs from Zap3 has suggested that it is expressed in skin, heart, mouse mammary gland, retina, eye ball, oviduct of 2-day pregnant female, testis and kidney and our work has also shown its expression in 7.5 dpc of mouse embryos. One human candidate (EMBL accession no: XM 085151) shows 74% identity to the mouse Zap3 and it is thought to be its human homologue. Given the presence of one UNK site in the human promoter it will be of interest to investigate whether the human ZAP3 has a nuclear localisation pattern and plays any role in the regulation of hTERT activity. One interesting finding is that one of the UNK sites where Zap3 is shown to bind (UNK-2 site) is located in the 5 0 untranslated region. New evidence from work with human TERT promoter suggests that the first two exons and intron 1 play a major role in downregulation of its promoter activity in telomerase-positive cells and are likely to be required for binding of transcription factors (Renauld et al., 2003). This may suggest that the Tert gene in human and mouse may employ their coding sequence and untranslated regions to recruit transcription factors as well as their promoter regions. 2.4. Overexpression of Zap3 in murine ES cells leads to reduction in mTert expression, telomerase activity and results in telomere shortening To investigate the function of Zap3 in murine ES cells we stably transfected the Zap3 cDNA construct (see Section 2.3)

1515

and selected two clones Zap3/3 and Zap3/5 for further analysis. The real time RT-PCR assays showed that Tert expression was reduced more than 50% in these two clones compared to the mock transfected cells (see Fig. 5a). This considerable downregulation of mTert expression was accompanied by a reduction in the activity of the telomerase holoenzyme complex as measured by the telomerase repeat amplification protocol (TRAP) (see Fig. 5b). The reduction in telomerase activity also had an effect on telomere shortening. Using fluorescence in situ hybridization combined with flow cytometry (Flow-FISH) we observed progressive reduction of the total telomere fluorescence signal during time in culture for the Zap3 transfected ES cells (see Fig. 5c). In contrast, mock transfected cells showed a slight increase in total telomere fluorescence signal with time in culture (see Fig. 5c). Liu et al. (2000) also reported ES cells heterozygous for the mTert disruption showed telomere shortening and this suggests that Tert is limiting and necessary for telomerase function. Telomere shortening was also reported in murine haematopoietic stem cells after two rounds of serial transplantation which indicates that proliferating cells such as those present in tissues with self-renewal capacity and ES cells do shorten their telomeres after each cell division and need normal telomerase activity to maintain telomere length (Allsopp et al., 2001). We did notice variation in total telomere fluorescence between Zap3 transfected clones and also the control ES cells and this could be due to slightly different population doublings that the cells have to go through before clonal derivation especially if the gene of interest has an effect on cell growth. Similar variations in telomere length of different clones have been reported by Amit et al. (2000) for human ES cells. We also performed karyotype analysis at three different cell passages corresponding to PD130, PD190 and PD250. Statistical analysis also revealed that the Zap3 transfected clones showed a significant increase in aneuploidy compared to the control in the early passage, PD130 (c2test P!0.05: see Fig. 5d). The significance of aneuploidy in the later passages is hard to interpret since both control and Zap3 transfected clones showed significant increase in aneuploidy with increasing passage number (c2-test P!0.05: see Fig. 5d). To examine chromosome stability in early and late passage ES cells we carried out FISH analysis on metaphase spreads (see Fig. 5e). We noticed a significant increase in the number of signal free ends in Zap3 tranfected clones in comparison to the control clone at both early passage and late passage (c2-test P!0.01). In addition a significant increase in the number of signal free ends was observed with time in culture for the Zap3 transfected clones (c2-test P! 0.01), but not for the control ES cells. We also noticed some chromosome fusion occurring in Zap3 transfected clones but not the control ES cells (see Fig. 5f). We investigated the ability of these cells to grow in culture and ability to resist induced apoptosis. We observed

1516

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

Fig. 5. Zap3 overexpression leads to downregulation of mTert expression, reduction in telomerase activity, telomere length and chromosome abnormalities. (a) Zap3 overexpression causes downregulation of mTert expression. Real time RT-PCR results are presented as Tert/Gapdh ratios. This ratio is normalised to 1 for the vector and the clones are calculated with respect to the vector. Standard deviations were determined by bootstraping (1000 replicates). (b) Zap3 overexpression causes reduction in telomerase activity. Telomerase activity was measured using the TRAP-ELISA assay as described in Experimental procedures. Three different protein concentrations were tested from each cell extract. Relative telomerase activity (RTA) was calculated with respect to internal standard (IS) as outlined in Experimental procedures. The results are represented as mean of three independent measurements. (c) Zap3 overexpression causes shortening of telomeres. Telomere fluorescence signal was measured in arbitrary units using Flow-Fish at PD130, PD190 and PD250. The results are represented as meanGSEM of three independent measurements. (d) Increased aneuploidy in Zap3 transfected ES cells. (e) Zap3 overexpression results in increased numbers of signal free ends (SFE) in late passages. SFE ends were estimated carrying out FISH on metaphase spreads of ES cells from early (PD130) and late passages (PD 250). (f). Chromosomal abnormalities in Zap3 transfected clones. Fused chromosome is indicated by a black arrow.

that the overexpression of Zap3 and subsequent reduction in telomerase activity did indeed inhibit cell growth to some degree (Fig. 6a), however, the degree of apoptosis was substantially no different to that of cells containing only vector (see Fig. 6b). The cells analysed in this study

have only undergone approximately 250 population doublings and it would probably take a lot more cell divisions to detect any apoptotic cell death caused by extensive telomere shortening given the long telomeres in inbred strains from which the ES cells have been derived.

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

1517

Fig. 7. Overexpression of Zap3 reduces the haematopoietic commitment of ES cells. Embryoid bodies (EBs) were allowed to differentiate in the absence of LIF for 6 days and hematopoietic commitment assessed by counting the proportion of embryoid bodies producing mixed colonies of neutrophils, erythrocytes, macrophages and megakaryocytes in the CFU-GEMM assay as described in Experimental procedures. Results are expressed as mean of three independent experiments, each of three replicates.

Fig. 6. Overexpression of Zap3 slows down cell growth and causes changes in cell cycle. (a) Overexpression of Zap3 slows down cell growth. 50,000 live cells were plated in triplicates and recounted after 2 and 5 days using Trypan blue exclusion method. (b) Zap3 transfected cells do not show increased apoptosis in culture. The percentage of apoptotic cells was measured in the absence of protein kinase C inhibitor bis-indolylmaleimide I (0 mM) and presence of inhibitor (2 and 5 mM). Student t-test at 1% threshold level revealed no significant statistical differences between Zap3 and vector transfected cells both in presence or absence of protein kinase C inhibitor. (c) Overexpression of Zap3 causes changes in cell cycle. Cells were synchronised by serum starvation and stained with propidium iodide and outlined in Experimental procedures. There were significant differences between percentages of Zap and vector transfected cells in G0/G1 (Students T-test: P%0.01 for Zap3/3 cells/vector cells and P%0.01 for Zap3/5 cells/vector cells) and S phase (P%0.01 for Zap3/3 cells/vector cell and P%0.01 for Zap3/5 cells/vector cells).

By analysing the cell cycle status of these cells we noticed that a higher percentage of Zap3 transfected cells tend to be in G0/G1 rather than S phase although G2/M is largely unchanged (see Fig. 6c). These results are similar to

published literature which shows that disruption of telomerase activity in murine ES cells, immortal murine fibroblasts and normal human fibroblasts results in decreased cell proliferation, altered cell cycle and disruption of the 3 0 overhangs (Boklan et al., 2002; Niida et al., 2000; Masutomi et al., 2003). Finally, we investigated the ability of these cells to contribute to the development of haematopoietic activity during embryoid body differentiation using the CFU-GEMM assay and observed that the percent haematopoietic commitment at days 4–6 of EB development is reduced when compared to that of control cells carrying only the vector (Fig. 7). Since each positive ‘burst’ results from the proliferation of a single haematopoietic stem cell (HSC) or myeloid progenitor cell induced by the cytokines in the assay mixture and there are a number of possible explanation for this effect: (i) the ES cells are less capable of differentiating into HSC; (ii) the HSC do not proliferate sufficiently to give observable bursts. To distinguish between these two possibilities we measured the levels of Flk-1 expression, a marker of early endothelial and haematopoietic stem cells derived from differentiating ES cells and did not observe any significant difference in the numbers of Flk-1 positive haematopoietic precursor cells occurring after day 3 of embryoid body differentiation of the Zap3 or vector transfected clones (approx. 23% Flk-1 positive; data not shown). This would seem to support the view that overexpression of Zap3 protein and the concomitant decrease in telomerase activity does not affect the ability of ES cells to differentiate, at least down an early haematopoietic pathway, and that the observed effects are more likely to be due to their limited proliferation capacity. Similar results have been reported by studies of telomerase deficient haematopoietic stem cells which have shown reduced colony forming abilities and long-term repopulating capacity (Samper et al., 2002; Lee et al., 1998).

1518

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

In conclusion we would like to emphasise that the overexpression of Zap3 is sufficient to slow down cells growth, reduce telomere activity and telomere length, cause changes in cell cycle and ability of these cells to differentiate along haematopoietic lineages in vitro. These results are consistent with a role for nucleoprotein Zap3 in downregulating telomerase activity during differentiation of ES cells. However, to ensure that these effects are not caused indirectly through some other effects of Zap3 overexpression, this work should be repeated in telomerase deficient cell lines.

the promoter construct and the mixture was allowed to stand at ambient temperature (15 min) and then added to the cultured cells. The above procedure was repeated for each promoter construct in triplicate. The cells were trypsinised, washed with phosphate buffered saline and analysed for GFP mean fluorescence intensity using a Cytometer (Coulter) 24 h after transfection. The GFP mean fluorescence intensity was expressed as relative intensity with respect to that of the promoterless construct, pEGFP1. Stable transfections were performed in the same way as the transient transfections (see above). The transfected colonies were selected by G418 (400 mg/ml) for 10 days. More than 50 colonies were pooled from each plate.

3. Experimental procedures 3.1. Materials Cell culture reagents and fetal bovine serum were purchased from Gibco, BRL. 3.2. Cell culture The murine embryonic stem cell (CGR8) were routinely passaged and maintained in a suitable medium (GMEM) with 0.25% NaHCO3, 100 mM non-essential amino acids, 2 mM L-glutamine and 0.1 mM pyruvate, 0.1 M 2-mercaptoethanol and 10% fetal calf serum) with leukaemia inhibitory factor (10 ng/ml). In order to form embryoid bodies, ES cells were cultured in hanging drops (10 ml) at a concentration of 3!104 cells/ml in the presence of LIF for 48 h in a humidified 5% CO2 atmosphere. ES cell aggregates were harvested into a Petri dish (103 aggregates/10 ml ES cell culture medium lacking LIF) and allowed to differentiate for up to 6 days. Medium was replaced every 2 days. Haematopoietic colony forming unit (CFU-GEMM) assays were performed every 2 days to monitor differentiation of ES cells. 3.3. Reporter constructs Various lengths DNA fragments from the promoter and 5 0 untranslated region of the mTert gene (EMBL accession number AF 121949) were amplified by PCR from the mouse genomic DNA and inserted into the multiple cloning site of the promoterless GFP vector, pEGFP1 (Clontech). Site directed mutagenesis of the core promoter construct was carried out using a PCR based approach (Stratagene). 3.4. Promoter assays Transient transfections of GFP promoter constructs were performed using Fugene reagent (Boehringer Mannheim) following manufacturer’s instructions. Murine ES cells (CGR8) were plated at a density of 200,000 cells/well in six well plates 1 day before the transfection. Fifteen microliters of Fugene reagent were diluted in 485 ml of serum free CGR8 medium. This mixture was added to 5 mg of

3.5. Nuclear extract preparation Extracts of nuclear proteins from CGR8 cells and differentiating embryoid bodies at days 2, 4 and 6 were prepared as previously described (Schreiber et al., 1989). The protein concentration was measured using the Bradford assay and all extracts were stored at K80 8C. 3.6. Electrophoretic mobility shift assays (EMSA) EMSA were performed using the BandShift Kit (Amersham Pharmacia Biotech). Synthetic complementary oligonucleotides were labelled with [g-32P] dATP using T4 polynucleotide kinase. DNA bind reactions were performed according to manufacturer’s instructions. In brief, 5 mg of nuclear extract from either undifferentiated cells (CGR8) or embryoid bodies at days 2, 4 or 6 of differentiation was incubated with 1 mg of poly (dI-dC) in a 20 ml reaction volume containing 20 mM Tris–HCl (pH 7.5), 100 mM sodium chloride, 1 mM dithithreitol, 10% glycerol, 0.05% NP-40 and 5 mM magnesium chloride. 1!105 cpm of the 32 P-labelled probe was added and the reaction was incubated on ice for 20 min. Molar excess (100-fold) of unlabelled oligonucleotides or oligonucleotides containing the desired mutations were used as competitors. When antibodies were employed for supershift assays, the nuclear extract was preincubated with 0.2–0.4 mg of antibody for 20 min at 4 8C. The reactions were separated on 10% polyacrylamide gels. The following pairs of oligonucleotides were used: GC-box1: 5 0 -GTGCGCCCCCTTTCGTTACTCC-3 0 , 5 0 -GGAGTAACGAAAGGGGGCGCAC-3 0 , UNK-1: 5 0 -AACACATCCAGCAACCACT-3 0 , 5 0 -AGT GGTTGCTGGATGTGTT-3 0 , E-box1: 5 0 -CGGGGAACACACCTGGTCCTCA-3 0 , 5 0 -TGAGGACCAGGTGTGTTCCCCG-3 0 , GC-box2: 5 0 -GCTGCGACCCCGCCCCTTCC-3 0 , 5 0 -GGAAGGGGCGGGGTCGCAGC-3 0 , GC-box3: 5 0 -TGAATCCCGCCCCTTCCTCC-3 0 , 5 0 -GGAGGAAGGGGCGGGATTCA-3 0 , UNK-2: 5 0 -GTTCCCAGCCTCATCTTTTT-3 0 , 5 0 -AAAAAGATGAGGCTGGGAAC-3 0 , E-box2: 5 0 -GTTCCCGCACGTGGGAGGCC-3 0 , 5 0 -GGCCTCCCACGTGCGGGAAC-3 0 (see Fig. 1a).

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

3.7. Yeast one hybrid screening Yeast one hybrid screening was carried out using the One-Hybrid system from Clontech (Palo Alto, California). In brief, six copies of the UNK-2 site (see Fig. 1a) were inserted into the multiple cloning site of pHiSi-1 using the EcoRI and XbaI restriction enzymes. The pHiSi-1 containing sic copies of UNK-2 site was digested with XhoI and used to transform the yeast strain YM4271. The yeast colonies with integrated reporter construct were tested for background expression by plating on SD/KHis plates supplemented with different concentrations of 3-AT (3amino-1,2,4-triazole). The optimal concentration of 3-AT for inhibiting background expression was 7.5 mM. A mouse 7 dpc embryo Matchmaker cDNA library was purchased from Clontech and amplified using manufacturer’s instructions. DNA was extracted from the library using the HiSpeed Maxi Plasmid Purification Kit. A large-scale transformation of the DNA from the above library into the yeast reporter strain was carried out using the manufacturer’s instructions and the mixture was plated out on 30 SD/KHis/KLeu/C7.5 mM 3-AT square plates (150 mm). The plates were incubated at 30 8C for 6–8 days. The largest colonies were restreaked on secondary plates (SD/KHis/K Leu/C7.5 mM 3-AT). Plasmid DNA was isolated from positive yeast clones and used to transform E. coli. The sequences of the clones were compared to EMBL and Prosite databases. For electrophoretic mobility shift assays the yeast clones were grown into liquid culture (SD/K His/KLeu) and protein extract was obtained using manufacturer’s instructions. 3.8. Overexpression of Zap3, Nhp2 and RhoC into ES cells using the pDsRed2-C1 vector DNA fragments from the mouse Zap3(EMBL accession number AB0033168), Nhp2 (EMBL accession number BQ126798) and RhoC (EMBL accession number MMRHOCR) were amplified from day 4 embryoid bodies using the following primers: Zap3EXPF1: 5 0 AGATCTGACAAGTTGGATGGCTTGAG, Zap3EXPR1: 5 0 -GAATTCATATATTTGGTCCGATTGA, Nhp2EXPF1: 5 0 -AGATCTACAGAGGCTGATGTGAA, Nhp2EXPR1: 5 0 -GAATTCACCAAGAGCCTTTCAATAGA, RhoCEXPF1: 5 0 -AGATCTATGGCTGCGATCCGAAAGAA, RhoCEXPR1: 5 0 -GAATTCAGAATGGGACAGCCCCTCC. The fragments were inserted into the multiple cloning site of the vector containing the red fluorescent protein (pDsRed2-C1; Clontech) using the BglII and EcoRI enzymes. These cDNAs will be expressed as fusions with the C-terminus of DsRed2 proteins since they have been inserted in the same reading frame of pDsRed2 and there are not intervening stop codons. Transfections into ES cells carrying the mTert core promoter-GFP was carried out using the Fugene Transfection kit (Boehringer Mannheim) according to manufacturer’s instructions. The expression of fusion

1519

proteins was monitored by flow cytometry and localisation by fluorescence microscopy. Overexpression of Zap3, Nhp2 and RhoC was confirmed by semi-quantitative RT-PCR. 3.9. RT-PCR analysis About 500 EBs were selected each day and total RNA was extracted using RNAzol (Biotech) and reverse transcribed using AMV reverse transcriptase (Promega) and random oligo hexamers following the manufacturer’s instructions. The amount of cDNA into each sample was normalised using actin as a control (actin 5 0 -AGGGGCCGGACTCATCGTACTC-3 0 and actin 3 0 -GTGACGAGGCCCAGAGCAAGAG-3 0 ). This information was used to equilibrate the amount of cDNA in each sample amplified in subsequent reactions with primers for Zap3(Zap3F: 5 0 -GTGGCAAGAGAAATATTCAT-3 0 and Zap33R: 5 0 CCTCCGCTAACGGTCGTCTT-3 0 ) and mTert(mTert3: 5 0 -ACCCAGGATGTACTTTGTTAAGG-3 0 and mTert4: 5 0 -AGCAAACAGCTTGTTCTCCA-TGTC-3 0 ). PCR products were run on 2% agarose gels and stained with ethidium bromide. Results were assessed on the presence or absence of the appropriate size PCR products. RNA controls were included to monitor genomic contamination. 3.10. Telomeric repeat amplification protocol (TRAP) TRAP reactions were carried out using TeloTAGGG Telomerase PCR ELISA plus (Roche) following the manufacturer’s instructions. Embryonic stem cells and embryoid bodies were washed in PBS, and directly transferred into eppendorf tubes containing 100 (l CHAPS lysis buffer. Several concentrations of protein extracts from each sample (0.1, 1 and 10 ng) were used in the PCR reaction to ensure that the assay is in the linear range given that ES cells show a relatively high level of telomerase activity (for each sample 3 PCR reactions were performed). Telomerase specific products were determined by the presence of a 6 bp incremental DNA ladder on a 12% polyacrylamide gel stained with ethidium bromide TRAP products were quantified by a colorimetric reaction at 450 nm (using manufacturer’s instruction) and a Titertek Multiskan plate reader. In all assays, the CHAPS lysis buffer was used as a negative control for both PCR and telomerase reactions. Heat denatured samples were used as negative controls. Relative telomerase activity (RTA) was calculated using the following formula: RTA Z fðAS K ASO Þ=AS;IS =ATS8 K ATS8;0 =ATS8;IS Þg !100 where AS ASO AS,IS ATS8 ATS8,0

absorbance absorbance absorbance absorbance absorbance

of sample of heat treated control of internal standard of control template of lysis buffer

1520

ATS8

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

IS absorbance of internal standard (IS) of the control template (TS8)

3.11. CFU-GEMM assay CFU-GEMM assays were performed according to manufacturer’s instructions (Stem Cell Technologies; Vancouver, Canada). In brief approximately 100 EBs in a total volume of 0.3 ml were added to 3 ml of Methocult medium yielding triplicate cultures of 1.1 ml each. Methocult containing EBs was dispensed into three 35 mm dishes using a 16 g blunt-end needle. Two plates were then placed into a 100 mm covered dish containing a third uncovered dish containing 3 ml of sterile water. Cultures were then placed in a 37 8C incubator maintained with 5% CO2 and O95% humidity. The concentration of cytokines in Methocult is as follows: IL-3 (10 ng/ml); IL-6 (10 ng/ml), SCF (50 ng/ml), Epo (3 units/ml). After 10 days in culture, the plates were assayed by counting the number of EBs surrounded by a corona or ‘burst’ of differentiated cells. Such ‘bursts’ represent haematopoietic colony forming units and the percentage of these increases markedly on day 4 of the differentiation protocol described above which is indicative of the differentiation of the ES cells. 3.12. Cell cycle analysis Cell cycle analysis was performed using the CycleTest Plus DNA reagent kit (Becton Dickinson, San Jose, CA) in accordance with manufacturers’ instructions. Cells were first treated with serum free medium for 12 h and left to recover for 48 h in serum supplemented medium. Cells were harvested by trypsinisation and counted (haemocytometer). 500,000 cells were centrifuged (400 g, 5 min) then after complete removal of supernatant, 250 ml of solution A (trypsin buffer) was added with gentle mixing. The suspension was incubated at room temperature (10 min) then 200 ml of solution B (trypsin inhibitor/ RNAse buffer) was added with gentle mixing. After a further 10 min at room temperature, 200 ml of cold (2–8 8C) solution C (propidium iodide stain) was added and the sample was analysed by flow cytometry (Becton Dickinson FACSvantage) measuring FL2 area vs total counts. The data was analysed using ModFit 3 (Becton Dickinson) to generate percentages of cells in G0/G1, S and G2/M phases. 3.13. Apoptosis assay Many cell types, including embryonic stem cells undergo apoptosis in response to inhibition of protein kinase C by bis-indolylmaleimide II (Quinland et al., 2003). Cells undergoing apoptosis can be enumerated using the Annexin V-FITC apoptosis detection kit (BD

biosciences-Pharmingen) in which FITC conjugated Annexin V binds to the externalised phosphatidyl serine characteristic of apoptotic cells. The protocol was carried out in accordance with manufacturers’ instructions and briefly comprises the following. Cells were harvested by trypsinisation, washed twice with ice cold phosphate buffered saline and counted. 1!105 cells are suspended in 100 ml of 1!binding buffer (supplied) then 5 ml of annexinV-FITC and 5 ml of propidium iodide solution are added. The mixture is vortexed gently and incubated 15 min at room temperature in the dark. 400 ml 1!binding buffer is added and the cells are analysed by flow cytometry (FACS Caliber, Becton Dickinson). 3.14. Cell proliferation assay Cells were harvested by trypsinisation and live cells were counted using Trypan Blue (Sigma) exclusion. 50,000 live cells were added to each well of a six-well gelatine coated plate (Iwaki) containing 3 ml of murine CGR8 medium (detailed above) and leukaemia inhibitory factor (10 ng/ml). Cells were permitted to grow for 2 and 5 days then were harvested by trypsinisation, washed with phosphate buffered saline and cell numbers were estimated using Trypan Blue exclusion method. 3.15. Mean telomere fluorescence by flow-fish The mean length of telomeric repeats was measured by flow cytometry following the hybridisation of a fluorescently conjugated peptide nucleic acid (PNA) probe (Rufer et al., 1998). Briefly, cells were harvested by trypsinisation, washed with phosphate buffered saline with 0.1% Bovine serum albumin (BSA) then counted. 3!105 cells (in triplicate) were resuspended in a hybridisation mix comprising formamide (70% aqueous solution; deionised), 20 mM Tris, pH 7.0, 1% (W/V) BSA and FITC conjugated PNA telomeric probe (Perseptive biosystems, Framingham MA) to a final concentration of 0.3 mg/ml. The mixture was denatured (80 8C/10 min) then allowed to hybridise at room temperature (2 h, dark) then the cells were collected by centrifugation and washed twice with wash buffer comprising formamide (70% aqueous solution, deionised), 10 mM Tris, pH 7.0, 0.1% (W/V) BSA and 0.1%(V/V) Tween 20 collecting the cells by centrifugation (3000 rpm/5 min). A further wash was performed using phosphate buffered saline with 0.1% (W/V) BSA, 0.1% Tween 20 with centrifugation (2000 rpm/5 min). The cells were resuspended in 300 ml phosphate buffered saline with 0.1% (w/v) BSA plus RNAse at a final concentration of 10 mg/ml. Propidium iodide was added at a final concentration of 0.06 mg/ml, the sample was incubated 2–4 h at room temperature then analysed by flow cytometry measuring FL1 fluorescence intensity vs total counts.

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

1521

3.16. Real time RT-PCR

Appendix. Supplementary material

Quantitative real time RT-PCR was carried out using the Quantitect SYBR Green RT-PCR kit (Qiagen) according to manufacturer’s instructions. Ct values were measured for five different sample concentrations (1, 10, 20, 50, 100 ng) prepared in triplicates and compared to corresponding Ct values for Gapdh which was used as a housekeeping gene. Regression analysis was used to estimate relative amount of gene per sample.

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.mod.2004. 07.005.

3.17. Pulsed field gel electrophoresis. For preparation of highest quality genomic DNA, cells were embedded in 0.65% low-melting agarose plugs at a density of 107 cells/ml before deproteination by proteinase K treatment (Petersen et al., 1998; Sitte et al., 1998). The DNA was digested until completion with Hinf I (60U per plug; Boehringer Mannheim) at 37 8C. Plugs were analysed in a 1.0% agarose gel by pulsed field gel electrophoresis (CHEF-DRIII-SYSTEM, Biorad) at 3.5 V/cm for 17 h with a switching time of 2–10 s in 0.5!TBE (4.5 mM Tris, 4.5 mM boric acid, 0.1 mM EDTA, pH 8.0). 3.18. Karyotype analysis of ES cells Chromosome preparations were made using stardard cytogenetics techniques and a 16 h colcemid/BrdU mitotic arrest step. The karyotype of ES cells was determined by standard G-banding procedure. 3.19. Fish analysis on metaphase spreads The analysis was carried out essentially as described by Surralles et al. (1999). A FITC labelled (CCCTAA)3 PNA was used as a probe. 3.20. Statistical analysis Unless indicated otherwise, statistical analysis was carried out using c2-test without Yates’ correction and results expressed as one-tailed P-value.

Acknowledgements We would like to thank Ms Trudy Horton for helping us with flow cytometry analysis and Jaclyn C. Barel and Karen A. Thompson for help with cytogenetics analysis. This work was supported by the Wellcome Trust, the Leukaemia Research Fund and Life Knowledge Park.

References Allsopp, R.C., Vaziri, H., Patterson, C., Goldstein, S., Younglai, E.V., Futcher, A.B., et al., 1992. Telomere length predicts replicative capacity of human fibroblasts. Proc. Natl Acad. Sci. USA 89, 10114–10118. Allsopp, R.C., Cheshier, S., Weissman, I.L., 2001. Telomere shortening accompanies increased cell cycle activity during serial transplantation of haematopoietic stem cells. J. Exp. Med. 193, 917–924. Amit, M., Carpenter, M.K., Inokuma, M.S., Chiu, Y.P., Harris, C.P., Waknitz, M.A., et al., 2000. Clonally derived human embryonic stem cell lines maintain pluripotenct and proliferative potential for prolonged periods of culture. Dev. Biol. 227, 271–278. Armstrong, L., Lako, M., Lincoln, J., Cairns, P.M., Hole, N., 2000. mTert expression correlates with telomerase activity during the differentiation of murine embryonic stem cells. Mech. Dev. 97, 107–116. Bechter, O.E., Eisterer, W., Dlaska, M., Kuhr, T., Thaler, J., 2002. CpG island methylation of the hTERT promoter is associated with lower telomerase activity in B-cell lymphocytic leukemia. Exp. Hematol. 30, 26–33. Blackburn, E.H., 1991. Structure and function of telomeres. Nature 350, 569–573. Blasco, M.A., Funk, W.D., Villeponteau, B., Greider, C.W., 1995. Functional characterization and developmental regulation of mouse telomerase RNA. Science 269, 1267–1270. Bodnar, A.G., Ouellette, M., Frolkis, M., Holt, S.E., Chiu, C.-P., Morin, G.B., et al., 1998. Extension of life-span by introduction of telomerase into normal human cells. Science 279, 349–352. Boklan, J., Nanjangud, G., MacKenzie, K.L., May, C., Sadelain, M., Moore, M.A.S., 2002. Limited proliferation and telomere dysfunction following telomerase inhibition in immortal human fibroblasts. Cancer Res. 62, 2104–2114. Broccoli, D., Young, J.W., DeLange, T., 1995. Telomerase activity in normal and malignant hematopoietic cells. Proc. Natl Acad. Sci. USA 9, 9082–9086. Carninci, P., Shibata, Y., Hayatsu, N., Sugahara, Y., Shibata, K., Itoh, M., et al., 2000. Normalization and subtraction of cap-trapper-selected cDNAs to prepare full-length cDNA libraries for rapid discovery of new genes. Genome Res. 10, 1617–1630. Chiu, C.P., Dragowska, W., Kim, N.W., Vaziri, H., Yui, J., Thomas, T.E., et al., 1996. Differential expression of telomerase activity in hematopoietic progenitors from adult human bone marrow. Stem Cells 14, 239–248. Greenberg, R.A.A., Allsopp, R.C., Chin, L., Morin, G.B., DePinho, R.A., 1998. Expression of mouse telomerase reverse transcriptase during development, differentiation and proliferation. Oncogene 16, 1723–1730. Greider, C., 1994. Mammalian telomere dynamics: healing, fragmentation shortening and stabilization. Curr. Opin. Genet. Dev. 4, 203–211. Harle-Bachor, C., Boukamp, P., 1996. Telomerase activity in the regenerative basal layer of the epidermis inhuman skin and in immortal and carcinoma-derived skin keratinocytes. Proc. Natl Acad. Sci. USA 93, 6476–6481. Harley, C.B., 1991. Telomere loss: mitotic clock or genetic time bomb?. Mutat. Res. 256, 271–282.

1522

L. Armstrong et al. / Mechanisms of Development 121 (2004) 1509–1522

Harrington, L., McPhail, T., Mar, V., Zhou, W., Oulton, R., Bass, M.B., et al., 1997. A mammalian telomerase-associated protein. Science 275, 973–977. Hiyama, K., Hiray, Y., Kyoizumi, S., Akiyama, M., Hiayma, E., Piatyszek, M.A., et al., 1995. Activation of telomerase in human lymphocytes and hematopoietic progenitor cells. J. Immunol. 155, 3711–3715. Hiyama, E., Hiayma, K., Tatsumoto, N., Shay, J.W., Yokoyama, T., 1996. Int. J. Oncol. 9, 453–458. Hole, N., Graham, G.J., Menzel, U., Ansell, J.D., 1996. A limited temporal window for the derivation of multilineage repopulating hematopoietic progenitors during embryonal stem cell differentiation in vitro. Blood 88, 1266–1276. Hou, M., Wang, X.B., Popov, N., Zahng, A., Zhao, X., Zhou, R., et al., 2002. The histone deacetylase inhibitor trichostatin A derepresses the telomerase reverse transcriptase (hTERT) gene in human cells. Exp. Cell Res. 274, 25–34. Kyo, S., Takakura, M., Taira, T., Kanaya, T., Itoh, H., Yutsudo, M., et al., 2000. Sp1 cooperates with c-Myc to activate transcription of the human telomerase reverse transcriptase gene (hTERT). Nucleic Acids Res. 28, 669–677. Lee, Y., Blasco, M.A., Gotlieb, G.J., Horner, J.W., Greider, C., dePinho, A.A., 1998. Essential role of mouse telomerase in highly proliferative organs. Nature 392, 569–574. Liu, Y., Snow, B.E., Hande, M.P., Yeung, D., Erdman, N.J., Wakeham, A., et al., 2000. The telomerase reverse transcriptase is limiting and necessary for telomerase function in vivo. Curr. Biol. 10, 1459–1462. Martin-Riviera, L., Herrera, L., Albar, J.P., Blasco, M.A., 1998. Expression of mouse telomerase catalytic subunit in embryos and adult tissues. Proc. Natl Acad. Sci. USA 95, 10471–10476. Masutomi, K., Yu, E.Y., Khurts, S., Ben-Porath, I., Currier, J.L., Metz, G.B., et al., 2003. Telomerase maintains telomere structure in normal human cells. Cell 114, 241–253. Meyerson, M., Counter, C.M., Eaton, E.N., Ellisen, L.W., Steiner, P., Caddle, S.D., et al., 1997. hEST2, the putative human telomerase catalytic subunit gene, is up-regulated in tumor cells and during immortalization. Cell 90, 785–795. Misawa, K., Nosaka, T., Morita, S., Kaneko, A., Nakahata, T., Asano, S., Kitamura, T., 2000. A method to identify cDNAs based on localization of green fluorescent protein fusion products. Proc. Natl Acad. Sci. USA 97, 3062–3066. Morin, G.B., 1989. The human telomere terminal transferase enzyme is a ribonucleoprotein that synthesizes TTAGGG repeats. Cell 59, 521–529. Moyzis, R.K., Buckingham, J.M., Crams, L.S., Dani, M., Deaven, L.L., Jones, M.D., et al., 1988. A highly conserved repetitive DNA sequence, (TTAGGG)n, present at the telomeres of human chromosomes. Proc. Natl Acad. Sci. USA 85, 6622–6626. Niida, H., Shinkai, Y., Hande, M.P., Matsumoto, T., Takehara, S., Tachibana, M., et al., 2000. Telomere maintenance in telomerasedeficient mouse embryonic stem cells: characterization of an amplified telomeric DNA. Mol. Cell. Biol. 20, 4115–4127. Nomoto, K., Maekawa, M., Sugano, K., Ushiama, M., Fukayama, N., Fujita, S., Kakizoe, T., 2002. Methylation status and expression of human telomerase reverse transcriptase mRNA in relation to hypermethylation of the p16 gene in colorectal cancers as analyzed by bisulfite PCR-SSCP. Jpn. J. Clin. Oncol. 32, 3–8. Nozawa, K., Maehara, K., Isobe, K.-I., 2001. Mechanism for the reduction of telomerase expression during muscle cell differentiation. J. Biol. Chem. 276, 22016–22023.

Oh, S., Song, Y.-H., Kim, U.-J., Yim, J., Kim, T.K., 1999a. In vivo and in vitro analyses of Myc for differential promoter activities of the human telomerase (hTERT) gene in normal and tumor cells. Biochem. Biophys. Res. Commun. 263, 361–365. Oh, S., Song, Y., Yim, J., Kim, T.K., 1999b. The Wilms’ tumor 1 tumor suppressor gene represses transcription of the human telomerase reverse transcriptase gene. J. Biol. Chem. 274, 37473–37478. Oh, S., Song, Y.H., Yim, J., Kim, T.K., 2000. Identification of Mad as a repressor of the human telomerase (hTERT) gene. Oncogene 19, 1485–1490. Petersen, S., Saretzki, G., von Zglinitcki, T., 1998. Preferential accumulation of single-stranded regions in telomeres of human fibroblasts. Exp. Cell Res. 25, 152–160. Quinland, L.R., Faherty, S., Kane, M.T., 2003. Phospholipase C and protein kinase C involvement in mouse embryonic stem—cell proliferation and apoptosis. Reproduction 126, 121–131. Renauld, S., Bosman, F.T., Benhattar, J., 2003. Implication of the exon region in the regulation of the human telomerase reverse transcriptase gene promoter. Biochem. Biophys. Res. Commun. 300, 47–54. Rufer, N., Dragowska, W., Thornbury, G., Roosnek, E., Lansdorp, P.M., 1998. Telomere length dynamics in human lymphocyte subpopulations measured by flow cytometry. Nat. Biotechnol. 16, 743–747. Samper, E., Frenandez, P., Eguia, R., Martin-Riviera, L., Bernard, A., Blasco, M.A., Aracil, M., 2002. Long-term repopulating ability of telomerase-deficient murine haematopoietic stem cells. Blood 99, 2767–2775. Schreiber, E., Matthias, P., Muller, M.M., Shaffner, W., 1989. Rapid detection of octamer binding proteins with ‘mini-extracts’, prepared from a small number of cells. Nucleic Acids Res. 17, 6419. Shay, J.W., 1997. Telomerase in human development and cancer. J. Cell. Phys. 173, 266–270. Sitte, N., Saretzki, G., von Zglinicki, T., 1998. Accelerated telomere shortening in fibroblasts after extended periods of confluency. Free Radic. Biol. Med. 24, 885–893. Stewart, S.A., Ben-Porath, I., Carey, V.C., O’Connor, B.F., Hahn, W.C., Weinberg, R.A., 2003. Erosion of the telomeric single-strand overhang at replicative senescence. Nat. Genet. 33, 492–496. Surralles, J., Hande, M.P., Marcos, R., Landsdorp, P.M., 1999. Accelerated telomere shortening in the human inactive X chromosome. Am. J. Hum. Genet. 65, 1617–1622. Sutherland, H.G.E., Mumford, G.K., Newton, K., Ford, L.V., Farrall, R., Dellaire, G., et al., 2001. Large-scale identification of mammalian proteins localized to nuclear sub-compartments. Hum. Mol. Genet. 10, 1995–2011. Takakura, M., Kyo, S., Sowa, Y., Wang, Z., Yatabe, N., Maida, Y., et al., 2001. Telomerase activation by histone deacetylase inhibitor in normal cells. Nucleic Acids Res. 29, 3006–3011. Tzukerman, M., Shachaf, C., Ravel, Y., Braunstein, I., Cohen-Barak, O., Yalon-Hacohen, M., Skorecki, K.L., 2000. Identification of a novel transcription factor binding element involved in the regulation by differentiation of the human telomerase (hTERT) promoter. Mol. Biol. Cell 11, 4381–4391. Watson, J.D., 1972. Origin of concatemeric T7 DNA. Nat. New Biol. 239, 197–201. Wen, J., Yu-Sheng, C., Bacchetti, S., 1998. The human telomerase catalytic subunit hTERT: organization of the gene and characterization of the promoter. Hum. Mol. Genet. 7, 1137–1141. Wright, W.E., Piatyszek, M.A., Rainey, W.E., Byrd, W., Shay, J.W., 1996. Telomerase activity in human germline and embryonic tissues and cells. Dev. Genet. 18, 173–179.