**3. Results**

#### **3.1 Isolation of 5'-flanking regions of human telomere-associated protein-encoding genes**

Previously, we isolated and characterized 5'-flanking regions of the human *TERT* and *TERC* genes (Zhou et al., 2009; Uchiumi et al., 2010a). In this study, those of different human telomere-associated protein-encoding genes were obtained by PCR and inserted into the MCS of the pGL4-basic (pGL4[luc 2.10]) vector. Putative transcription-factor binding elements were found by TF-SEARCH analysis. As summarized in Fig. 1, c-Ets/Elk1, Sp1/GC-box, CREB, OCT, p300, SRY, GATA, E2F, NF-κB/c-Rel, CCAAT-box, and other motifs are located within 300-bp from the 5'-upstream region of the cDNAs. Although all of these telomere-associated protein factors are commonly involved in the maintenance of telomeres, a rigid rule in the order of the *cis*-elements could not be found in their core promoter regions. However, one or more Sp1/GC-box elements are located in 5'-upstream regions of the *DKC1*, *RAP1*, *TANK1*, *TIN2*, *TPP1*, *TRF1*, *TRF2*, *TERT*, and *TERC* genes, but not in the *POT1* and *TANK2* genes. Similar to the 5'-flanking region of the *WRN* gene, all of the isolated DNA fragments have no obvious TATA-box like sequences except the 5' flanking region of the *TERC* gene (Uchiumi et al., 2010a).

#### **3.2 Effect of Rsv on the promoter activities of 5'-flanking regions of the shelterinencoding genes**

The natural compound Rsv is known to have life-span promoting properties in yeast and metazoans by affecting the insulin-signaling cascade (Fröjdö et al., 2008). In order to examine the effect of Rsv on the isolated 5'-upstream regions of the shelterin encoding genes, Luc assays were performed. Luc expression plasmids which contained 5'-flanking regions of various telomere maintenance factor-encoding genes were transfected into HeLa-S3 cells by the DEAE-dextran based multiple transfection method (Uchiumi et al., 2010a). Luc activities of reporter plasmid-transfected cells were normalized to that of the pGL4-PIF1 transfected cells, because PIF1 has been suggested to have a negative effect on telomere elongation in yeast cells (Schulz & Zakian, 1994), and it has been shown that the change in the PIF1 promoter activity is largely unaffected after treatment with Rsv (Uchiumi et al., 2011). As shown in Table 2, treatment with Rsv (10 μM) for 24 h augmented Luc activities from the cells transfected with Luc reporter plasmids. Apparent positive responses to the Rsv treatment of the 5'-flanking regions of the *TERT* and *TERC* genes were observed, consistent with the activation of telomerase by Rsv in HeLa-S3 cells (Uchiumi et al., 2011). Although no obvious GC-box like elements are found in the 300-bp 5'-upstream regions of the *POT1* and *TANK2* genes (Fig. 1), Luc activities of these plasmid-transfected cells increased 2.53- and 1.69-fold, respectively, by Rsv treatment.

#### **3.3 Effect of 2DG on the promoter activities of 5'-flanking regions of the shelterinencoding genes**

2DG is known to affect life span by its CR mimetic effect on various species (Roth et al., 2001). We previously observed that treatment with 2DG induces telomerase activity along with transcriptional activation of the *TERT* and *WRN* genes in HeLa-S3 cells (Zhou et al., 2009). Therefore, we examined the effect of 2DG on the promoter activities of shelterinencoding genes. Although most of the Luc activities of cells transfected with shelterin promoter-Luc expression constructs were diminished by 2DG, the treatment induced

**3.1 Isolation of 5'-flanking regions of human telomere-associated protein-encoding** 

**3.2 Effect of Rsv on the promoter activities of 5'-flanking regions of the shelterin-**

**3.3 Effect of 2DG on the promoter activities of 5'-flanking regions of the shelterin-**

2DG is known to affect life span by its CR mimetic effect on various species (Roth et al., 2001). We previously observed that treatment with 2DG induces telomerase activity along with transcriptional activation of the *TERT* and *WRN* genes in HeLa-S3 cells (Zhou et al., 2009). Therefore, we examined the effect of 2DG on the promoter activities of shelterinencoding genes. Although most of the Luc activities of cells transfected with shelterin promoter-Luc expression constructs were diminished by 2DG, the treatment induced

The natural compound Rsv is known to have life-span promoting properties in yeast and metazoans by affecting the insulin-signaling cascade (Fröjdö et al., 2008). In order to examine the effect of Rsv on the isolated 5'-upstream regions of the shelterin encoding genes, Luc assays were performed. Luc expression plasmids which contained 5'-flanking regions of various telomere maintenance factor-encoding genes were transfected into HeLa-S3 cells by the DEAE-dextran based multiple transfection method (Uchiumi et al., 2010a). Luc activities of reporter plasmid-transfected cells were normalized to that of the pGL4-PIF1 transfected cells, because PIF1 has been suggested to have a negative effect on telomere elongation in yeast cells (Schulz & Zakian, 1994), and it has been shown that the change in the PIF1 promoter activity is largely unaffected after treatment with Rsv (Uchiumi et al., 2011). As shown in Table 2, treatment with Rsv (10 μM) for 24 h augmented Luc activities from the cells transfected with Luc reporter plasmids. Apparent positive responses to the Rsv treatment of the 5'-flanking regions of the *TERT* and *TERC* genes were observed, consistent with the activation of telomerase by Rsv in HeLa-S3 cells (Uchiumi et al., 2011). Although no obvious GC-box like elements are found in the 300-bp 5'-upstream regions of the *POT1* and *TANK2* genes (Fig. 1), Luc activities of these plasmid-transfected cells

flanking region of the *TERC* gene (Uchiumi et al., 2010a).

increased 2.53- and 1.69-fold, respectively, by Rsv treatment.

Previously, we isolated and characterized 5'-flanking regions of the human *TERT* and *TERC* genes (Zhou et al., 2009; Uchiumi et al., 2010a). In this study, those of different human telomere-associated protein-encoding genes were obtained by PCR and inserted into the MCS of the pGL4-basic (pGL4[luc 2.10]) vector. Putative transcription-factor binding elements were found by TF-SEARCH analysis. As summarized in Fig. 1, c-Ets/Elk1, Sp1/GC-box, CREB, OCT, p300, SRY, GATA, E2F, NF-κB/c-Rel, CCAAT-box, and other motifs are located within 300-bp from the 5'-upstream region of the cDNAs. Although all of these telomere-associated protein factors are commonly involved in the maintenance of telomeres, a rigid rule in the order of the *cis*-elements could not be found in their core promoter regions. However, one or more Sp1/GC-box elements are located in 5'-upstream regions of the *DKC1*, *RAP1*, *TANK1*, *TIN2*, *TPP1*, *TRF1*, *TRF2*, *TERT*, and *TERC* genes, but not in the *POT1* and *TANK2* genes. Similar to the 5'-flanking region of the *WRN* gene, all of the isolated DNA fragments have no obvious TATA-box like sequences except the 5'-

**3. Results** 

**encoding genes** 

**encoding genes** 

**genes** 

Fig. 1. Promoter regions of the human genes encoding telomere-associated proteins or shelterin protein factors. PCR-amplified 5'-flanking regions of these genes, which were inserted upstream of the *Luciferase* gene of the pGL4-basic vector (pGL4[luc 2.10]), are shown. Transcription start sites (or 5'-end of cDNAs) are designated +1. The TF-SEARCH program (http://www.cbrc.jp/research/db/TFSEARCH.html) was performed and putative transcription-factor binding-elements (score > 85) are shown schematically.

Characterization of 5'-Flanking Regions of Various

pGL4-PIF1 pGL4-PIF1

pGL4-RTEL pGL4-RTEL

pGL4-DKC1 pGL4-DKC1

pGL4-POT1 pGL4-POT1

pGL4-RAP1 pGL4-RAP1

pGL4-TANK1 pGL4-TANK1

pGL4-TANK2 pGL4-TANK2

pGL4-TIN2 pGL4-TIN2

pGL4-TPP1 pGL4-TPP1

pGL4-TRF1 pGL4-TRF1

pGL4-TRF2 pGL4-TRF2

pGL4-TERT pGL4-TERT

pGL4-TERC pGL4-TERC

**4. Discussion** 

**mimetic drugs** 

Student's *t*-test (\*p<0.05, \*\*p<0.01, \*\*\*p<0.005).

Human Telomere Maintenance Factor-Encoding Genes 591

1.000 + 0.206 1.000 + 0.230

1.800 + 0.802 4.011 + 0.917

1.136 + 0.111 8.767 + 4.556

0.030 + 0.008 0.139 + 0.065

0.993 + 0.247 5.456 + 1.411\*

0.047 + 0.012 0.272 + 0.110

0.012 + 0.003 0.066 + 0.011\*\*\*

0.130 + 0.038 0.686 + 0.273

0.604 + 0.151 3.211 + 0.237\*\*\*

0.853 + 0.131 2.178 + 0.408\*\*

0.173 + 0.073 0.378 + 0.036\*

0.586 + 0.094 1.844 + 0.498\*

0.651 + 0.120 1.878 + 0.426\*\* 1.000 + 0.141 1.000 + 0.148

2.550 + 0.648 6.651 + 1.958\*

2.560 + 0.265 6.698 + 0.921\*\*\*

0.027 + 0.010 0.102 + 0.015\*\*\*

2.201 + 0.236 4.977 + 0.749\*\*\*

0.106 + 0.012 0.567 + 0.150\*

0.006 + 0.002 0.033 + 0.032

0.213 + 0.023 0.474 + 0.093\*\*

0.751 + 0.099 5.721 + 1.302\*

1.355 + 0.279 3.442 + 0.567\*\*

0.232 + 0.022 0.693 + 0.244

1.707 + 0.316 3.456 + 0.963\*

0.897 + 0.119 2.516 + 0.507\*\*

Reporter Relative Luc activity














Table 3. Effect of 2-deoxy-D-glucose (2DG) on promoter activities of telomere-associated genes in HeLa-S3 cells Various reporter plasmids were introduced into HeLa-S3 cells by multiple DEAE-dextran method transfections. After 4 h of transfection, the culture medium was discarded and changed to 2DG-containing (4 and 8 mM) or non-containing medium. Cells were harvested after 24 h (4 mM) or 19 h (8 mM) of the 2DG treatment, then Luc assays were performed. Relative values represent Luc activities compared with that of the pGL4- PIF1 transfected cells. Results show means + S.D. from three independent samples (N=3). Significance of differences between control and 2DG treated cells were analyzed by

**4.1 The promoter regions of the shelterin-encoding genes coordinately respond to CR** 

In the present study, 5'-flanking regions of different human telomere-associated protein factor-encoding genes were isolated, and these Luc reporter plasmids were used for transient transfection assays. The shelterin- or telomere-associated protein-encoding genes, including *TERT*, *TERC*, *DKC1*, and double-stranded break repair protein-encoding genes,

2DG (mM) 4 8


Table 2. Effect of Resveratrol (Rsv) on promoter activities of telomere-associated genes in HeLa-S3 cells Various reporter plasmids were introduced into HeLa-S3 cells by multiple DEAE-dextran method transfections. After 4 h of transfection, the culture medium was discarded and changed to Rsv-containing or non-containing medium. Cells were harvested after 24 h of treatment, then Luc assays were performed. Relative values represent Luc activities compared with that of the pGL4-PIF1 transfected cells. Results show means + S.D. from three independent samples (N=3). Significance of differences between control and Rsv treated cells were analyzed by Student's *t*-test (\*p<0.05, \*\*p<0.01, \*\*\*p<0.005).

relatively positive values compared to that of the pGL4-PIF1-transfected cells (Table 3). Similar to the response to Rsv (Table 2), the *TERT* and *TERC* promoters were activated by the 2DG treatment. The increase in relative promoter activity (compared with that of the pGL4-PIF1-transfected cells) after 2DG (8 mM) treatment was significant for the *RTEL*, *DKC1*, *POT1*, *RAP1*, *TANK1*, *TIN2*, *TPP1*, and *TRF1* promoters (Table 3). These results suggest that the CR mimetic compound 2DG affects the balance of gene expression to protect telomeres.

activity Fold

1.000 + 0.088 1.00

2.667 + 0.326 1.23

2.390 + 0.325\*\* 1.88

0.045 + 0.008\*\* 2.53

1.372 + 0.164\* 1.84

0.107 + 0.014 1.54

0.0099 + 0.00230 1.69

0.190 + 0.013 1.48

0.714 + 0.115\* 1.54

1.189 + 0.104\*\*\* 1.83

0.224 + 0.013\*\*\* 1.61

1.532 + 0.081\*\*\* 1.93

1.096 + 0.067\* 1.97

1.000 + 0.033

2.170 + 0.119

1.271 + 0.117

0.018 + 0.003

0.746 + 0.023

0.069 + 0.023

0.0059 + 0.00155

0.128 + 0.022

0.463 + 0.032

0.648 + 0.078

0.139 + 0.005

0.794 + 0.042

0.557 + 0.142

Reporter Rsv (10 μM) Relative Luc














treated cells were analyzed by Student's *t*-test (\*p<0.05, \*\*p<0.01, \*\*\*p<0.005).

Table 2. Effect of Resveratrol (Rsv) on promoter activities of telomere-associated genes in HeLa-S3 cells Various reporter plasmids were introduced into HeLa-S3 cells by multiple DEAE-dextran method transfections. After 4 h of transfection, the culture medium was discarded and changed to Rsv-containing or non-containing medium. Cells were harvested after 24 h of treatment, then Luc assays were performed. Relative values represent Luc activities compared with that of the pGL4-PIF1 transfected cells. Results show means + S.D. from three independent samples (N=3). Significance of differences between control and Rsv

relatively positive values compared to that of the pGL4-PIF1-transfected cells (Table 3). Similar to the response to Rsv (Table 2), the *TERT* and *TERC* promoters were activated by the 2DG treatment. The increase in relative promoter activity (compared with that of the pGL4-PIF1-transfected cells) after 2DG (8 mM) treatment was significant for the *RTEL*, *DKC1*, *POT1*, *RAP1*, *TANK1*, *TIN2*, *TPP1*, and *TRF1* promoters (Table 3). These results suggest that the CR mimetic compound 2DG affects the balance of gene expression to

pGL4-PIF1 pGL4-PIF1

pGL4-RTEL pGL4-RTEL

pGL4-DKC1 pGL4-DKC1

pGL4-POT1 pGL4-POT1

pGL4-RAP1 pGL4-RAP1

pGL4-TANK1 pGL4-TANK1

pGL4-TANK2 pGL4-TANK2

pGL4-TIN2 pGL4-TIN2

pGL4-TPP1 pGL4-TPP1

pGL4-TRF1 pGL4-TRF1

pGL4-TRF2 pGL4-TRF2

pGL4-TERT pGL4-TERT

pGL4-TERC pGL4-TERC

protect telomeres.


Table 3. Effect of 2-deoxy-D-glucose (2DG) on promoter activities of telomere-associated genes in HeLa-S3 cells Various reporter plasmids were introduced into HeLa-S3 cells by multiple DEAE-dextran method transfections. After 4 h of transfection, the culture medium was discarded and changed to 2DG-containing (4 and 8 mM) or non-containing medium. Cells were harvested after 24 h (4 mM) or 19 h (8 mM) of the 2DG treatment, then Luc assays were performed. Relative values represent Luc activities compared with that of the pGL4- PIF1 transfected cells. Results show means + S.D. from three independent samples (N=3). Significance of differences between control and 2DG treated cells were analyzed by Student's *t*-test (\*p<0.05, \*\*p<0.01, \*\*\*p<0.005).
