| Nucleic Acids Research | Pages |
Structure and function of the human Werner syndrome gene promoter: evidence for transcriptional modulation
Introduction
Materials and Methods
Human WRN promoter
Cell culture
RNAse protection assay
Luciferase assay constructs
5[prime] deletion assay
Results
Structure of the human WRN promoter and mapping of the 5[prime] end of the mRNA
Functional analysis of the human WRN promoter
Decreased WRN promoter activity in cells from WS patients
Discussion
Acknowledgements
References
Structure and function of the human Werner syndrome gene promoter: evidence for transcriptional modulation
DDBJ/EMBL/GenBank accession no. AF032113
ABSTRACT
INTRODUCTION
Werner syndrome (WS) is a rare autosomal recessive disorder that has been characterized as a `caricature of aging' (1) and as a `segmental progeroid syndrome' (2). Affected individuals exhibit an outward appearance of accelerated aging, with features such as premature graying and thinning of hair, skin atrophy and regional atrophy of subcutaneous fat. By middle age, WS individuals also prematurely develop many disorders commonly associated with advanced age, including bilateral ocular cataracts, type 2 diabetes mellitus, hyperlipidemia, gonadal atrophy and osteoporosis. In WS patients, these age-related disorders develop ~20 years ahead of normal populations. WS patients also develop several forms of arteriosclerosis, and a variety of benign and malignant neoplasms (1,3).
Primary WS fibroblast cultures have very limited replicative life spans, compared to age-matched controls (4-6). The S phase has been shown to be prolonged both in skin fibroblast cultures (7) and lymphoblastoid cell lines (LCLs) (8). Somatic cells from WS patients have a hypermutator phenotype at the levels of both chromosomes and genes (9-11). Error-prone DNA ligation of transfected linearized plasmids has been observed in LCLs from WS patients (12). Taken together, these observations suggest that WS may be a member of a family of genomic instability syndromes, such as Bloom (13,14), Fanconi (15) and Cockayne syndromes (16), as well as various forms of xeroderma pigmentosum (17).
The initial mapping of the WS gene (WRN) to chromosome 8p (18) was confirmed by homozygosity mapping (19). Physical and genetic maps of the region were constructed, and several haplotypes were delineated (20-23). WRN was subsequently identified by positional cloning (DDBJ/EMBL/GenBank accession no. L76937).">http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?db=n&form=6&uid=L76937&dopt=g">L76937). Fifteen distinct mutations in the WRN gene have been reported in WS patients (24-26).
The protein predicted from the WRN cDNA encodes a 1432 amino acid protein with regional homology to RecQ helicases from various species. The WRN helicase contains seven helicase domains, including an ATPase domain, which are common to many helicases (27). The WRN protein belongs to the DEAH (Asp-Glu-Ala-His) subfamily of DNA and RNA helicases, as does Escherichia coli RecQ (28). Escherichia coli RecQ has DNA helicase activity and ATPase activity, and can translocate on single-stranded DNA in the 3[prime]->5[prime] direction (29). Likewise, recombinant human WRN protein has been shown to be a 3[prime]->5[prime] ATP-dependent DNA helicase (30,31). A 3[prime]->5[prime] exonuclease has been speculated based on the protein structure of the N-terminal domain of the WRN protein (32,33). Also, a nuclear localization signal in the C-terminal domain of WRN (amino acids 1370-1375 NKRRCF) (34) has been recently demonstrated.
DNA helicases have been implicated in a number of DNA transactions requiring unwinding of a duplex. Processes such as DNA replication (35) and repair (36,37) are thought to be carried out by heteromultimeric complexes that contain helicases. Nucleotide excision repair appears to be coupled with transcription, and some mutant helicases responsible for DNA instability syndromes may impair lesion recognition and/or lesion removal of damaged nucleotides during transcription (36,37). Examples include the ERCC2 helicase, which complements xeroderma pigmentosum D, and its yeast homologue RAD3 (38), as well as ERCC3, which complements xeroderma pigmentosum B and group B Cockayne syndrome, and its yeast homologue RAD25 (39-42). In E.coli, RecQ helicase is involved in the initial step of recombinational DNA repair (43).
The extent to which WRN is subject to transcriptional regulation is unknown. As one approach to this question, we investigated the structure of its promoter. While several of our findings are consistent with its categorization as a constitutive `housekeeping' promoter, we have also observed a region that could play a novel role in the regulation of transcription.
MATERIALS AND METHODS
Human WRN promoter
Human PAC clones were obtained by PCR of the arrayed library (Genomic Systems Inc., St Louis, MO) using the primers for human WRN exon 1 (24) in order to select PAC clones bearing the WRN promoter region. Two positive PAC clones, 13 226 and 13 227, were digested with various restriction enzymes and hybridized to human WRN exon 1. A 5.5 kb HindIII DNA fragment, which hybridized to human WRN exon 1, was subcloned into pBlueScript KS(-) and sequenced.
Cell culture
HeLa cells, SV40 transformed human fibroblasts (GM649), human umbilical cord endothelial cells (CRL1998), human aortic smooth muscle cells (CRL1999) and SV40 transformed WS fibroblasts (AG11395) were obtained from the ATCC (Rockville, MD). Cells were maintained in Dulbecco's modified Eagle's Medium containing 10% heat-inactivated (55°C × 30 min) fetal bovine serum, 100 U/ml penicillin G and 100 µg/ml of streptomycin.
LCLs established by infection with Epstein-Barr virus were from the International Registry of Werner Syndrome, University of Washington, Seattle (Directors, G.M.Martin and J.Oshima). Cells were maintained in RPMI medium containing 10% heat-inactivated fetal bovine serum, 100 U/ml penicillin G and 100 µg/ml of streptomycin.
RNAse protection assay
A 675 bp PstI-NaeI fragment was subcloned into pBlueScript KS(-) at the PstI-EcoRV site. An RNA probe was synthesized using T3 RNA polymerase for the antisense probe and T7 RNA polymerase for the sense probe. An aliquot of 50 µg of total RNA was isolated from human endothelial cells and human smooth muscle cells as described previously (44). The ribonuclease protection assay was performed as described (45).
Luciferase assay constructs
Six constructs containing varied lengths of putative human WRN promoter sequences, pGL-BK, pGL-RS, pGL-PS, pGL-RH, pGL-BH and pGL-SH, were generated by subcloning 1.6 kb HindIII-SphI (BK), 637 bp SmaI-SphI (RS), 486 bp PstI-SphI (PS), 826 bp SmaI-NaeI (RH), 675 bp PstI-NaeI (BH) and 189 bp SphI-NaeI (SH) fragments, respectively, from the 5.5 kb HindIII fragment into a pGL3 basic vector which contains a luciferase reporter gene (Promega Inc., Madison, WI).
Human hexokinase II proximal promoter (hHK) (nt -964 to +106), which contains a TATA and CAAT box as well as a GC-rich region with eight Sp-1 binding sites, was subcloned into a luciferase vector, pXP1 (ATCC #37576) (46). A TATA-less promoter, human lipoprotein lipase promoter (hLPL) (nt -519 to +44), was also subcloned into pXP1 (47). Both pXP1-HK and pXP1-hLPL were provided by Dr Samir S. Deeb, University of Washington. SV40 promoter driven luciferase reporter construct, pGL3p, was obtained from Promega Inc. (Madison, WI).
5[prime] deletion assay
HeLa cells or SV40 transformed human fibroblasts (1 × 105) were plated in a 35 mm dish for 48 h. LCLs (1 × 106) were placed in 15 ml polypropylene tubes. An aliquot of 2 µg of pGL DNA containing the putative WRN upstream fragments was co-transfected with 0.02 µg of pRL-CMV (cytomegalovirus promoter driven Renilla luciferase), which served as an internal standard, using 10 µl of lipofectin (Gibco BRL) and 200 µl of Opti-MEM (Gibco BRL) for 12 h. The transfection medium was then aspirated and 2 ml of DMEM with 10% FBS (or RPMI with 10% FBS for LCLs) were added. Thirty-six hours later, the cells were washed three times with phosphate buffered saline (PBS) and scraped in 250 µl of lysis buffer. Luciferase activity was determined according to the manufacturer's instructions.
RESULTS
Structure of the human WRN promoter and mapping of the 5[prime] end of the mRNA
A 5.5 kb HindIII fragment which hybridized to exon 1 contained ~1.6 kb of promoter sequence. The 5[prime] end of the cDNA (or transcriptional start site) was determined by 5[prime] RACE using various combinations of primers designed from the cDNA (26). We did not find a classical TATA box or CAAT box within 826 bp of a SmaI-NaeI fragment, which contains 17 bp of the 5[prime] end from the previously published human WRN cDNA sequence (DDBJ/EMBL/GenBank accession no. L76937) (Fig.
Figure 1. Nucleotide sequence of the human WRN promoter. -> indicates an upstream transcription initiation site. Ý indicates the start of L76937. Restriction sites used for the 5[prime] deletion assay, Sp1 binding consensus sequence (GGGCGG), ETF binding core sequence (CCCC) in forward orientation, and GCF binding consensus sequence are indicated. We then performed an RNAse protection assay, using a 726 bp probe containing a 675 bp PstI-NaeI fragment, including the initial 17 bp of L76937 (Fig. Figure 2. Mapping of the human WRN upstream transcription start site. CRL1998, human umbilical cord endothelial cell; CRL1999, human aortic smooth muscle cell. An aliquot of 50 µg of total RNA was analyzed by RNAse protection assay, using antisense or sense probe. The sizes of protected fragments and markers are indicated. The human WRN promoter shares two characteristics of housekeeping genes described by Razin and Kafri (48): a (G+C) content of >50% and a CpG/GpC ratio >60%. This region contains multiple Sp1 binding site sequences (consensus GGGCGG) (49-51). It also contains 13 core binding sequences for the nuclear factor, ETF (core sequence GGGG) in the forward orientation, which specifically stimulates transcription from TATA-less promoters (52) (Fig.
Functional analysis of the human WRN promoter
Human WRN promoter activity was analyzed using luciferase as a reporter. Six promoter constructs were initially tested in two different cell lines: three constructs (BK, RS and PS) utilize the upstream start site only, one utilizes the downstream start site (SH), and two constructs include both start sites (RH and BH). The luciferase activities driven by these various constructs in two different host cell lines are summarized in Figure
Figure 3. Deletion analysis of WRN promoter in epithelial cells and in fibroblasts. The background activity (the activity of luciferase with no promoter) was subtracted from the total luciferase activity. The luciferase activity was then normalized by an internal standard, a co-transfected Renilla luciferase activity driven by a CMV promoter (Dual luciferase, Promega Inc., Madison, WI). Black bars indicate the normalized WRN luciferase activity in HeLa epithelial cells; gray bars indicate such results in GM649 fibroblasts. The values represent the averages and standard deviations of four independent experiments. In both SV40 transformed fibroblast cell line GM649 and HeLa cells which are derived from human cervical carcinoma, the constructs containing the downstream start site gave higher activity than the ones containing the upstream start site only. The average differences were 6.8-fold in GM649, and 8.1-fold in HeLa cells. Unlike tissue-specific genes, in which promoter activities generally decrease as the 5[prime] end of the promoter is progressively deleted, luciferase activities were similar or were slightly increased as the 5[prime] regions of the promoter were deleted. This was the case for both start sites. The observed slight increases in reporter gene activity suggest the presence of a negative regulatory binding site in this region. One such candidate is a binding site for the transcriptional repressor, GC factor (GCF) (consensus GCGGGGC) (53) at -882 bp. Aside from the GCF binding consensus sequence, analysis using the TESS data base (used for the identification of transcription factor binding sites) did not show a perfect alignment with any other known transcriptional repressor.
Decreased WRN promoter activity in cells from WS patients
In order to assess the possibility of an auto-loop mechanism for the WRN promoter, we carried out luciferase reporter assays in somatic cells from patients homozygous for WRN mutations. We did these studies on LCLs from a WS patient and a normal control (both members of the LGS pedigree), on SV40 virus transformed WS fibroblasts (AG11395) and on control SV40 transformed fibroblasts (GM649). The WRN mutations in the LGS family and in AG11395 result from a single nucleotide substitution resulting in a change from amino acid 369 CGA (Arg) to TGA (Stp) (25,26). This mutation results in a truncated product that eliminates the acidic repeat region, the helicase region, and the C-terminal region of WRN protein.
The luciferase activity of a transfected RH construct was decreased to 20% of controls in WS LCL cells and to 31% of the controls in SV40 transformed fibroblasts from WS (Fig.
| Figure 4. Diminished WRN promoter activity in somatic cells from WS patients. (A) Luciferase activity of promoter construct RH (Fig. 3) was determined in transfected somatic cells from control subjects (WRN +/+) (black bars) and from WS patients (WRN -/-) (shaded bars). Left set shows the activities in LCLs derived from the LGS pedigree; right set shows the activities in SV40 fibroblasts derived from a control individual (GM649) and a WS patient (AG11395). RH luciferase activity was normalized to the activity of co-transfected CMV-driven Renilla luciferase. Activities of WS cells are shown relative to controls (activity = 1.0). The values represent the averages and standard deviations of three independent experiments. (B) Luciferase activity was determined using six different WRN promoter constructs in fibroblast cell lines from a normal individual (GM649; black bars) and a WS patient (AG11395; shaded bars). The activity levels were normalized to co-transfected Renilla luciferase activity. The values represent the averages and standard deviations of three independent experiments. |
![]() |
In order to determine the specificity of reduction in WRN promoter activity among WS cells, we also examined luciferase activities of three other promoter constructs in SV40 transformed WS fibroblast cells (Fig.
Figure 5. The activities of various promoters in WS and control fibroblasts. hHKP (human hexokinase II promoter), hLPLP (human lipoprotein lipase promoter) and SV40P (SV40 promoter) were tested in normal SV40 transformed fibroblasts (GM649; black bars) and in SV40 transformed WS fibroblasts (AG11395; shaded bars). The activity levels were normalized to co-transfected Renilla luciferase activity. The values represent the averages and standard deviations of three independent experiments. Some eukaryotic promoters do not contain a classical TATA or CAAT box. The activation of these TATA-less promoters relies on a transcriptional initiation complex other than the TATA-binding TFIID (51). TATA-less promoters that appear to be expressed `constitutively' (expressed in all cell types and presumed to be stable) have been referred to as `housekeeping' genes. It is clear, however, that for some so-called `housekeeping' genes, their levels of expression can be modulated by external stimuli such as growth factors or oxidative stress. Examples include glucose-6-phosphate dehydrogenase (54), dihydrofolate reductase (55), the insulin receptor (56), the epidermal growth factor receptor (57) and the c-ras-Harvey proto-oncogene (58). In this study, we show that the human WRN gene is another so-called housekeeping gene. The human WRN promoter shares similarities with the promoters of several other helicases. For example, the WRN promoter, as well as the promoters of ERCC3 and its mouse homologue, all have no apparent TATA or CAAT boxes (59). The tissue distributions of ERCC3 and WRN mRNAs are also similar. Both are expressed constitutively at low levels in all tissues examined with the exception of the testis, which exhibits a relatively high expression. ERCC2, by contrast, has a classical promoter with a TATA box, a reverse CAAT box, and a GC box (38), indicating that a TATA-less promoter is not a general feature of helicases thought to be involved in the repair of DNA damage. We observed that several features of `housekeeping' genes are present in the WRN promoter. Housekeeping genes generally have a CpG island of some 200-1000 bp spanning the promoter region, frequently extending into the first exon (60). These promoters are not only rich in G+C, but are characterized by a ratio of CpG/GpC that is >0.6 (48). The 5[prime] end of the WRN gene falls well within these values. CpG islands in housekeeping genes are generally unmethylated in both somatic and germline cells (60). We have not determined the methylation status of the 5[prime] end of the WRN gene. Although the sequence suggests constitutive expression, the luciferase assays in this study revealed that transcriptional activity can be altered by the presence or absence of functional WRN protein. Our analysis shows that the WRN gene has two transcriptional start sites. All three cell lines (HeLa cells, SV40 fibroblasts and LCLs) initiate transcription most frequently with the downstream start site. Sp1 (known to be abundant in HeLa cells) (49) and other binding sites between -170 and +19 could be responsible for the observable difference in the use of transcriptional initiation. We were, however, unable to demonstrate the usage of the two different transcriptional start sites by conventional northern analysis because of the relatively large mRNA size of WRN (the major form of WRN cDNA is 5189 bp). In addition, cDNA library screening and 3[prime] RACE of the WRN gene gave us two additional minor forms of WRN cDNA which utilize polyadenylation sites ~100 and 200 bp 3[prime] to the site of the major form. The length of the poly(A) is not known for each form. Typically, 5[prime] deletion analysis reveals the binding of potential positive transcriptional factors. In the WRN gene promoter, however, activity was increased as the 5[prime] end underwent deletion, suggesting that negative regulators may bind near the transcriptional start sites. What might be the significance of these observations for normal WRN function and the pathogenesis of WS? It is conceivable, for example, that the putative negative regulator is inactivated via some protein-protein interaction in response to certain forms of DNA damage, thus permitting enhanced expression. It is also possible that this domain may come into play under circumstances of enhanced recombinational repair (61,62), such as may occur in the testis, where expression is relatively high. With this in mind, the severe and premature testicular atrophy and the associated rapid decline in fertility seen in WS patients is of special interest (1). Some nuclear proteins have been shown to be autoregulated. An example is the nuclear protein, Fos, which is involved in both positive and negative regulation of its own promoter (63,64). When WRN promoter activity was examined in WS cells, transcription of constructs, particularly the ones containing the downstream start site, was dramatically reduced compared to control cells. This was seen in two different cell types, LCLs and SV40 fibroblasts. It should be pointed out, however, that the control lines were not in genetically matched background, which leaves the possibility that the specific reduction of WRN promoter activity is limited to the WS cell lines tested in the study. Although luciferase activity does not directly measure the actual transcriptional rate in cells, it is generally proportional to transcriptional activity. Reduced transcription in cells lacking functional WRN protein suggests the presence of a positive regulatory loop. A positive transcriptional modulation could potentially enable cells to induce the expression levels of helicase protein (and, presumably, a family of interacting proteins) more effectively when needed. While such processes as post-translational modifications (phosphorylation, glycosylation, etc.) or shifts in cellular compartmentalization can be expected to result in more rapid recruitment of the WRN helicase and related proteins, regulation at the transcriptional level may have evolved to produce a more sustained response to DNA damage. Although we are at a very early stage of research on WRN protein function(s), two recent publications raise the question of roles for the WRN helicase protein in, respectively, transcription-coupled repair (65) and mismatch repair (66). Reduced expression of mutated WRN mRNA (67) could be due to its rapid degradation (68) as well as its reduced transcriptional rate. The positive transcriptional modulation could possibly be caused by either direct binding of WRN helicase to the WRN promoter or an indirect effect of WRN helicase. If the mechanism of positive transcriptional modulation is direct or is part of a complex binding of WRN protein to its own promoter, the decrease in WRN promoter activity in WS is expected in all WS cells regardless of the location of mutation. All mutations so far identified in WS patients cause the altered intracellular localization of the truncated WRN protein due to the loss of the nuclear localization signal at the C-terminus (34). Such events would be expected to lead to functional null mutations. A set of experiments has been attempted to rescue the reduced WRN promoter activity in WS cells by transient co-transfection of WRN expression vector with the WRN promoter constructs. Resumption of WRN promoter activity was not observed, however (data not shown). There are several possible explanations for the inability to rescue promoter activity by transient transfection experiments. First, the expression of WRN wildtype protein in the co-transfection experiments could not be controlled at the same level as endogenous expression. Differences in the stoichiometry of the transfected versus endogenous WRN protein expression may therefore have influenced the results. Second, overexpression of WRN protein alone may not be sufficient for transcription. The expression of other genes involved in normal promoter function could also have been altered in WS cells. Third, transiently transfected DNA in some cases does not acquire a physiologically spaced nucleosome structure. Some transcription factors require the tertiary chromatin structure for correct binding (69). The transiently transfected WRN promoter construct may not form the necessary structure required for optimum transcription. Several functional domains within the WRN helicase have been proposed, including a 3[prime]->5[prime] ATP-dependent helicase domain (30,31) and a putative nuclease domain (32,33). The present study now raises the possibility that a function of WRN may include information relevant for its transcriptional modulation. Therefore, understanding of WRN promoter regulation may shed light on normal WRN protein functions, and on the cellular responses to WRN mutations. We are grateful to Mr Charles E. Ogburn and Drs Samir S. Deeb, Wei-Shuing Yang and Sam Li for their technical assistance. This work is supported by R01 AG14446 (J.O.), AFAR Research Grant (J.O.) and R24 CA78088 (G.M.M.). The nucleotide sequence reported in this paper has been submitted to the GenBankTM/EMBL Data Bank with accession number AF032113.
DISCUSSION
ACKNOWLEDGEMENTS
REFERENCES
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 20 Jul 1998
Copyright©Oxford University Press, 1998.
![]()
CiteULike
Connotea
Del.icio.us What's this?
![]()
![]()

![]()
![]()
![]()
C. Grandori, K.-J. Wu, P. Fernandez, C. Ngouenet, J. Grim, B. E. Clurman, M. J. Moser, J. Oshima, D. W. Russell, K. Swisshelm, et al.
Werner syndrome protein limits MYC-induced cellular senescence
Genes & Dev.,
July 1, 2003;
17(13):
1569 - 1574.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
H. Liu and E. M. Johnson
Distinct proteins encoded by alternative transcripts of the PURG gene, located contrapodal to WRN on chromosome 8, determined by differential termination/polyadenylation
Nucleic Acids Res.,
June 1, 2002;
30(11):
2417 - 2426.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
M. Fry
The Werner Syndrome Helicase-Nuclease--One Protein, Many Mysteries
Sci. Aging Knowl. Environ.,
April 3, 2002;
2002(13):
re2 - 2.
[Abstract]
[Full Text]
[PDF]
![]()
![]()
![]()

![]()
![]()
![]()
L. Wang, C. E. Ogburn, C. B. Ware, W. C. Ladiges, H. Youssoufian, G. M. Martin, and J. Oshima
Cellular Werner Phenotypes in Mice Expressing a Putative Dominant-Negative Human WRN Gene
Genetics,
January 1, 2000;
154(1):
357 - 362.
[Abstract]
[Full Text]
![]()
This Article ![]()
![]()
Abstract
![]()
Print PDF (186K)
![]()
Alert me when this article is cited
![]()
Alert me if a correction is posted
![]()
Services ![]()
![]()
Email this article to a friend
![]()
Similar articles in this journal
![]()
Similar articles in ISI Web of Science
![]()
Similar articles in PubMed
![]()
Alert me to new issues of the journal
![]()
Add to My Personal Archive
![]()
Download to citation manager
![]()
Search for citing articles in:
ISI Web of Science (11)
![]()
Request Permissions ![]()
Commercial Re-use Guidelines
for Open Access NAR Content
![]()
Google Scholar ![]()
![]()
Articles by Wang, L.
![]()
Articles by Oshima, J.
![]()
Search for Related Content
![]()
PubMed ![]()
![]()
PubMed Citation
![]()
Articles by Wang, L.
![]()
Articles by Oshima, J.
![]()
Social Bookmarking ![]()
![]()
What's this?

