WO1997020942A1 - Hsd17b1 promoter, enhancer, silencer and use thereof - Google Patents

Abstract

The invention concerns human 17β-hydroxysteroid dehydrogenase type 1 (HSD17B1) gene regulation. Specifically, the invention concerns the regulation of 1.3 kb 17HSD type 1 mRNA transcription. More specifically the invention concerns HSD17B1 promoter, enhancer and silencer regions and their use in the identification of trans-acting agents capable of up-regulating or down-regulating estrogen production.

Description

HSD17B1 PROMOTER. ENHANCER. SILENCER AND USE THEREOF

TECHNICAL FIELD OF THE INVENTION

The present invention is directed to human 17ß-hydroxysteroid dehydrogenase ( 17HSD) type 1 gene (HSD17B1 previously also called EDH17B2) regulation. Specifically, the invention is directed to the regulation of 1.3 kb 17HSD type 1 mRNA transcription. More specifically the invention is directed to HSD17B1 promoter, enhancer and silencer regions and their use in the identification of trans-acting agents capable of up-regulating or down-regulating estrogen production. 17ß-Hydroxysteroid dehydrogenases ( 17HSDs) catalyze the interconversion of 17-ketosteroids and 17ß-hydroxysteroids. such as estrone and estradiol, and androstenedione and testosterone. So far five different types of 17HSD have been cloned and the isoenzymes have been found to differ from each other in substrate specificity as well as in tissue distribution and subcellular localization.

Human 17ß-hydroxysteroid dehydrogenase type 1 ( 17HSD type 1) catalyzes primarily the reductive reaction from the low-activity estrone to the biologically more active form, estradiol. It is essential for estradiol production in ovarian granulosa cells and it is also highly expressed in the human placenta, which is the major source of estradiol during pregnancy. In addition to the steroidogenic tissues, 17HSD type 1 is present in some estrogen target cells, such as breast and endometrial epithelial cells, in which it takes part in regulation of estradiol supply for estrogen receptor locally. Thus, the action of 17HSD type 1, together with other factors, may lead to accumulation of estradiol and consequently affect cell proliferation rate. Multiple evidence has shown that excessive estradiol stimulation increases susceptibility to endometrial (Thomas, D.B., Cancer 62: 1755- 1767 ( 1988); King. R.J.B., Contraception 43:527-542 ( 1991)) and breast cancer (Vihko, R.. Apter. D., CRC. Crit. Rev. Oncol./Hematol. 9: 1- 16 ( 1989)). Altogether 17HSD type 1 has a significant role in the biosynthesis of estradiol in certain steroidogenic and peripheral tissues and probably in regulation of the estrogen response in estrogen-dependent tissues, both normal and malignant. 17HSD type 1 is encoded by the gene HSD17B1. which has been localized to the Ioci 17q 12-21 (Winqvist. R. et al., Hum Genet. 85:473-476 ( 1990)). Two transcription start points in the HSD17B1 gene result in two major 17HSD type 1 mRNAs, 1.3 kb and 2.3 kb in size (LuuThe. V. et al., Mol. Endocrinol. 4:268-275 ( 1990)). The longer species is constitutively expressed in several tissues and cell lines, whereas the amount of 1.3 kb 17HSD type 1 mRNA correlates with the concentration of 17HSD type 1 protein (Poutanen, M. et al., Cancer Res. 45- 897-900 ( 1992)). The amount of 1.3 kb mRNA responds also to several stimuli such as progestins, cAMP, retinoic acids (RAs) and growth factors, rather than the amount of the longer transcript.

The region from -78 to +9 in the HSD17B1 gene, with respect to the cap site for the 1.3 kb mRNA. contains a sequence typical of a TATA-box and a GC-rich area. 30 nucleotides in size (LuuThe, V. et al., Mol. Endocrinol. 4:268-275 ( 1990);

Peltoketo. H. et al., Eur. J. Biochem. 209:459-466 ( 1992)). Further, the fragment prepared from the region surrounding the cap site for the 1.3 kb transcript (fragment -303/+9) has previously been verified to produce a transcript whose initiation site is the same as that of the 1.3 kb 17HSD type 1 mRNA (Peltoketo. H. et al., Genomics 23:250-252 ( 1994)).

Although several factors affecting 17HSD type 1 expression have been identified, the regulatory mechanisms controlling HSD17B1 gene expression are poorly understood.

SUMMARY OF THE INVENTION The present invention provides isolated nucleic acid molecules capable of regulating transcription of 1.3 kb 17ß-hydroxysteroid dehydrogenase ( 17HSD) type 1 mRNA. In one aspect, the promoter region with crucial subsequences is described. The region from -78 to +9 in the HSD17B1 gene, with respect to the cap site for the 1.3 kb mRNA, was found by the present inventors to be capable of promoting the transcription and therefore to be able to act as a basal promoter. By the invention, a binding of Spl or similar transcription factor to the position from -43 to -52 is needed for the entire function of the HSD17B1 promoter, but on the other hand, mutation of the AP-2 binding site at the position from -53 to -62 results in increase of reporter gene expression. In addition, isolated nucleic acid molecule capable of enhancing HSD17B1 gene expression are provided. The HSD17B1 enhancer is localized within bases -661 to -392 as calculated from the transcription start site for the 1.3 kb 17HSD type 1 mRNA. The inventors have discovered that the enhancer increases, in both orientations, promoter activity more than 200-fold in certain cell lines. Deletions from either the 5'-region from -659 to -550 or from the 3'-region from -392 to -549 of the enhancer abolished its activity, which indicates the importance of both of the halves for enhancer function. Detailed chaiacterization of the enhancer area further demonstrated that subelements from -544 to - 528 and from -589 to -571 can be bound by factor( s) not identified so far and that the elements are essential for the enhancer activity. In addition, double mutation of the nucleotides at the positions -480 and -486 led to dramatic decrease in the enhancer activity making evident the importance of these nucleotides. Thymidine at the position -435 is also needed to achieve the full enhancer activity. On the other hand, change of adenine to guanine at the position -452 increased the enhancer activity by 40%.

By the invention, the function of the HSD17B1 enhancer is not dependent on promoter type. Moreover, the enhancer increases transcription whether linked adjacent to a promoter or when placed upstream at a distance of 2 kb or more. Thus, by the invention, an isolated HSD17B1 enhancer is provided which is localized within bases -661 and -392 of the sequence disclosed on page 10 (SEQ ID NO 1). In particular, the present invention is directed to isolated nucleic acid molecules at least 16 bases in length which are capable of enhancing gene transcription and have a nucleotide sequence that is contained within (and can include) nucleotides -661 and -392 of the sequence disclosed on page 10. Preferred isolated nucleic acid molecules of the present invention which are capable of enhancing gene transcription will also include the retinoic acid response element at -503 to -487 (discussed below) and will thus be at least 88 nucleotides in length. Further, 5 nucleotides extending up to -784 can be included without having a significant detrimental effect on transcriptional

enhancement.

In a further aspect, isolated nucleic acid molecules capable of repressing HSD17B1 gene expression are provided. The HSD17B1 silencer is localized within the bases -391 to -78. with the region from - 113 to -78 being essential for silencer function. Further, site-directed double mutagenesis of the nucleotides at the position -99 and - 100 or -101 and - 102 abolish the function of the silencer in choriocarcinoma cell lines, pointing to the fundamental role of the region from -99 to - 102. a binding site for GATA-2 and GATA-3 regulatory factors, in function of the silencer. The silencer of the present invention counteracts enhancer activity but not basal promoter

In particular, the present invention is directed to isolated nucleic acid molecules at least 36 bases in length which are capable of repressing gene transcription, wherein said nucleic acid molecules include nucleotides -113 to -78 in the sequence disclosed on page 10 and have a nucleotide sequence which is contained within (and can include ) nucleotides -391 and -78 of the sequence disclosed on page 10. Preferably, in addition to nucleotides from - 113 to -78. the nucleic acid molecules of the present invention which are capable of repressing gene expression will also include at least a part of the region from - I 14 to -391. which can increase the repressing effect. Specifically, the invention relates to the intact and mutated forms of HSD17B1 promoter, HSD17B1 enhancer region containing a functional retinoic acid response element and HSD17B1 silencer region and to their uses e.g. in the screening of drugs

The present invention is further directed to recombinant DNA molecules, including vectors such as expression plasmids, which contain a nucleic acid molecule, intact or mutated, of the present invention Preferably, the nucleic acid molecules of the present invention will, together with the promoter, be operably linked to a reporter gene. The recombinant constructs of the present invention may be introduced into a host cell, preferably a eukaryote cell, to assay reporter gene expression Preferred eukaryotic cells include choriocarcinoma cell lines, granulosa cells, breast cancer cell lines, prostate carcinoma cell lines and kidney cell lines.

Thus, in a further aspect, the present invention provides a screening assay for identifying trans-acting agents capable of affecting HSD17B1 gene expression. The screening assay involves ( 1) providing a host cell transfected with a recombinant nucleic acid molecule containing an HSD17B1 transcriptional regulatory element of the present invention ( i.e., enhancer and/or silencer), a promoter, and a reporter gene, wherein the transcriptional element and promoter are operably linked to the reporter gene; (2) administering a candidate HSD17B1 trans-acting agent to the transfected host cell; and (3) determining the effect on reporter gene expression. Thus, by the invention, drugs and ligands can be identified capable of up-regulating or down-regulating 17HSD type 1 expression, leading to an increase or decrease of estrogen production.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 illustrates the progressive 5'-end deletion analysis of the HSD17B1 promoter. In each panel, bars 1-7 show chloramphenicol acetyl transferase (CAT) expression driven by the nested deletion fragments, whereas bar 8 shows the background CAT activity of the vector pCAT-BY. In JEG-3 panel the results represent mean ± SEM from four independent experiments in each of which duplicate samples were analyzed. In the JAR panel the results represent mean ± range from two independent experiments in each of which duplicate samples were analyzed.

Figures 2, panels A and B. illustrate the localization of the HSD17B1 enhancer. In panel A of figure 2. bat 1 shows the basic activity of TK promoter resulting from the vector pBLCAT2 and bars 2- 15 show the TK promoter activity when it is linked to the region ranging from -859 to - I 13 in the HSD17B1 gene The results represent mean ± SEM from six independent experiments in each of which duplicate samples were analyzed In panel B of figure 2, bar 1 shows the basal activity of SV40 promoter, whereas bars 2 and 3 show SV40 promoter activity when linked to the region from -764/-392 at a distance of about 2.0 kb in the vector pCAT-Promoter. The results represent mean ± range from two independent experiments in each of which duplicate samples were analyzed. Figure 3 illustrates the cell-specificity of the HSD17B1 enhancer. The first bar in each group shows the basal TK promoter activity, whereas the second and third ones show the TK promoter activities when the region from -764 to -392 was linked to it in different orientations. The results represent mean ± SEM from six independent experiments in JEG-3 cells, three experiments in JAR cells and mean ± range from two independent experiments in BT-20. T-47D, MCF-7, PC-3 and CV-1 cells. In each experiment, duplicate samples were analyzed. The results from different cell lines are not directly comparable with each other because the expression of the reference reporter gene, the ß-galactosidase. varies from one cell line to another.

Figure 4, panels A and B, illustrate the interaction between HSD17B1 RARE and RXR_α/RAR_α extract. In panel A of Figure 4, lane 1 shows the binding between wt-RARE and RXR_α/RAR_α complex. As indicated in lane 2 to lane 6, unlabelled oligonucleotide competitors were included in the binding reaction at a 100-fold molar excess over probe . These competitors included wt-RARE (AGGACAggagaAGGTCA), m 1 -RARE (ACGACTggagaAGGTCA), m2-RARE ( AGGACAggagaAAGTCG), m3-RARE (ACGACTggagaAAGTCG) and unrelated

DNA (for the whole sequences, see Example 4. ). The position of the complex is indicated by the arrow at the left. In panel B of figure 4 interaction of the labelled oligonucieotides ßRARE01 , wt-RARE and m3-RARE with RXR_α/RAR_α protein extract is shown. In each group, a free probe has been applied in lane 1 ; a probe incubated with RXRα/RARα extract in lane 2, a probe incubated with RXR_α/RAR_α extract and a 100-fold molar excess of cold probe in lane 3, and a probe incubated with RXR_α/RAR_α extract and a 100-fold molar excess of cold unrelated DNA in lane 4. The position of the complex is indicated by the arrow at the left.

Figure 5. panels A and B, illustrate the functional studies of the HSD17B1 RARE by reporter gene analyses. The first of each of the adjacent bars represents reporter gene expression in the absence of at-RA and the second one. the reporter gene expression in the presence of 1.0 μM at-RA. In panel A of Figure 5. the effect of at-RA administration on HSD17B1 enhancer activity in JEG-3 and T-47D cells is shown. The results represent mean ± range from two independent experiments in JEG-3 cells and mean ± SEM from three independent experiments in T-47D cells. In each experiment, duplicate samples were analyzed. In panel B of Figure 5 the effect of mutations in the HSD17B1 RARE on retinoic acid induction in T-47D cells is shown. pBL(-661/-392)CAT2 and pBL(-392/-661)CAT2 represent intact constructs and pBLm3(-661/-392)CAT2 and pBLm3(-392/-661)CAT2 mutated constructs The results represent mean ± SEM from three independent experiments, in each of which duplicate samples were analyzed. Figure 6 illustrates the reporter gene analysis of the HSD17B1 promoter fragments -859/+9, - 113/+9 and -78/+9 linked to pCAT-EY vectors. In each cell line, the first bar shows basal reporter gene expression of the pCAT-EY vector and the second, third and fourth bars show the promoter activities of the fragments -859/+9. -113/+9 and -78/+9 with SV40 enhancer, respectively. The results represent mean ± range from two independent experiments, in each of which duplicate samples were analyzed. The results from different cell lines are not directly comparable with each other because the expression of the reference reporter gene, the ß-galactosidase, varies from one cell line to another.

Figure 7 illustrates the localization of a silencer element in the HSD17B1 gene Bar 1 shows the basal activity of pCAT-B Y vector and bar 2 the promoter activity of the fragment -659/+9. Bars 3 to 5 depict the activities of proximal promoter fragments -228/+9, -113/+9 and -78/+9 when linked to HSD17B1 enhancer, respectively. The results represent the mean ± SEM from three independent experiments in each of which duplicate samples were analyzed. Figures 8A and 8B illustrate the effect of at-RA administration on CAT expression. In particular, Figure 8A illustrates the effect of at-RA administration on CAT expression by the constructs pCAT-BY-659 and pCAT-BY-En-78. Construct pCAT-BT-659 contains the intact HSD17B1 promoter fragment -659/+9, whereas in construct pCAT-BY-En-78 the HSD17B1 enhancer fragment -661/-392 has been linked to proximal promoter fragment -78/+9. The results represent the mean ± range from two independent experiments in each of which duplicate samples were analyzed. Figure 8B illustrates the effect of at-RA administration on CAT expression by constructs pCAT-BY-659, pCAT-BY-En-228, pCAT-BY-En- 113 and pCAT-BY-En-78. Construct pCAT-BY-659 contains the intact HSD17B1 promoter fragment -659/+9, whereas in constructs pCAT-BY-En-228, pCAT-BY-En- 113 and pCAT-BY- En-78 the HSD17B1 enhancer fragment -661/-392 has been linked to proximal promoter fragments -228/+9. - 113/+9 and -78/+9 respectively. The results represent the mean ± range from two independent experiments in each of which duplicate samples were analyzed. Figure 9, panels A, B and C, illustrate the interaction between the labelled

HSD17B1 promoter fragment (from -69 to -36. named as HSD-AP-2/Sp1 ) and nuclear extracts from various ceil lines. In panel A of Figure 9, lanes 1 to 3 show binding between the HSD17B1 promoter fragment and nuclear extract from JAR, JEG-3 and T-47D cells, respectively. Lanes 4 and 5 depict binding between the fragment and HeLa nuclear extract which is known to be rich in Sp 1 factor, and commercial AP-2 protein extract. In panel B. lane 1 shows a free probe, and lane 2 the binding between the wt-fragment and nuclear extract from JAR-cells. As indicated in lane 3 to lane 9, unlabeiled ohgonucleotide competitors were included in the binding reaction at a 100-fold molar excess over probe. These competitors included wt-HSD17B1 promoter fragment (HSD-AP-2/Sp 1 ), HSD17B1 promoter fragment containing AP-2 binding site (HSD-AP-2), HSD17B1 promoter fragment containing mutated AP-2 binding site (HSD-mAP-2), consensus AP-2 binding site (AP-2 consensus), HSD17B1 promoter fragment containing Sp1 binding site (HSD-Sp1), HSD17B1 promoter fragment containing mutated Sp1 binding site (HSD-mSp1), and consensus Sp1 binding site (Sp 1 consensus) (for the whole sequences, see Table 1). The position of the complexes, are indicated by the arrows at the left. In panel C lane 1 shows the free probe and lanes 2 and 9 interaction of the labelled HSD17B1 promoter fragment HSD-AP-2/Sp 1 and nuclear extracts from JAR and JEG-3 cells, respectively. Lanes 3 to 5 (JAR) and 7 to 9 (JEG-3) demonstrate the formation of supershift complexes after administration of antibodies against AP-2, Sp1, and GATA-3 factors to the reaction mixture. The position of the complexes are indicated by the arrow at the left. Complex 1 , binding between AP-2 factor and the HSD17B1 promoter fragment; Complex 2, binding between Sp1 factor and the HSD17B1 promoter fragment; Supershift 1 , binding between AP-2 factor, AP-2 antibody and the HSD17B1 promoter fragment: Supershift 2, binding between Sp1 factor. Sp1 antibody and the HSD17B1 promoter fragment. Panel D of Figure 9 illustrates the reporter gene analysis of the HSD17B1 promoter fragment -97/+9 linked to pCAT-EY vector and of analogous constructs, in which AP-2 or Sp1 binding site has been mutated. In both, JAR and JEG-3, cell lines the first bar shows basal reporter gene expression of the pCAT-EY vector and the second, third, fourth and fifth bars show the promoter activities of the fragments -97/+9, -78/+9. AP-2 binding site mutated

-97/+9, and Sp1 -binding site mutated -97/+9, with SV40 enhancer, respectively. The results represent mean ± range from two independent experiments, in each of which duplicate samples were analyzed.

Figure 10 A illustrates interactions between the end-labeled sense strand of the HSD17B1 enhancer region (from -661 to -392) and nuclear extracts prepared from

JAR and JEG-3 choriocarcinoma cells and from T-47D and MCF-7 breast cancer cells as achieved using DNase I footprinting analysis ( for details, see Example 7). Bovine serum albumin (BSA) has been used for controlling unspecific binding, and the lanes marked G+A depict results of Maxam-Gilbert reactions. Positions of the protected nucleotides. which have been replaced with nonsense sequence in functional analyses ( see panel B) are shown with vertical ban on right Panel B of Figure 10 illustrates significance of the footprinted areas and some single nucleotides to the function of the HSD17B1 enhancer analyzed by reporter gene assays. Eight differently mutated HSD17B1 enhancer elements were generated either by replacing the footprinted areas with nonsense sequence [FP1(-), FP2(-) and FP(3-)] or by replacing the nucleotides at the positions -615, -486, -480, -452 and -435 with those located in the HSD17BP1 gene (tor details, see Example 7). The intact or mutated HSD17B1 enhancers were then linked adjacent to a thymidine kinase promoter-reporter gene (pBLCAT4) constructs, which were further transfected into JEG-3 cells. Reporter gene expression of each construct as compared to the one of intact fragment ( 100%) is represented by bars . The construct containing an intact HSD17B1 enhancer linked to thymidine kinase promoter is marked as [pBL( -66 l/-392)-CAT4]. Markings of the other constructs depict the mutations introduced to the enhancer element.

Figure 11, panels A and B, illustrate the interaction between the labelled HSD17B1 silencer fragment (from - 114 to -77. named here as HSD-GATA ) and nuclear extracts from various cell lines. In panel A of Figure 11, lane 1 shows a free probe, lane 2 binding between the HSD17B1 silencer fragment and nuclear extract from JEG-3 cells. As indicated in lane 3 to lane 6, unlabelled oligonucleotide competitors were included in the binding reaction at a 100-fold molar excess over probe. These competitors included wt-HSD17B 1 silencer fragment (HSD-GATA), HSD17B1 silencer fragment containing mutated GATA-binding site (HSD-mGATA), consensus GATA binding site (GATA consensus), and ßRARE01 as an unspecific competitor (for the whole sequences, see Table 1). The position of the complexes, are indicated by the arrows at the left. In panel B. in case of each cell line the first lane, lanes 1, 4, 7 and 10 show interaction of the labelled HSD17B1 silencer fragment. HSD-GATA, and nuclear extracts from JEG-3, JAR, T-47D and BT-20 cells, respectively. The second and third lanes in each case, lanes 2 and 3 (JEG-3), 5 and 6 (JAR). 8 and 9 (T-47D), and 11 and 12 (BT-20) demonstrate formation of supershift complexes after administration of antibodies against GATA-2 and GATA-3 factors to the reaction mixture. The position of the complex is indicated by the arrow at the left Complex 1, binding between GATA-3 and the HSD17B1 promoter fragment. Complex 2, binding between GATA-2 and the HSD17B1 promoter fragment. Supershift 1, binding between GATA-3. antibody against GATA-3 and the HSD17B1 promoter fragment; Supershift 2, binding between GATA-2, antibody against GATA- 2 and the HSD17B1 promoter fragment Panel C of Figure 11 illustrates the functional analysis of the silencer element by reporter gene analysis. In both. JAR and JEG-3 cell lines the first bar shows basal reporter gene expression of the pCAT-EY vector and the second and third bars show the promoter activities of the fragments - 113/+9 and -78/+9, respectively. Fourth and fifth bars demonstrate the fragments - 113/+9, in which binding site for GATA-factors has been mutated. The results represent mean ± range from two independent experiments, in each of which duplicate samples were analyzed.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides isolated nucleic acid molecules capable of regulating transcription of the 1.3 kb 17ß-hydroxysteroid dehydrogenase ( 17HSD) type 1 mRNA. In particular, isolated nucleic acid molecules coding an HSD17B1 promoter, enhancer and silencer are provided. Further provided are methods for screening tram-acting agents capable of affecting HSD17B1 gene expression

The DNA sequence of the 5'-region of the HSD17B1 gene is provided below. The translation initiation codon, transcription start site, the enhancer region with the retinoic acid response element (RARE) and the silencer region with the essential parts for the silencer function are indicated. In addition, binding sites for GATA, Sp1 and AP-2 transcription factors have been marked. The promoter, enhancer and silencer regions and their use in screening assay are discussed in detail below.

-

r

The region from -78 to +9 in the HSD17B1 gene, with respect to the cap site for the 1.3 kb mRNA was found by the present inventors to be a target for the transcription initiation machinery needed for the expression of the 1.3 kb transcript.

When located upstream a reporter gene, the present inventors have discovered that the

-78 to +9 region in the HSD17B1 gene is, by itself ( i.e., without the presence of additional promoter elements), capable of promoting transcription of the 1.3 kb 17HSD type l mRNA and therefore is able to act as a basal promoter. The present inventors have characterized a binding site for a Sp1 transcription factor, binding of it is needed for the entire function of the HSD17B 1 promoter. As described itemized in Example 6 and Figure 9D below, mutation of the binding site for Spl led to absence of Sp1 binding and decreased reporter gene expression in JEG-3 and JAR choriocarcinoma cells. On the other hand, mutation of the AP-2 binding site increased reporter gene expression in JEG-3 and JAR cells to 260% and 120% respectively of that of nonmutated reporter gene construct, as shown in the Figure 9D.

By the invention, the present inventors have isolated an HSD17B1 enhancer and silencer which affect basal promoter function. In particular, as described in detail in Example 1 below, recombinant DNA constructs were prepared for reporter gene analyses to localize the enhancer and silencer regions.

The HSD17B1 enhancer was localized within the bases -661 to -392 as shown in Figure 2A. The inventors have discovered that the enhancer increases, in both orientations, promoter activity more than 200-fold in certain cell lines. Deletions from either the 5'-region from -659 to -550 (Figure 1 ) or from the 3'-region from -392 to -549 (Figure 2A) of the enhancer abolished its activity, which indicates the importance of both of the halves for enhancer function. The activity of fragment -764A549 was minor compared with that of fragments -764/-392 and -661/-392, which demonstrates that the region from -659 to -550 cannot function alone as an enhancer. The HSD17B1 enhancer increased transcription efficiency in connection with its own promoter or with thymidine kinase and SV40 promoters. Accordingly, the function of the

HSD17B1 enhancer is not dependent on promoter type. Moreover, the enhancer increases transcription whether linked adjacent to a promoter or when placed upstream at a distance of 2 kb or more. Indeed, the present inventors have demonstrated that the

HSD17B1 enhancer increases reporter gene expression more than 20-fold in spite of a distance 2 kb from the promoter. Authenticity of the HSD17B1 enhancer was further supported by results which showed that the enhancer was not functional in cell lines not expressing the 17HSD type 1 protein. Thus, by the invention, an isolated HSD17B1 enhancer is provided which is localized within bases -661 and -392 of the sequence disclosed on page 10. In particular, the present invention is directed to isolated nucleic acid molecules at least 16 bases in length which are capable of enhancing gene transcription and have a nucleotide sequence that is contained within (and can include) nucleotides -661 and -392 of the sequence disclosed on page 10. Specific isolated nucleotide molecules of the present invention include, but are not limited to. the following with the 5' and 3' ends indicated in parenthesis (see the sequence disclosed on page 10): a 270 base nucleic acid molecule (5' end: -661; 3' end:-392); a 268 base nucleic acid molecule (5' end: -659; and 3' end: -392); a 200 base nucleic acid molecule (5' end: -661 and 3' end:-462); a 200 base nucleic acid molecule (5' end: -591 and 3' end:-392), a 50 base nucleic acid molecule (5' end: -574 and 3' end:-525). Preferred isolated nucleic acid molecules of the present invention which are capable of enhancing gene transcription will also include the retinoic acid response element at -503 to -487 (discussed below ) and will thus be at least 88 nucleotides in length. It will be understood in the art that the present invention is further directed to all other isolated nucleic acid molecules at least 16, preferably at least 30, 50, 88, 100, 125, 150, 175 or 200 nucleotides in length and which fall within (and can include) the nucleotides at positions -661 and -392 of the sequence disclosed on page 10. From the examples below, it will be understood in the art that the nucleic acid molecules of the present invention can enhance transcription in either orientation (See Figure 2A). The inventors have further discovered that 5' nucleotides extending up to -784 (see the sequence disclosed on page 10) can be included without having a significant detrimental effect on

transcriptional enhancement. Thus, further nucleic acid molecules of the present invention include those which extend up to (and can include) -784 as shown in the sequence disclosed on page 10, provided that such molecules also meet the constraints indicated above.

The HSD17B 1 enhancer contains at least three areas bound by protein or proteins as explained in Example 7 and Figure 10 A B below. Replacing two of the regions, from -544 to -528 and from -589 to -571 with nonsense sequence, results in greatly decreased enhancer activity pointing to the importance of these regions in enhancer function (Figure 10B). Double mutation of the nucleotides at the positions

-480 and -486 also leads to significantly reduced enhancer function, indicating that the nucleotides are critical for enhancer activity. Because the single mutation of the nucleotide at the position -480 or substitution the area from -495 to -485 with a nonsense sequence lessens the function of the enhancer only to some extent, simultaneous mutation of the positions -480 and -486 may be needed to abolish the enhancer activity. Finally, replacement of the nucleotide at the positions -435 results in decreased, and replacement of the nucleotide -452 in increased reporter gene expression, suggesting that these nucleotides are involved in binding of regulatory factors modulating the enhancer function. It has been discovered by the present inventors that the nucleic acid molecules of the present invention which enhance transcription also include a retinoic acid response element (RARE) (see -503 to -487 in the sequence disclosed on page 10). The activity of the HSD17B1 enhancer was induced by all-trans- retinoic acid (at -RA) in JEG-3 and T-47D cells. It has previously been shown that at -RA induces 1.3 kb 17HSD type 1 mRNA expression in T-47D cells in a time- and dose-dependent manner (Reed, M.J. et al., Endocrinology 135:4-9 ( 1994)). T-47D cells contain RA receptors RAR_{α, -β} and _-γ, which form heterodimers with retinoid X receptors RXR_{α, -β} and _-γ after activation by the ligands at - or 9-cis - RAs. The hall sites of receptors RAR_{α, -β} and _-γ, which form heterodimers with retinoid X receptors RXR_{α, -β} and _-γ after activation by the ligands at - or 9-cis - RAs. The half sites of the RARE in the HSD17B1 enhancer shown in the sequence disclosed on page 10 are separated by five nucleotides, which is typical for the binding sites of the RAR/RXR heterodimers. In the present invention it was demonstrated that the double-stranded oligonucleotide bound the complex from RAR _α/RXR_α extract similarly to the reference oligonucleotide which contained RARE of the RAR_ß gene (Figure 4B) An excess of unlabelled intact HSD17B1 oligonucleotide competed efficiently with the labelled oligonucleotide for binding to the complex (Figures 4A and 4B) Introduction of mutation into the right half-site or simultaneously to both half-sites of the RARE abolished its capability to compete for binding (Figure 4A). On the other hand, mutation of the left half-site removed the binding capacity only partially. Taken together, the present data show that the right half-site of the HSD17B1 RARE, which consists of a perfect RARE consensus sequence, may bind the RARrχ/RXRrχ complex more efficiently than the left site with one mismatch and that the mutations introduced were more crucial for the right half-site of the RARE.

The DNA sequence of the enhancer region (-661 to -392) is set forth herein below (SEQ ID NO 2). The retinoic acid response element and elements and nucleotides whose change modulate enhancer activity, are indicated

The HSD17B1 silencer was localized within bases -391 to -78 as shown in Figure 7. To test their hypothesis concerning the existence of a silencer within bases -391 to -78, the present inventors fused the enhancer element (-661 to -392) in front of the proximal promoter fragments -228/+9, - 113/+9 and -78/+9 in pCAT-B Y vectors Reporter gene analysis in JEG-3 cells showed that CAT expression generated from the construct pCAT-BY-En-78 was 4-fold greater than that from the intact promoter approximately 20-fold increase in transcription efficiency in T-47 cells. Thus, the inventors have discovered that the region from -391 to -78 contains a silencer and that the region from - 113 to -78 is essential for silencer function. In connection with the SV40 enhancer, shortening of the HSD17B1 promoter from - 113 to -78 led to significant increase in reporter gene expression in the tested breast, prostate and choriocarcinoma cell lines as well as in monkey kidney cell line. On the other hand, shortening of the HSD17B1 promoter did not increase reporter gene transcription significantly in human choriocarcinoma. JAR, the human prostate carcinoma, PC-3, and the monkey kidney. CV-1, cell lines when the fragments were not linked to any enhancer. Hence, the silencer appears mainly to counteract the HSD17B1 and SV40 enhancers, not the HSD17B1 basal promoter.

The inventors have characterized a binding site for GATA-factors at the position from - 102 to -99 in the silencer. Double-stranded oligonucleotide 38 base pan in size responding to the element from 114 to -77 is capable to bind proteins, which are recognized by antibodies against GATA-2 or GATA-3 regulatory factor (Figure 11B) Mutation of the nucleotides -99 and - 100 simultaneously abolish the binding partially (Figure 11 A) and mutations at the positions -99 and - 100 or - 101 and - 102 increase reporter gene expression about two fold in JEG-3 cells and slightly in JAR cells (Figure 11 C). In connection with SV40 enhancer, mutated HSD17B1 gene fragment from - 113 to +9 drives transcription at least as efficiently as the fragment from -78 to +9, while intact - 113/+9 -fragment leads to significantly lower reporter gene expression (Figure 11C). These results strongly suggest that GATA-factors are involved in modulating HSD17B1 gene expression.

Thus, by the invention, an isolated HSD17B1 silencer is provided which is localized within bases -391 to -78 of the sequence disclosed on page 10 In particular. the present invention is directed to isolated nucleic acid molecules at least 36 bases in length which are capable of repressing gene transcription, wherein said nucleic acid molecules include nucleotides - 113 to -78 in the sequence disclosed on page 10 and have a nucleotide sequence which is contained within (and can include) nucleotides -391 and -78 of the sequence disclosed on page 10. Specific isolated nucleotide molecules of the present invention include, but are not limited to. the following with the 5' and 3' ends indicated in parenthesis ( see the sequence disclosed on page 10) a 314 base nucleotide molecule (5' end - 391. 3' end -78). and a 36 base nucleotide molecule (5' end - 113, 3' end -78). Preferably, in addition to nucleotides from - 113 to -78, the nucleic acid molecules of the present invention which are capable of repressing gene transcription will also include at least a part of the region from - 114 to -391. This is because the present inventors have discovered that including region - 114 to -391 can increase the repressing effect of the silencer. It will be understood in the art that the present invention is further directed to all other nucleic acid molecules at least 36. more preferably at least 50, 75, 100, 125, 150, 175 or 200 nucleotides in length which include -113 to -78 in the sequence disclosed on page 10 and which fall within ( and can include) the nucleotides -391 and -78 in the sequence disclosed on page 10.

The DNA sequence of the silencer region is set forth below. The essential part of the sequence for the silencer function and consensus binding site for GATA-factors are indicated.

invention can simply be read from the sequence disclosed on page 10 following the constraints indicated above. Of course, since the nucleotide sequence is known, routine methods are available for producing such nucleic acid molecules synthetically

(see, for example, Synthesis and Application of DNA and RNA, S A Narang, ed., 1987, Academic Press, San Diego, CA). Alternatively, such isolated nucleic acid molecules of the present invention can be generated as follows. The HSD17B1 gene promoter region from -859 to +9 is obtained by amplification using the polymerase chain reaction (PCR). The amplified fragment is then inserted into an appropriate plasmid (such as, for example pCAT™ (Promega, Madison, WI)). Nested deletion plasmids are then generated using the commercially available "Erase-a-Base" System (Promega, Madison, WI) as described in Henikoff (Gene 28 :351-359, ( 1984)). (See Example 1 below for further details.) Thus, only routine experimentation would be required to generate any of the isolated nucleic acid molecules of the present invention which are capable of enhancing or silencing gene expression.

As indicated, the nucleic acid molecules of the present invention include promoter and cis-acting enhancer and /or silencer elements capable of affecting gene transcription For simplicity, these isolated nucleic acid molecules of the present invention are referred to below as "HSD17B1 transcriptional regulatory elements or transcriptional elements". As indicated, to determine the effect of a transcriptional element of the present invention on gene expression, nested deletion reporter plasmids can be generated containing a transcriptional element of the present invention linked in front of the chloramphenicol acetyltransferase (CAT) reporter gene. Such recombinant molecules of the present invention actually generated by the inventors include transcriptional elements inserted, in both orientations, into the XbaI site of pBLCAT2 vector (Luckow, B., Schutz, G., Nucleic Acids Res. 15:5490 ( 1987)). As a result. pBL(-859/-637)CAT2, pBL(-637/-859)CAT2, pBL(-859/-549)CAT2, pBL(-549/-859)CAT2, pBL(-859/-392)CAT2, pBL(-392/-859)CAT2, pBL(-859/-113)CAT2, pBL(-113/-859)CAT2, pBL(-764/-549)CAT2, pBL(-549/-764)CAT2, pBL(-764/-392)CAT2, pBL(-392/-764)CAT2, pBL(-661/-392)CAT2 and pBL(-392/-661)CAT2 were obtained. Further, pBLCAT2 vector was modified using NdeI and EcoO 109 digestion (Jonat C et al., Cell 62:1189- 1204 ( 1990)) to remove AP- 1 binding site from the vector and to the modified form, pBLCAT4, the intact and mutated -661/-392 fragments were linked into HindIII and XbaI sites. As a results pBL(-661/-392)CAT4, pBL(-661/-392)CAT4-FP1(-),pBL(-661/-392)CAT4-FP2(-),pBL(-661/-392)CAT4-FP3(-), pBL(-661/-392)CAT4-( -615T->C), pBL(-661/-392)CAT4-(-480C->G), pBL(-661/-392)CAT4-( -480C->G,-486G->A), pBL(-661/-392)CAT4-(-452A->G), and pBL(-661/-392)CAT4-(-435T->A) were obtained. Markings of the single or double mutated constructs depict the mutations introduced to the enhancer element. In the constructs pBL(-661/-392)CAT4-FP1 (-), -FP(-), and -FP3(-), the regions from -495 to -485, from -544 to -528 and from -589 to -571, respectively, were replaced with the sequence (GGGTTTCCCAAA)_n.

In addition, the PCR-amplified fragment -764/-392 was inserted into the XbaI site of pCAT™-Promoter vector (Promega, Madison, WI) in both orientations. Thus, the constructs pCAT-P-(-764/-392) and pCAT-P-(-392/-764) were created. Further. the reporter gene piasmids containing the HSD17B1 enhancer linked to the proximal promoter area were constructed by inserting the PCR-amplified fragment -661/-392 with a HmdIII site at the 5'-end and a SαcII site at the 3'-end into pCAT-BY-228, pCAT-BY- 113 and pCAT-BY-78. The resulting constructs were named pCAT-BY- En-228, pCAT-BY-En-113 and pCAT-BY-En-78. Analogous constructs, pCAT-EY-228, pCAT-EY- 113 pCAT-EY-97 and pCAT-EY-78 were also generated. Those constructs contained, in addition to pCAT-BY backbone, an SV40 enhancer element. Finally, pCAT-EY-113-m1GATA and pCAT-EY-113-m2GATA were created by introducing mutations from TA to AT to the nucleotides - 102 and - 101 and from TC to AG to the nucleotides - 100 and -99 pCAT-EY-97-mSp1 and pCAT-EY-97-mAP-2 were generated by replacing the two Gs at the positions -44 and -45 with TT, and the two Cs at the positions -60 and -61 with TT. respectively.

By the invention, a recombinant DNA molecule containing a transcriptional element of the present invention is used to transiently transfect an appropriate cell line such as, for example, human choriocarcinoma cell lines (JEG-3 and JAR ), the human prostate carcinoma cell line PC-3, or the monkey kidney cell line CV- 1, all of which are available from the American Type Culture Collection, or human granulosa cells In addition to using the CAT system for reporter gene analyses, the hGH transient expression system can also be used (Selden et al., Mol. Cell Biol.6:3173-3179

( 1986)) or other systems that are based on the expression of ß-galactosidase (An et al , Mol. Cell Biol. 2:1628- 1632 ( 1982)) and xanthine-guanine phophoribosyl transferase (Chu et al. Nucleic Acids Res. 13.2921-2930 (1985)).

A transcriptional element of the present invention may be inserted into an appropriate vector in accordance with conventional techniques, including blunt-ending or staggered-ending termini for ligation. restriction enzyme digestion to provide appropriate termini, filling in of cohesive ends as appropriate, alkaline phosphatase treatment to avoid undesirable joining, and ligation with appropriate ligases.

Techniques for such manipulations are disclosed by Maniatis, T., et al., infra, and are well known in the art. Clones containing a transcriptional element of the present invention may be identified by any means which specifically selects for HSD17B1 enhancer or silencer region DNA such as. for example by hybridization with an appropriate nucleic acid probe(s) containing a sequence complementary to all or part of the transcriptional element. Oligonucleotide probes specific for a transcriptional element of the present invention can be designed simply by reference to the sequence disclosed on page 10 above. Techniques for nucleic acid hybridization and clone identification are disclosed by Maniatis, T., et al., (In. Molecular Cloning, A

Laboratory Manual, Cold Spring Harbor Laboratones, Cold Spring Harbor, NY ( 1982)) and by Hames, B.D., et al., (In Nucleic Acid Hybridization, A Practical Approach, IRL Press, Washington DC ( 1985)). To facilitate the detection of the desired clone containing a transcriptional element of the present invention, the above-described nucleic acid probe may be labelled with a detectable group. Such detectable groups can be any material having a detectable physical or chemical property. Such materials have been well-developed in the field of nucleic acid hybridization and in general most any label useful in such methods can be applied to the present invention.

Particularly useful are radioactive labels, such as ³²P, ³H, ¹⁴C, ³⁵S, ¹²⁵ I, or the like. Any radioactive label may be employed which provides for an adequate signal and has a sufficient half-life. The oligonucleotide may be radioactively labeled, for example, by "nick-translation" by well-known means, as described in, for example, Rigby, P.J.W., et al., J. Mol. Biol. 113: 237 ( 1977) and by T4 DNA polymerase replacement synthesis as described in, for example. Deen. K.C., et al., Anal Biochem 135:456 ( 1983). Alternatively, polynucieotides are also useful as nucleic acid hybridization probes when labelled with a non-radioactive marker such as biotin. an enzyme or a fluorescent group. See, for example. Leary. J.J ., et al., Proc Natl. Acad. Sci. USA 80:4045 ( 1983), Renz, M., et al., Nucl. Acids Res 12:3435 ( 1984), and Renz, M., EMBO J. 6:817 ( 1983).

As used herein, "heterologous protein" is intended to refer to a peptide sequence that is heterologous to the transcriptional regulatory elements of the invention. A skilled artisan will recognize that, if desired, the teaching herein will also apply to the expression of genetic sequences encoding the 17HSD type 1 protein by such transcriptional regulatory elements. The reporter genes for use in the screening assay described below can code for either the 17HSD type 1 protein or a heterologous protein. Alternatively, detection of reporter gene expression can be at the mRNA level, such as, for example, detection of the 1 .3 kb 17HSD type 1 mRNA.

To express a reporter gene under the control of the transcriptional regulatory elements of the invention, the gene must be "operably-linked" to the regulatory element. An operable linkage is a linkage in which a desired sequence is connected to a transcriptional or translational regulatory sequence (or sequences) in such a way as to place expression (or operation) of the desired sequence under the influence or control of the regulatory sequence.

Two DNA sequences (such as a reporter gene and a promoter region sequence linked to the 5' end of the reporter gene) are said to be operably linked if induction of promoter function results in the transcription of the reporter gene and if the nature of the linkage between the two DNA sequences does not ( 1) result in the introduction of a frame-shift mutation (if reporter protein activity is necessary for detection of reporter gene expression), (2) interfere with the ability of the expression regulatory sequences to direct reporter gene expression, or (3) interfere with the ability of reporter gene to be transcribed by the promoter region sequence. Thus, a promoter would be operably linked to a DNA sequence if the promoter were capable of affecting transcription of that DNA sequence.

In a similar manner, a transcriptional regulatory element of the present invention that enhances or represses gene expression may be operably-linked to such a promoter. Exact placement of the element in the nucleotide chain is not critical as long as the element is located at a position from which the desired effects on the operably linked promoter may be revealed. A nucleic acid molecule, such as DNA. is said to be "capable of expressing" a polypeptide if it contains expression control sequences which contain transcriptional regulatory information and such sequences are operable linked to the nucleotide sequence which encodes the polypeptide. For the complete control of gene expression, all transcriptional and translational regulatory elements (or signals) that are operably linked to a heterologous gene should be recognizable by the appropriate host By "recognizable" in a host is meant that such signals are functional in such host.

The HSD17B1 transcriptional regulatory elements of the present invention. obtained through the methods described above, and preferably in a double-stranded form, may be operably linked to a heterologous gene (such as a reporter gene), preferably in an expression vector, and introduced into a host cell, preferably a eukaryote cell, to assay reporter gene expression. Preferred eukaryotic cells include chonocarcinoma cell lines, granulosa cells, breast cancer cell lines, prostate carcinoma cell lines and kidney cell lines.

As indicated, the enhancer and silencer elements of the present invention are not promoter specific. Thus, the enhancer and silencer transcriptional elements of the invention may be operably linked to any promoter that is functional in the desired host cell. Similarly, the promoter of the invention may be linked to any other regulatory elements. A wide variety of transcriptional and translational regulatory sequences can be employed, operable linked to a transcriptional element of the invention, depending upon the nature of the eukaryotic host. In eukaryotes. where transcription is not linked to translation, such control regions may or may not provide an initiator methionine (AUG) codon. depending on whether the operably linked heterologous sequence contains such methionine. Such regions will, in general, include a promoter region sufficient to direct the initiation of RNA synthesis in the host cell. Promoters from heterologous mammalian genes that encode an mRNA product capable of translation are preferred, and especially, strong promoters such as the promoter for actm, collagen, myosin, etc., can be employed provided they also function as promoters in the host cell, and provided that their function is also capable of being control by the desired positive or suppressor of the invention.

As is widely known, translation of eukaryotic mRNA is initiated at the codon that encodes the first methionine. For this reason, it is preferable to ensure that the linkage between a eukaryotic promoter and a reporter gene does not contain any intervening codons that are capable of encoding methionine. The presence of such codons results either in a formation of a fusion protein (if the AUG codon is in the same reading frame as the DNA encoding the heterologous protein) or a frame-shift mutation (if the AUG codon is not in the same reading frame as the reporter gene).

If desired, a fusion product of a reporter protein may be constructed. For example, the sequence coding for the reporter protein may be linked to a signal sequence which will allow secretion of the protein from, or the compartmentahzation of the protein in, a particular host Such signal sequences may be designed with or without specific protease sites such that the signal peptide sequence is amenable to subsequent removal Alternatively, the native signal sequence for this protein may be used.

The transcriptional regulatory elements of the invention can be selected to allow for repression or activation, so that expression of the operably linked reporter genes can be modulated. Translational signals are not necessary when it is desired to express antisense RNA sequences or to assay reporter gene expression via mRNA detection.

If desired, the non-transcribed and/or non-translated regions 3' to the reporter gene can be obtained by the above-described cloning methods. The 3'-non-transcribed region may be retained for its transcriptional termination regulatory sequence elements; the 3'-non-transiated region may be retained for its translational termination regulatory sequence elements, or for those elements that direct polyadenylation in eukaryotic cells. Where the native expression control sequences signals do not function satisfactorily host cell, then sequences functional in the host cell may be substituted.

To transform a mammalian cell with the DNA construct of the invention many vector systems are available, depending upon whether it is desired to insert the reporter gene product into the host cell chromosomal DNA, or to allow it to exist in an extrachromosomal form. If the reporter gene and an operably linked promoter are introduced into a recipient eukaryotic cell as a non-replicating DNA (or RNA) molecule, which may either be a linear molecule or, more preferably, a closed covalent circular molecule that is incapable of autonomous replication, reporter gene expression may occur through the transient expression of the introduced sequence.

Genetically stable transformants may be constructed with vector systems, or transformation systems, whereby the reporter gene is intergrated into the host chromosome. Such integration may occur de novo within the cell or, in almost preferred embodiment, be assisted by transformation with a vector that functionally inserts itself into the host chromosome. Vectors capable of chromosomal insertion include, for example, retroviral vectors, transposons or other DNA elements which promote integration of DNA sequences in chromosomes, especially DNA sequence homologous to a desired chromosomal insertion site.

Cells that have stably integrated the introduced DNA into their chromosomes are selected by also introducing one or more markers that allow for selection of host cells which that the desired sequence. For example, the marker may provide biocide resistance, e.g., resistance to antibiotics, or heavy metals, such as copper, or the like. The selectable marker gene can either be directly linked to the reporter gene, or introduced into the same cell by co- transfection. In another embodiment, the introduced sequence is incorporated into a plasmid or viral vector capable of autonomous replication in the recipient host Any of a wide variety of vectors may be employed for this purpose, as outlined below. Factors of importance in selecting a particular plasmid or viral vector include: the ease with which recipient cells that contain the vector may be recognized and selected from those recipient cell which do not contain the vector; the number of copies of the vector which are desired in a particular host; and whether it is desirable to be able to "shuttle" the vector between host cells of different species.

Preferred eukaryotic piasmids include those derived from the bovine papilioma virus, vaccinia virus and SV40. Such plasmids are well known in the art and are commonly or commercially available. For example, mammalian expression vector systems in which it is possible to cotransfect with a helper virus to amplify plasmid copy number, and. integrate the plasmid into the chromosomes of host cells have been described (Perkins. A.S. et al., Mol Cell Biol. 3: 1123 ( 1983), Clontech, Palo Alto, California). Particularly preferred are vectors derived from pCAT-Basic, pCAT-Enhancer, and pCAT-Promoter vectors (Promega, Madison, WI).

Once the vector or DNA sequence containing the construct(s) is prepared for expression, the DNA construct(s) is introduced into an appropriate host cell by any of a variety of suitable means, including transfection, electroporation or delivery by liposomes. DEAE dextrin, calcium phosphate, and preferably, the transfection reagent DOTAP, may be useful in the transfection protocol.

After the introduction of the vector in vitro , recipient cells are grown in a selective medium, that is. medium that selects for the growth of vector-containing cells. Expression of the reporter gene results in the production mRNA and. if desired, reporter protein. According to the invention, this expression can take place in a continuous manner in the transformed ceils, or in a controlled manner. If desired, in in vitro culture, the reporter protein is isolated and punfied in accordance with conventional conditions, such as extraction, precipitation, chromatography, affinity chromatography, electrophoresis, or the like Alternatively, levels of reporter protein expression can be assayed according to conventional protein assays, such as, for example, the CAT expression system.

The HSD17B1 transcriptional regulatory elements of the present invention (i.e., isolated nucleic acid molecules capable of promoting, enhancing and/or repressing gene expression) are useful for screening drugs, ligands, and/or other trans-acting agents to determine which are capable of affecting estrogen production. Human 17B-hydroxysteroid dehydrogenase type 1 (17HSD type 1) catalyzes the reductive reaction of estrone to the biologically more active form, estradiol The 17HSD type 1 enzyme is highly expressed in the human placenta and the ovary and. in addition, in certain estrogen target cells, such as breast epithelial cells By the invention, trans-acting factors can be identified by their ability to up-regulate or down-regulate 17HSD type 1 expression thereby leading to an increased or decreased production of estradiol from estrone. As used herein, by "HSD17B1 trans-acting agent" is intended a drug, ligand, or other compound capable interacting, either directly or indirectly, with a HSD17B1 transcriptional regulatory elements of the present invention to enhance or repress gene expression. Such HSD17B1 trans-acting elements which interact directly with a transcriptional regulatory element of the present invention include those, which, for example, bind directly to the element and either enhance or repress gene expression HSD17B1 trans-acting agents which interact indirectly with a transcriptional regulatory element of the present invention include those which, for example, bind to and induce activity of a second trans-acting agent (e.g., a receptor molecule) which itself then, either alone or complexed to the first trans-acting agent, binds to the element and either enhances or represses gene expression.

An example of an HSD17B1 trans-acting agent which indirectly interacts with transcriptional regulatory elements of the present invention is all trans-retinoic acid (at-RA). As discussed in detail in Example 4 below, the present inventors have shown that at-RA induces 17HSD type 1 1.3-kb mRNA expression in breast cancer T-47D cells, leading to increased formation of estradiol from estrone. As indicated, the enhancer element of the present invention includes a retinoic acid response element (RARE) at -503 to -487. RARs and RXRs are ligand-inducible trans-regulators that control transcription initiated from the promoters of retinoic acid (RA) target genes by interacting with cis-acting RAREs. Using Electrophoretic Mobility Shift Assays (EMSAs), the inventors have shown that the double-stranded wt-RARE binds the hRXR_a/hRAR_a heterodimer complex in vitro. Naturally occurring retinoids which act as a "switch" by binding to the RAR and RXR ligand binding domains (LBDs) an activating RAR function include at-RA, 3,4-didehydroretinoic acid and 9-cis RA (Kastner and Chambon, TIBS 17:427-433 ( 1992)). Synthetic retinoids capable of activating RARs include Am80 (Hashimoto et al., Biochem. Biophys. Res. Commun.

166: 1300- 1397 ( 1990); CD417 (Bernard et al., Biochem. Biophys. Res. Commun. 186:977-983 ( 1992)); and CD666 (Bernard et al., Biochem. Biophys. Res. Commun. 186:977-983 ( 1992)). Others include BMS453 and BMS411 (Chen et al., EMBO J. 14 (6): 1187-1197 ( 1995)). Retinoids play a major role in a variety of developmental processes, cell differentiation and proliferation (Lotan, R., Biochim. Biophys. Acta

605:33-91 (1980)). The vitamin A-derived retinoids and their derivatives have been shown to inhibit carcinogenesis. including estrogen-dependent breast cancer

(Lippman. S.M. et al., Cancer Treat. Rep. 71:391-405 (1987)). Further retinoids have been shown to down-regulate estrogen receptor expression and strongly inhibit proliferation of estrogen-dependent breast cancer cell lines (van der Burg, B. et al.,

Mol. Cell Endrocrinol. 97: 149- 157 ( 1993); Rubin. M. et al., Cancer Res. 54:6549- 6556 ( 1994)). Hence, in the present invention it was unexpectedly found that at -RA up-regulates 17HSD type 1 expression, leading to increased formation of estradiol from estrone, which contrasts with the action of RAs as antiproliferative agents. Thus. in one aspect, the invention provides a screening assay for determining whether any given naturally occurring or synthetic retinoid is capable of up-regulating or down- regulating 17HSD type 1 expression, leading to an increase or decrease of estradiol production.

It will be understood in the art that the assay of the present invention can be used not only for screening retinoids. but any candidate HSD17B1 trans-acting agent to determine whether it affects HSD17B1 gene expression. The screening assay involves ( 1) providing a host cell tiansfected with a recombinant nucleic acid molecule containing an HSD17B1 transcriptional regulatory element of the present invention, a promoter, and a reporter gene, wherein the transcriptional element and promoter are operably linked to the reporter gene, (2) administering a candidate HSD17B1 trans-acting agent to the transfected host cell, and (3) determining the effect on reporter gene expression.

Suitable and preferred host cells, transfection methods, expression vectors. promoters, reporter genes, candidate HSD17B1 trans-acting agents, and amounts of administration for use in the screening method of the present invention are described above in detail and in the Examples below and will be known in the art Suitable and preferred HSD17B1 transcriptional regulatory elements include the isolated nucleic acid molecules described above in detail which are capable of enhancing or repressing gene expression. As indicated, using the above-described screening assay, the present inventors determined that a candidate HSD17B1 trans-acting agent. at-RA, when administered to JEG-3 (a human chonocarcinoma cell line) and T-47D (a human breast cancer cell line) cells, which were transfected with a recombinant plasmid containing the HSD17B1 transcriptional regulatory element -661 to -392 (see the sequence disclosed on page 10) in either orientation, a TK promoter and a reporter gene, increased the activity of the HSD17B1 enhancer 2- to 3-fold in JEG-3 cells and 3- to 4- fold in the T-47D cells (see Fig.5).

The following examples are merely intended to illustrate the present invention and not in any way limit its scope. EXAMPLE 1

Preparation of the plasmid constructs for reporter gene analyses

The expression plasmids pCAT-BY and pCAT-EY were derived from pCAT™-Basic and pCAT™-Enhancer vectors (Promega. Madison. WI), respectively, by replacing the original polylinker region HindIII - SphI - PstI - SalI - Acc I - XbaI with HindIII - SphI - PstI - NsiI - SaclI - XmaIII - NheI - SmaI - SacI - BglII - XbaI.

The HSD17B1 gene promoter region from -859 to + 9. in respect to the transcription start site for the 1.3 kb 17HSD type 1 mRNA, was obtained by polymerase chain reaction (PCR) amplification using primers to which recognition sites for SmaI and XbaI had been added. The amplified fragment was then inserted into pCAT-EY using the SmaI and XbaI sites, resulting in the construct pCAT-EY-859. All the nested deletion plasmids. pCAT-EY-659, pCAT-EY-550, pCAT-EY-381, pCAT-EY-228 pCAT-EY- 113. pCAT-EY-97 and pCAT-EY-78. were derived from this parent plasmid using the procedure developed by Henikoff (Gene 28:351-359. 1984) (Erase-a-Base System, Promega. Madison, WI). The analogous deletion series in the pCAT-BY-backbone was obtained by cutting off the deletion fragments from the pCAT-EY-backbone by Hind III and Xb a I followed by transfer into pCAT-BY.

Reporter gene plasmids used to localize the enhancer element of the HSD17B1 gene contained the fragments -859/-637, -859/-549. -859/-392, -859/-113, -764/-549. -764/-392 and -661/-392 linked to thymidine kinase promoter. The fragments, including the recognition sites for XbaI-SacII at the 5'-end and for SmaI-XbaI at the 3' -end, were amplified by PCR, after which they were inserted, in both orientations, into the XbaI site of pBLCAT2 vector (Luckow, B., Schütz, G., Nucleic Acids Res. 15:5490 ( 1987)) . As a result, pBL( -859/-637)CAT2, pBL(-637/-859)CAT2, pBL( -859/-549)CAT2 , pBL( -549/-859)CAT2, pBL( -859/-392)CAT2, pBL( -392/-859)CAT2, pBL(-859/-113 )CAT2, pBL( - 113/-859)CAT2, pBL(-764/-549)CAT2, pBL(-549/-764)CAT2, pBL(-764/-392)CAT2, pBL(-392/-764)CAT2, pBL(-661/-392)CAT2 and pBL(-392/-661)CAT2 were obtained.

In addition, the PCR-amplified fragment -764/-392 was inserted into the XbaI site of pCAT -Promoter vector (Promega, Madison, WI) in both orientations. Thus, the constructs pCAT-P-(-764/-392) and pCAT-P-(-392/-764) were created.

The reporter gene piasmids containing the HSD17B1 enhancer linked to the proximal promoter area were constructed by inserting the PCR-amplified fragment -661/-392 with a HindIII site at the 5 '-end and a SacII site at the 3'-end into pCAT-BY-228, pCAT-BY-113 and pCAT-BY-78. The resulting constructs were named pCAT-BY-En-228, pCAT-BY-En- 113 and pCAT-BY-En-78. The overlap-extension technique described by Ho et al. (Gene 77:51-59 ( 1989)) was used to introduce a mutation into the retinoic acid response element (AGGACAggagaAGGTCA -> ACGACTggagaAAGTCG) in the fragment -661/-392. The mutated fragment was inserted into the XbaI site of pBLCAT2 in both orientations and the two resulting constructs were called pBLm3(-661/-392)CAT2 and pBLm3(-392/-661)CAT2. Further, pBLCAT2 vector was modified using NdeI and EcoOW9 digestion

(Jonat. C. et al., Cell 62: 1189- 1204 ( 1990)) to remove AP- 1 binding site from the vector and to the modified form, pBLCAT4, the intact and mutated -661/-392 fragments were linked into HindIII and Xbal sites. As a results pBL(-661/- 392)CAT4, pBL(-661/-392)CAT4-FP1 (-),pBL(-661/-392)CAT4-FP2(-),pBL(-661/- 392)CAT4-FP3(-), pBL(-661/-392)CAT4-(-615T->C), pBL(-661/-392)CAT4-(- 480C->G), pBL(-661/-392)CAT4-(-480C->G.-486G->A), pBL(-661/-392)CAT4-(- 452A->G), and pBL(-661/-392)CAT4-( -435T->A) were obtained. Markings of the single or double mutated constructs depict the mutations introduced to the enhancer element using overlap extension methods as described above In the constructs pBL(-661/-392)CAT4-FP1(-), -FP(-), and -FP3(-), the regions from -495 to -485, from -544 to -528 and from -589 to -571. respectively, were replaced with the sequence (GGGTTTCCCAAA)_n. Finally, pCAT-EY- 1 13-m l GATA and pCAT-EY- 113- m2GATA were created by introducing mutations from TA to AT to the nucleotides - 102 and - 101 and from TC to AG to the nucleotides - 100 and -99 pCAT-EY-97-mSp1 and pCAT-EY-97-mAP-2 were generated by replacing the two Gs at the positions -44 and -45 with TT, and the two Cs at the positions -60 and -61 with TT, respectively.

The plasmid constructs were transformed in JM 109 and DH 5a bacterial cell lines and the plasmids were purified by use of a kit from QIAGEN Inc. (Chatsworth, CA) according to the instructions of the manufacturer. The constructs pCAT-EY-859. pBL(-661/-392)CAT2, pBL(-392/-661)CAT2, pCAT-BY-En-228, pCAT-BY-En- 113, pCAT-BY-En-78, pBLm3(-661/-392)CAT2 and pBLm3(-392/-661)CAT2 and all the other constructs containing mutations what so ever, were verified by sequencing, using either Sequenase (United States Biochemical Corp., Cleveland, OH) or a Sequencing Kit (Pharmacia Biotech. Sweden) according to the instructions of the manufacturer. At least two individual preparations of each plasmid were mixed together for transfection experiments. EXAMPLE 2

Progressive 5'-end deletion analysis of the HSD17B1 promoter and localization of an

HSD17B1 enhancer

In order to characterize the function of the HSD17B1 gene promoter, a series of 5'- deletion mutants from the HSD17B1 promoter were linked in front of the chloramphenicol acetyl transferase (CAT) reporter gene, resulting in the constructs from pCAT-BY-859 to pCAT-BY-78 as described in Example 1. The longest fragment contained, in respect of the transcription start point for the 1.3 kb 17HSD type 1 mRNA, the region from -859 to 4-9. which is the last nucleotide before the translation initiation codon. The contracts were then transiently transfected into a transformed human placental cytotrophoblast cell line JEG-3 and into a

choriocarcinoma cell line JAR, which both express 17HSD type 1.

Human chonocarcinoma cell lines (JEG-3 and JAR), the breast cancer cell lines (T-47D, BT-20 and MCF-7). the human prostate carcinoma cell line PC-3 and the monkey kidney cell line CV-1 were obtained from the American Type Culture Collection (Rockville. MD)) and were maintained according to the instructions of the supplier For reporter gene analyses, cells were plated onto 60 mm dishes in amounts of 7.5 × 10⁵ (JEG-3 and JAR), 12.0 × 10⁵ (T-47D), 5.0 × 10⁵ (BT-20), 8.0 × 10⁵

(MCF-7), or 5.5 × 10⁵ (PC-3 and CV- 1 ) 20-36 hours before transfection 4.0-5.0 μg of each plasmid construct together with 1.0 or 2.0 μg ß-galactosidase control vector pCMVß (Clontech Laboratories Inc.. Palo Alto, CA) were then transiently transfected into the cells using the transfection reagent DOTAP (5.0-6.0 μg/ml). After 18 hours the media were replaced and the cells were cultured for further 48-52 hours before collection

First, the harvested cells were subjected to four freeze-thaw cycles (freezing in dry ice/ethanol for 5 minutes and thawing at 31°C for 3 minutes) in 100-200 μl of 0.25 M Tns-HCl (pH 7.8). The treatment was followed by heat-inactivation at 65°C for 20 min., after which the cloramphenicol acetyl transferase (CAT) activity of the samples was measured by fluor diffusion assay (Neumann. J.R.. et al., Biotechniques 5:444-447 ( 1987); Eastman, A., Biotechniques 5 73132 ( 1987)). The ß-galactosidase activities were determined by measuring an increase in optical density at 420 nm, using o-mtrophenol-ß-D-galactopyranoside as a substrate, according to the method of Rosenthal ( Methods Enzymol. 152:704-720 ( 1987)) (ß-galactosidase enzyme assay, Promega, Madison, WI). ß-Galactosidase activity was used as the marker of transfection efficiency to normalize the CAT activity. Protein concentrations were measured in samples prepared for CAT and ß-galactosidase assays by the Bio-Rad protein assay (Bio-Rad Laboratories, Richmond. CA). CAT expression was described as picograms per mg of protein or nanograms per mg protein resulting from comparison of the CAT activity in samples with the CAT standard curve.

The results are shown in Figure 1. It can be seen that in JEG-3 cells reporter gene expression increased dramatically when the promoter fragments contained the region from -659 to -550, indicating that this region contains important element(s) for transactivation, an enhancer or a part of it. On the other hand, extension of the fragment to -859 resulted in 50% lower reporter gene expression than that driven from fragment -659/+9. In JAR cells the transcriptional activities of deleted constructs, except for the two longest constructs pCAT-BY-859 and pCAT-BY-659, were higher than in JEG-3 cells. However, in contrast to JEG-3 cells, reporter gene expression of the construct pCAT-BY-659 did not differ significantly from that of pCAT-B Y-550 in JAR cells.

The result of 5'-deletion analysis of the HSD17B1 promoter in JEG-3 cells pointed to an enhancer element in the characterized region. Have various fragments between positions -859 and - 113 were linked in both orientations in front of the thymidine kinase (TK) promoter (Figure 2A) and the constructs were transiently transfected into JEG-3 cells. The longest fragment ranging from -859 to - 113 significantly increased TK promoter activity, and this result further supports the idea of there being an enhancer in the legion. By narrowing the region from both ends, the enhancer element could be limited to between bases -661 and -392. In both orientations the enhancer increased reporter gene expression more than 200-fold compared with the basal activity of the TK promoter in JEG-3 cells. A critical element for enhancer activity was demonstrated to be situated between -549 and -392, as deletion of this region destroyed enhancer activity almost completely. On the other hand, extension of the region from -661 to -859 decreased CAT expression 64% and 49% when the fragment was linked to the TK promoter in the original and opposite orientations, respectively. Thus, the region from -859 to -661 reduced expression of the reporter gene both when linked to HSD17B1 promoter in the pCAT-BY-backbone (Figure 1) and when linked to HSD17B1 enhancer in the pBLCAT2-backbone (Figure 2A). Hence, the lowered CAT expression most probably did not result from an artificially generated silencer at the border of inserted fragments and vectors, but from the region from -859 to -661. In addition to the TK promoter, the fragment from -764 to -392 was connected to a reporter gene vector containing an SV40 promoter. In spite of a distance of 2 kb from the promoter to the cloned fragment, the enhancer increased CAT expression over 20-fold in JEG-3 cells, in both orientations (Figure 2B). The results indicated that the enhancer is also functional when it is not linked adjacent to a promoter.

EXAMPLE 3

Cell-specificity of the HSD17B1 enhancer

To investigate cell-specificity of the enhancer, the constructs pBL(-764/-392)CAT2 and pBL(-392/-764)CAT2, described in Example 1, were transfected into the chonocarcinoma cell lines JEG-3 and JAR, the breast cancer cell lines BT-20, T-47D and MCF-7, the prostate cancer cell line PC-3 and the monkey kidney cell line CV- 1 as described in Example 2.

The results are presented in Figure 3. It can be seen that the enhancer acted most efficiently in JEG-3 and JAR cells and to a clearly lesser extent in BT-20 and T- 47D cells. The enhancer also had a significant effect on reporter gene expression in T-47D cells, when the shorter fragment -661/-392 was linked to the TK promoter. The fragment -661/-392 increased TK promoter activity 12-fold in T-47D cells (Figures 5A and 5B) while the effect of fragment -764Λ392 was only 4-fold (Figure 3). This was not only due to the length of the fragment (see also the corresponding situation in

JEG-3 cells, Figure 2A), but also due to the conditions in which the cells were cultured. While the experiments shown in Figure 3 were performed in the presence of inactivated fetal calf serum, the experiments depicted in Figures 5A and 5B, were performed with twice-stripped serum, which led to reduction of basal thymidine kinase activity of the vector in the latter experiment. In MCF-7, PC-3 and CV- 1 cells. which do not express detectable amounts of 17HSD type 1 protein or 1.3 kb mRNA, the HSD17B1 enhancer element did not display any activity (Figure 3) EXAMPLE 4

The HSD17B1 enhancer contains a retinoic acid response element

A putative retinoic acid response element (RARE), AGGACAggagaAGGTCA, is located at the region between -503 to -487 in the HSD17B1 enhancer. The putative RARE contains two direct repeats of the half-site consensus sequence PuGGTCA with one mismatch in one of the half-sites. The half-sites are spaced five nucleotides apart, which is typical for the binding-site of the RAR/RXR heterodimers, the complexes activated by all-trans- and 9-cm-RA. To test whether this element can bind RA receptors in vitro, the double-stranded oligonucleotide wt-RARE, containing the putative RARE, was incubated with a RXR_α/RAR_α extract.

For electrophoretic mobility shift assays (EMSAs) double-stranded fragments with 5'-protruding ends were used.

Wt-RARE. 5'-GCTGAAAAGAGGACAGGAGAAGGTCAGAGAGACG-3 ', and its mutated analogues

m1-RARE, 5'-GCTGAAAAGACGACTGGAGAAGGTCAGAGAGACG-3', m2-RARE, 5'-GCTGAAAAGAGGACAGGAGAAAGTCTGAGAGACG-3' and m3-RARE, 5'-GCTGAAAAGACGACTGGAGAAAGTCTGAGA-3' corresponded to the position from -512 to -479 in the HSD17B1 gene. In addition, the fragment ßRARE01 , 5'-GGGTAGGGTTCACCGAAAGTTCACTCGCTCC-3', corresponding to the region containing RARE in the hRARß gene (Vivanco Ruiz, M.M. et al., EMBO J. 10:3829-3833 (1991)) was used as a positive control in RA receptor binding studies.

The doubled-stranded fragments with 5'-protruding ends were labelled with [a- ³²P] dCTP using Klenow fragment of DNA polymerase I. The EMSAs were performed as follows; 10-80 ng RXR_α/RAR_α heterodimer protein extract was first incubated on ice for 10 min. in a binding buffer containing 15% glycerol. 20 mM Hepes pH 7.9, 50 mM NaCl, 5 mM MgCl₂, 0.1 mM EDTA, 1 mM DTT, 0.1 mM

PMSF, 50 ng poly(dI-dC)/μl, 0.05% NP-40 and 50 ng BSA/μl. Five nanograms of a probe was added to the reaction mixture and the incubation was further continued on ice for 20 min. For the competition experiments, a 100-fold excess of specific or unspecific oligonucleotide was added together with the probe The DNA-protein complexes were separated on a pre-run ( 175 V, 20 min.) 5 polyacryiamide gel (ratio of acrylamide to bisacrylamide 29: 1) containing 0.25 x TBE. Gels were

electrophoresed for 4-5 hours at 175 V, dried and exposed to Kodak X-AR films (Eastman Kodak, Rochester, NY) for 4-6 hours.

The EMSAs demonstrated that wt-RARE bound proteins specifically from the RXR_α/RAR_α extract and that only weak or no competition was shown by the mutated oligonucleotides m1-RARE. m2-RARE and m3-RARE. or unrelated DNA (Figure 4A ) Furthermore, as shown in figure 4B, wt-RARE bound RXR_α/RAR_α complex in a similar manner to ßRARE01 , the oligonucleotide which contained a RARE of the hRARß gene and is known to bind RXR_α/RAR_α heterodimer (Vivanco Ruiz, M M., et al., EMBO J. 10:3829-3833 ( 1991) ) The mutated oligonucleotide m3-RARE did not bind RXR_α/RAR_α extract at all. In addition to EMSA, reporter gene analyses were performed with RA administration to investigate the function of the RARE JEG-3 and T-47D cells were maintained and transfected as described in Example 2. After transfection, the medium was replaced with one containing dextran-coated charcoal (DCC)-treated FCS (5%) and the cells were further cultured for four hours before RA administration. To induce the cells with RA, the medium was replaced with one containing DDC-FCS (5%) and 1.0 uM at-RA and the cells were cultured for 48 hours before harvest.

Induction by 1 μM at -RA increased the activity of the HSD17B1 enhancer linked to TK promoter 3- to 4-fold in T-47D ceils and 2- to 3-fold in JEG-3 ceils (Figure 5A). The introduction of mutations to the RARE (AGGACAggagaAGGTCA

-> ACGACTggagaAAGTCG) did not only abolish the response to RA completely, but also decreased the enhancer activity to 10% of that for uninduced samples when retinoic acid was administrated (Figure 5B). This indicates that the enhancer may contain another element or that the introduction of mutations can create a new one. which mediates an opposite RA effect either directly or indirectly. On the other hand, the pBLCAT2 vector used in the experiments consists of a pUC-backbone which contains a binding site for activator protein- 1 (AP-1) (Kushner, P.J. et al., Mol. Endocrinol 8:405-407 ( 1994)). While RARs are able to form complexes with proteins binding the AP- 1 site, it cannot be excluded that the repression detected is an artificial phenomenon. The AP-1 site of the vector backbone did not affect the intact RARE in the HSD17B1 enhancer, for deletion of the AP- 1 site from the pBLCAT2- vector did not modulate RA induction of the intact enhancer region in T-47D cells (data not shown).

EXAMPLE 5

Localization of a silencer element in the HSD17B1 gene

Two of the three shortest fragments of the HSD17B1 promoter, fragments - 113/+9 and -78/+9, were inserted in the vectors pCAT-EY and pCAT-BY. which contained the CAT reporter gene with and without SV40 enhancei, respectively. To test whether such short fragments are able to act as basal promoters, the constructs were transfected into several cell lines When the fragment -78/+9, was inserted in pCAT-BY vector, low but detectable CAT expression was observed in JAR (Figure 1), PC-3 and CV- 1 cells (data not shown). In pCAT-EY vector, as shown in Figure 6. fragment -78/+9 drove significant CAT expression in BT-20. PC-3, CV- 1 and JEG-3 cells The results suggested that fragment -78/+9 was enough for the binding of RNA polymerase II transcription premitiation complexes.

Surprisingly, CAT expression from the construct pCAT-EY-78 was 30- 100% higher than from pCAT-EY-113 in all the cell lines tested (Figure 6). This, together with the observation that the HSD17B1 enhancer (-661 to -392) increased reporter gene expression significantly more efficiently when linked to the TK promoter (Figure 2) than when linked to its own promoter, as in the fragment -659/+9 (Figure 1), pointed to the presence of a silencer element between positions -392 and -78 To test this hypothesis, the enhancer element (-661 to -392) was fused in front of the proximal promoter fragments -228 /+9, - 113/+9 and -78/+9 in pCAT-BY vectors. Reporter gene analysis in JEG-3 cells showed that CAT expression generated from the construct pCAT-BY-En-78 was 4-fold greater than that from the intact promoter construct pCAT-BY-659 (Figure 7) In addition, fragment -78/+9 displayed 2-fold transcriptional activity compared with fragment - l 13/+9 when linked to pCAT-BY-En-vector. Altogether, the results indicated that the region from -392 to -78 contains a silencer and that the region from - 113 to -78 is essential for the repression of

HSD17B1 gene transcription. Because the region from -113 to -78 decreased transcription activity in the context of several different backbones, it excludes the possibility that the silencer element was artificially created by linking the fragments together or that the function of the enhancer element is dependent on the distance between the HSD17B1 promoter and enhancer.

To investigate the effect of deletion of fragment -392/-78 on HSD17B1 gene regulation in T-47D cells, the cells were transfected with the constructs pCAT-BY-659 and pCAT-BY-En-78 with and without RA administration While the intact fragment

-659/+9 generated minor reporter gene expression, CAT expression was notably, 20-fold, higher when the HSD17B1 enhancer was connected adjacently to the proximal HSD17B1 promoter. The results are presented in Figure 8. It can be seen that no significant increase in the promoter activity of the intact fragment was observed after at -RA administration. On the other hand, a clear 3-fold induction by RA was detected when the construct pCAT-BY-En-78 was tested. This also demonstrated

functioning of the RARE, not only when connected to TK promoter, but also in connection to the HSD17B1 promoter.

Figure 8B illustrates the el feet of at-RA administration on CAT expression by constructs pCAT-B Y-659. pCAT-B Y-En-228. pCAT-BY-En- 113 and pCAT-B Y-En- 78 Construct pCAT-BY-659 contains the intact HSD17B1 promoter fragment

-659/+9, whereas in constructs pCAT-BY-En-228. pCAT-BY-En- 113 and pCAT-BY- En-78 the HSD17B1 enhancer fragment -661/-392 has been linked to proximal promoter fragments - 228/+9. - 113/+9 and -78/+9 respectively. The results represent the mean ± range from two independent experiments in each of which duplicate samples were analyzed. The results in Figure 8B show that deletion of the fragment -392/-79 results in an approximately 20 fold increase in transcription efficiency in T-47D cells. Deletion of the fragments -392/-229 and -392/- 114 increases reporter gene expression also, three to eight fold, which suggest that those regions can participate in repression of transcription.

EXAMPLE 6

Detailed characterization of HSD17B1 promoter: role of Sp1 and AP-2 binding sites The proximal promoter, the region from -78 to 4-9 contains consensus sequences for Sp1 and AP-2 transcription factors, for example. In order to study whether the consensus sequences can bind proteins important for the function of HSD17B1 gene promoter, the fragment called HSD-AP-2/Sp1 from -69 to -36 was firstly incubated with various nuclear protein extracts (for the whole sequences, see Table 1). The extracts included nuclear extracts from JAR, JEG-3 and T-47D cells and additionally from HeLa cells, which are known to be rich in Sp1 factor. The nuclear extracts were prepared as described by Ausubel, F.M. et al. (In Current Protocols in Molecular Biology, John Wiley & Sons, Inc. ( 1994)) according to Dignam, J.D. et al. (Nucl. Acids Res. 11:1475- 1489 ( 1983)). In addition to the cell line nuclear extracts, an AP-2 protein extract (Promega, Madison, WI) was used. Briefly, the double stranded oligonucleotide was labeled with [a-³²P] dCTP using Klenow fragment of DNA polymerase I (Boehringer Mannheim, Mannheim, Germany) and the EMSAs were performed as described above with some modifications 4.0-8.0 mg nuclear extract was first incubated at RT for 10 min. in a 20 ml of binding buffer containing 10% glycerol, 20 mM Hepes (pH 7.9), 50 mM NaCl, 5 mM MgCl₂, 0.1 mM EDTA. 1 mM DTT. 0.1 mM PMSF, 0.05% NP-40 and 2.0 mg poly(dl-dC). 0.25-0.5 ng of a labeled probe was added to the reaction mixture and the incubation was further continued at RT for 20 min. The DNA-protein complexes were separated on a pre-run (175 V, 20 min.) 5% polyacrylamide gel (ratio of acrylamide to bisacrylamide 29:1) containing 0.5 x TBE. Gels were electrophoresed for 3-4 hours at 175 V, dried and exposed to Kodak X-AR films (Eastman Kodak. Rochester, NY) for 12-24 hours.

As depicted in the Figure 9A. the fragment formed two separate complexes with the nuclear extracts prepared from JAR, JEG-3. T-47D and HeLa cells. The lower complex, Complex 1, had drifted equally to the one formed from AP-2 extract, whereas the minor complex. Complex 2, formed with JAR, JEG-3 and T-47D extracts had drifted similarly to the major complex formed with HeLa cell extract. Specificity of the binding of the factors to their motifs was further characterized by adding unlabelled oligonucleotides (for the whole sequences, see Table 1) to the binding reactions. A 100-fold excess of competitor was added together with the probe to the EMSA reactions and the result have been shown in the Figure 9B. The HSD17B1 promoter fragment containing an AP-2 binding site like sequence (HSD-AP-2) prohibited the formation of the Complex 1 as efficiently as the AP-2 consensus fragment, whereas the mutated analog, HSD-mAP-2 was unable to do that. In the same way, HSD-Sp1 and Sp1 consensus fragments hindered formation of the Complex 2, while a HSD-mSp1 fragment was without any effect. These results strongly suggested that AP-2 and Sp1 or similar factors bind to HSD17B1 proximal promoter which was further verified using specific antibodies against AP-2 and Sp1.

For the supershift assays, 1.0-2.0 ml of appropriate TransCruz™ Gel Supershift antibody (Santa Cruz Biotechnology Inc ) was added to the reaction mix subsequent to addition of P-labeled probe, and the reaction was then continued at 4∞C for 60 min. Complexes 1 and complexes 2 formed either from JAR or JEG-3 nuclear extract were supershifted by anti-AP-2 and anti-Sp1 , respectively, whereas neither complex could be influenced by anti-GATA-3, used as an unspecific antibody here (Fig. 9C).

Role of the binding of Sp1 and AP-2 or similar factors for the function of HSD17B1 promoter was further characterized by reporter gene analysis. As shown in the Figure 9D, firstly, there were no significant difference in the reporter gene expressions derived from fragments -78/+9 or -97/+9 in JAR or JEG-3 cells. In both of the cell lines, mutation of AP-2 binding site resulted in increased promoter activity, which was 1.2-fold in JAR and 2.6-fold in JEG-3 cells. These results together with EMSAs suggested that AP-2 or similar factor is bound by HSD17B1 gene promoter and that the factor decreases function of the promoter On the contrary, mutation of the Sp1 binding site led to drastic decrease of the reporter gene expression indicating that binding of Sp1 or similar factor is essential for full function of the HSD17B1 gene promoter. EXAMPLE 7

Detailed characterization of the HSD17B1 enhancer; localization of essential elements

In order to study interactions between nuclear proteins and the HSD17B1 enhancer, end-labeled sense strand of the -661/-392 fragment was incubated with nuclear extracts prepared from JAR and JEG-3 chonocarcinoma cells and from T-47D and MCF-7 breast cancer cells. Briefly, 75 mg different nuclear extracts or 50 mg BSA were firstly incubated for 10 minutes in RT with 4 mg of poly (dl-dC) in the D50 buffer (final concentrations being 10 mM Tris-HCl, pH 7 5, 10% glycerol. 0.5 mM DTT, 0.5 mM EDTA and 50 mM NaCl) containing 0.04% NP-40 and 2 mM PMSF 40 000 cpm of ³²P-labelled -661/-392 fragment was then added to the reaction mixture following by further incubation for 20 minutes. Next, the binding mixtures were subjected for I minute to DNase I digestion, in which 0.2 U DNase I for BSA-sample and 2.0 and 3.0 U for other samples were used Digestion was stopped with DNase I stop solution (Pharmacia, Biotech. Sweden) after which the reaction mixtures were purified by phenol/chloroform extractions and ethanol precipitation. Purified samples were then dissolved into FP-loading dye (Pharmacia Biotech, Sweden) and applied to 6% polyacrylamideurea gel (Sequagel™, National Diagnostics, Atlanta, GA). Finally, gels were electrophoresed for 2 hours at 1500 V, dried and exposed to Kodak X-AR films (Eastman Kodak, Rochester. NY) for 24 hours As a result, three distinct areas had been protected with the nuclear extracts used (Figure 10A). Instead, no apparent differences in protection between 17HSD type 1 expressing (JEG-3. JAR, T-47D) and non-expressing (MCF-7) cell lines, or between chonocarcinoma and breast cancer cell lines, were detected. Thus the result indicated that the three protected areas may be important for the function of the enhancer, but may not determine the cell-specificity of the enhancer.

Importance of the protected areas was further examined by reporter gene analyses. The reporter gene constructs contained thymidine kinase promoter and the enhancer region, which was mutated as described above in the Example 1. In addition to the three protected region, effect of five single nucleotides to function of enhancer was investigated. The five nucleotides were chosen according to the differences between the HSD17B1 enhancer and the analogous area in the HSD17BP1 gene. henceforth called as a "pseudoenhancer". The HSD17BP 1 gene is situated beside the HSD17B1 gene, but its protein product, if any. is unknown. The 5' regions of the HSD17B1 gene and HSD17BP1 gene share 98% similarity (LuuThe. V. et al., Mol. Endocrinol. 4:268-275 ( 1990); Peltoketo. H. et al., Eur. J. Biochem. 209.459-466 ( 1992)), but the pseudoenhancei has only 20% of the activity of the HSD17B1 enhancer. Altogether, there are five dissimilar nucleotides in the HSD17B1 enhancer and pseudoenhancer at the positions -615, -486, -480, -452 and -435 as calculated according the HSD17B1 enhancer. These nucleotides in the HSD17B1 enhancer were individually changed to those ones situating in the pseudoenhancer. except the guanine at the position -486, which was changed together with cytosine at the position -480, as described in the Example 1. Figure 10B depicts, how replacing of the protected area from -544 to -528 (FP2) with a nonsense sequence practically abolished the enhancer activity completely. After replacing of the region from -589 to -571 with a nonsense sequence and after simultaneous mutation of the nucleotides at the positions -486 and -480 also less than 20% of the original enhancer activity was left. Because the single mutation of the nucleotide at the position -480 or substitution the area from -495 to -485 (FP1 ) with a nonsense sequence lessened the function of the enhancer only to some extent, simultaneous mutation of the positions -480 and -486 may be needed to abolish the enhancer activity. The results together strongly suggest that those regions and nucleotides bind factors essential for the function of the enhancer. The results are also in line with the results showing that both halves of the enhancer are needed for the function of it. Finally, replacement of the nucleotide at the positions -435 results in decreased, and replacement of the nucleotide -452 in increased reporter gene expression, suggesting that these nucleotides are also involved in binding of re g u l atory factors modu l at in g the e n h ance r fu n c t ion .

EXAMPLE 8

Detailed characterization of the HSD17B1 silencer; role of GATA-binding site

As described in the Example 5. results achieved using deleted reporter gene constructs pointed to the presence of a silencer containing an essential part for function of it between the nucleotides - 113 and -78 The region was found to comprise a consensus binding site for so called GATA-transcription factors, of which GATA-2 and GATA-3 are known to be abundantly expressed in JEG-3 cell regulating the expression of the human glycoprotein a-subunit gene, for example (Steger, D.J et al., Mol. Cell . Biol. 14-5592-5602 ( 1994)). As illustrated in the Figure 11A, the double- stranded fragment - 114/-77, named as HSD-GATA, formed two complexes with factors in the nuclear extract prepared from JEG-3 cells. The analogous non labeled fragment and the fragment containing the GATA consensus sequence prevented the formation of the complexes indicating, that the binding between proteins and DNA was specific and that proteins binding GATA-site are involved in formation of the complexes. The mutated - 114/-77 fragment. HSD-m1GATA. decreased the formation of the complexes to some extent suggesting, that the mutation introduced did not restrain binding of proteins completely. The unspecific competitor. ßRARE1 fragment, was not able to disturb the binding reactions, which thus further indicated specificity of them (For complete sequences, see Table 1.)

Figure 11B depicts that in addition to JEG-3 nuclear extract, proteins in JAR and T-47D, but not in BT-20. nuclear extracts were also able to form the two complexes with the HSD-GATA fragment. Administration of GATA-2 or GATA-3 antibodies to the reaction mixtures led to shift of the complexes indicating that the binding factors are GATA-2 and GATA-3 or very similar factors recognized by the antibodies. Absence of the supershift 2 from the JAR panel in the Figure 11B is most probably due to the lower concentration the GATA-2 factor in the JAR cells than in

JEG-3 and T-47D cells.

Introduction of mutations into the GATA-site of the pCAT-EY-113 construct was further used to verify the role GATA-binding proteins in repressing function of HSD17B1 promoter Figure 11C illustrates how the mutated constructs pCAT-EY-113m 1 GATA and pCAT-EY-113m2GATA led as high reporter gene expression as the pCAT-EY-78 in JEG-3 cells. Similar tendency could be detected also in JAR cells, even though the differences between reporter gene expressions were marginal as compared to those in JEG-3 cells. Altogether, the results demonstrated in the panels A, B and C argue that the region from - 102 to -99 is involved in binding GATA-2 and GATA-3 or very similar factors and that binding of the factors reduce function of

HSD17B1 promoter.

EXAMPLE 9

Screening of trans-acting agents modulating the HSD17B1 gene function In the Examples 6-8, it has been described some strategies how trans-acting factors affecting HSD17B1 gene function can be screened and identified. Those methods include use of reporter gene analyses with intact, deleted, and mutated gene fragments as well as binding assays (DNase I footprinting, EMSA and supershift EMSA and in addition to those, methylation and uracil interference assays and UV crosslinking of proteins to nucleic acids) with intact and mutated fragments. Reporter gene analyses, in which the gene fragment is linked together with a promoter or enhancer when necessary, and a reporter gene, are first used to sublocalize the areas affecting gene expression (see also Examples 2 and 5). The described reporter gene system is also used to screen extracellular agents, such as synthetic and naturally occuring retinoic acids, which affects function of the HSD17B1 gene and

consequently estradiol production Briefly, a host cell line is transiently or stably transfected with a reporter gene construct containing elements from the HSD17B1 gene. The cell line is treated with the agent or agents to be examined and effects of it or them to reporter gene. CAT for example, expression is measured (see also the Example 4). Several agents can be administred simultaneously to investigate synergistic effects. While DNA regions needed for mediating the signals from extracellular agent(s) have been localized, factors binding the DNA elements and thus mediating the signals are identified as described herein below. The system is especially useful for screening properties of various retinoic acids, whether an investigated form modulates the function of the HSD17B1 gene and thus possibly estradiol-dependent cell growth, and if so, what kind of specific ligand-receptor complex is needed for that.

When a factor hasbeen detectably bound to the HSD17B1 gene fragment in the bindingassays, the factor/DNA-complex can be cut out from a gel, separate from each other and the protein is digested to small peptides. which are fractionated and partially sequenced. Corresponding DNA sequences are then used for preparing a probe by PCR and the probe is then used for screening cDNA libraries prepared from the cell line or tissue, in which a corresponding regulatory element has been found to be functional. General protocols (see e.g. Maniatis, T. et al. above) are used in screenings of the libraries.

In addition to that, oligonucleotides containing the regulatory sequences described herein before, are biotin-labeled, after which the end-labeled fragment is incubated with a nuclear extract containing a searched factor. A protein-DNA-complex is then bound by streptavidin tetramer and next the protein/DNA-fragment/streptavidin ternary complex is trapped by biotin resin. The column is washed and finally the protein of interest is eluted from the resin.

Double-stranded DNA fragments containing specific binding sites can also be linked covalently to a solid support, such as cyanogen bromide activated agarose resin. Proteins from possibly partially purified nuclear extract is then allowed to bind to the DNA, unspecifically bound proteins are rinsed out and finally the bound factors are eluted using increased salt concentrations.

Alternatively, a λgt11 cDNA library prepared from the cell line containing a desired factor, is screened using labeled fragment containing possibly several copies about the identified binding sites. Potentially positive clones, which are not recognized by unspecific DNA-fragments. are rescreened, enriched, purified and isolated. Finally the fragments encoding the searched factor(s) are transferred into plasmids. in which they are sequenced and their properties are further characterized. While a clone containing a sequence encoding the binding protein has been trapped, it can further be use for catching of the proteins, which associate by protein-protein interactions to the protein bound by the specific DNA.

A method for screening of the associating proteins is the two-hybrid system of Clontech (Palo Alto, California), in which a vector containing a sequence for the DNA-binding protein is used to generate a fusion of a GAL4 DNA-binding domain and the protein binding to DNA. In addition to that, random collection of cDNAs are cloned into a vector containing a sequence for generation of fusion protein with a GAL4 activation domain. Both hybrid proteins are then expressed in the same cell and interaction of them results in ss-galactosidase expression in the system.

The disclosure of all references, patents and patent applications, if any, recited herein are hereby incorporated by reference in their entirety. (

Claims

CLAIMS 1. A nucleic acid molecule which enhances HSD17B1 gene expression and wherein said nucleic acid molecule comprises at least 16 nucleotides of SEQ ED

No 1.

2. The nucleic acid molecule according to claim 1 wherein said nucleic acid molecule comprises nucleotides 222 to 491 of SEQ ID No 1 or a functionally active fragment thereof.

3. The nucleic acid molecule according to claim 1 wherein said nucleic acid molecule comprises nucleotides 339 to 355 of SEQ ID No 1.

4. The nucleic acid molecule according to claim 1 wherein said nucleic acid molecule comprises nucleotides 294 to 312 of SEQ ID No 1.

5. The nucleic acid molecule according to claim 1 wherein said nucleic acid molecule comprises nucleotides 339 to 355 and nucleotides 294 to 312 of SEQ ID No 1.

6. The nucleic acid molecule according to any one of claims 1 to 5 wherein said nucleic acid molecule comprises a retinoic acid response element characterized by the sequence spanning nucleotides 380 to 396 of SEQ ID No 1.

7. The nucleic acid molecule according to any one of claims 1 to 6 wherein said nucleic acid molecule comprises nucleotides 397 to 403 of SEQ ID No 1.

8. The nucleic acid molecule according to any one of claims 1 to 7 wherein said nucleic acid molecule comprises thymidine at the position 448 of SEQ ID No 1.

9. The nucleic acid molecule according to any one of claims 2 to 8 wherein said nucleic acid molecule further comprises 5'-nucleotides extending up to nucleotide 99 of SEQ ID No 1.

10. The nucleic acid molecule according to claims 1 or 2 which comprises a mutation at the position 431 of SEQ ID No 1.

11. The nucleic acid molecule according to claim 10 wherein adenine at the position 431 of SEQ ID No 1 is substituted by guanine.

12. A nucleic acid molecule which represses HSD17B1 gene expression and wherein said nucleic acid molecule comprises at least 36 nucleotides of SEQ ID

No 1.

13. The nucleic acid molecule according to claim 12, wherein said nucleic acid molecule comprises nucleotides 492 to 805 of SEQ ID No 1.

14. The nucleic acid molecule according to claim 12, wherein said nucleic acid molecule comprises nucleotides 770 to 805 of SEQ ID No 1.

15. The nucleic acid molecule according to claim 14. wherein said nucleic acid molecule further comprises nucleotides at least part of nucleotides 769 to 492 of

SEQ ID No 1.

16. The nucleic acid molecule according to any on of claims 12 to 15, wherein said nucleic acid molecule comprises nucleotides 784 to 781 of SEQ ID No 1.

17. The nucleic acid molecule according to claim 16 which comprises a mutation at any of the positions from 784 to 781 of SEQ ID No 1.

18. The nucleic acid molecule according to claim 17 wherein thymidine at the position 781 of SEQ ID No 1 is substituted by adenine and adenine at the position

782 of SEQ ED No 1 is substituted by thymidine.

19. The nucleic acid molecule according to claim 17 wherein thymidine at the position 783 of SEQ ID No 1 is substituted by adenine and cytidine at the position 784 of SEQ ID No 1 is substituted by guanine.

20. The nucleic acid molecule according to any one of claims 1- 11, wherein aid nucleic acid molecule further comprises a silencer.

21. The nucleic acid molecule according to any one of claims 12- 19, wherein said nucleic acid molecule further comprises an enhancer.

22. A nucleic acid molecule which promotes HSD17B1 gene expression and wherein said nucleic acid molecule comprises nucleotides 840 to 831 of SEQ ID

No 1.

23. The nucleic acid molecule according to claim 22 wherein said nucleic acid molecule further comprises nucleotides 830 to 821 of SEQ ID No 1.

24. A nucleic acid molecule which promotes HSD17B1 gene expression and wherein said nucleic acid molecule comprises nucleotides 805 to 891 of SEQ ID No 1.

25. The nucleic acid molecule according to claims 23 or 24. which comprises a mutation at any of the positions from 830 to 821 of SEQ ID No 1.

26. The nucleic acid molecule according to claim 25, wherein cytidines at the positions 822 and 833 of SEQ ID No 1 are substituted by thymidines.

27. A method for screening transacting agents capable of affecting HSD17B1 gene expression, wherein said method comprises

(1) providing a host cell, wherein said host cell has been transfected with a recombinant nucleic acid molecule containing an HSD17B1 transcriptional regulatory element, a promoter, and a reporter gene, and wherein said transcriptional regulatory element and promoter are operably linked to the reporter gene;

(2) administering a candidate HSD17B1 trans-acting agent to the transfected host cell; and

(3) determining the effect of said agent on reporter gene expression.

28. A method according to claim 27, wherein said HSD17B1

transcriptional regulatory element comprises an HSD17B1 enhancer and an

HSD17B1silencer.

29. A method according to claims 27 or 28. wherein the agent is a naturally occunng or synthetic retinoid.

30. A method for screening trans-acting agents capable of enhancing

HSD17B1 gene expression, wherein said method comprises

(1) providing a host cell, wherein said host cell has been transfected with a recombinant nucleic acid molecule containing an HSD17B1 enhancer, a promoter, and a reporter gene, and wherein said enhancer and promoter are operably linked to the reporter gene;

(2) administering a candidate HSD17B1 trans-acting agent to the transfected host cell; and

(3) determining the effect of said agent on reporter gene expression.

31. A method according to claim 30, wherein said HSD17B1 enhancer comprises the nucleic acid molecule according to any one of claims 1 to 11.

32. A method for screening trans-acting agents capable of repressing HSD17B1 gene expression, wherein said method comprises

(1) providing a host cell, wherein said host cell has been transfected with a recombinant nucleic acid molecule containing an HSD17B1 silencer, a promoter, and a reporter gene, and wherein said silencer and promoter are operably linked to the reporter gene;

(2) administering a candidate HSD17B1 trans-acting agent to the transfected host cell ; and

(3) determining the effect of said agent on reporter gene expression.

33. A method according to claim 32, wherein said HSD17B1 silencer comprises the nucleic acid molecule according to any one of claims 12 to 19.

34. A recombinant DNA molecule comprising the nucleic acid molecule according to any one of claims 1 to 19.

35. A recombinant DNA molecule according to claim 34, wherein said recombinant DNA molecule further comprises a promoter operably linked to a reporter gene.

36. A vector comprising a nucleic acid molecule according to claims 34 or 35.

37. The vector according to claim 36, wherein said vector is a plasmid.

38. A plasmid selected from the group consisting of pBL(-859/-637)CAT2, pBL(-637/-859)CAT2, pBL( -859/-549)CAT2, pBL(-549/-859)CAT2, pBL( -859/- 392)CAT2, pBL(-392/-859)CAT2. pBL(-859/-113)CAT2, pBL(-113/-859)CAT2. pBL(-764/-549)CAT2, pBL(-549/-764)CAT2, pBL(-764/-392)CAT2, pBL(-392/- 764)CAT2, pBL(-66l/-392)CAT2 and pBL(-392/-661)CAT2

39 A plasmid selected from the group consisting of pBL(-661/-392)CAT4, pBL(-661/-392)CAT4-FP1 (-),pBL( -661/-392)CAT4-FP2(-),pBL(-661/-392)CAT4-FP3(-), pBL(-661/-392)CAT4-(-615T->C), pBL(-661/-392)CAT4-(-480C->G), pBL(-661/-392)CAT4-(-480C->G,-486G->A), pBL(-661/-392)CAT4-(-452A->G), and pBL(-661/-392)CAT4-(-435T->A).

40. A plasmid selected from the group consisting of pCAT-P-(-764/-392), pCAT-P-(-392/-764), pCAT-BY-228 , pCAT-BY- 113, pCAT-BY-78, pCAT-B Y-En-228, pCAT-BY-En- 113, pCAT-BY-En-78, pCAT-EY-228, pCAT-EY-113, pCAT-EY-97, pCAT-EY-78, pCAT-EY- 113-m 1 GATA, pCAT-EY-113-m2GATA, pCAT-EY-97-mSp1 and pCAT-EY-97-mAP-2.

41. A host cell transfected by the recombinant DNA molecule of any one of claims 34 and 35.

42. The host cell according to claim 41, wherein said host cell comprises a eukaryotic cell.

43. The host cell according to claim 42, wherein said eukaryotic cell is selected from the group consisting of a choriocarcinoma cell, a prostate carcinoma cell, a breast cancer cell, a kidney cell and a granulose cell.

44. Use of an HSD17B1 promoter in the identification of a transacting agent capable of affecting estrogen production.

45. Use according to claim 44, wherein said HSD17B1 promoter comprises the nucleic acid molecule according to any one of claims 22-26.

46. Use of an HSD17B1 enhancer in the identification of a transacting agent capable of up-regulating estrogen production.

47. Use according to claim 46, wherein said HSD17B1 enhancer is the nucleic acid molecule according to any one of claims 1- 11.

48. Use of an HSD17B1 silencer in the identification of a trans-acting agent capable of down-regulating estrogen production.

49. Use according to claim 48, wherein said HSD17B1 silencer is the nucleic acid molecule according to any one of claims 12- 19.