| Human Molecular Genetics | Pages |
©1999 Oxford University Press |
A common polymorphic allele of the human luteinizing hormone [beta]-subunit gene: additional mutations and differential function of the promoter sequence
Introduction
Results
PCR amplification and sequence analysis of the V-LH[beta] promoter
Expression and deletion analysis of the wt- and V-LH[beta] promoter constructs in vitro
Effects of sex steroids on the LH[beta] promoter activity
Effects of pulsatile GnRH administration on the LH[beta] promoter activity
Phorbol ester, forskolin and 8-bromo-cAMP stimulation of the LH[beta] promoter activity
Discussion
Materials And Methods
Amplification and sequencing of the V-LH[beta] promoter
Plasmid preparations
Cell cultures, transient transfections, luciferase and [beta]-galactosidase assays
Trans-activation of the wt- and V-LH[beta] promoter
Computer programs and statistical analysis
Acknowledgements
References
A common polymorphic allele of the human luteinizing hormone [beta]-subunit gene: additional mutations and differential function of the promoter sequence
Received April 30, 1999; Revised and Accepted July 20, 1999
A common genetic variant (V) of the human luteinizing hormone (LH) [beta]-subunit gene was recently discovered. The V-LH molecules have higher bioactivity in vitro, but shorter half-life in circulation, which apparently is related to the alterations of LH function observed in individuals homo- and heterozygous for the V-LH[beta] allele. We have now studied whether additional mutations in the V-LH[beta] promoter sequence could contribute to the altered physiology of the LH variant molecules. The 661 bp 5[prime]-flanking region of the V-LH[beta] gene, retrieved from human genomic DNA by PCR, contained eight single-nucleotide changes, as compared with the wild-type (wt) LH[beta] promoter. The finding was consistent in DNA samples of different ethnic groups. Reporter constructs with various lengths of the wt- and V-LH promoter sequences, driving the firefly luciferase reporter gene, were transfected into an immortalized mouse pituitary cell line, L[beta]T2, known to express the endogenous LH[beta] gene, and into a non-endocrine human embryonic kidney cell line, HEK 293. Basal expression levels of the V-LH[beta] promoter constructs were on average 36% higher in L[beta]T2 cells (P < 0.001; n = 29), and 40% higher in HEK 293 cells (P < 0.001; n = 16), as compared with the respective wt sequences. Numerous qualitative and quantitative differences were found between the two cell lines in responses of the two promoter sequences to stimulation with 12-O-tetradecanoylphorbol-13-acetate, forskolin, 8-bromo-cAMP, progesterone and gonado- tropin-releasing hormone. In conclusion, the V-LH[beta] promoter has higher basal activity, and differs in response to hormonal stimulation, as compared with the wt-LH[beta] promoter. The altered promoter function of the V-LH[beta] gene provides evidence for differences in regulation of the wt- and V-LH[beta] genes, which may contribute to the differences observed in pituitary-gonadal function between carriers of the two LH[beta] alleles. The findings also suggest a novel evolutionary mechanism whereby polymorphic changes resulting in altered bioactivity of a gene product may be compensated for by additional mutations in the cognate promoter sequence, changing transcription of the same gene.
INTRODUCTION
The two pituitary gonadotropins, luteinizing hormone (LH) and follicle-stimulating hormone (FSH), are essential for gonadal function. LH and FSH, together with thyroid-stimulating hormone (TSH) and chorionic gonadotropin (CG), belong to the family of glycoprotein hormones. Structurally, they are heterodimers, consisting of two dissimilar subunits, the common (C) [alpha]-subunit and the hormone-specific [beta]-subunit (1), and both are glycosylated to varying extents. The intact heterodimer structure is required for biological activity. The two LH subunits (i.e. C[alpha] and [beta]) are encoded by separate genes, the former in chromosome 6 and the latter in chromosome 19 (2).
A common genetic variant (V) of LH, due to two point mutations in the LH[beta] gene, one in codon 8 (TGG->CGG; Trp->Arg), and the other in codon 15 (ATC->ACC; Ile->Thr), was discovered, recently (3-5). The two mutations alter the biological function of the V-LH molecule, which is higher in vitro, but has shorter half-life in circulation, as compared with the wild-type (wt)-LH (6,7). The kinetics of action of the secretory peaks of V-LH, with high potency but short duration, is thus different from that of wt-LH. Immunological measurements have shown that in heterozygous individuals, serum contains roughly equal amounts of wt- and V-LH (3,6,8). Hence, the hormone encoded by the V-LH[beta] gene plays an important role in the overall LH activity in circulation. Although the mean level of immunoreactive LH in circulation of V-LH[beta] homozygotes is only slightly lower than in wt or heterozygous individuals (9), the qualitative differences in the LH secretory peaks (see above) may be of functional significance. In accordance, homo- or heterozygous expression of the V-LH[beta] allele has clear physiological and patho- physiological consequences, as evidenced by slight but significant alterations in ovarian steroidogenesis (8), and delayed progression of puberty and gain of height in boys (10). In addition, the V-LH[beta] allele appears to protect obese women from developing symptoms of polycystic ovarian syndrome (PCOS) (11), and disturbances of the menstrual cycle have been reported in women homozygous for the V-LH[beta] gene (4,5,12). Moreover, since the carrier frequency of the V-LH[beta] allele ranges from 0 to 50% in various populations, with an average of 15-20% in Caucasians (9), additional information on the biology of the V-LH[beta] gene is of clinical importance. We hypothesized that, besides the two point mutations in the coding sequence of the V-LH gene, additional alterations could exist in its regulatory sequences. Indeed, a total of eight point mutations were identified within the first 661 nucleotides of the 5[prime]-flanking sequence of the V-LH[beta] gene. In comparison with the wt promoter, that of V-LH[beta] was found to be transcriptionally more active and showed qualitative differences in response to hormonal regulation.
RESULTS
PCR amplification and sequence analysis of the V-LH[beta] promoter
PCR was used to amplify a 661 bp fragment covering nucleotides -8 to -668 of the V-LH[beta] promoter region, in relation to the translation initiation codon. In subjects homozygous for the V-LH[beta] gene, eight nucleotide changes were found as compared with the respective promoter sequence of the wt-LH[beta] gene (13,14), at bases -238, -276, -489, -490, -504, -506, -525 and -552 (Fig. 1). When the nucleotide sequence of the V-LH[beta] promoter was compared with those of the wt-LH[beta] and of hCG[beta] genes (13,14), it was apparent that four alterations at positions -238 (A->G), -276 (G->A), -489 (C->A) and -490 (T->A) differed both from the wt-LH[beta] promoter and those of the hCG[beta] genes. The nucleotide change at position -552 (C->T) was the same as found in the respective sequences of hCG[beta]1, hCG[beta]5 and hCG[beta]7, but different from those of wt-LH[beta], hCG[beta]3 and hCG[beta]8. In position -525 (T->G), the altered nucleotide differed from the respective wt-LH[beta] promoter sequence, but was similar to that of the hCG[beta] genes. The two additional nucleotide changes, at positions -504 (T->A) and -506 (T->C), differed from the promoter sequences of wt-LH[beta], hCG[beta]5, hCG[beta]6 and hCG[beta]7, but were the same as present in hCG[beta]1, hCG[beta]3 and hCG[beta]8 (13,14).
Figure 1. Alignment of the 5[prime]-flanking sequences of the human wt- and V-LH[beta], and hCG[beta] genes. Eight single nucleotide changes (shaded) were found in the V-LH[beta] promoter as compared with that of the wt-LH[beta] promoter. The numbers indicate base pairs upstream of the LH[beta] translation start site. The three-letter codes above the nucleotide sequence indicate the first LH[beta] exon. The empty space indicates that the sequence is not available.
No additional changes were found in the intronic or 3[prime]-flanking regions of the V-LH[beta] gene. The same nucleotide changes were found in the LH[beta] promoter sequence in DNA samples obtained from V-LH homozygotes of several ethnic groups (data not shown), which demonstrates that they belong with nucleotide changes of the V-LH[beta] coding sequence to the same variant allele.
Expression and deletion analysis of the wt- and V-LH[beta] promoter constructs in vitro
The expression levels of the 661 bp V-LH[beta] promoter construct (Fig. 2) were, in L[beta]T2 cells, 35.6 ± 2.7% (mean ± SEM; P < 0.001; n = 29) and, in human embryonic kidney (HEK) 293 cells, 40.4 ± 2.7% (P < 0.001; n = 16) higher than those of the corresponding wt-LH[beta] promoter construct (Fig. 3). Effects of deletions at the 5[prime] and 3[prime] ends on LH[beta] promoter activity were studied using DNA fragments consisting of bases -8/-299 or -300/-668, respectively (Fig. 2). In L[beta]T2 cells, both the wt(-8/-299) and M2(-8/-299) plasmids were more active than the respective full-length constructs, and the M2( 8/-299) plasmid displayed higher activity than the respective wt construct (Fig. 4). These findings indicate the presence of negative regulatory elements in the distal part of the full-length promoter fragment, i.e. beyond nucleotide -300, and suggest that these elements are suppressed by the mutations in the V-LH[beta] promoter sequence. The latter finding was not made in transfections of HEK 293 cells. The two 3[prime]-deleted wt- and V-mutants, spanning bases -300/-668 (Fig. 2), displayed ~50% decreased luciferase activity in both cell types transfected (Fig. 4), indicating the presence of positive cis-acting elements in the proximal part of the promoter.
Figure 2. Structures of the LH[beta] promoter/luciferase (luci) gene constructs, i.e. promoterless, wt(-8/-668), V(-8/-668), M2(-8/-668), M6(-8/-668), wt(-8/-299), M2(-8/-299), wt(-300/-668) and M6(-300/-668), from top to bottom. The numbers at the ends of the promoter parts refer to base pairs in relation to translation initiation site of the LH[beta] subunit gene. The dashed line represents deletion between nucleotides -8 and -299. The asterisks in the V/M (variant) sequences indicate the number and relative positions of the point mutations in comparison to the wt sequence.
Figure 3. The relative expression levels (mean ± SEM) of the wt- and V-LH[beta] promoter constructs in L[beta]T2 (n = 29) and HEK 293 cells (n = 16). In each case, the mean expression level of the wild-type construct was taken as 100%. *** P < 0.001, versus wt.
Figure 4. Expression of the different deletion mutants of the human LH[beta] promoter/luciferase reporter genes in cultured L[beta]T2 (A) and HEK 293 cells (B). The mean level of expression of the wild-type construct wt(-8/-668) was assigned the value of 100%, and the activities of the other constructs are expressed relative to this. The results are the means ± SEM of three to eight experiments in triplicate. * P < 0.05; ** P < 0.01; *** P < 0.001 versus the wt(-8/-668) construct. The different letters above the bars indicate that these groups differ from each other at least at P < 0.05.
To study effects of the individual mutated nucleotides on LH[beta] promoter activity, two mutants were used: M2(-8/-668) with two nucleotide changes at -238 and -276, and M6(-8/-668) with six nucleotide changes at -489, -490, -504, -506, -525 and -552 (Fig. 2). In L[beta]T2 and HEK 293 cells both constructs displayed higher promoter activity than the wt(-8/-668) promoter control (Fig. 4), indicating that both the proximal two and distal six mutations contribute to the overall increased activity of the V-LH[beta] promoter.
Effects of sex steroids on the LH[beta] promoter activity
Progesterone increased the activity of the V-LH[beta] promoter by 24.6 ± 4.0% (P < 0.01) in L[beta]T2 cells and by 16.3 ± 0.1% in HEK 293 cells (P < 0.01; data not shown), but not that of the wt-LH[beta] gene. Estradiol and 5[alpha]-dihydrotestosterone (DHT) did not affect the activities of either of the two promoters (Fig. 5).
Figure 5. Regulation of the human LH[beta] promoter activity in cultured L[beta]T2 cells. The cells were transfected using the 661 bp wt- (A) or variant-LH[beta] (B) promoter/luciferase reporter constructs, and the CMV-[beta]-galactosidase plasmid, to control for transfection efficiency. After 24 h, the cells were cultured for another 24 h in serum-free medium in the absence (control) or presence of DHT (20 nmol/l), estradiol (E; 2 nmol/l) or progesterone (P; 100 nmol/l). The cells were harvested and the lysates were subjected to luciferase and [beta]-galactosidase measurements. The results shown are the luciferase/[beta]-galactosidase activity ratios in relation to the control samples (mean 100%). The results are the means ± SEM of three to four independent experiments in triplicate. *** P < 0.001 versus control.
Effects of pulsatile GnRH administration on the LH[beta] promoter activity
L[beta]T2 cells transfected with the wt- and V-promoter/luciferase reporter constructs were exposed to six 10 nmol/l GnRH pulses of 15 min duration, with 90 min interpulse interval. Figure 6 shows that GnRH increased the luciferase activity 2 to 3.5-fold in cells expressing the wt construct and 3 to 4-fold in those with the mutant construct, although the fold increase of luciferase activity over baseline by GnRH appeared to be similar with both promoter types.
Figure 6. Effect of pulsatile GnRH treatment on human LH[beta] subunit promoter activity in transfected L[beta]T2 cells. The cells were transfected with the wt or variant promoter/luciferase reporter constructs, and subsequently treated with six pulses of GnRH (10 nmol/l) for 15 min with 90 min intervals. The control samples were treated with similar pulses of medium. The data represent the means ± SEM of five independent experiments in triplicate. *** P < 0.001 versus control (no GnRH).
Phorbol ester, forskolin and 8-bromo-cAMP stimulation of the LH[beta] promoter activity
The phorbol ester TPA had an effect on function of the LH[beta] promoter in L[beta]T2 cells by significantly increasing the luciferase activity evoked by the full-length wt and variant promoters, as well as by constructs M2(-8/-668), M6(-8/-668), wt(-8/-299) and M2(-8/-299) (Fig. 7). In contrast, with constructs wt(-300/-668) and M6(-300/-668), the promoter activity was decreased after TPA treatment (Fig. 7). When tested in HEK 293 cells, TPA increased the expression level of the wt-LH[beta] promoter construct, but not that of the variant promoter (Fig. 8). Forskolin (50 µmol/l) and 8-bromo-cAMP (100 µmol/l) suppressed the expression of the V-LH[beta] promoter, but not that of wt-LH[beta] in L[beta]T2 cells (Fig. 8). In HEK 293 cells, forskolin increased the expression level of the wt- and V-LH[beta] promoter construct (Fig. 8). Upon forskolin and 8-bromo-cAMP stimulation, it was observed that also the expression of the control plasmid, cytomegalovirus (CMV)/[beta]-galactosidase, was increased. Therefore, in these experiments, the luciferase data were corrected for transfection efficiency by protein contents of the cell lysates. Although these data indicate differences between function of the two promoter sequences, their limited nature (with respect to dose- and time-dependence) preclude strong conclusions on their physiologic significance.
Figure 7. TPA stimulation of human LH[beta] promoter activity in designated deletion mutants, driving the luciferase reporter gene in transfected L[beta]T2 cells. The cells were transfected with the different deletion constructs (Fig. 2), and cultured for 24 h. Thereafter, the cells were cultured for an additional 24 h in serum-free medium in the absence (control, open bars) or presence (filled bars) of 1 µmol/l TPA. In each case the mean of the control level was assigned a value of 100%. The bars represent the means ± SEM of three to four independent experiments in triplicates. * P < 0.05; ** P < 0.01; *** P < 0.001 versus control. The different letters above the filled bars indicate significant differences (P-value at least <0.01) between various constructs in response to TPA.
Figure 8. Regulation of the human LH[beta] promoter activity in transfected L[beta]T2 (A and B) and HEK 293 (C and D) cells by forskolin (50 µmol/l), and 8-bromo-cAMP (100 mmol/l) (A and B) or TPA (100 mmol/l) (C and D). The cells were cotransfected with the 661 bp wt (A and C) or variant (B and D) promoter/luciferase reporter and pCMV-[beta]-galactosidase plasmids. After 24 h, the cells were incubated further in serum-free medium, in the absence (control) or presence of forskolin, 8-bromo-cAMP or TPA. After 24 h, the cells were harvested and the cell lysates were collected for measurement of luciferase and [beta]-galactosidase activities. Since the latter was increased by forskolin, the final correction of the luciferase activity was done per mg protein in the cell lysates. The results shown are the ratios of luciferase activity/mg protein in relation to control samples (mean 100%). The results represent the means ± SEM of three to five independent experiments in triplicate. ** P < 0.01; *** P < 0.001 versus control.
DISCUSSION
Numerous mutations have recently been discovered in the gonadotropin receptor genes (15-17), but those of the ligand hormones are still rare (16,18). No naturally occurring mutations of the common [alpha]-subunit gene have yet been described. It is possible that such a mutation would be lethal in the human, due to simultaneous blockade of synthesis of hCG, TSH and gonadotropins, although a mouse model with a disrupted [alpha]-subunit gene appeared to be viable (19). One inactivating LH[beta] mutation has been described, where a single base substitution in exon 3 of the gene changed codon 54 from glutamine to arginine (20). The male patient presented with hypogonadism, which was found to be due to synthesis of an LH form that was immunologically active but totally devoid of bioactivity. Consequently, the affected homozygous male had an absence of Leydig cells and a lack of spontaneous puberty. There was impaired androgen production and infertility in male heterozygotes of his family, but the female heterozygotes had normal sexual development and fertility.
We have previously described a healthy woman with an immunologically anomalous form of LH (3), which was not detectable by a monoclonal antibody (Mab) directed against a specific epitope in the intact [alpha]/[beta] dimer. Her FSH and TSH levels were normal, which indicated that there is likely a structural alteration in her LH[beta] subunit that affects the interaction with the Mab. The LH[beta] subunit gene of this subject was sequenced, and two mutations in the N-terminal region were identified (21,22). The first mutation in codon 8 (TGG->CGG) changes tryptophan to arginine, and the second in codon 15 (ATC->ACC) changes isoleucine to threonine. In addition, the latter change introduces an extra glycosylation signal for oligosaccharide attachment to Asn13. Analysis of genetic origin of the V-LH[beta] allele confirmed an autosomal recessive mode of inheritance in the proband's family (3). Suganuma et al. (7) recently reported that the mutation in codon 8 (Trp->Arg) may be more important for the anomalous immunoreactivity of the V-LH molecule.
Very recently, our survey of occurrence of V-LH in various populations demonstrated that it is a common polymorphism with carrier frequency varying from 0 to >50% (9,22). There are several reports of association of V-LH with female menstrual disturbances, infertility and subfertility (4,12; D. Cramer and I. Huhtaniemi, unpublished data). In addition, we have found in healthy women association of V-LH with elevated serum levels of estradiol, testosterone and sex-hormone binding globulin in the follicular phase of the menstrual cycle (8), and in boys with delayed progression of puberty (10). In obese women with PCOS, the V-LH[beta] allele occurs with significantly lesser frequency than in control subjects (11). Thus, there is ample evidence that the action of LH in subjects homo- or heterozygous for the V-LH[beta] allele differs from that of individuals with wt-LH. In accordance, V-LH has elevated bioactivity in vitro, but significantly shorter half-life in circulation, as compared with wt-LH (6,7; L. Joshi, P. Manna and I. Huhtaniemi, unpublished data). It is therefore apparent that the kinetics of LH action after pulses of wt- and V-LH differs and could explain the differences observed in gonadal function. As a further mechanism of the observed differences in LH action, we hypothesized that additional changes could be found in structure and function of the V-LH[beta] promoter.
In this study, we isolated and sequenced the 661 nucleotide region (-8/-668) of the V-LH[beta] gene, and within this sequence, eight point mutations were found. In transient transfection studies, the activity of the mutant LH[beta] promoter was ~40% higher than that of the wild-type gene. Since V-LH has significantly shorter half-life in circulation (see above), the higher transcriptional activity of the V-LH[beta] gene may thus provide, as hypothesized, a compensatory mechanism to maintain the largely normal LH levels that are measured in individuals homo- or heterozygous for the V-LH[beta] allele (3,6,8,9). Due to its increased in vitro bioactivity, but decreased half-life in circulation (6), the short but highly active pulses of V-LH could cause the changes observed in gonadal function (see above).
The LH[beta] and CG[beta] subunit genes are clustered together and form an array of tandem and inverted copies in chromosome 19 (2). The LH[beta] and CG[beta] genes are highly homologous (~94%) in both their coding and non-coding regions. However, they possess unique patterns of tissue expression and in locations of their transcriptional start sites and regulatory regions (23-25). The LH[beta] gene is expressed in gonadotropes, and it is more dependent upon GnRH than on the activin/inhibin system.Many studies have focused on GnRH as the primary regulator of the gonadotropin genes (26-29). The regulation of gonadotropin gene expression is influenced by the pattern of GnRH exposure, and some degree of GnRH input is necessary for expression of each of the gonadotropin subunit genes. The individual gonadotropin genes respond differentially to the frequency of GnRH pulses (26); LH[beta] mRNA rises preferentially by rapid frequency GnRH stimulation, whereas FSH[beta] responds well to a lower frequency GnRH pulses. When pulsatile GnRH is applied to pituitary cells, LH[beta] gene transcription is increased (27-29), indicating the presence of GnRH-responsive elements in the LH[beta] promoter. In our study, when GnRH was applied in pulsatile fashion to the L[beta]T2 cells expressing wt- and V-LH[beta] promoter-driven reporter genes, both of them responded similarly, when the enhanced basal activity was taken into account. Hence, of the various factors that regulate LH[beta] expression, GnRH action was not altered by the mutations present in the V-LH[beta] promoter.
The regulatory and hormone-responsive elements of the LH[beta]/CG[beta] genes are not as well characterized as those of the C[alpha] promoter (29). cAMP regulates transcription of the human [alpha]-subunit gene, and an 18 bp repeated palindromic sequence in its 5[prime]-flanking region is responsible for this effect (30-32). Also theratLH[beta] gene is cAMP responsive through several complex protein interactions within its promoter/enhancer sequence (33). The 661 bp human wt and V-LH[beta] promoters were not found to contain exact tandem repeat palindromic cyclic response element (CRE; TGACGTCA) (34), or AP-1 [TPA-REs; TGA(G/C)T(C/A)A] recognition sites (35), although there were several sequences in both promoters with one mismatch (Table 1). Luciferase activity, driven by the V-LH[beta] promoter, was reduced in L[beta]T2 cells by forskolin (an agent increasing intracellular cAMP) and 8-bromo-cAMP (a cAMP analogue), but not that driven by the corresponding wt promoter. These are apparently indirect responses to cAMP, through cis-acting elements present in the V-LH[beta], but not in the wt promoter. In HEK 293 cells, forskolin treatment resulted in increased expression level of both wt- and V-LH[beta] promoters. The different responses found between the two cell types could be due to differences in trans-acting regulatory elements.
Table 1. Comparison of putative transcription factor binding sites in the human wt- and V-LH[beta] promoters
| wt-LH[beta] promoter (-8/-668) | V-LH[beta] promoter (-8/-668) | Consensus sequence for comparison | Reference |
| TGAGGTCA (-232/-225) | TGAGGTCA (-232/-225) | CRE binding site TGACGTA | (34) |
| TGAGTCT (-240/-234) | TGGGTCT (-240/-234) | AP-1 binding sites (TPA-REs) [TGA(G/C)T(C/A)A] | (35) |
| TGGCTTC (-495/-489) | TGGCTAA (-495/-489) | ||
| TGAGTTA (-630/-624) | TGAGTTA (-630/-624) | ||
| CCAGATGCCC (-502/-511) | CCTGGTGCCC (-502/-511) | GATA-1 binding sites [(G/C)(G/C/A)NGAT(A/G/T)G(G/C/T)(G/C/T)] | (36) |
| CCAGATGCCC (-502/-511) | CCTGGTGCCC (-502/-511) | GATA-2 binding sites [(G/C/T)(A/G/C)NGAT(A/G)G(G/C/T)(G/C/T)] | (36) |
| TGGCCATGT (-136/-128) | TGGCCATGT (-136/-128) | SF-1 binding site TGACCTTGT | (37-39) |
| TGGCCTTGC (-66/-58) | TGGCCTTGC (-66/-58) | ||
| CGCCCCCGG (-119/-111) | CGCCCCCCGG (-119/-111) | NGFI-A (Egr-1) binding site CGCCCCCGC | (39,40) |
| CGCCCCCAC (-57/-49) | CGCCCCCAC (-57/-49) | ||
| TAATCC (-525/-520) | GAATCC (-525/-520) | Ptx-1 binding site TAATCC | (40,41) |
| TAATCC (-104/-109) | TAATCC (-104/-109) | ||
| GGGCGG (-562/-557) | GGGCGG (-562/-557) | Sp1 binding site GGGCGG | (42,43) |
Protein kinase C has been established to participate in LH synthesis and secretion, and it is involved in actions of GnRH upon these processes (44). In this study, TPA affected function of the LH[beta] promoter in L[beta]T2 cells by increasing the luciferase activity of the full-length wt and V promoter constructs, as well as those deleted at the 5[prime] end [wt(-8/-299) and M2(-8/-299)]. In contrast, with the 3[prime] end deletion constructs [wt(-300/-668) and M6(-300/-668)], the promoter activity was suppressed by TPA treatment. This indicates that there is a weak inhibitory element between -668 and -299 bp of the LH[beta] promoter, and the region between -299 and -8 bp is probably responsible for the TPA stimulation. In HEK 293 cells, TPA increased the expression level of the wt-LH[beta] promoter construct, but not that of the variant promoter.
When the TFSEARCH and the TRANSFAC Database (45) were used to search for transcription factor binding sites, several of them (AP-1, GATA-1, GATA-2; Table 1) were found to have differences between the wt- and V-LH[beta] promoter sequences. GATA-1 and GATA-2 binding sites of the wt-LH[beta] promoter were closer to the respective consensus sequences (36) than those of the V-LH[beta] promoter. In contrast, one of the Ptx-1 binding sites (-526/-521) had one mismatch in the variant promoter (Table 1). By and large, the search for differences in putative transcription factor recognition sites revealed no apparent alterations that would have explained the functional differences of the two promoter sequences.
In conclusion, the present data clearly demonstrate that the V-LH[beta] promoter is more active than that of the corresponding wt gene, and there are clear-cut qualitative differences between them in response to hormonal stimuli. Admittedly, these comparative data on functional responses of the wt- and V-LH[beta] promoters remain partly descriptive in nature and are hard to fit into a generalized picture. Likewise, some of the functional differences between the two cell lines lack specific explanation. Nevertheless, the differences between the two promoters are in good agreement with the documented functional differences between wt- and V-LH proteins. The differences in pituitary-gonadal function between individuals homozygous for the wt-LH[beta] allele and carriers of the V-LH[beta] gene may be explained by the qualitative and quantitative differences in responses of the two promoters to hormonal stimuli. The current observations also propose a novel evolutionary mechanism whereby polymorphic changes in the coding sequence of a gene could be compensated for by additional mutations in the corresponding promoter sequence. Whether genotyping for the LH[beta] allele can be used as a diagnostic or prognostic factor for pathologies related to gonadal function (e.g. PCOS, infertility, hormone-dependent cancer and disorders in pubertal maturation) remains to be studied.
MATERIALS AND METHODS
Amplification and sequencing of the V-LH[beta] promoter
The 5[prime]-flanking region (nucleotides -8 to -668, in relation to translation initial codon) of the V-LH[beta] gene was retrieved from human genomic DNA by PCR. The primers were designed based on the known sequence of the human LH[beta] promoter and selected with specific mismatches in order to discriminate between LH[beta] and the highly homologous hCG[beta] genes (13,14). The upstream primer, 5[prime]-CGGGGGCAAGACACGCA (17mer), extended from position -668 to -652 of the wt-LH[beta] subunit promoter, and the downstream primer, 5[prime]-GCATCCCCTGCCTCGTGTAT-3[prime] (20mer), from position -8 to -27. The amplified DNA thus spanned a 661 bp region of the LH[beta] promoter. Genomic DNA was purified by a non-enzymatic isolation method (46) and amplified in a total reaction volume of 50 µl, containing Taq polymerase (1 U), deoxynucleotide triphosphates (0.4 mmol/l of each), and both forward and reverse primers (0.5 µmol/l) in final buffer containing KCl (50 mmol/l), Tris-HCl (10 mmol/l, pH 8.8), Triton X-100 (0.1%) and MgCl2 (1.5 mmol/l). Thirty-five PCR cycles were performed as follows: denaturation (96°C, 1 min), annealing (68°C, 1 min) and extension (72°C, 2 min). The Taq DNA polymerase (DyNAZyme II; Finnzymes OY, Espoo, Finland) was added after the initial denaturation step (5 min). DNA from the PCR reaction was purified by using electrophoresis on an agarose gel and then following the QiaEx II Agarose Gel Extraction Protocol (Qiagen, Hilden, Germany). The PCR primers were used as sequencing primers with the dideoxy chain termination method. Three forward primers and five reverse primers, with specific mismatches as compared with the hCG[beta] sequences (13,14), were designed to amplify fragments of different lengths, and to cover the promoter region of the LH[beta] gene. The genotypes were confirmed by sequencing from several individual PCR runs using samples of individuals of different ethnic origins.
Plasmid preparations
The 661 bp promoter region of the LH[beta] gene, amplified by PCR, was purified by electrophoresis on agarose gel and with the QiaEx II Agarose Gel Extraction kit. Initially, these 661 bp promoter regions of the LH[beta] gene, consisting of the respective sequences of the wt- and V-LH promoters, were inserted into the pGEM-T vector (Promega, Madison, WI), and then subcloned into the promoterless firefly luciferase reporter vector. All the LH[beta] promoter deletion mutants were constructed by using specific restriction endonuclease recognition sites of this sequence. To check the effects of the different mutant nucleotide sequences on expression levels of the luciferase reporter gene in L[beta]T2 and HEK 293 cells (see below), three expression constructs of the mutant promoter were prepared (Fig. 2). The first [V(-8/-668)] contained the 661 bp (-8/-668) promoter fragment with all of the eight mutations (-238, -276, -489, -490, -504, -506, -525, -552). Construct M2(-8/-668) contained two mutations at positions -238 and -276, and construct M6(-8/-668) carried six mutations at nucleotides -489, -490, -504, -506, -525 and -552, in relation to the LH[beta] translation start site.
To further study the cis-elements of the LH[beta] promoter, another pair of expression constructs were designed to contain a 5[prime] end deletion of either the wt [wt(-8/-299)] or variant [M2(-8/-299)] sequence (Fig. 2). The other pair of constructs were truncated at the 3[prime] end, i.e. wt(-300/-668) and M6(-300/-668) (Fig. 2). The structural identity of the LH[beta] promoter/luciferase constructs was verified by specific restriction endonuclease digestion analyses and sequencing. All DNA constructs were purified by Qiagen Plasmid Purification kit and dissolved into sterile distilled deionized H2O for transfection experiments. The CMV promoter-driven [beta]-galactosidase gene was co-transfected in all experiments as the internal control of transfection efficiency (47).
Cell cultures, transient transfections, luciferase and [beta]-galactosidase assays
To study the basal activities of the wt- and V-LH[beta] promoter sequences, a novel immortalized mouse pituitary cell line, L[beta]T2, known to express the LH[beta] subunit gene (48), was cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco BRL, Paisley, UK) with 4.5% glucose, 10% heat-inactivated fetal calf serum (FCS; Bioclear, Berkshire, UK), penicillin (50 × 103 U/l; Biological Industries, Haemek, Israel), and streptomycin (50 mg/l; Biological Industries) in a humidified 5% CO2 incubator at 37°C.
The gene constructs were transfected into L[beta]T2 cells using lipofectamine (Gibco BRL, Life Technologies, Gaithersburg, MD). Before transfections (18-24 h), the L[beta]T2 cells were plated into six-well tissue culture plates (Greiner, Frickenhansen, Germany) at 1 × 106 cells per well in 2 ml of the appropriate complete growth medium. The L[beta]T2 cells were cotransfected with 2 µg of the LH[beta] promoter/luciferase gene construct and 1 µg of the internal control plasmid, CMV-[beta]-galactosidase (pCMV-lac). Lipofectamine (10 ml) was used in each transfection reaction, according to the manufacturer's instructions. Forty-eight hours after starting the transfection, the L[beta]T2 cells were washed twice with phosphate-buffered saline (PBS, pH 7.5; Gibco BRL), harvested by scraping off into 60 µl of buffer containing 150 mmol/l NaCl, 40 mmol/l Tris-HCl and 1 mmol/l EGTA, pH 7.5, at 25°C. After centrifugation at 5000 g (5 min, 4°C), the cell pellets were resuspended in extraction buffer (100 mmol/l potassium phosphate, 1 mmol/l EGTA, 1 mmol/l DTT and 0.2% Saponin, pH 7.8) at 25°C, followed by three cycles of freezing-thawing.
A non-endocrine cell line, HEK 293 cells, was used as the control of gonadotrope specificity of the expression. The HEK 293 cells were cultured in DMEM/F12 (Gibco BRL), supplemented with 10% heat-inactivated FCS and 100mg/l gentamycin or penicillin (50 × 103 U/l)/streptomycin (50 mg/l), and transfected by the electroporation method (49). A total of ~3 × 106 HEK 293 cells in serum- and antibiotic-free medium were electroshocked with 280 V/960 µFD in Genepulser cuvettes (0.4 cm gap; Bio-Rad, Richmond, CA). Thereafter, the electroporated cells were plated in 6 cm diameter culture dishes (Greiner), and incubated for 48 h before harvesting and measuring the luciferase and [beta]-galactosidase activities (as above). All the transfection experiments were carried out in triplicate, and in at least three independent experiments.
Luciferase activity was measured using the 125I Luminometer (Bio-Orbit, Turku, Finland). The [beta]-galactosidase activity was measured by spectrophotometry, in order to correct the luciferase activities measured for transfection efficiency (luciferase/[beta]-galactosidase). The results are reported as means ± SEM of at least three independent experiments.
Trans-activation of the wt- and V-LH[beta] promoter
To investigate the regulation of the LH[beta] promoter activity, 1 × 106 L[beta]T2 cells were transfected by lipofectamine (see above) with 2 µg of one of the chimeric LH[beta] gene constructs and 1 µg pCMV/[beta]-galactosidase. Twenty-four hours later, the media were changed into serum-free medium containing designated concentrations of estradiol (2 nmol/l), progesterone (100 nmol/l) or DHT (20 nmol/l). All the steroids (purchased from Sigma, St Louis, MO) were dissolved as stock solutions at high concentration of absolute ethanol, then diluted with the culture medium. The final ethanol concentration was adjusted to 0.01% in all samples and controls. Following a 24 h culture in the presence of one of the steroids, the cells were harvested as described above. The pulsatile treatment with gonadotropin-releasing hormone (GnRH; Sigma) was carried out in L[beta]T2 cells transfected with wt- and V-LH[beta] promoter constructs. The cells were challenged with six pulses of 10 nmol/l GnRH of 15 min duration and with interpulse interval of 90 min, after a change of fresh medium. GnRH was dissolved, and the L[beta]T2 cells were cultured, in DMEM containing 10% charcoal-treated FCS (50-52). For experiments using the phorbol ester TPA (0.1 or 1 µmol/l; Sigma) and forskolin (50 µmol/l; Sigma), the agents were added in the medium 24 h before harvesting and measurement of reporter gene activity.
Computer programs and statistical analysis
The alignment program (DNASTAR, Madison, WI) was used for DNA sequence comparison. TFSEARCH: Searching Transcription Factor Binding Sites (v.1.3; Yutaka Akiyama, Kyoto University Real World Computing Partnership, Japan) and the TRANSFAC Database (-ASCII flat file release 3.3; GBF-Branuschweig, Germany) were used to search for transcription binding sites.
The data are expressed as means ± SEM. Statistically significant differences between groups were determined by one-factor ANOVA (a Macintosh version of the super-ANOVA Program; Abacus Concepts, Berkeley, CA), followed by Duncan's new multiple range test. A P-value <0.05 was considered statistically significant.
ACKNOWLEDGEMENTS
Dr Matti Poutanen (Department of Physiology, University of Turku) is acknowledged for helpful discussions. We thank Dr P.L. Mellon (University of California San Diego) for donation of the L[beta]T2 cells. This work was supported by grants from the Academy of Finland and the Sigrid Jusélius Foundation.
REFERENCES
+To whom correspondence should be addressed. Tel: +358 2 3337579; Fax: +358 2 2502610; Email: ilpo.huhtaniemi{at}utu.fi
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Humaidan Are endogenous LH levels during ovarian stimulation for IVF using GnRH analogues associated with the probability of ongoing pregnancy? A systematic review Hum. Reprod. Update, May 1, 2006; 12(3): 325 - 326. [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Ahda, J. Gromoll, A. Wunsch, K. Asatiani, M. Zitzmann, E. Nieschlag, and M. Simoni Follicle-Stimulating Hormone Receptor Gene Haplotype Distribution in Normozoospermic and Azoospermic Men J Androl, July 1, 2005; 26(4): 494 - 499. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. De Placido, C. Alviggi, A. Perino, I. Strina, F. Lisi, A. Fasolino, R. De Palo, A. Ranieri, N. Colacurci, A. Mollo, et al. Recombinant human LH supplementation versus recombinant human FSH (rFSH) step-up protocol during controlled ovarian stimulation in normogonadotrophic women with initial inadequate ovarian response to rFSH. A multicentre, prospective, randomized controlled trial Hum. Reprod., February 1, 2005; 20(2): 390 - 396. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Jiang, M.-L. Savontaus, H. Simonsen, C. Williamson, R. Mullenbach, J. Gromoll, N. Terwort, M. Alevizaki, and I. Huhtaniemi Absence of the genetic variant Val79Met in human chorionic gonadotropin-beta gene 5 in five European populations Mol. Hum. Reprod., October 1, 2004; 10(10): 763 - 766. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. V. Vasilyev, M. A. Lawson, D. Dipaolo, N. J. G. Webster, and P. L. Mellon Different Signaling Pathways Control Acute Induction versus Long-Term Repression of LH{beta} Transcription by GnRH Endocrinology, September 1, 2002; 143(9): 3414 - 3426. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Gazvani, P. Pakarinen, P. Fowler, S. Logan, and I. Huhtaniemi Lack of association of the common immunologically anomalous LH with endometriosis Hum. Reprod., June 1, 2002; 17(6): 1532 - 1534. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Jiang, T. Lamminen, P. Pakarinen, J. Hellman, P. Manna, R. J. Herrera, and I. Huhtaniemi A novel Ala-3Thr mutation in the signal peptide of human luteinizing hormone {beta}-subunit: potentiation of the inositol phosphate signalling pathway and attenuation of the adenylate cyclase pathway by recombinant variant hormone Mol. Hum. Reprod., March 1, 2002; 8(3): 201 - 212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. R. Manna, L. Joshi, V. N. Reinhold, M. L. Aubert, N. Suganuma, K. Pettersson, and I. T. Huhtaniemi Synthesis, purification and structural and functional characterization of recombinant form of a common genetic variant of human luteinizing hormone Hum. Mol. Genet., February 1, 2002; 11(3): 301 - 315. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. De Placido, A. Mollo, C. Alviggi, I. Strina, M. T. Varricchio, A. Ranieri, N. Colacurci, A. Tolino, and M. Wilding Rescue of IVF cycles by HMG in pituitary down-regulated normogonadotrophic young women characterized by a poor initial response to recombinant FSH Hum. Reprod., September 1, 2001; 16(9): 1875 - 1879. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Bergendah and J. D. Veldhuis Is there a physiological role for gonadotrophin oligosaccharide heterogeneity in humans?: III. Luteinizing hormone heterogeneity: a medical physiologist's perspective Hum. Reprod., June 1, 2001; 16(6): 1058 - 1064. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.H. Nilsson, M. Kaleva, H. Virtanen, A.M. Haavisto, K. Pettersson, and I.T. Huhtaniemi Disparate response of wild-type and variant forms of LH to GnRH stimulation in individuals heterozygous for the LH{beta} variant allele Hum. Reprod., February 1, 2001; 16(2): 230 - 235. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. P. N. Themmen and I. T. Huhtaniemi Mutations of Gonadotropins and Gonadotropin Receptors: Elucidating the Physiology and Pathophysiology of Pituitary-Gonadal Function Endocr. Rev., October 1, 2000; 21(5): 551 - 583. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||