The Human MC4R Promoter

RESEARCH DESIGN AND METHODS

Human fetal brain RNA preparation and 5′RACE.

Transcriptional start site determination by 5′ rapid amplification of cDNA ends (5′RACE) was performed on total human fetal brain RNA using the GeneRacer kit (Invitrogen, Carlsbad, CA). First-strand cDNA was synthesized using specific primer MC4R-BR (5′ TCCACTGCAATTGAAAGCAG 3′) and avian myeloblastosis virus–reverse transcriptase. We amplified 5′ cDNA ends with MC4R-BR or MC4R-AR primer (5′ CATGGGTGAATGCAGATTCT 3′), followed by a nested PCR with MC4R-AR or MC4R-OR (5′ GCAGGAGAAGTGTGCATCCCA 3′). The PCR products were subcloned in pcDNA3.1/V5-His TOPO TA vector (Invitrogen) and analyzed by enzymatic digestion using a frequent cutter restriction enzyme (HphI) to eliminate duplicate clones. We then systematically sequenced 2 clones presenting with the same restriction map (16 clones sequenced).

Alignment of human and mouse proximal promoter.

Alignments of mouse and human 5′untranslated regions (UTRs) and regions 500 bp upstream of the main transcription start site were performed and analyzed with MacVector software (Accelrys, San Diego, CA), TRANSFAC (the Transcription Factor Database), and manual searches.

Cloning of 5′ flanking MC4R luciferase reporter vectors.

Fragments of the 5′ region of the human MC4R gene were obtained by restriction digestion of a bacterial artificial chromosome clone isolated from a human genomic bacterial artificial chromosome library (Research Genetics; Invitrogen). Fragments were ligated upstream of the Firefly luciferase gene in the plasmid pFOXLuc1 (a gift from Michael German, University of California San Francisco [UCSF]).

Cell transfection and luciferase assays.

Mouse hypothalamic GT1-7 cells were provided by Richard Weiner (UCSF). Neuro2A, HEK293, and GT1-7 cells were transfected using Effectene reagent (Qiagen, Valencia, CA), and NIH3T3 cells were transfected with Lipofectamine Reagent (Invitrogen). Cells were cotransfected with MC4R promoter constructs and a plasmid containing the Renilla luciferase to assess for transfection efficiency. Forty-eight hours after transfection, cells were harvested and Firefly and Renilla luciferases were assessed using the Stop & Glo kit (Promega, Madison, WI). Firefly luciferase activity was normalized to Renilla luciferase activity, and results were expressed relative to the result obtained with the promoterless vector.

Human subjects and phenotypes.

Two cohorts were screened for mutations in the MC4R coding sequence and minimal promoter. The first group was comprised of 266 children (aged 12.8 ± 0.8 years) from a previously described cohort (14). The protocol was approved by Cochin Ethics Committee, and informed consent was obtained from subjects and parents.

The second population was comprised of 165 unrelated, severely obese adults (BMI 40–85.3 kg/m² [median 48.8]). This study was approved by the UCSF Committee on Human Research, and informed consent was obtained from subjects.

Mutation screening of the MC4R minimal promoter and 5′UTR.

Mutation screening was performed by direct sequencing of the coding region (9), the 5′UTR, and the minimal promoter of MC4R. This region was amplified from lymphocyte-derived genomic DNA using primers MC4R-250 (5′ GCTATAGGTACCCTTGAAAACTTAAAAAGG GA 3′) and MC4R PromRev1 (5′ GCTATACTCGAGAATTCCAGTGTCCCCC TGA 3′). Sequencing was performed on a ABI3700 automated DNA Sequencer (Applied Biosystems, Foster City, CA) and analyzed using Sequencher software (Gene Codes, Ann Arbor, MI).

RESULTS

Determination of the human MC4R transcriptional start site.

We determined transcriptional start sites of the MC4R gene from human fetal brain RNA by 5′RACE (Fig. 1). The nucleotide sequences of all the clones were colinear with the genomic sequence in the 5′ end of the coding region of human MC4R, indicating that no introns were present in the 5′UTR. Analysis of human fetal brain by 5′RACE indicates that the MC4R gene possesses two major transcriptional start sites, at 426 and 139 bp upstream of the translation initiation codon, and multiple minor transcription start sites, ranging from 213 to 366 bp (Fig. 2). We designated the transcription start site of the longest transcript as nucleotide “+1”. A TATA-like box and classical CAAT sequence (in an antisense orientation) were found at 33 and 94 bp upstream of this major transcription start site, respectively.

In silico analysis of the proximal human MC4R promoter region reveals the presence of E-boxes, cAMP response element binding protein, GATA and signal transducers and activators of transcription sites, homeoboxes, and a response element for the neural zinc finger family of transcription factors (Fig. 2).

Comparative genomic analysis of the mouse (15) and human MC4R promoter regions indicates that only the signal transducers and activators of transcription site, neural zinc finger response element, and CAAT box are conserved between species. In addition, this comparative analysis reveals the presence of a 30-bp sequence starting with the putative CAAT box that is 100% conserved between human and mouse (Fig. 2). This sequence is also present in the putative MC4R promoter of the rat and pig and is highly specific for the MC4R promoter, since BLAST analysis does not reveal the presence of another similar sequence in all four genomes. This sequence contains a potential binding site (AttGGTCA) for monomeric nuclear receptors, such as thyroid receptor or SF1 (steroidogenic factor 1). However, we did not observe any effect of SF1 on MC4R promoter activity (data not shown).

Characterization of the essential promoter region for MC4R gene transcription.

A series of progressive 5′ deletions of the human MC4R promoter, each extending to +10 bp of 5′UTR, were tested for transcriptional activity in nonneuronal cell lines (NIH 3T3 and HEK293) and in neuronal cell lines (Neuro 2A and GT-1-7) (Fig. 3). GT1-7 cells express the MC4R (16). A 2.5-kb fragment of human MC4R promoter stimulates transcription in all of the cell lines, with the highest relative level of expression in HEK 293 cells (Fig. 3). Serial deletions from 2.5 kb to −130 bp did not diminish the promoter activity in any of the cell lines (Fig. 3), while deletion of 80 additional base pairs abolished the transcriptional activity of the promoter in NIH 3T3, Neuro 2A, and GT1-7 cells and reduced it in HEK 293 cells. Therefore, we defined the presence of a minimal core promoter between −130 and −50 bp. However, this minimal promoter does not recapitulate tissue specificity because we observed a basal transcription activity in all cell lines tested. This result suggests that the regulation of the MC4R gene expression involves at least an additional repressor elsewhere in the sequence. Because they would lead to an increase in MC4R transcription, mutations in such tissue-specific repressors are unlikely to be a cause of severe obesity in humans.

Are MC4R promoter mutations implicated in human obesity?

To determine whether mutations in essential regulatory regions of the MC4R are a significant cause of severe obesity, we screened two populations of severely obese subjects for mutations in the coding region and in the minimal MC4R promoter defined above (Table 1). The frequency of MC4R mutations, other than known polymorphisms, was 3.6% (95% CI 0.78–6.49) in the adult population and 1.5% (0.04–2.97) in the child population. Both results are consistent with previous results obtained in this laboratory and with the literature. We detected three variants in the 5′UTR of MC4R. The −178 A/C single nucleotide polymorphism (SNP) was present at a frequency of 5.57% (3.4–7.73) and has been found previously at the same frequency in nonobese control populations (17). Two rare variants (A/G and G/T at 176 and 360 bp upstream of the ATG, respectively) were found in two different subjects. In contrast, no genetic variation was observed in the minimal promoter of MC4R in the 431 obese subjects of this cohort.

DISCUSSION

We have used comparative genomic tools and classical promoter studies to begin to elucidate the essential regulatory mechanisms underlying human MC4R gene expression. The major transcriptional site, according to our 5′RACE result, is located 426 bp upstream of the start of translation, where sequence analysis indicates the presence of a consensus CAP site. The DNA sequence upstream of this site contains a CAAT box (in an antisense orientation) and a nontypical TATA box. This is in contrast to the mouse Mc4r promoter that lacks a classic TATA box (15). In addition to the mouse Mc4r promoter, the 5′ flanking regions have been described for two melanocortin receptor subtypes: the human MC1R (18) and human and mouse Mc2r (19,20). Neither of these contain a TATA box nor a CAAT box. While several studies reported that most GPCR promoters are TATA-less and GC-rich (21,22), it appears that the human MC4R promoter does not share these characteristics.

Cross-species comparison reveals the presence of an identical 30-bp sequence following the putative CAAT box in the human, mouse, rat, and pig MC4R promoter. Our results confirm that this sequence is part of the minimal region that is necessary for basal MC4R promoter activity. Further characterization of trans-acting factors interacting with this cis element may lead to the discovery of novel candidate genes for human obesity.

While mutations in the MC4R gene are well established as an infrequent cause of obesity in humans (3–10), it is plausible that common SNPs in regulatory regions of MC4R could also contribute to obesity. Indeed, among multiple studies searching for linkage to obesity phenotypes, some have provided suggestive evidence of linkage between BMI (23) and body fat (24,25) on chromosome 18q21, flanking the MC4R region. These reports suggest that in addition to the rare severe obesity–causing variants in the MC4R coding region, common genetic variants/SNPs in the regulatory regions of the MC4R gene may be contributing to the risk of developing obesity, whereas in our study, the frequencies of mutations in the coding sequence of MC4R and of common genetic variants in the 5′UTR were concordant with previous reports, our results indicate that genetic variants in the essential/minimal region of the MC4R promoter are not a common monogenic cause of human obesity. Further epidemiologic studies are necessary to rule out any effect of the common SNPs found in the promoter on obesity-related phenotypes.