The microphthalmia transcription factor (Mitf) controls expression of the ocular albinism type 1 gene: link between melanin synthesis and melanosome biogenesis

Abstract

Melanogenesis is the process that regulates skin and eye pigmentation. Albinism, a genetic disease causing pigmentation defects and visual disorders, is caused by mutations in genes controlling either melanin synthesis or melanosome biogenesis. Here we show that a common transcriptional control regulates both of these processes. We performed an analysis of the regulatory region of Oa1, the murine homolog of the gene that is mutated in the X-linked form of ocular albinism, as Oa1's function affects melanosome biogenesis. We demonstrated that Oa1 is a target of Mitf and that this regulatory mechanism is conserved in the human gene. Tissue-specific control of Oa1 transcription lies within a region of 617 bp that contains the E-box bound by Mitf. Finally, we took advantage of a virus-based system to assess tissue specificity in vivo. To this end, a small fragment of the Oa1 promoter was cloned in front of a reporter gene in an adeno-associated virus. After we injected this virus into the subretinal space, we observed reporter gene expression specifically in the retinal pigment epithelium, confirming the cell-specific expression of the Oa1 promoter in the eye. The results obtained with this viral system are a preamble to the development of new gene delivery approaches for the treatment of retinal pigment epithelium defects.

FIG. 1.

Conservation of Oa1 regulatory region during evolution. (A) Sequence comparison between the murine and human Oa1 genomic regions upstream of the ATG codon (in bold). Bioinformatic analysis of the murine Oa1 promoter identified putative binding sites for known transcription factors. Binding sites that are conserved in the murine and human sequences are indicated. The large arrow indicates the transcription start site in the mouse gene. Open arrows indicate restriction sites used for the generation of construct pOa−562/+55. (B) The first 68 bp upstream of the ATG codon of the murine Oa1 gene were compared by Clustal methods to the same genomic region in the human, rat, and fugu genomes. Bases that are conserved in the mouse and other species are shaded. The E-box is boxed.

FIG. 2.

Promoter analysis of Oa1 gene. (Left) Murine Oa1 promoter fragments were cloned upstream of the firefly luciferase reporter gene in the pGL-3-Basic vector. A schematic representation of putative transcription factor binding sites is shown for the longest element analyzed. (Right) Firefly luciferase activities were normalized with the R. reniformis luciferase that was cotransfected in each transfection experiment. The means and standard deviations from at least seven independent experiments are shown. RLU, relative light units.

FIG. 3.

Mitf transactivation of Oa1 promoter. (A) Cotransfection experiments in NIH 3T3 cells with pOa−3373/+55 construct and increasing doses of Mitf expression vector showing dose response to Mitf. (B) Mutagenesis of two bases in the E-box at position −28 (pOa−3373/+55mut1) completely abolishes the response to Mitf in vitro. Mutagenesis of two bases in the E-box at position −1972 (pOa−3373/+55mut2) does not impair the response to Mitf. (C) Wild-type and mutagenized promoter constructs were analyzed in different pigmented cell lines. The luciferase activity measured for the pOa−3373/+55 wild-type construct was considered to be 100%. (D) RT-PCR analysis of Oa1 expression in Mitf mutant mice (Mitf^mi-enu198) showing that Oa1 transcription depends on Mitf in the eye, as was also found for Tyr and Trp1, but not for Trp2.

FIG. 4.

Mitf binds the E-box close to the transcription start site of the Oa1 gene. (A) Sequences of oligonucleotides used for EMSAs. E-boxes are shown in bold, and the introduced mutations are shown in lowercase and underlined. (B) Gel shift analysis was performed with a B16-F10 murine melanoma nuclear extract. Competition experiments were performed by adding a 50-fold excess of unlabeled homologous (WT) or mutated oligonucleotides. The arrow indicates the supershift detected in the presence of an anti-Mitf monoclonal antibody. (C) Gel shift analysis of human OA1 promoter performed with nuclear extracts of MNT-1 human melanoma cells. Competition experiments were performed by adding a 50-fold excess of unlabeled homologous (WT) or mutated oligonucleotides. The arrow indicates the supershift detected in the presence of an anti-Mitf monoclonal antibody.

FIG. 5.

Chromatin immunoprecipitation. Chromatin immunoprecipitation was performed with material from 501mel cells with the primers specified in Materials and Methods. A region residing in the first intron was amplified as a control.

FIG. 6.

Tissue-specific expression driven by Oa1 promoter in vivo. C57/BL6 wild-type mice were injected subretinally with either AAV2/5-Oa1-EGFP (A), AAV2/5-CMV-EGFP (B), or AAV2/5-Oa1mut-EGFP (C). While the CMV promoter drives expression in RPE and photoreceptor cells (B, asterisk), the Oa1 promoter drives specific expression in RPE cells only (A, arrow), not in photoreceptor cells (A, asterisk). Mutagenesis of the Mitf binding site reduces the expression driven by the Oa1 promoter in the RPE and drives a low expression level in the photoreceptor layer (C, arrowheads). onl, outer nuclear layer (photoreceptor cells).