Mitf is a master regulator of the v-ATPase, forming a control module for cellular homeostasis with v-ATPase and TORC1

Abstract

The v-ATPase is a fundamental eukaryotic enzyme that is central to cellular homeostasis. Although its impact on key metabolic regulators such as TORC1 is well documented, our knowledge of mechanisms that regulate v-ATPase activity is limited. Here, we report that the Drosophila transcription factor Mitf is a master regulator of this holoenzyme. Mitf directly controls transcription of all 15 v-ATPase components through M-box cis-sites and this coordinated regulation affects holoenzyme activity in vivo. In addition, through the v-ATPase, Mitf promotes the activity of TORC1, which in turn negatively regulates Mitf. We provide evidence that Mitf, v-ATPase and TORC1 form a negative regulatory loop that maintains each of these important metabolic regulators in relative balance. Interestingly, direct regulation of v-ATPase genes by human MITF also occurs in cells of the melanocytic lineage, showing mechanistic conservation in the regulation of the v-ATPase by MITF family proteins in fly and mammals. Collectively, this evidence points to an ancient module comprising Mitf, v-ATPase and TORC1 that serves as a dynamic modulator of metabolism for cellular homeostasis.

Keywords: Gut; MITF; Melanocytes; TFEB; TORC1; v-ATPase.

Fig. 1.

Mitf is evolutionarily conserved and functions in gut. (A) Evolutionary conservation of fly and vertebrate MiT factors as assessed using the Maximum Likelihood method. The tree is to scale with branch lengths measured in number of changes per site (0.5). Monophyletic groups representing orthologous gene families are shown with brackets on the right. MITF and TFE3 clustering is strongly supported. Extra whole genome duplications occurred in the fish branch, generating duplicate copies in D. rerio. (B) Mitf gene and reagents. The genes PIP4K, Dyrk3, and Caps and four Mitf mRNA isoforms are shown (Flybase). The bHLHZip region is marked in red. Double slashes (//) in the Mitf intron 1 reflect a ∼20 kb genomic region with repeated DNA. Df(4)TZ^DC, Mitf^KO [=Df(4)TZ] removes Mitf, Dyrk3 and Caps. Mitf rescue constructs span the Mitf DNA except for part of intron 1 (magenta). Arrowheads show sites of Mitf^TZ1 and Mitf^TZ2 nonsense mutations. dsRNA transgenes Mitf^RNAi-TZ1-3, Mitf^RNAi-TZ4-8 and Mitf^RNAi-VDRC (KK113614) target exons shown by linked arrows or bracket in blue. The Mitf^2.2-GFP reporter contains 2.2 kb of upstream DNA and 5′UTR driving nuclear GFP. (C–I) Gut and/or Malpighian tubules of L1–L3 larvae; anterior to the left; all panels show nuclear GFP (green; anti-GFP antibody); in panel F, Dlg (red; anti-Dlg antibody) marks cell membranes. (C) Mitf^2.2-GFP expression in wild-type (WT) L3 midgut (MG), hindgut (HG) and Malpighian tubules (MT). (D–I) 4Mbox-GFP expression in wild type (E,F) and mutants (G–I). (D) Diagram of 4Mbox-GFP. (E) 4Mbox-GFP expression is present in L3 hindgut and Malpighian tubules, but not midgut. (F) Higher magnification of 4Mbox-GFP in hindgut shows expression in cells of the gut lining. (G–I) 4Mbox-GFP expression is Mitf dependent. All panels show live larvae; anterior to the left. (G) 4Mbox-GFP expression is reduced in midgut, hindgut and Malpighian tubules of L1 Mitf^TZ2/Df(4)TZ; residual weak expression is likely due to stop-codon read-through or residual maternal product. Inset shows expression in wild type. (H,I) In byn-Gal4 UAS-Mitf^RNAi L2 larvae, 4Mbox-GFP expression is lost in the hindgut but not Malpighian tubules, as expected for byn-Gal4; (H) byn-Gal4 UAS-Mitf^RNAi-TZ1-3 UAS-Mitf^RNAi-TZ4-8; (I) byn-Gal4 UAS-Mitf^RNAi-VDRC.

Fig. 2.

Mitf controls expression of v-ATPase subunit genes. Summary and validation of microarray analysis. (A) nub-Gal4 UAS-Mitf L3 wing discs expressing exogenous Mitf in the wing pouch (red; anti-Mitf antibody). DIC of same disc showing aberrant morphology and cell death (arrows), and a control nub-Gal4 disc. (B) As compared to wild-type (WT) tissues, 543 genes were upregulated and 359 genes were downregulated (≥1.4 fold; P≤0.01) by exogenous (exog) Mitf in wing discs; whereas 897 genes were downregulated and 898 upregulated (≥1.4 fold; P value ≤0.01) in mutant (mut) hindgut (HG)+Malpighian tubules (MT). (C) 85 genes were upregulated in wing (by ectopic expression, ect) and downregulated in mutant hindgut+Malpighian tubules. (D) 33 Vha genes encode the 15 subunits of the v-ATPase (known as V1A–V1H for the V1 complex and V0a–V0e for the V0 complex; M8.9 and AC45 are accessory subunits called AP1 and AP2 in mammals). Mitf controls the 15 Vha genes shown in red; one for each component of the holoenzyme. (E,F) Validation of microarray findings in wing discs. (E) en-Gal4 Vha26-lacZ^ET and en-Gal4 UAS-Mitf Vha26-lacZ^ET wing discs showing upregulation of Vha26-lacZ^ET expression (green; anti-β-Gal antibody) where Mitf protein is present (red; anti-Mitf antibody). (F) Ser-Gal4 Vha16-1-GFP^GT and Ser-Gal4 UAS-Mitf Vha16-1-GFP^GT wing discs showing upregulation of Vha16-1-GFP^GT (green; anti-GFP antibody) where Mitf protein is present (red; anti-Mitf antibody). (G,H) Downregulation of Vha genes in mutant and RNAi larvae. (G) Vha68-2-lacZ^ET and Vha68-2-lacZ^ET Mitf^TZ2/Df(4)TZ larval hindgut and Malpighian tubules show downregulation of enhancer trap expression (red; anti-β-Gal antibody) when Mitf is reduced. (H) In byn-Gal4 UAS-GFP 2xUAS-Mitf^RNAi-TZ Vha68-2-lacZ^ET L3 larvae, GFP expression (green; anti-GFP antibody) marks hindgut cells with RNAi co-expression (left panel); enhancer trap Vha68-2-lacZ^ET expression (red; anti-β-Gal antibody) is absent in the hindgut, but present in midgut and Malpighian tubules (right panel).

Fig. 3.

Exogenous Mitf affects holoenzyme activity and lysosomal pathway in vivo. Genotypes are as marked around panels. (A) Vha loss-of-function alleles dominantly suppress (right) wing defects caused by Mitf overexpression (middle). (B–D) Vha55–GFP expression increases in the presence of Mitf protein. (B) Diagram showing that the stability of the Vha55–GFP fusion protein correlates with the amount of assembled v-ATPase. (C,D) Panels show the dorsal region of dpp-Gal4 UAS-Vha55-GFP eye discs where dpp-Gal4 drives robust expression. Discs are stained for GFP (green; anti-GFP antibody) and the cell membrane marker Dlg (red; anti-Dlg antibody). The fusion protein is detected in the presence (D) but not in the absence (C) of Mitf. (E) Strong Lysotracker staining (red) is present in the posterior half (top) of the en-Gal4 UAS-Mitf wing disc, where Mitf is expressed (green; anti-Mitf antibody). (F) Reduction of lysosomal factors suppresses Mitf-induced wing defects. (G,H) Effect of Mitf overexpression on lysosomal vesicles in the wing. (G) Disc expressing Rab7–YFP ubiquitously but Mitf only in the posterior region shows increased Rab7–YFP (green; anti-GFP antibody) where Mitf is present (red; anti-Mitf antibody). Inset shows Mitf- and Rab7-positive punctae near the nucleus (DAPI) of a posterior wing-disc cell. (H) Confocal Z-sections along midline of dpp-gal4 UAS-spin-GFP wing discs without (left) or with (right) UAS-Mitf. Spin-GFP expression (green; anti-GFP antibody) expands (cell membranes in red; anti-Dlg antibody) in the presence of exogenous Mitf (not stained in this disc). (I,J) Effect of Mitf loss on lysosomal vesicles in the gut. Overall Rab7–YFP (I) or Spin–GFP (J) staining (green; anti-GFP antibody) increases in the hindgut from byn-Gal4 UAS-Mitf^RNAi-TZ L3 larvae. Punctate staining is also visibly increased as compared to byn-Gal4 controls. (K,L) In vivo assessment of lysosomal function using the lysosomal activity sensor tub-GFP-Lamp1 (green; anti-GFP antibody) in gain-of-Mitf wing disc (K) and loss-of-Mitf hindgut (L). (K) Mitf overexpression (right) increases the level of GFP only slightly over Gal4-only control (left), indicating efficient GFP degradation. (L) Loss-of-Mitf dramatically increases GFP expression (right) as compared to control (left), indicating a failure of protein degradation.

Fig. 4.

Mitf directly regulates Vha gene expression in the fly. Mitf-binding site identification and testing. (A) Consensus sequence of the 10-bp M-box binding site enriched in 15 Mitf-regulated Vha genes. (B) Evolutionarily conserved sites are found in 13 Mitf-dependent Vha genes, but are absent from 18 non-regulated Vha loci. The remaining two Mitf-regulated loci (not included in *) contain evolutionarily conserved M-box sequences matching the 8-bp core consensus. (C) ChIP confirms binding of Mitf–Myc protein to M-box-containing Vha regulatory regions (from wing discs and salivary glands cells expressing exogenous Mitf–Myc). (D–I) Analysis of Vha-GFP reporters containing wild-type or M-box-mutated regulatory genomic DNA. (D,G) Diagrams of the reporters Vha68-2-GFP (D) and Vha13-GFP (G), including M-box sequences and mutated base pairs (red). Upside-down triangles mark the positions of the sites. The hsp70 basal promoter (orange box) was used for the Vha68-2 intronic regulatory region; whereas the endogenous promoter with the 5′-flanking genomic fragment was sufficient for Vha13. (E,H) For both Vha68-2 and Vha13 reporters, GFP (nuclear) expression is strongly induced by endogenous Mitf in midgut, hindgut and Malpighian tubules, but is severely reduced when the sites are mutated. (F,I) Quantification of GFP expression levels for wild-type and mutant Vha68-2 (F) and Vha13 (I) reporters. ***P<0.0001 (one-tailed Student's t-test).

Fig. 5.

Mitf, v-ATPase and TORC1 form a homeostatic feedback loop. (A) Diagram of wild-type and truncated proteins lacking the Rag-binding or the S-rich regions. (A′) Panels show staining for Mitf proteins (red; anti-Myc or anti-HA antibody) and nuclear β-galactosidase (green; anti-β-Gal antibody). Genotypes are as marked above. Notice that the full length and the truncated Mitf^DelC1 proteins are cytoplasmic, whereas the truncated Mitf^DelN1 protein is nuclear. (B) Exogenous Mitf protein driven by ey-Gal4 in salivary gland cells is detected mainly in the cytoplasm. RNAi-mediated knockdown of the negative TORC1 regulator gig enhances the cytoplasmic, punctate expression of Mitf. Conversely, knockdown of the positive TORC1 regulator RagA-B shifts Mitf localization to the nucleus. See supplementary material Fig. S3A,A′ for quantification. (C,C′) Knockdown of VhaSFD (or Vha68-2; not shown) shifts wild-type Mitf protein localization from the cytoplasm (left panels) to the nucleus (right panels). Quantification of nuclear localization is shown in C′ as a percentage of total cells scored (Y-axis). **P<0.01; ***P<0.0001 (one-tailed Student's t-test). (D) Exogenous, nonfunctional Mitf^DelR mutant protein is distributed throughout the nucleus and cytoplasm in salivary gland cells (left panels); co-expression of wild-type Mitf shifts the balance of protein localization to the cytoplasm (right panels). (E) Western blot shows that the level of pS6K increases in the presence of exogenous Mitf, but decreases if the level of a v-ATPase component is reduced (UAS-VhaSFD^RNAi). Numbers show the relative band intensity compared to a Gal4-only control set as 1; a similar blot showed a 1:2.3:0.5 ratio (not shown). Total Drosophila (d)S6K does not change significantly; GADPH was used as loading control. (F) The diagram shows the consequences of the regulatory relationship among loop components. The Mitf/v-ATPase/TORC1 regulatory loop maintains the activity levels of these important metabolic regulators within an optimal range in response to intrinsic and environmental influences.

Fig. 6.

MITF controls v-ATPase subunit genes in human cells. (A,B) Boxplots of correlation (Kendall) coefficients between MiT genes (TFE3, TFEB, TFEC and MITF) and ATP6 genes in 24 melanoma cell lines (A) and 331 melanoma tumors (B). The box represents the 25–75th percentiles, and the median is indicated. The whiskers show the 10–90th percentiles. (C,C′) Direct binding of MITF to ATP6 genes in 501Mel cell line, including examples of MITF ChIP-Seq peaks at selected v-ATPase genes (C) and summary of results from ChIP-Seq analysis (C′); see supplementary material Table S6B for details. (D–D″) ATP6 gene regulation in response to downregulation of MITF by siRNA in 501Mel cells. (D) Cells were transfected with control siRNA (siCTR) or siRNA against MITF (siMITF) for 72 h before harvesting and analyzed by western blotting with the indicated antibody. (D′) Level of MITF mRNA (qPCR readings normalized to TBP) in siCTR or siMITF transfected 501Mel cells. Treatment with siMITF results in ∼60% decrease in MITF mRNA. (D″) Level of ATP6 mRNAs (qPCR readings normalized to TBP) in siMITF versus siCTR transfected 501Mel cells. Results are mean±s.e.m., n=3. (E) Upregulation of TORC1 activity by MITF in two MITF-inducible melanoma lines. Western blots show that the level of pS6K increases in response to MITF induction. Proteins detected by antibody are listed on the right. Numbers show relative band intensity compared to control set as 1. EV, empty vector control. Actin protein was used for loading control.