The Lin28/let-7 axis regulates glucose metabolism

Abstract

The let-7 tumor suppressor microRNAs are known for their regulation of oncogenes, while the RNA-binding proteins Lin28a/b promote malignancy by inhibiting let-7 biogenesis. We have uncovered unexpected roles for the Lin28/let-7 pathway in regulating metabolism. When overexpressed in mice, both Lin28a and LIN28B promote an insulin-sensitized state that resists high-fat-diet induced diabetes. Conversely, muscle-specific loss of Lin28a or overexpression of let-7 results in insulin resistance and impaired glucose tolerance. These phenomena occur, in part, through the let-7-mediated repression of multiple components of the insulin-PI3K-mTOR pathway, including IGF1R, INSR, and IRS2. In addition, the mTOR inhibitor, rapamycin, abrogates Lin28a-mediated insulin sensitivity and enhanced glucose uptake. Moreover, let-7 targets are enriched for genes containing SNPs associated with type 2 diabetes and control of fasting glucose in human genome-wide association studies. These data establish the Lin28/let-7 pathway as a central regulator of mammalian glucose metabolism.

Figure 1. Lin28a Tg and iLIN28B Tg mice are resistant to obesity and diabetes and Lin28a is physiologically required for normal glucose homeostasis

(A) Aged wild-type (left) and Lin28a Tg mice (right) fed a normal diet, at 20 weeks of age. (B) Percentage body fat and (C) lean mass as measured by DEXA. (D) Weight curve of mice fed a HFD containing 45% kcals from fat. (E) Glucose tolerance test (GTT) and (F) Insulin tolerance test (ITT) of mice on HFD. (G) Liver histology of mice fed HFD. (H) Human LIN28B mRNA expression in a mouse strain with dox inducible transgene expression (named iLIN28B). (I) Mature let-7 expression in gut, spleen, liver, muscle and fat. (J) Kinetics of fed state glucose change after induction. (K) GTT and (L) ITT under normal diets. (M) iLIN28B growth curve under HFD. (N) GTT after 14 days of HFD and induction. (O) GTT and (P) ITT of Myf5-Cre; Lin28a^fl/fl mouse. Controls for Lin28a Tg mice are WT. Controls for iLIN28B Tg mice carry only the LIN28B transgene. Controls for muscle knockout mice are Lin28a^fl/fl mice. The numbers of experimental animals are listed within the charts.

Figure 2. iLet-7 mice are glucose intolerant

(A) let-7g and let-7a qRT-PCR in tissues of dox induced iLet-7 mice (n = 3) and controls (n = 3). (B) Reduced size of induced animals. (C) iLet-7 growth curve for males and females. (D) Fed state glucose in iLet-7 mice induced for 5 days. GTTs performed on mice fed with either (E) normal diet or (F) HFD. (G) ITT on normal diet. (H) Insulin production during a glucose challenge. (I) GTT of LIN28B/Let-7 compound heterozygote mice before (blue) and after (red) induction with dox. (J) Area under the curve (AUC) analysis for this GTT. Controls for iLet-7 Tg mice carry either the Let-7 or Rosa26-M2rtTa transgene only. The numbers of experimental animals are listed within the charts.

Figure 3. Insulin-PI3K-mTOR signaling is activated by Lin28a/b and suppressed by let-7

(A) Western blot analysis of Lin28a protein expression in C2C12 myoblasts infected with control pBabe or Lin28a overexpression vector, and mouse ESCs, with tubulin as the loading control. (B) Quantitative PCR for let-7 isoforms in C2C12 myoblasts, normalized to sno142, after Lin28a overexpression. (C) 2-deoxy-D-[³H] glucose uptake assay on 3-day-differentiated C2C12 myotubes with and without Lin28a overexpression, treated with DMSO, the PI3K inhibitor LY294002, and the mTOR inhibitor rapamycin for 24 hrs. (D) Western blot analysis of the effects of Lin28a overexpression on PI3K-mTOR signaling in C2C12 myoblasts, under serum-fed (fed), 18 hr serum starved (SS) or insulin-stimulated (Ins) conditions. Insulin stimulation was performed in serum-starved myoblasts with 10 μg/mL insulin for 5 min. Prior to insulin stimulation, serum-starved myoblasts were treated with either DMSO or 20 ng/mL rapamycin for 1 hr. (E) Western blot analysis of the effects of let-7f or control miRNA on PI3K-mTOR signaling in C2C12 myoblasts under serum-fed (fed), 18 hr serum starved (SS) or insulin-stimulated (Ins) conditions. (F) Western blot analysis of the effects of LIN28B induction by dox on PI3K-mTOR signaling in quadriceps muscles in vivo (n = 3 iLIN28B Tg mice and 3 LIN28B Tg only mice). (G) Insr and p-4EBP1 protein levels in wild-type and Lin28a muscle-specific knockout adults.

Figure 4. Lin28a/b and let-7 regulate genes in the insulin-PI3K-mTOR pathway

(A) Shown are the numbers of conserved let-7 binding sites within 3′UTRs found using the TargetScan algorithm. (B) Putative let-7 binding sites in 16 genes of the insulin-PI3K-mTOR pathway and in Lin28a/b. (C) 3′UTR luciferase reporter assays performed to determine functional let-7 binding sites. Bar graphs show relative luciferase reporter expression in human HEK293T cells after transfection of mature let-7f duplex normalized to negative control miRNA. Shown also are mutations in the seed sequence of the let-7 binding sites for INSR, IGF1R and IRS2. (D) Western blot analysis of Lin28a, Irs2, and tubulin in C2C12 myoblasts with and without Lin28a overexpression. (E) Western blot analysis of LIN28B, total IRS2, INSR and TUBULIN in HEK293T cells with either let-7f transfection or shRNA knockdown of LIN28B.

Figure 5. mTOR is required for Lin28a’s effects on growth and glucose metabolism in vivo

(A) Rapamycin (left 2 mice) and vehicle (right 2 mice) treated wild-type and Lin28a Tg mice shows relative size differences. (B) Curves showing relative growth (normalized to weight on first day of treatment) for mice treated from 3 weeks to 6.5 weeks of age. Blue and red represent wild-type and Lin28a Tg mice, respectively. Solid and dotted lines represent vehicle and rapamycin treated mice, respectively. Growth was measured by several other parameters: (C) weight, (D) crown-rump length or height, and (E) tail width. (F) GTT performed after 2 doses of vehicle or (G) rapamycin. (H) ITT performed after 1 dose of vehicle or (I) rapamycin. Controls for Lin28a Tg mice are WT. The numbers of experimental animals are listed within the charts.

Figure 6. let-7 target genes are associated with Type 2 diabetes mellitus and a model of the Lin28/let-7 pathway in glucose metabolism

mRNA expression of Igf2bp and Hmga family members in (A) C2C12 with and without Lin28a overexpression and in (B) 3T3 cells with and without LIN28A or LIN28B overexpression. (C) Western blot of NIH 3T3 cells with Lin28a overexpression showing Lin28a and Igf2bp1/2/3 protein levels (n = 3 biological replicates). (D) Igf2bp2 and Igf2bp3 mRNA in Lin28a Tg muscle. (E) Model of Lin28/let-7 pathway in glucose metabolism.