PTEN loss in the Myf5 lineage redistributes body fat and reveals subsets of white adipocytes that arise from Myf5 precursors

Abstract

The developmental origin of adipose tissue and what controls its distribution is poorly understood. By lineage tracing and gene expression analysis in mice, we provide evidence that mesenchymal precursors expressing Myf5--which are thought to give rise only to brown adipocytes and skeletal muscle--also give rise to a subset of white adipocytes. Furthermore, individual brown and white fats contain a mixture of adipocyte progenitor cells derived from Myf5(+) and Myf5(neg) lineages, the number of which varies with depot location. Subsets of white adipocytes originating from both Myf5(+) and Myf5(neg) precursors respond to β(3)-adrenoreceptor stimulation, suggesting "brite" adipocytes may also have multiple origins. We additionally find that deleting PTEN with myf5-cre causes lipomatosis and partial lipodystrophy by selectively expanding the Myf5(+) adipocyte lineages. Thus, the spectrum of adipocytes arising from Myf5(+) precursors is broader than previously thought, and differences in PI3K activity between adipocyte lineages alter body fat distribution.

Figure 1. Deleting PTEN in the Myf5⁺ lineage causes severe combined lipomatosis and partial lipodystrophy

(A) Anatomy of a 6-week-old PTEN^myf5cKO mutant (right) and a littermate control (left). PTEN^myf5cKO mice have a horse-collar-like growth and overall torpedo shape. (B) Lateral view of a PTEN^myf5cKO mouse (bottom panel) and a control (top panel). (C) Macroscopic images of control and PTEN^myf5cKO mouse. Black arrow indicates iBAT region; white arrow indicates iWAT. White dashed circles show axillary WAT (top panels). Vertebral WAT is indicated with a black dashed circle. A star indicates the trapezius muscle. (D) Macroscopic images of iBAT (scale bar = 5mm) (E) Macroscopic images of rWAT (black arrow). (F) Fat mass relative to body weight (top panel) and total fat mass (bottom panel) for the indicated tissues in 6-week-old PTEN^myf5cKO mice (black bars) and controls (white bars) (n=13; Bars represent mean± SEM; T-test; ***, p<0.001). (G) Representative images of mesenteric fat in control (left panels) and PTEN^myf5cKO mouse (right panels). (H) Representative images of perigonal WAT (black arrow). See also Figure S1.

Figure 2. PTEN-deficient fats have large adipocytes and more total cells

(A) H&E images of iBAT, iWAT, rWAT and quadriceps from a control and PTEN^myf5cKO mouse (40×). (B) Nuclei density per mm² of iBAT, sBAT and cBAT (n=5, 6 w old). (C) rWAT adipocytes cell diameter from 6 week old mice (n=7) (D) Images of H&E stained E14.5 control and PTEN^myf5cKO embryo sections (10×). Brown fat is marked with an arrow in control. In the mutant, iBAT (1), sBAT (2) and cBAT (3) precursor pools are enlarged. (E) Detail of embryonic BAT precursors at E14.5 and E17.5. Lipid droplets can be seen forming prematurely in the mutants by E17.5 (arrow). (F) Total genomic DNA purified iBAT in 6 and 12 week old control (white bars) and PTEN^myf5cKO (black bars) (n=8). (G) Total genomic DNA from rWAT (n=8). Bars represent mean± SEM. T-test; *, p<0.05, **, p< 0.01, ***, p<0.001). See also Figure S2.

Figure 3. PTEN^myf5cKO mice completely lack PTEN in WAT and BAT

(A) IHC and H&E stains on serial sections of E14.5 BAT precursor for PTEN and phospho-Akt-S473 in control and PTEN^myf5cKO mice. (B) PTEN, phospho-Akt^T308, and phospho-Akt^S473 levels in lysates from iBAT, iWAT, rWAT, triceps, and trapezius. See also Figure S3. (C) Proliferation curve of primary BAT preadipocytes (n=3;. Points represent mean± SEM. T-test; *, p< 0.05; **, p<0.01). (D) Primary BAT preadipocytes were differentiates and stained with ORO at different stages. (E) Tissue mRNA expression of PTEN (n=5–6; Bars represent mean± SEM. T-test; ***, p<0.001).

Figure 4. A subset of white adipocytes trace to the Myf5 lineage

(A–C) YFP (A), Ucp1 (B), and Prdm16 (C) mRNA expression in tissues (n=5) prepared from myf5-cre;R26R-YFP mice (Tri:triceps; Q: quadriceps; G: gastrocnemius; H: heart; L: liver; S:spleen; K: kidney). In (A) YFP level detected in each tissue from mice lacking Cre was subtracted. Bars represent mean± SEM. (D) Images of the indicated depots from a R26R-LacZ and a myf5-cre;R26R-LacZ mouse after X-Gal staining. (E) β-galactosidase activity in tissues from negative control (R26R-LacZ) and from myf5-cre;R26R-LacZ or PTEN^myf5cKO-LacZ mice. Counterstained with NFR. Liver and spleen at 40X; others at 63X. See also Figure S3. (F) qRT-PCR analysis of Cidea, Prdm16, Zic1, Ucp1, HoxC8, HoxC9 and DPT in whole iBAT and rWAT. (n=8; Horizontal line indicates the mean. *, p<0.05; **, p< 0.01).

Figure 5. The number of Myf5-lineage derived adipocyte progenitor cells in each depot varies with its anatomical location

(A) Macroscopic (top) and microscopic (40×-bottom) images of SVF cultures from R26R-LacZ and myf5-cre;R26R-LacZ mice from the indicated depots and stained with X-Gal. (B) Positive and negative β-gal cell quantification shown in (A). (C) Positive and negative β-gal activity in only the individual lipid containing cells (see Figure S5A). (D) Macroscopic (top) and microscopic (40×-bottom) images of SVF cultures from PTEN^myf5cKO and PTEN^myf5cKO-LacZ mice stained with X-Gal. (E) Positive and negative β-gal cell quantification shown in (D). (F) Positive and negative β-gal activity in only the individual lipid containing cells (see Figure S5B). (G) APCs were purified from the myf5-cre;R26R-YFP (wild type) fat depots. Yellow bars: YFP⁺ APCs; white bars: YFP^neg. (n=9). (H) qRT-PCR of Zic1, prdm16, HoxC8 and HoxC9 mRNA in YFP⁺ and YFP^neg APCs purified from the iBAT and rWAT (n=3) of myf5-Cre;R26R-YFP mice. ND: not detectable. Bars represent mean± SEM; ***, p<0.001. See also Figure S4.

Figure 6. Deleting PTEN with myf5-cre, but not prolonged CL316,243 treatment, selectively expands the Myf5⁺ adipocyte lineages

(A) APCs were purified from myf5-cre;R26R-YFP mice treated with PBS or CL316,243 for a week (n=3). Yellow bars: YFP⁺ APCs; white bars: YFP^neg. (B) APCs were purified by FACS from the PTEN^myf5cKO-YFP fat depots (n=4). Yellow bars represent YFP⁺ APCs; white bars are YFP^neg. (C) Total number of YFP⁺ (yellow bars, top) and YFP^neg cells (white bars, bottom) in each depot of myf5-cre;R26R-YFP and PTEN^myf5cKO-YFP (cKO) mice (n=4). (D) qRT-PCR of Zic1 and prdm16 mRNA in YFP⁺ and YFP^neg APCs purified from the iBAT and rWAT (n=3) of PTEN^myf5cKO-YFP mice. ND indicates not detectable. (E) qRT-PCR of HoxC8 and HoxC9 mRNA as in (D). Bars represent mean± SEM; T-test; *, p<0.05, **, p<0.01, ***, p<0.001. See also Figure S5.

Figure 7. Model of adipose tissue development based on lineage analysis and conditional PTEN deletion with Myf5-Cre

(1) Myf5-expressing progenitor cells give rise to a classical brown adipocytes as well as a subset of white adipocytes in iWAT and rWAT. Alternatively, Myf5-Cre expressing progenitor cells, distinct from those which give rise to classical BAT, could give rise to adipocytes in iWAT and rWAT– indicated by the question mark. (2) Myf5^neg progenitors of unknown origin give rise to ingWAT and pgWAT. (3) Many adipose depots contain a mixed population of adipocyte progenitor cells arising from both Myf5-Cre⁺ and Myf5-Cre^neg precursors. The Myf5-lineage significantly contributes to the adipocyte population in iWAT and rWAT, while in ingWAT and pgWAT the significance of the resident Myf5⁺ cells is unknown. Notably, the multilocular “brite fat” cells residing in ingWAT at ambient temperature, or induced after prolonged exposure to CL316,243 (not shown in figure), do not trace to the Myf5-Cre lineage. (4) PTEN loss in Myf5⁺ precursors expands the Myf5⁺ preadipocyte pool. (5) Upon differentiation, the PTEN-deficient adipocyte lineages accumulate excess lipid, resulting in overgrowth. (6) The growing Myf5⁺ adipocyte lineages restrict development of Myf5^neg adipocyte lineages, resulting in the selective expansion of fats exclusively derived from Myf5⁺ precursors.