Th-POK regulates mammary gland lactation through mTOR-SREBP pathway

Abstract

The Th-inducing POK (Th-POK, also known as ZBTB7B or cKrox) transcription factor is a key regulator of lineage commitment of immature T cell precursors. It is yet unclear the physiological functions of Th-POK besides helper T cell differentiation. Here we show that Th-POK is restrictedly expressed in the luminal epithelial cells in the mammary glands that is upregulated at late pregnancy and lactation. Lineage restrictedly expressed Th-POK exerts distinct biological functions in the mammary epithelial cells and T cells in a tissue-specific manner. Th-POK is not required for mammary epithelial cell fate determination. Mammary gland morphogenesis in puberty and alveologenesis in pregnancy are phenotypically normal in the Th-POK-deficient mice. However, Th-POK-deficient mice are defective in triggering the onset of lactation upon parturition with large cellular lipid droplets retained within alveolar epithelial cells. As a result, Th-POK knockout mice are unable to efficiently secret milk lipid and to nurse the offspring. Such defect is mainly attributed to the malfunctioned mammary epithelial cells, but not the tissue microenvironment in the Th-POK deficient mice. Th-POK directly regulates expression of insulin receptor substrate-1 (IRS-1) and insulin-induced Akt-mTOR-SREBP signaling. Th-POK deficiency compromises IRS-1 expression and Akt-mTOR-SREBP signaling in the lactating mammary glands. Conversely, insulin induces Th-POK expression. Thus, Th-POK functions as an important feed-forward regulator of insulin signaling in mammary gland lactation.

Fig 1. Th-POK is expressed in the mammary luminal epithelial cells and is essential to pup survival.

(A) Wild-type (WT) females were mated with Th-POK knockout (KO) males and KO females were mated with WT males. Percent of survived pups nursed by either WT or KO dams (N = 30) within 48 hours after birth. (B) Heterozygote pups born to WT females were fostered to WT (N = 7) or KO (N = 7) dams at the time of parturition. Kaplan-Meier survival analysis of the pups. (C) Weight gain of the pups fostered to WT (N = 7) or KO (N = 5) dams. (D) Immunostaining of Th-POK on mammary gland sections from WT and KO virgin mice. Scale bar: 50μm. (E) Western blot analysis of Th-POK protein levels in isolated mammary epithelial cells from WT and KO virgin mice. (F) Double staining of Th-POK (red) with luminal cell marker cytokeratin 8 (K8) or basal cell marker αSMA (green) on virgin mammary gland sections. Scale bars: 25μm. (G) RT-qPCR analyses of expression of Th-POK, luminal cell marker cytokeratin 8 (K8) and basal cell marker cytokeratin 14 (K14) in luminal (Lin^-, CD24⁺, CD29^lo) and basal (Lin^-, CD24⁺, CD29^hi) mammary epithelial cells FACS-isolated from virgin mice (N = 4). (H) Immunostaining of Th-POK on mammary gland sections from WT and KO mice at lactation day 2 (L2). Scale bar: 50μm. (I) Double staining of Th-POK (red) with luminal cell marker cytokeratin 8 (K8) or basal cell marker αSMA (green) on mammary gland sections at L2. Scale bars: 25μm. (J) RT-qPCR analyses of expression of Th-POK, K8 and K14 in luminal and basal mammary epithelial cells FACS-isolated from mice at L2 (N = 4). Data are presented as mean ± SEM. **P < 0.01, ***P < 0.001. (K) Western blot analysis of Th-POK expression in mammary glands at different stages. (L and M) RT-qPCR (L, N = 3) and western blot (M) analyses of Th-POK expression in isolated mammary epithelial cells at different stages. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, compared to virgin.

Fig 2. Impaired milk lipid secretion in Th-POK-deficient mammary glands.

(A) Carmine-stained whole-mounted mammary glands from WT and KO mice at pregnancy day 17.5 (P17.5) or lactation day 2 (L2). Scale bar: 2mm. (B) Hematoxylin-and-eosin-stained sections of mammary glands from WT and KO mice at P17.5 or L2. Scale bar: 100μm. (C and D) Milk was collected from fourth mammary glands following oxytocin stimulation at lactation day 2. Milk volume (WT = 11, KO = 8) (C) and milk triacylglycerol (TAG) concentration (WT = 10, KO = 6) (D) were compared. (E) Staining of MFGs from WT and KO mice. Scale bars: 100μm. (F) Number of MFGs from WT and KO mice (N = 5, five fields/mice). (G) Size distribution of MFGs from WT and KO mice (N = 5). (H) Lumen diameters of mammary glands from WT and KO mice at P17.5 or L2 (N = 3, five fields/mice). (I) Triacylglycerol (TAG) and nonesterified fatty acid (NEFA) concentrations in the mammary epithelial cells at L2 (WT = 4, KO = 4). (J) RT-qPCR analyses of genes regulating lipolysis and triacylglycerol synthesis in mammary glands (N = 4 mammary glands from individual mice) at lactation day 2. (K) Sections of mammary glands from WT and KO mice at P17.5 or L2 were stained with anti-Perilipin2 (red). The luminal surface of secretory alveoli was visualized with wheat germ agglutinin (green). CLDs are marked by arrows. Scale bars: 25μm. (L) Quantification of CLD size in mammary glands from WT and KO mice at P17.5 or L2 (N = 3, five fields/mice). Data are presented as mean ± SEM. Statistic analyses were performed with two-way Anova followed by Bonferroni's multiple comparison test (H and K). *P < 0.05, **P < 0.01, ***P < 0.001. n.s.: not significant.

Fig 3. Precocious involution in Th-POK knockout mammary glands.

(A) Hematoxylin-and-eosin-stained sections of mammary glands from 12-week virgin mice and mice at lactation day 9. Scale bar: 100μm. (B) GSEA data showing the enrichment of apoptosis signature in mammary glands at involution day 2, compared to those at lactation day 9. (C) GSEA data showing marginal enrichment of apoptosis signature in KO mammary gland compared to the WT mammary gland at lactation day 1. (D) Immunostaining of cleaved caspase-3 on mammary gland sections from WT and KO mice at pregnancy day 17.5 (P17.5) and lactation day 1 (L1). Scale bar: 50μm. (E) Quantitative analysis of cleaved caspase-3 staining in (D). Data are represented as the number of cleaved caspase-3-positive cells per field (N = 3, five 20× fields/mice). (F) GSEA data showing the enrichment of IL6-JAK-STAT3 signature in mammary glands at involution day 2, compared to those at lactation day 9. (G) GSEA data showing enrichment of IL6-JAK-STAT3 signature in KO mammary gland compared to the WT mammary gland at lactation day 1. (H) RT-qPCR analyses of expression of Cebpd and Socs3 in mammary glands from WT and KO mice at lactation day 2 (N = 4). (I) RT-qPCR analyses of expression of Cebpd and Socs3 in Th-POK-expressing HC11 cells. Data are presented as mean ± SEM. *P < 0.05, ***P < 0.001. ES: Enrichment score. NES: normalized enrichment score. FDR: false discovery rate.

Fig 4. Impaired lipid secretion in Th-POK-deficient mammary glands is attributed mainly to intrinsic defects in the mammary epithelial cells.

Wild-type or Th-POK-deficient mammary epithelial cells were implanted into clear fat pads of wild-type or Th-POK-deficient mice. Reconstituted mammary glands were harvested at lactation day 2. (A) Hematoxylin-and-eosin-stained sections of reconstituted mammary glands at lactation day 2. Scale bar: 100μm. (B) Sections of reconstituted mammary glands at lactation day 2 were stained with anti-Perilipin2 (red). The luminal surface of secretory alveoli was visualized with wheat germ agglutinin (green). Scale bars: 25μm. (C) Quantification of CLD size in transplanted mammary glands from WT and KO mice (N = 3, five fields/mice). Data are presented as mean ± SEM. Statistic analyses were performed with two-way Anova followed by Bonferroni's multiple comparison test. **P < 0.01. n.s.: not significant.

Fig 5. Th-POK deficiency impairs SREBP pathway in lactating mammary glands.

(A-C) GSEA analyses of genes regulating fatty acid synthesis (A) and oxidation (B) and SREBP target genes (C) in KO mammary gland compared to the WT mammary gland at lactation day 1. ES: Enrichment score. NES: normalized enrichment score. FDR: false discovery rate. (D) Immunostaining of SREBP1 on mammary gland sections from WT and KO mice at lactation day 1. Scale bar: 100μm. (E) Score of nuclear SREBP1 in WT or KO mammary epithelial cells at lactation day 1 (N = 3, five fields/mice). (F) RT-qPCR analyses of SREBP target gene expression in mammary glands (N = 4 mammary glands from individual mice) at lactation day 2. (G) RT-qPCR analyses of SREBP target gene expression in isolated mammary epithelial cells (N = 3) from WT and KO mice at lactation day 2. (H) Western blot analysis of Th-POK expression in HC11 mammary epithelial cells treated with prolactin, dexamethasone or insulin. (I) RT-qPCR analyses of Th-POK expression in HC11 cells treated with or without insulin (N = 3). (J) RT-qPCR analyses of SREBP target gene expression in Th-POK-expressing HC11 cells treated with or without insulin (N = 3). (K) RT-qPCR analyses of SREBP target gene expression in primary WT and KO mammary epithelial cells treated with or without insulin (N = 2). Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig 6. Th-POK regulates insulin-induced activation of mTOR pathway.

(A) SREBP signature is significantly correlated to mTORC1 signature in lactating mammary glands. Black dots: lactating wild-type mammary glands; red dots: lactating knockout mammary glands. (B and C) GSEA analyses of mTORC1 gene signature (B) and genes upregulated by mTOR (C) in KO mammary gland compared to the WT mammary gland at lactation day 1. ES: Enrichment score. NES: normalized enrichment score. FDR: false discovery rate. (D) Western blot analyses of Akt-mTOR pathway in mammary glands (4 biological replicates) from WT and KO mice at lactation day 2. (E) Immunostaining of pmTOR and pS6 on mammary gland sections from WT and KO mice at lactation day 1. Scale bar: 100μm. (F) Score of pmTOR and pS6 in WT or KO mammary epithelial cells at lactation day 1 (N = 3, five fields/mice). Data are presented as mean ± SEM. *P < 0.05, ***P < 0.001. (G) Western blot analyses of Akt-mTOR pathway in Th-POK-expressing HC11 cells treated with insulin.

Fig 7. Th-POK regulates mTOR pathway through IRS-1.

(A) Western blot analyses of Akt-mTOR pathway in mammary glands (4 biological replicates) from WT and KO mice at lactation day 2. (B) Immunostaining of IRS-1 on mammary gland sections from WT and KO mice at lactation day 1. Scale bar: 100μm. (C) Score of IRS-1 in WT or KO mammary epithelial cells at lactation day 1 (N = 3, five fields/mice). (D) RT-qPCR analyses of genes in Akt-mTOR pathway in mammary glands (N = 4) at lactation day 2. (E) RT-qPCR analyses of genes in Akt-mTOR pathway in isolated mammary epithelial cells (E, N = 3) from WT and KO mice at lactation day 2. (F) Western blot analyses of Akt-mTOR pathway in Th-POK-expressing HC11 cells treated with insulin. (G) RT-qPCR analyses of genes in Akt-mTOR pathway in HC11 cells expressing Th-POK (N = 3). (H) Schematic diagram of predicted Th-POK binding sites and regions amplified with corresponding PCR primers to detect ChIP products in Irs1 locus (left). Lysates from FLAG-tagged Th-POK-expressing HC11 cells were prepared for the ChIP assay with anti-FLAG antibody (N = 3) (right). HC11 cells transduced with empty vector was used as control. Actb (β-actin) was used as negative control. (I) Western blot of analyses of Akt-mTOR pathway in Th-POK-expressing, IRS-1 knock-down HC11 cells. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001. n.s.: not significant.