Folliculin regulates cell-cell adhesion, AMPK, and mTORC1 in a cell-type-specific manner in lung-derived cells

Abstract

Germline loss-of-function BHD mutations cause cystic lung disease and hereditary pneumothorax, yet little is known about the impact of BHD mutations in the lung. Folliculin (FLCN), the product of the Birt-Hogg-Dube (BHD) gene, has been linked to altered cell-cell adhesion and to the AMPK and mTORC1 signaling pathways. We found that downregulation of FLCN in human bronchial epithelial (HBE) cells decreased the phosphorylation of ACC, a marker of AMPK activation, while downregulation of FLCN in small airway epithelial (SAEC) cells increased the activity of phospho-S6, a marker of mTORC1 activation, highlighting the cell type-dependent functions of FLCN. Cell-cell adhesion forces were significantly increased in FLCN-deficient HBE cells, consistent with prior findings in FLCN-deficient human kidney-derived cells. To determine how these altered cell-cell adhesion forces impact the lung, we exposed mice with heterozygous inactivation of Bhd (similarly to humans with germline inactivation of one BHD allele) to mechanical ventilation at high tidal volumes. Bhd(+/-) mice exhibited a trend (P = 0.08) toward increased elastance after 6 h of ventilation at 24 cc/kg. Our results indicate that FLCN regulates the AMPK and mTORC1 pathways and cell-cell adhesion in a cell type-dependent manner. FLCN deficiency may impact the physiologic response to inflation-induced mechanical stress, but further investigation is required. We hypothesize that FLCN-dependent effects on signaling and cellular adhesion contribute to the pathogenesis of cystic lung disease in BHD patients.

Keywords: AMPK; BHD; Cell–cell adhesion; folliculin; mTOR; pneumothorax.

Figure 1.

FLCN primarily localizes to the cytoplasm and does not regulate proliferation of lung‐derived cells. (A) Nuclear and cytoplasmic fractions were prepared from SAEC, HEK293, and HeLa cells. FLCN localized primarily to the cytoplasm of all three cell types. CREB was used as a control for the nuclear fraction and GAPDH as a control for the cytoplasmic fraction. Equal amounts of protein in each fraction were loaded. Each panel represents a separate gel. (B) FLCN was downregulated in HBE and SAEC cells using shRNA. Downregulation of FLCN was confirmed by immunoblot. (C) No obvious change in cellular morphology was observed. (D) No difference in cell doubling was observed over 96 h between HBE and SAEC cells expressing control shRNA versus FLCN shRNA.

Figure 2.

Impact of FLCN downregulation on cellular signaling in lung‐derived cells. (A, B) HBE cells with FLCN downregulation or control shRNA (three biologic replicates per shRNA cell line) were immunoblotted for proteins and phospho‐proteins in pathways known to be regulated by FLCN in other cellular lineages, including AMPK (phospho‐ACC), mTORC1 (phospho‐S6), autophagy (p62/sequestosome1), mitochondrial biogenesis (Cox4), and Rho Kinase (phospho‐cofilin). *P < 0.05 (C, D) SAEC cells with FLCN downregulation or control shRNA (three biologic replicates per shRNA cell line) were immunoblotted for the same panel of proteins and phospho‐proteins as in Figure 2A. *P < 0.05

Figure 3.

FLCN differentially regulates TGF‐beta target gene transcription in lung‐derived cells. mRNA from HBE (A) and SAEC (B) cells (three biologic replicates per shRNA cell line and three technical replicates) expressing FLCN shRNA or control shRNA was analyzed by qRT‐PCR for five TGF‐beta targets. *P < 0.05

Figure 4.

FLCN downregulation increases cell–cell adhesion in HBE cells. (A) HBE and SAEC cells with FLCN downregulation or control shRNA were grown as clusters in adherence‐free conditions for 96 h and then sheared using a standardized protocol (see Methods). (B) Immunoblot was used to compare levels of cell adhesion proteins in SAEC and HBE cells with control shRNA or FLCN shRNA cells grown in adherent monolayers or in detached conditions.

Figure 5.

Bhd haploinsufficiency does not cause airspace enlargement at 5 months of age. (A) Representative Gills‐stained images of inflated lung sections from Bhd^+/+ and Bhd^+/− mice. (B) Mean chord length and alveolar area in Bhd^+/− mice compared with littermate control Bhd^+/+ mice (n = 5 per group). Data are presented as mean ± SD.

Figure 6.

(A) Interstitial edema in Bhd^+/+ and Bhd^+/− mice after 6 h of mechanical ventilation. (B) One mouse from the Bhd^+/+ cohort had areas of hemorrhage after 6 h mechanical ventilation.

Figure 7.

Bhd^+/− mice do not exhibit higher lung injury parameters following ventilation at high tidal volume. Bhd^+/− and Bhd^+/+ littermate controls (n = 5 per group) were subjected to ventilation at high tidal volume (24 cc/kg) for 6 h. Bhd^+/− mice exhibited a trend toward increased lung elastance after ventilation (P = 0.08 at 6 h) compared with Bhd^+/+ mice (A). Scatter plot of elastance measurements after 6 h of mechanical ventilation in Bhd^+/− mice compared with Bhd^+/+ mice (B). BAL protein levels (C), BAL total leukocyte and macrophage numbers (D) and BAL IL‐6 (E) were measured following 6 h of mechanical ventilation. Data are presented as mean ± SD.

Figure 8.

Mechanical ventilation in Bhd^+/− does not cause airspace enlargement. (A) Representative H&E‐stained images of inflated lung sections from Bhd^+/+ and Bhd^+/− mice after 6 h of ventilation (lower and higher magnifications of representative fields are shown, scale bars are 100 µm). (B) Mean chord length in Bhd^+/− mice compared with littermate control Bhd^+/+ mice after 6 h of ventilation (n = 5 per group). Data are presented as mean ± SD.