LKB1 controls inflammatory potential through CRTC2-dependent histone acetylation

Highlights

• •Loss of LKB1 sensitizes cells to diverse inflammatory stimuli
• •LKB1 regulates inflammatory responses via CRTC2-CREB signaling
• •Loss of LKB1 promotes increased CRTC2-dependent H3K27ac at inflammatory gene loci
• •Increased histone acetylation boosts inflammatory potential in LKB1-deficient cells

Summary

Deregulated inflammation is a critical feature driving the progression of tumors harboring mutations in the liver kinase B1 (LKB1), yet the mechanisms linking LKB1 mutations to deregulated inflammation remain undefined. Here, we identify deregulated signaling by CREB-regulated transcription coactivator 2 (CRTC2) as an epigenetic driver of inflammatory potential downstream of LKB1 loss. We demonstrate that LKB1 mutations sensitize both transformed and non-transformed cells to diverse inflammatory stimuli, promoting heightened cytokine and chemokine production. LKB1 loss triggers elevated CRTC2-CREB signaling downstream of the salt-inducible kinases (SIKs), increasing inflammatory gene expression in LKB1-deficient cells. Mechanistically, CRTC2 cooperates with the histone acetyltransferases CBP/p300 to deposit histone acetylation marks associated with active transcription (i.e., H3K27ac) at inflammatory gene loci, promoting cytokine expression. Together, our data reveal a previously undefined anti-inflammatory program, regulated by LKB1 and reinforced through CRTC2-dependent histone modification signaling, that links metabolic and epigenetic states to cell-intrinsic inflammatory potential.

Introduction

Liver kinase B1 (LKB1) is a serine/threonine kinase with key regulatory roles in metabolism, cell polarity, cell growth, and proliferation.¹ Loss of LKB1 function (via mutations in the serine/threonine kinase 11 [STK11] gene) is a common occurrence in human cancer,^2,3,4,5,6 while heterozygous germline mutations in STK11 predispose humans and mice to the development of Peutz-Jeghers syndrome (PJS),^7,8,9,10 an autosomal dominant disease characterized by the development of hamartomatous polyps in the gastrointestinal (GI) tract.¹¹ In addition to GI issues, PJS patients carry a 93% cumulative risk of developing GI, breast, pancreatic, and gynecological cancers by age 65—a cancer risk similar to that in BRCA1/2 carriers.¹² Loss of LKB1 function is among the most common genetic mutations in human epithelial cancers, including non-small cell lung cancer (NSCLC), where it is frequently co-mutated with KRAS,^13,14,15 pancreatic cancer,^6,16 and cervical cancer.^4,17 LKB1 mutations are associated with poor patient outcomes and resistance to both conventional chemotherapy and immune checkpoint inhibitors (ICIs).^13,18,19,20 Despite these observations, how LKB1 mutations lead to the development of PJS polyps and malignant tumors remains poorly understood, limiting treatment options and increasing patient mortality.

LKB1 regulates the activity of a series of downstream kinases belonging to the AMP-activated protein kinase (AMPK) and AMPK-related kinase (ARK) families. Many of the tumor suppressor functions of LKB1 have been attributed to its negative regulation of the mechanistic target of rapamycin (mTOR) complex 1 (mTORC1) signaling pathway, mediated in part by LKB1-dependent phosphorylation and activation of AMPK.^21,22,23 However, despite elevated mTORC1 activity in LKB1 mutant tumors,^21,24 the mTORC1 inhibitor rapamycin has shown modest effects on the growth of pre-existing GI polyps in pre-clinical PJS mouse models.^25,26,27 Genetic evidence also indicates that both mTOR and AMPK are dispensable for polyposis in PJS,^28,29 suggesting that other ARK family members contribute to the tumor suppressor functions of LKB1. Salt-inducible kinases (SIKs) are one such family, beyond AMPK, with the potential to mediate the tumor suppressor functions of LKB1—silencing SIKs promotes lung tumor development similar to LKB1 loss,^30,31,32 although the mechanism(s) of tumor suppression by SIKs remain to be defined.

More recent work has associated deregulated inflammation as a common feature of LKB1 mutant tumors. PJS mouse models and patient samples exhibit hallmarks of inflammation, including increased infiltration of immune cells (i.e., T cells, macrophages, and neutrophils), elevated levels of inflammatory cytokines (i.e., IL-6 and IL-11), and the activation of signal transducer and activator of transcription 3 (STAT3) in GI polyp tissues.^28,29 The expression of cyclooxygenase-2, which synthesizes inflammatory prostaglandins, is also elevated in PJS polyps.³³ Elevated IL-6 and Janus kinase (JAK)-STAT signaling is also a common signature of LKB1- and SIK-deficient NSCLC.^32,34,35 Evidence suggests that the inflammatory environment fostered by LKB1 mutant tumors promotes immune invasion, in part through recruitment of suppressive myeloid cells (i.e., CD11b⁺Ly6G⁺ neutrophils)^34,35 and contributes significantly to tumor progression. Inhibiting aberrant inflammation via genetic ablation of Il6, dosing animals with IL-6 neutralizing antibodies, or treatment with JAK inhibitors can reduce the growth of LKB1 mutant tumors,^28,29,35 suggesting that inflammation is a driver—rather than byproduct—of LKB1-dependent tumorigenesis.

As deregulated inflammation is a common feature of tumors promoted by LKB1 mutations, a common etiology linking LKB1 loss and increased inflammatory potential is likely; however, the mechanism(s) by which LKB1 loss promotes deregulated inflammation has remained undefined. Through evaluation of signaling networks downstream of LKB1, we have identified aberrant epigenetic programming—instilled by cAMP response element binding protein (CREB)-regulated transcription coactivator 2 (CRTC2)-CREB signaling downstream of SIKs—as a cell-intrinsic regulator of inflammatory potential, revealing a previously unappreciated anti-inflammatory pathway controlled by LKB1.

Results

Loss of LKB1 sensitizes cells to inflammatory stimuli

Loss of the tumor suppressor LKB1 is associated with deregulated cell proliferation, cell growth, and survival.^1,36 However, recent work across diverse tissue types has identified heightened inflammatory signatures as features of LKB1-deficient cells.^28,29,32,35 To establish a tractable platform to assess the role of LKB1 in control of tissue inflammation, we engineered LKB1-deficient mouse embryonic fibroblasts (MEFs), which respond to diverse inflammatory stimuli by producing acute inflammatory cytokines such as IL-6.²⁹ Gene set enrichment analysis (GSEA) of differentially expressed genes between LKB1-deficient MEFs versus their control counterparts (measured by RNA sequencing [RNA-seq]) revealed enrichment of several pathways previously associated with LKB1 loss (i.e., DNA repair, epithelial-mesenchymal transition [EMT], KRAS signaling) and enrichment of genes associated with inflammatory processes, including IL-6-JAK-STAT signaling and cytokine-cytokine receptor interactions (Figures 1A and S1A). Genes encoding prominent members of the IL-6 (i.e., Il6, Il11, Clcf1, and Lif), chemokine (i.e., Cxcl12, Cxcl2, Cxcl5), and tumor necrosis factor (TNF) (i.e., Tnfsf11/RANKL) superfamilies were significantly increased in LKB1-null MEFs at baseline (Figures 1B and S1B). Similarly, PJS polyps also display elevated expression of Il6 and Il11, which are associated with chronic gastric inflammation and GI tumor development,^29,37,38,39 as well as Cxcl2, a proinflammatory chemokine involved in neutrophil recruitment.²⁹ These data corroborate observations of increased inflammatory cytokine expression in LKB1-deficient NSCLC cells,^28,32,35 validating our model…