Lung Adenocarcinoma Distally Rewires Hepatic Circadian Homeostasis

Abstract

The circadian clock controls metabolic and physiological processes through finely tuned molecular mechanisms. The clock is remarkably plastic and adapts to exogenous "zeitgebers," such as light and nutrition. How a pathological condition in a given tissue influences systemic circadian homeostasis in other tissues remains an unanswered question of conceptual and biomedical importance. Here, we show that lung adenocarcinoma operates as an endogenous reorganizer of circadian metabolism. High-throughput transcriptomics and metabolomics revealed unique signatures of transcripts and metabolites cycling exclusively in livers of tumor-bearing mice. Remarkably, lung cancer has no effect on the core clock but rather reprograms hepatic metabolism through altered pro-inflammatory response via the STAT3-Socs3 pathway. This results in disruption of AKT, AMPK, and SREBP signaling, leading to altered insulin, glucose, and lipid metabolism. Thus, lung adenocarcinoma functions as a potent endogenous circadian organizer (ECO), which rewires the pathophysiological dimension of a distal tissue such as the liver. PAPERCLIP.

Copyright © 2016 Elsevier Inc. All rights reserved.

Figures

Figure 1. Lung adenocarcinoma rewires the circadian hepatic transcriptome

A) DNA microarray analysis was performed using mouse liver total RNA from ZT 0, 4, 8, 12, 16 and 20. Using JTK_cycle, genes selected to be circadian at a p-value <0.01 are displayed as heat maps for WT and lung tumor-bearing (TB) livers. Left panels display circadian genes exclusively in WT mice and right panels show genes with more robust oscillation in TB mice. B) Pie charts indicate actual numbers of circadian genes that oscillate exclusively in WT, TB or BOTH conditions. C) Top 10 gene ontology (GO) terms for biological process were identified by DAVID pathway analysis tool, based on a 0.01 p-value cutoff. D) Phase analysis of WT and TB-specific oscillating gene expression profiles. E) Phase analysis of ‘BOTH’ genes that remain circadian in WT and TB mice.

Figure 2. The circadian metabolome is reorganized by lung cancer

A) Heat maps displaying oscillating metabolites as determined by JTK_cycle (p-value <0.05) in WT and TB mice. Left panels display circadian metabolites exclusively in WT liver and right panels show metabolites with more robust oscillation in TB mice. B) Two-way ANOVA analysis using a p-value cutoff of 0.05 reveals metabolites that are responsive to lung tumors, circadian time point, or both. Numbers of oscillating metabolites using JTK_cycle are indicated from WT, TB or BOTH categories. C) Phase analysis was performed using JTK_cycle to identify the phase of peak metabolite expression. D) Oscillating metabolites are displayed based on biological sub-pathway, including amino acid, carbohydrate, cofactors, lipids, nucleotides, peptides and xenobiotics. E) Examples of energetic metabolites that are dampened in TB mice. NAD+= nicotinamide adenine dinucleotide; ATP= adenosine 5′-triphosphate.

Figure 3. The circadian clock is unaffected by lung adenocarcinoma

A) Heatmap for ‘BOTH’ category genes that are unaltered in expression between WT and TB mice. B) GO pathway analysis using biological process for ‘BOTH’ oscillating genes. C) BMAL1 protein phosphorylation by western and circadian expression of the clock genes, Bmal1, Clock, Rev-erbα, Dbp, Per2 and Cry1, as determined by quantitative real-time PCR (RT-PCR). D) Locomotor activity analysis for WT and TB mice, as calculated by the free-running period (Tau) in dark/dark (D/D) conditions. An n=12 WT and n=12 TB mice were used for behavioral analysis. E) Food intake of WT and TB mice shown over a 48-hour period (left panel). Total food intake was normalized to body weight of each animal. An n=7 WT and n=8 TB mice were used for indirect calorimetry analysis. F) VCO2/VO2 is shown as the respiratory exchange ratio (RER) for WT and TB mice over a 48-hour period. Average RER is quantified during the light and dark phases (right panel). Error bars indicate standard error of mean (SEM). Significance was calculated using Student’s T test and * and ** indicate p-value cutoffs of 0.001 and 0.0001, respectively.

Figure 4. Lipid metabolism is altered in TB mice

A) Western blot analysis for phospho-AMPK (Thr 172) and total AMPK in WT and TB mice, at the indicated circadian times. Quantification of pAMPK/total AMPK signaling is shown as a histogram. The ratio of AMP/ATP is shown over the circadian cycle and is elevated at all ZTs. B) Gene expression as determined by RT-PCR and protein expression of SREBP1c in WT and TB mice. Precursor (P) indicates uncleaved protein and mature (M) shows cleaved SREBP1 protein. C) Gene expression by RT-PCR was performed for Fasn, Acaca, and Elovl6 in WT and TB mice over the indicated ZTs. D) Levels of fatty acids and fatty acid esters as determined by metabolomics analysis for myristate, linolenate, palmitoleate and eicosapentaenoate (EPA). E) Gene expression of Srebp2 and its target genes Lss, Hmgcs1 and Pmvk as determined by RT-PCR. F) Total cholesterol levels in WT and TB mice over the circadian cycle were determined by metabolomics analysis. Error bars indicate SEM. Significance was calculated using Student’s T test and * indicates a p-value cutoff of 0.05.

Figure 5. Lung tumor-induced inflammation in the liver

A) Serum samples from WT and TB mice were assayed in an unbiased, multiplexed cytokine array platform and displayed as a heat map. Red indicates increased levels of cytokines and green indicates decreased cytokine levels in TB normalized to WT mouse serum. Specific profiles of pro-inflammatory cytokines (IL-6, IL-1α and TNFα) and anti-inflammatory IL-10 are shown. B) Gene expression as profiled by RT-PCR is shown for Il6rα, Il1r1 and Tnfrsf1b. C)Stat3 gene expression as shown by RT-PCR. Phospho-STAT3 (Tyr 705) and total STAT3 protein levels over the circadian cycle in WT and TB mice. D) Gene expression profiles of Socs3, Socs1 and Socs7 by RT-PCR. E) Known STAT3 target genes were compared to our transcriptomics data. Heat map displays gene expression profiles in the TB-specific group normalized to WT, with red and green representing up and down-regulated genes, respectively. Additionally, to determine enrichment of STAT3 target genes in WT and TB, Fisher’s exact test was used. The ** indicates the odds that the overlap of 39 genes in TB over random is 1.716 and the p-value is 0.00358, which satisfies a p<0.005 threshold. For WT the odds ratio is 1.345 with a p-value of 0.209. F) Recruitment of p-STAT3 to the STAT binding element (SBE) in the Socs3 promoter or to the 3′ untranslated region (UTR) as determined by chromatin immunoprecipitation (ChIP). Error bars indicate SEM. Significance was calculated using Student’s T test and * indicates a p-value cutoff of 0.05.

Figure 6. Lung adenocarcinoma alters hepatic insulin signaling and glucose production

A) Western analysis of phospho-AKT (Ser 473), total AKT and total IRS1 in WT and TB mice over the circadian cycle. B) Serum insulin levels were measured by ELISA at ZT 8 and 16 in WT and TB mouse serum. Insulin levels at ZT 16 are statistically significant as indicated by # (p-value = 0.053, using Student’s T-test). C) Insulin tolerance test (ITT) in WT and TB mice. D) Overnight fasting glucose levels in WT and TB mice. E) Glucose tolerance test (GTT) in overnight fasted WT and TB mice. F) Gluconeogenic gene expression profile of Pepck (Pck1) by RT-PCR was done in livers of WT and TB mice. G) Levels of phosphoenolpyruvate (PEP) were determined by metabolomics analysis from livers of WT and TB mice. H) RT-PCR of glycolytic gene expression of L-PK (Pklr) and GK (Gck) over the circadian cycle. I) Gene expression of lactate dehydrogenases Ldha and Ldhc in WT and TB mice by RT-PCR. J) Levels of pyruvate were determined by metabolomics in livers of WT and TB mice. Error bars indicate SEM. Significance was calculated using Student’s T test and * indicates a p-value cutoff of 0.05.

Figure 7. Lung adenocarcinoma distally rewires circadian hepatic metabolism

Schematic overview depicting the effects of the tumor macroenvironment on circadian hepatic metabolism. Our results show that lung tumors, acting through the inflammatory STAT3-Socs3 axis, operate to distally rewire circadian transcription and metabolism by acting as an endogenous circadian organizer (ECO). This manifests in loss of hepatic insulin signaling, glucose intolerance, and deregulated lipid metabolism through the AMPK/SREBP pathway.