Metabolomic profiling in a Hedgehog Interacting Protein (Hhip) murine model of chronic obstructive pulmonary disease

Abstract

Genetic variants annotated to the hedgehog interacting protein (HHIP) are robustly associated with chronic obstructive pulmonary disease (COPD). Hhip haploinsufficiency in mice leads to increased susceptibility towards the development of emphysema following exposure to chronic cigarette smoke (CS). To explore the molecular pathways which contribute to increased susceptibility, we performed metabolomic profiling using high performance liquid chromatography tandem mass spectroscopy (LC/MS-MS) on plasma, urine, and lung tissue of Hhip ^+/- heterozygotes and wild type (Hhip ^+/+) C57/BL6 mice exposed to either room-air or CS for six months. Univariate comparisons between groups were made with a combined fold change ≥2 and Student's t-test p-value < 0.05 to denote significance; associations with mean alveolar chord length (MACL), a quantitative measure of emphysema, and gene-by-environment interactions were examined using empiric Bayes-mediated linear models. Decreased urinary excretion of cotinine despite comparable plasma levels was observed in Hhip ^+/- heterozygotes; a strong gene-by-smoking association was also observed. Correlations between MACL and markers of oxidative stress such as urinary methionine sulfoxide were observed in Hhip ^+/- but not in Hhip ^+/+ mice. Metabolite set enrichment analyses suggest reduced antioxidant capacity and alterations in macronutrient metabolism contribute to increased susceptibility to chronic CS-induced oxidative stress in Hhip haploinsufficiency states.

Figure 1

Urinary and plasma cotinine levels by genotype. Urinary cotinine (normalized for creatinine concentration) relative to plasma cotinine in Hhip ^+/− heterozygotes (left panel) and Hhip ^+/+ wild type mice (right panel) subjected to chronic cigarette smoke. In Hhip ^+/+ wild type mice, a strong inverse correlation between urine and plasma cotinine levels exist (Pearson rho = −0.89, p-value = 0.04) whereas in Hhip ^+/− heterozygotes, no correlation was found (Pearson rho = 0.12, p-value = 0.88). The best fit line is plotted in blue while the 95% confidence interval is plotted in dark gray.

Figure 2

Association between plasma C56:10 (panel A) and C58:10 (panel B) triacylglycerol (TAG) and lung mean alveolar chord length (MACL) by experimental condition. A strong correlation between C56:10 TAG (Panel A, Pearson’s rho = 0.97, p-value = 7.22 × 10^–3) and C58:10 TAG (Panel B, Pearson’s rho = 0.97, p-value = 7.14 × 10^–3) and MACL is observed Hhip ^+/+ wild type exposed to room air which is not observed in Hhip ^+/+ wild type exposed to chronic cigarette smoke or in Hhip ^+/− heterozygotes exposed to either experimental condition. The best fit line is plotted in blue while the 95% confidence interval is plotted in dark gray.

Figure 3

Association between urinary thiamine/creatinine ratio and lung mean alveolar chord length (MACL). A significant association between urinary thiamine/creatinine was noted in Hhip ^+/− heterozygotes exposed to chronic cigarette smoke (CS) (lower right panel, Pearson’s rho = 0.99, p-value = 1.67 × 10^–3). No association was observed in Hhip ^+/− mice exposed to room air or in Hhip ^+/+ wild type mice in either experimental condition. The best fit line is plotted in blue while the 95% confidence interval is plotted in dark gray (except Hhip ^+/+, room air, where wide 95%CI exceeds panel borders).

Figure 4

Association between urinary methionine sulfoxide/creatinine ratio and lung mean alveolar chord length (MACL). A significant association between urinary methionine sulfoxide/creatinine was noted in Hhip ^+/− heterozygotes exposed to chronic cigarette smoke (CS) (lower right panel, Pearson’s rho = 0.99, p-value = 3.94 × 10^–3). No association was observed in Hhip ^+/− mice exposed to room air or in Hhip ^+/+ wild type mice in either experimental condition. The best fit line is plotted in blue while the 95% confidence interval is plotted in dark gray.

Figure 5

Metabolite set enrichment analysis based on differentially expressed metabolites identified in plasma (a) and urine (b) from Hhip ^+/− heterozygotes exposed to chronic cigarette smoke relative to mice exposed to room air. Metabolite sets significant at an FDR < 0.05 are denoted with a blue star.