Inflammation and Metabolism of Influenza-Stimulated Peripheral Blood Mononuclear Cells From Adults With Obesity Following Bariatric Surgery

Abstract

Background: Obesity dysregulates immunity to influenza infection. Therefore, there is a critical need to investigate how obesity impairs immunity and to establish therapeutic approaches that mitigate the impact of increased adiposity. One mechanism by which obesity may alter immune responses is through changes in cellular metabolism.

Methods: We studied inflammation and cellular metabolism of peripheral blood mononuclear cells (PBMCs) isolated from individuals with obesity relative to lean controls. We also investigated if impairments to PBMC metabolism were reversible upon short-term weight loss following bariatric surgery.

Results: Obesity was associated with systemic inflammation and poor inflammation resolution. Unstimulated PBMCs from participants with obesity had lower oxidative metabolism and adenosine triphosphate (ATP) production compared to PBMCs from lean controls. PBMC secretome analyses showed that ex vivo stimulation with A/Cal/7/2009 H1N1 influenza led to a notable increase in IL-6 with obesity. Short-term weight loss via bariatric surgery improved biomarkers of systemic metabolism but did not improve markers of inflammation resolution, PBMC metabolism, or the PBMC secretome.

Conclusions: These results show that obesity drives a signature of impaired PBMC metabolism, which may be due to persistent inflammation. PBMC metabolism was not reversed after short-term weight loss despite improvements in measures of systemic metabolism.

Keywords: bariatric surgery; immunometabolism; inflammation; influenza virus.

Figure 1.

Markers of serum inflammation are impaired with obesity. A, Florescent intensity z-score normalized data for serum markers from lean (n = 14), presurgery (n = 14), and postsurgery (n = 14) participants. Comparisons were made by nonpaired Wilcoxon (Lean–Pre, Lean–Post) and paired Wilcoxon (Pre–Post) tests with Benjamini–Hochberg (BH) correction to limit the false discovery rate. B, Multiplex assessment of interleukin 1RA, interleukin 6, epidermal growth factor, and hepatocyte growth factor serum protein concentrations determined by 5-parameter logistic curve found to be significant from (A). C, Mass spectrometry analyses of specific lipid metabolites including those related to inflammation resolution (D). Data are mean ± standard deviation, where BH adjusted P values are ⁺P < .10, *P < .05, **P < .01, ***P < .001, ****P < .0001. Abbreviations: 13-OxoOde, 13-oxo-9Z,11E-octadecadienoic acid (13-OxoOde); EET, eicosatrienoic acid; EGF, epidermal growth factor; FGF, fibroblast growth factor; G.CSF, granulocyte colony-stimulating factor; GM.CSF, granulocyte-macrophage colony-stimulating factor; HETE, hydroxyeicosatetraenoic acid; HDHA, hydroxydocosahexaenoic acid; HGF, hepatocyte growth factor; IFN, interferon; IL, interleukin; IP10, interferon gamma inducible protein-10; MCP, monocyte chemoattractant protein; MIG, monokine induced by interferon-γ; MIP, monocyte inflammatory protein; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor.

Figure 2.

Peripheral blood mononuclear cell (PBMC) metabolism is impaired with obesity. Resting and A/California/07/2009pdm H1N1 influenza–stimulated PBMCs from lean (n = 10–13), presurgery (n = 13–14), and postsurgery (n = 12–13) participants were analyzed by extracellular flux analysis on the Agilent Seahorse XFe96. Cells were plated at 200 000 cells per well in seahorse media. A, Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were measured in response to the mitochondrial stress test. B and C, Calculations for basal OCR and basal ECAR (B) and OCR:ECAR ratio and adenosine triphosphate production (C). D, Pearson correlation of unstimulated PBMC basal OCR:ECAR by body mass index (BMI) and influenza-stimulated PBMC basal OCR:ECAR by BMI. Data represent mean ± standard deviation (A–D) and values were normalized to cell number. Comparisons were made by nonpaired Wilcoxon and paired Wilcoxon tests with Benjamini–Hochberg correction: *P < .05, **P < .01, ***P < .001, ****P < .0001. Abbreviations: ATP, adenosine triphosphate; BMI, body mass index; ECAR, extracellular acidification rate; FCCP, carbonyl cyanide-4 (trifluoromethoxy) phenylhydrazone; OCR, oxygen consumption rate.

Figure 3.

The percentage of T-helper 2 (Th2) and select B-cell populations are impaired with obesity. Immune cell populations in resting and A/California/07/2009pdm H1N1 influenza–stimulated peripheral blood mononuclear cells (PBMCs) from lean (n = 10–13), presurgery (n = 13–14), and postsurgery (n = 12–13) participants. A, viSNE plots of PBMCs stained with the Human Immune Monitoring Kit and analyzed on the Fluidigm Helios 2.0 using mass cytometry by time of flight. B, Populations with significant changes based on clustering and biaxial gating (Supplementary Figure 1) for activated CD4 T cells, CD4 Th2 cells, central memory CD4 T cells, and terminal effector CD4 T cells. C, Significant changes in switched memory B cells, nonswitched memory B cells, and transitional B cells. Data are mean ± standard deviation. Comparisons were made by nonpaired Wilcoxon and paired Wilcoxon tests with Benjamini–Hochberg correction to limit the false discovery rate: ⁺P < .10, *P < .05, ***P < .001. Abbreviations: CM, central memory; DC, dendritic cell; NK, natural killer; PBMC, peripheral blood mononuclear cell; TCR, T cell receptor; TE, terminal effector; Th2, T-helper 2.

Figure 4.

Obesity increases interleukin 6 (IL-6) production in influenza-stimulated peripheral blood mononuclear cells (PBMCs). Conditioned media were used from unstimulated and A/California/07/2009pdm H1N1 influenza–stimulated PBMCs, which were assayed for 35 cytokine, chemokine, and growth factor markers by antibody multiplex. A, Florescent intensity (FI) z-score normalized data for conditioned media inflammatory markers from lean (n = 14), presurgery (n = 14), and postsurgery (n = 14) participants. Comparisons were made by nonpaired Wilcoxon and paired Wilcoxon tests with Benjamini–Hochberg corrected P values. Significance was defined as *P < .05 (see Supplementary Table 1). B, IL-6 FI z-score based on categorical groups (lean, presurgery, and postsurgery) from unstimulated and H1N1 influenza–stimulated conditioned media. Each dot represents a single individual ± standard deviation (SD). C, Spearman correlation of the IL-6 FI z-score from influenza-stimulated PBMC conditioned media by body mass index (kg/m²). Each dot represents a single individual ± SD. Abbreviations: BMI, body mass index; EGF, epidermal growth factor; FGF, fibroblast growth factor; FI, florescent intensity; G.CSF, granulocyte colony-stimulating factor; GM.CSF, granulocyte macrophage colony-stimulating factor; HGF, hepatocyte growth factor; IFN, interferon; IL, interleukin; IP10, interferon gamma inducible protein-10; MCP, monocyte chemoattractant protein; MIG, monokine induced by interferon-γ; MIP, monocyte inflammatory protein; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor.