Inflammation‐induced macrophage lysyl oxidase in adipose stiffening and dysfunction in obesity

Abstract

Obesity is associated with adipose tissue (AT) fibrosis with aggregation or crosslinking of collagen fibers.1, 2 Lysyl oxidase (LOX) gives rise to collagen crosslinking, while upregulated LOX is reported in AT of obese humans and mice.3, 4 Our data showed higher second harmonic generation signals as well as collagen crosslinks in ob/ob AT than wild-type (WT) AT (Figure 1A and B). In ob/ob AT, expression of LOX was highly upregulated among the LOX family, while a similar trend of LOX increment was also found in high-fat diet-induced obese mice (Figure 1C and S1). The ob/ob AT also exhibited higher LOX protein levels and enzymatic activity (Figure 1D and E). Increased AT stiffness was previously found in obese subjects non-invasively.5 Our direct measurements by atomic force microscopy (AFM) showed a similar trend by yielding higher effective Young's modulus (Eeff) in ob/ob mice and obese human subcutaneous AT (Figure 1F and G). With adipose tissue derived from obese subjects before and after weight loss surgery, we observed significantly attenuated LOX expression but not that of other LOX family members (Figure 1H), accompanied by reduced stiffness (Figure 1I). While obesity-induced physical property changes in AT may confine adipocyte functions,6 the origin and consequences of structural changes in AT as well as the augmented LOX during obesity await investigation. Immunofluorescence staining revealed increased LOX signals in ob/ob AT, predominantly expressed in crown-like structures and was highly associated with F4/80 signals (Figure 2A and B). To test whether obesity upregulated LOX in macrophages, we isolated peritoneal macrophages from WT and ob/ob mice and found that mRNA and protein levels and enzymatic activity of LOX were significantly augmented in ob/ob peritoneal macrophages (Figure 2C–E). Moreover, ob/ob peritoneal macrophages showed the ability to stiffen type I collagen gel, a mimicry of AT environment with its predominant extracellular matrix protein (Figure 2F). Conversely, our data demonstrated a reduction in LOX transcript levels in peripheral blood mononuclear cells (PBMCs) from human subjects 6 months after weight loss surgery, which was further reduced in PBMCs from subjects 12 months after surgery (Figure 2G). These results implied obesity was associated with increased macrophage LOX. Because LOX regulation by different cell types among distinct tissues was reported,7 we herein tested various cell types and stimuli to delineate the cause of LOX upregulation in obese AT (Figure S2A–F). Macrophage RAW 264.7 cells exhibited significant LOX induction by lipopolysaccharide (LPS; inflammation) but not by CoCl2 (hypoxia) or TGFβ (fibrosis) (Figure 3A). LPS or a combined inflammatory regimen increased pro-LOX and LOX proteins in RAW 264.7 cells but not the other cell types (Figure 3B and S2G and H). Importantly, LPS induced LOX enzymatic activity in RAW 264.7 cells (Figure 3C) and their ability to stiffen collagen gel; however, a LOX inhibitor, β-aminopropionitrile (BAPN) abolished this ability (Figure 3D). In mouse primary peritoneal macrophages, LPS and cytokine cocktail induced pro-LOX and LOX proteins and their enzymatic activity as well as the ability of macrophages to stiffen collagen gel (Figure S3). Collectively, these results implied increased LOX levels in macrophages caused by inflammatory stimuli resulted in AT stiffening. Thus, we hypothesized that obesity-associated inflammatory responses upregulate LOX and collagen crosslinking, leading to AT stiffening and dysfunction. Previous studies demonstrated that stiffened environment jeopardizes adipogenesis and mature adipocyte functions.8, 9 Based on the culture system and results established in Figure 3D, providing a bidirectional tool studying not only cell-to-matrix but also matrix-to-cell communications, we then tested whether LOX-stiffened environment mediated by inflamed macrophages changed adipocyte behavior. RAW 264.7 cells were firstly cultivated on collagen gel with LPS and BAPN. After a decellularization procedure of washing away RAW 264.7 cells and residual chemicals, differentiated 3T3-L1 adipocytes were then seeded on the macrophage-processed collagen gel (Figure 3E). Adipokine secretion and insulin sensitivity diminished in adipocytes seeded on LPS-treated macrophage-processed gel, but BAPN co-treatment during LPS administration restored these reductions (Figure 3F and G). These results implied that LOX, produced by inflamed macrophages, mediates substratum stiffness and subsequently affects adipocyte function. The involvement of LOX and macrophage in adipose tissue stiffening and metabolism was dissected with LOX inhibition by BAPN and macrophage depletion by clodronate. BAPN has shown a beneficial effect on glucose metabolism.4, 10 Ex vivo BAPN treatment significantly enhanced insulin-induced Akt phosphorylation and adiponectin expression, accompanied by decreased ob/ob AT stiffness (Figure 3H–J and S4). Consistently, in vivo BAPN treatment in ob/ob mice reduced AT stiffness (Figure 3K). In vivo clodronate treatment reduced pro-LOX and LOX proteins, LOX enzymatic activity, and Eeff in ob/ob gonadal AT (Figure 4A–C) on top of the overall improvements in metabolism previously reported. To further assess the involvement of macrophage LOX in AT stiffening and systemic metabolism in vivo, we adopted bone marrow transplantation to reconstitute LOX-knockdown bone marrow cells (BMCs, Figure S5A) in ob/ob mice. Eight weeks after transplantation, ob/ob mice receiving LOX-knockdown BMCs exhibited diminished Lox transcript and pro-LOX protein levels in gonadal AT (Figure 4D and E). Transplantation markedly suppressed LOX activity and reduced AT stiffness (Figure 4F and G). LOX knockdown in BMCs did not affect macrophage infiltration into AT, reflected by persisting macrophage markers, but improved AT functioning manifested by increased expression of adipocyte functional signature genes in ob/ob AT (Figure 4H). While LOX knockdown in BMCs had no impact on the body or AT weight (Figure S5B), a reduction in insulin area under curve and insulin resistance index during oral glucose tolerance test was discovered after transplantation (Figure 4I). These results suggested the involvement of macrophage LOX in obesity-induced AT stiffening and metabolic malfunction. In conclusion, we identified inflamed macrophages as key players in substratum stiffening by upregulating and releasing LOX, which further stiffened the environment where the cells were cultivated, leading to dysfunctional adipocytes. Macrophage depletion or LOX inhibition attenuated obesity-induced LOX and AT stiffening. Furthermore, ob/ob mice reconstituted with LOX-knockdown BMCs showed decreased AT stiffness and improved metabolism. Ongoing studies have gradually untangled underlying mechanisms of obesity, our study at least supports LOX derived from inflamed macrophages exerts AT stiffening and malfunction in obesity. This research also extends our perception of well-known chemical effects of macrophages to a newly explored mechanical impact in obesity. We thank Dr. Wan-Chun Chen, Dr. Hsiu-Kuan Lin, and Dr. Ching-Yi Liu at Department of Physiology, National Cheng Kung University for critical suggestions; Dr. Chia-Jung Li, Dr. Chia-Liang Yen, and Chang-Jung Chen at Institute of Clinical Medicine; the Core Laboratories of Center for Clinical Medicine Research at National Cheng Kung University Hospital; the Laboratory Animal Center, College of Medicine, National Cheng Kung University and Taiwan Animal Consortium; and Dr. Dean Tai and Ya-yun Ren of Histoindex Pte. Ltd. for technical supports. We thank Editage for language editing. This work was supported by grants from Ministry of Science and Technology (MOST-107-2320-B-006-063MY3 and MOST-107-2320-B-006-003), National Health Research Institutes (NHRI-EX107-10511SI), and National Cheng Kung University Top-Notch Project. The authors declare no conflict of interest. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts