Effects of tunable, 3D-bioprinted hydrogels on human brown adipocyte behavior and metabolic function

Abstract

Obesity and its related health complications cause billions of dollars in healthcare costs annually in the United States, and there are yet to be safe and long-lasting anti-obesity approaches. Using brown adipose tissue (BAT) is a promising approach, as it uses fats for energy expenditure. However, the effect of the microenvironment on human thermogenic brown adipogenesis and how to generate clinically relevant sized and functioning BAT are still unknown. In our current study, we evaluated the effects of endothelial growth medium exposure on brown adipogenesis of human brown adipose progenitors (BAP). We found that pre-exposing BAP to angiogenic factors promoted brown adipogenic differentiation and metabolic activity. We further 3D bioprinted brown and white adipose progenitors within hydrogel-based bioink with controllable physicochemical properties and evaluated the cell responses in 3D bioprinted environments. We used soft, stiff, and stiff-porous constructs to encapsulate the cells. All three types had high cell viability and allowed for varying levels of function for both white and brown adipocytes. We found that the soft hydrogel constructs promoted white adipogenesis, while the stiff-porous hydrogel constructs improved both white and brown adipogenesis and were the optimal condition for promoting brown adipogenesis. Consistently, stiff-porous hydrogel constructs showed higher metabolic activities than stiff hydrogel constructs, as assessed by 2-deoxy glucose uptake (2-DOG) and oxygen consumption rate (OCR). These findings show that the physicochemical environments affect the brown adipogenesis and metabolic function, and further tuning will be able to optimize their functions. Our results also demonstrate that 3D bioprinting of brown adipose tissues with clinically relevant size and metabolic activity has the potential to be a viable option in the treatment of obesity and type 2 diabetes.

Statement of significance: One promising strategy for the treatment or prevention of obesity-mediated health complications is augmenting brown adipose tissues (BAT), which is a specialized fat that actively dissipate energy in the form of heat and maintain energy balance. In this study, we determined how pre-exposing human brown adipose progenitors (BAP) to angiogenic factors in 2D and how bioprinted microenvironments in 3D affected brown adipogenic differentiation and metabolic activity. We demonstrated that white and brown adipogenesis, and thermogenesis were regulated by tuning the bioprintable matrix stiffness and construct structure. This study not only unveils the interaction between BAP and 3D physiological microenvironments, but also presents a novel tissue engineered strategy to manage obesity and other related metabolic disorders.

Keywords: Brown adipocytes; Obesity; Porosity; Stiffness; Tissue engineering.

Fig. 1.

Pre-exposure to angiogenic factors promoted the brown-adipogenic potential of human BAP. (A) qPCR analysis of Ucp1, Zic1, Vegfr1, Pgc1a and Hdac1; (B) Protein levels of UCP1, HDAC1, histone3, and α-actin as a loading control (n = 4 per group); (C) Oxygen consumption rate (OCR) by Seahorse XF analyzer; (D) Brown-specific miRNA levels of miR-30b, 193b, 365 and 378. U6 small nuclear RNA2 (RNU6–2) was used as a reference (n = 4–5 per group), *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 2.

The tunability of the bioprinted hydrogel constructs allows for varying physical properties and surface areas. (A) Multiple bioprinting patterns and crosslinking times creates different geometry and stiffness (Soft = UV crosslinking for 30 s, Stiff = UV crosslinking for 90 s, Solid = bioprinted constructs with solid structure, Porous = bioprinted constructs with porous structure); (B) Compressive modulus of the different hydrogel constructs; n = 5–6; **p < 0.01; (C) Typical SEM and optical micrographs.

Fig. 3.

Live/dead staining of the three types of hydrogel constructs shows the cell viability after induction. (A) Hydrogel constructs with WAP encapsulated; (B) Hydrogel constructs with BAP encapsulated; (C) Semi-quantitative measurement of cell viability based on Live/Dead images (n = 6).

Fig. 4.

Lipid production of white adipocytes across all constructs in IM and soft constructs in GM. (A) Bodipy staining for lipids and nuclear staining (scale bar, 100 μm); (B) Size and size distribution of the single lipid droplets; (more than 100 lipid droplets from at least three IF images); (C) Lipid density (calculated by comparing total lipid area with total cells in the images, n = 5–6 images); (D) qPCR analysis of white adipogenesis related genes expression. Relative gene expression is presented as normalized to 18S and expressed relative to soft, solid constructs in GM. (n = 3, *p < 0.05, **p < 0.01).

Fig. 5.

Staining and gene expression across all constructs in IM and soft constructs in GM. (A) Staining of brown adipose constructs. Bodipy staining for lipids, UCP-1 staining for a brown adipogenic marker, and nuclear staining; (B, C) qPCR analysis of brown adipogenesis and inflammation related genes expression. (B) Relative gene expression is presented as normalized to 18 s and expressed relative to soft, solid constructs with WA in GM. (n = 3, bars that do not share letters are significantly different from each other, **p < 0.01, # indicates p < 0.05 comparing to soft, solid constructs with WA, & indicates no significant difference comparing to stiff, solid constructs with BA); (C) Relative gene expression is presented as normalized to 18 s and expressed relative to soft, solid constructs with BA in IM. (n = 3, *p < 0.05).

Fig. 6.

Hydrogel stiffness and porosity altered metabolic function of 3D bioprinted BAP laden constructs. Roughly 5 × 10⁶ of WAP and BAP were bioprinted in differential bioink and bioprinting structure conditions. Glucose uptake was measured by using ³H-2-deoxy glucose (DOG) as a metabolic tracer. (A) Basal levels of 2-DOG uptake in WA and BA cultures; (n = 4, **p < 0.01, ***p < 0.001), (B) Oxygen consumption rate in BA cultures in stiff or stiff-porous hydrogel scaffolds (n = 4 per group). No cell control was used as a negative control. (C) 2-DOG uptake in the presence and absence of insulin (100 nM) for 90 min. Data normalized by protein levels and expressed as cpm/g protein/per 3D bioprinted construct (n = 4, *p < 0.05, **p < 0.01).