Aggregating in vitro-grown adipocytes to produce macroscale cell-cultured fat tissue with tunable lipid compositions for food applications

**Background:** Cultivated meat production has primarily focused on muscle tissue, but fat is critical for meat flavor, texture, and species-specific taste. Producing macroscale fat tissue is challenging due to mass transport limitations—oxygen and nutrients can only diffuse through several hundred microns of dense tissue. Current tissue engineering solutions like vascularization or perfusion channels are complex and difficult to scale. This study aimed to develop a simpler method: grow adipocytes in thin 2D layers with easy nutrient access, then aggregate them into bulk 3D fat tissue after maturation, eliminating the need for ongoing cell survival in the final food product.

**Methods:** Murine 3T3-L1 cells and primary porcine DFAT cells were cultured and differentiated into adipocytes. 3T3-L1s were induced for 2 days then maintained in lipid accumulation media for 15–30 days. Porcine DFAT cells underwent 4 days induction and 26 days accumulation. Adipocytes were harvested mechanically using cell scrapers, which aggregated cells into tissue-like masses. These were formed into 3D constructs using either sodium alginate (0.8% or 1.6% final concentration) or microbial transglutaminase (mTG, 3% final concentration) in 3D-printed cylindrical molds (6 mm diameter, 3 mm height). Unconfined uniaxial compression testing was performed to 50% strain at 1.667%/s. Lipidomics via LC-MS/MS characterized fatty acid compositions of triacylglycerides (TAGs) and phospholipids. Lipid supplementation experiments used Intralipid (soybean oil emulsion) at 0–1000 µg/ml. Native adipose tissues from mouse (7-day and 67-day), pig, cow, and chicken were used as comparators.

**Key Results:** After 15 days of differentiation, 81.6% (±5.5%) of 3T3-L1 cells were PPARγ-positive. Thirty-day adipocytes contained ~3 times more lipid per cell than 15-day adipocytes. Mean lipid droplet diameters were 8.1 µm (±0.04 SEM) at 15 days and 11.1 µm (±0.1 SEM) at 30 days. One 175 cm² culture flask yielded ~0.8 g of cell-scraped adipocytes. For mechanical properties, alginate-based cultured fat tissues showed mean compressive stresses of 11.1 kPa (±5.2) for 15-day and 12.8 kPa (±7.3) for 30-day adipocytes at 50% strain (p=0.7304). These values fell within the range of native animal fats: beef fat 20.2 kPa (±12.5), pork fat 35.6 kPa (±17.5), and chicken fat 6.2 kPa (±2.0). mTG-based tissues were weaker (0.5–0.7 kPa at 50% strain), closer to rendered lard (4.6 kPa ±1.4) and tallow (2.5 kPa ±0.9). Lipidomics revealed that unsupplemented murine adipocytes were enriched in 16:1 and deficient in 18-carbon FAs (18:1, 18:2) compared to native mouse fat. Intralipid supplementation (500–1000 µg/ml) dose-dependently shifted FA profiles toward native fat, increasing 18:1 and 18:2 while decreasing 16:1. Polyunsaturated FA levels in murine TAGs rose from 2.9% (±0.1) without Intralipid to 12.3% (±2.7) with 1000 µg/ml Intralipid, similar to 7-day mouse fat. For porcine cells, in vitro and native TAG profiles were more similar, though in vitro samples lacked 20:4 (>10% difference) in phospholipids. Porcine cells cultured in 2% FBS showed over 2× the lipid accumulation but over 2× fewer cells compared to 20% FBS.

**Clinical Implications:** This study provides a proof-of-concept for scalable cultured fat production using a simple aggregation strategy that bypasses complex tissue engineering. The approach is cell-type and species-agnostic, demonstrated with both murine cell lines and primary porcine cells. The tunability of fatty acid composition via lipid supplementation offers a route to match the flavor and nutritional profiles of specific animal fats. The mechanical properties can be adjusted through binder selection and concentration. For food applications, this method could enable production of fat for cultivated meat or enhancement of plant-based meat alternatives. However, the study notes that structured cultured fat tissues may have lower caloric density due to dilution by binders (e.g., 1:1 mixing with alginate). Future work should include sensory evaluation panels, gas chromatography/mass spectrometry for volatile compounds, calorimetry for caloric content, and thermogravimetric analysis for cooking behavior. Scaling up will require transitioning from 2D flask culture to bioreactor systems such as fixed-bed or suspension bioreactors.