A Pathophysiologically Hypertrophic 3T3-L1 Cell Model-An Alternative to Primary Cells Isolated from DIO Mice

Abstract

Adipocyte hypertrophy in individuals with obesity is connected to alterations in adipocyte function. These pathophysiological changes are studied using animal models and adipose tissue engineering. However, knockdown, overexpression, and stimulation studies would benefit from an easily applicable cell model. Although several models (free fatty acids, glucose restriction, and long-term incubation) have previously been described, our evaluation demonstrated that they lack important features described for hypertrophic adipocytes found in obesity. Therefore, we aimed to develop a cell model depicting the pathophysiological state of adipocytes in obesity by applying novel approaches (insulin, macrophage supernatant, and Tnfα) using 3T3-L1 cells. To analyze changes in adipocyte phenotype and function, we detected the cell size, lipid accumulation, insulin sensitivity, cytokine/adipokine secretion, and expression of lipolytic enzymes. Combining long-term incubation with insulin and Tnfα co-stimulation, we found significantly increased cell size and lipid accumulation compared to 3T3-L1 adipocytes differentiated with standard protocols. Furthermore, these adipocytes showed significantly reduced insulin sensitivity, adiponectin secretion, and lipolytic enzyme expression, accompanied by increased IL6 and leptin secretion. In summary, the described cell model depicts pathophysiologically hypertrophic 3T3-L1 adipocytes. This model can be used for knockdown, overexpression, and stimulation studies, thereby serving as an alternative to primary cells isolated from DIO mice.

Keywords: adipocyte hypertrophy; adipose tissue; cell model; obesity.

Figure 1

Depiction of (patho)physiologically hypertrophic 3T3-L1 cell models. Using various approaches (I: Reduced Glucose; II: long-term incubation; III: FFAs; IV: Insulin; V: Long-term Incubation + Insulin; VI: long-term incubation + Insulin + M0, M1, or M2 THP1 macrophage supernatant; VII: long-term incubation + Insulin + Tnfα), 3T3-L1 cells were induced to become (patho)physiologically hypertrophic. Created in Biorender. Thor, D. (2025) https://BioRender.com/l205457 (accessed on 1 May 2025).

Figure 2

Changes in lipid accumulation of 3T3-L1 adipocytes after glucose restriction, stimulation with free fatty acids (FFAs, only 750 µM palmitate shown) or insulin, or after long-term incubation. 3T3-L1 cells were differentiated under various conditions (glucose restriction, FFA or insulin stimulation, long-term incubation; Figure 1) to induce hypertrophic 3T3-L1 adipocytes. Following this, 3T3-L1 adipocytes were stained with ORO for quantitative and qualitative lipid droplet analysis. As a control, standard differentiated 3T3-L1 cells (D10 w/o) were used. (A) Microscopic pictures of the applied conditions show differences in lipid accumulation (scale bar: 50 µm). (B) For quantification, ORO was eluted from adipocytes and normalized to D10 w/o (OD_{500 nm–620 nm} = 0.51 ± 0.03, n ≥ 3). For qualitative lipid droplet analysis, (C) droplet count (n ≥ 3), (D) average droplet size (n ≥ 3), and (E) lipid droplet size distribution were measured (n ≥ 3). Droplet count and average droplet size were normalized to D10 w/o (n_droplet = 4879 ± 348, A_droplet = 28.15 ± 2.81 µm²). The mean ± SEM of biological replicates is depicted. Significance was tested using a paired Student’s t test (B–D) or two-way ANOVA (E). * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 3

Phenotypical and functional alterations of 3T3-L1 adipocytes after insulin stimulation and/or long-term incubation. 3T3-L1 cells were differentiated into mature adipocytes until D10 or D13 of differentiation, either with the supplementation of insulin from D0 to D6 of differentiation or in the continuous presence of insulin (Figure 1). As a control, standard differentiated 3T3-L1 cells (D10 w/o) were used. The plasma membranes of 3T3-L1 cells were stained to (A) microscopically visualize variations in cell size (scale bar: 50 µm) and to analyze (B) average cell size (n = 5) and (C) cell size distribution (n = 5). When investigating insulin sensitivity, (D) acute insulin-stimulated glucose uptake (n = 5) and (E) p-AKT/AKT (n = 6) were detected and normalized to non-stimulated cells. Further, the expression of (F) InsR and Glut4 was determined and normalized to D10 w/o (n = 5). (G) To verify AT marker genes for obesity and, therefore, for hypertrophy, the gene expression of the adipokines Lep and AdipoQ and of the pro-inflammatory cytokine Il6 was analyzed using FATTLAS [32]. The expected changes in the expression of all three genes were observed in two independent datasets of visceral AT and one dataset of visceral adipocytes. The expression is shown as the ratio of obese/lean. Following this, the gene expression and secretion of (H) Il6/IL6 (n ≥ 3), (I) Lep/LEP (n ≥ 3), and (J) AdipoQ/ADIPO (n ≥ 3) were measured in all conditions. To analyze adipokine secretion, cells were serum-starved overnight followed by stimulation in serum-free media supplemented with insulin. Expression data were normalized to Actb and D10 w/o, and secretion data were normalized to cell count. The ΔCt and Ct_Actb of (F,H–J) are summarized in Supplementary Table S2. The mean ± SEM of biological replicates is given. Significant changes were tested using a paired Student’s t test (B,D–J) and two-way ANOVA (C). * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 4

Changes in the adipocyte function and viability of 3T3-L1 adipocytes differentiated with insulin and co-stimulated with insulin and M0, M1, or M2 THP-1 macrophage supernatant. 3T3-L1 adipocytes were differentiated under long-term insulin stimulation until D13, with co-stimulation of M0, M1, or M2 THP-1 macrophage supernatant from D10 to D13 (Figure 1). As a control, standard differentiated 3T3-L1 cells (D10 w/o) were used. To investigate acute insulin-stimulated signaling, (A) glucose uptake (n = 4) and (B) p-AKT/AKT (n = 7) were measured. Acutely insulin-stimulated cells were normalized against non-stimulated cells. Additionally, the gene expression of (C) InsR and Glut4 was detected and normalized to Actb and D10 w/o (n ≥ 3). To investigate adipokine expression and secretion (D), Il6/IL6 (n ≥ 4), (E) Lep/LEP (n ≥ 4), and (F) AdipoQ/ADIPO (n ≥ 4) were analyzed. In the secretion assays, adipocytes were starved and stimulated in serum-free media containing insulin (acute insulin stimulation) as indicated. Expression data were normalized to Actb and D10 w/o, and secretion data were normalized to cell count. Further, the effect on (G) apoptosis and (H) necrosis (n = 3, compared to Supplementary Figure S1L) was investigated. Apoptosis was normalized to D10 w/o (RLU = 1857 ± 533). Necrotic cells were normalized to total cell count. The ΔCt and CT_Actb of (C–F) are summarized in Supplementary Table S2. The mean ± SEM of biological replicates is depicted, and significance was determined using a paired Student’s t test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 5

The effect of Tnfα stimulation on the function and viability of long-term insulin-stimulated 3T3-L1 adipocytes. 3T3-L1 adipocytes were differentiated under long-term insulin stimulation until D13 of differentiation, with optional Tnfα stimulation from D10 to D13 (Figure 1). As a control, standard differentiated 3T3-L1 cells (D10 w/o) were used. When analyzing acute insulin-stimulated signaling, (A) glucose uptake (n ≥ 3) and (B) pAKT/AKT were determined (n = 4). Acute insulin-stimulated cells were normalized to non-stimulated cells. Additionally, (C) the expression of InsR and Glut4 was measured (n = 5) and normalized to Actb and D10 w/o. Adipokine gene expression and secretion were detected for (D) Il6/IL6 (n ≥ 3), (E) Lep/LEP (n ≥ 3), and (F) AdipoQ/ADIPO (n ≥ 3). In the secretion assays, adipocytes were starved and stimulated in serum-free media containing insulin (acute insulin) and/or Tnfα (acute Tnfα) as indicated. Corresponding expression values are shown in Supplementary Table S2. Expression was normalized to Actb and D10 w/o, and secretion data were normalized to cell count. Further, (G) apoptosis (n = 4) and (H) necrosis (n = 6, compared to Figure S2L) were measured. Apoptosis was normalized to D10 w/o (RLU = 1542 ± 419), and necrotic cells were normalized to total cell count. The ΔCt and Ct_Actb of (C–F) are summarized in Supplementary Table S2. The mean ± SEM of biological replicates is shown. Significance was determined using a paired Student’s t test. * p < 0.05, ** p < 0.01.

Figure 6

The impact of Tnfα and long-term insulin stimulation on 3T3-L1 adipocytes regarding the expression of additional genes differentially regulated in the adipocytes of lean and obese mice. For further validation of the pathophysiologically hypertrophic characteristics of the cell model (3T3-L1 D13 + Insulin + Tnfα), (A) genes involved in lipid deposition, lipolysis, and adipogenesis were investigated in RNAseq datasets of visceral AT and visceral adipocytes (vAdi). Only differentially expressed genes are shown as a ratio of obese/lean. (B–E) After differentiating 3T3-L1 adipocytes under long-term insulin stimulation until D13 of differentiation, with optional Tnfα stimulation from D10 to D13 (Figure 1), the gene and protein expressions of (B) ATGL (n ≥ 3), (C) HSL (n ≥ 3), and (D) PLIN1 (n ≥ 3) were investigated representatively. One representative Western blot per protein is depicted in (E), whole Western blots are provided in Supplementary Figure S3, and protein expression values are given in Supplementary Table S4. Expression values are shown in Supplementary Table S3. As a control, standard differentiated 3T3-L1 cells (D10 w/o) were used. Gene expression data were normalized to Actb and D10 w/o, and protein expression was normalized to β-actin and D10 w/o. Given is the mean ± SEM of biological replicates. Significant changes were tested using a paired Student’s t test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.