Isothermal microcalorimetry measures UCP1-mediated thermogenesis in mature brite adipocytes

Abstract

The activation of thermogenesis in adipose tissue has emerged as an important target for the development of novel anti-obesity therapies. Using multi-well isothermal microcalorimetry, we have demonstrated that mature murine brown and brite adipocytes produce quantifiable heat upon β₃-AR stimulation, independently of any anaerobic mechanisms. Additionally, in brite adipocytes lacking UCP1 protein, β₃-AR stimulation still induces heat production, albeit to a much lower extent than in their wildtype counterparts, suggesting that UCP1 is an essential component of adrenergic induced thermogenesis in murine brite adipocytes exvivo. Similarly, we could observe an increase in heat production in human-derived adipocytes (hMADS) upon β-AR stimulation. Collectively, these results establish the use of isothermal microcalorimetry as a sensitive and accurate technique for measuring thermogenic responses in intact mature brite adipocytes from murine and human origin.

Fig. 1. Methodological overview of isothermal microcalorimetry.

A brief methodological overview showing the adipocyte isolation protocol and the subsequent measurement of heat production in the isothermal microcalorimeter. Figure created in part with with BioRender.com.

Fig. 2. Isothermal microcalorimetry allows for long term measurement of basal heat production in mature inguinal brite adipocytes.

a Representative trace of the heat output in one microwell in the isothermal microcalorimeter; frictional heat output is shown as indicated; the steady state indicates the region used to calculate cumulative heat production. b, d Heat flow in μW. c, e Accumulated heat in μJ recorded from wells in triplicates containing serially diluted isolated mature brite adipocytes for 3 and 18 h from one representative preparation. f Normalized average accumulated heat (corrected for wells containing only medium) from wells containing isolated mature brite adipocytes over 1 h expressed as a function of dilution factor from four different preparations plated in triplicates from four independent adipocyte isolations (n = 4). The curves show mean value $\pm$ SEM. Linear regression analysis was performed to calculate the R² and P value.

Fig. 3. Brown adipocytes as a classical model to demonstrate thermogenesis.

a, f Representative traces of O₂ consumption over time in exvivo brown adipocytes from WT (solid line) and UCP1 KO mice (dashed line) on a C57BL6 genetic background measured with Oroboros Oxygraph-2K; stimulations are indicated on the graphs, “1” representing addition of oligomycin (2 µg/ml), “2” indicating the addition of Vehicle (black) or 1 µM CL316243 (red), and “3” representing FCCP (40 μM) additions. b, g Quantifications of O₂ consumption, normalized as percentage of basal respiration. The data is presented as a bar representing the mean value from independent adipocyte isolations represented by symbols in the scatter plot. Vehicle treated cells are shown in black and CL316243 treated cells are shown in red. Filled bars represent WT cells whereas empty bars denote cells derived from UCP1KO. Experiments were performed on three independent cell preparations for WT (n = 3) and 6 for KO mice (n = 6). Statistics was calculated using one-way ANOVA, followed by a Dunnett’s multiple comparison test. P value for WT Basal vs. FCCP is 0.0003 (***); WT Basal vs. CL316243 is 0.0006 (***), WT Basal vs. FCCP is 0.0040 (**), KO Vehicle Basal vs. FCCP is 0.0001 (***) and KO CL316243 Basal vs. FCCP is >0.0001 (****). Isothermal microcalorimetric measurements of c, h heat flow, d, i accumulated heat and e, j accumulated heat expressed as a percentage over basal for the first hour of the experiment for isolated mature brown adipocytes from UCP1 WT (filled line and bar) and KO (dashed line and empty bar) treated with vehicle (black) or 1 µM CL316243 (red) (n = 3). Statistics were calculated using paired t-test on the underlying raw data, P value for WT Vehicle vs. CL316243 = 0.0089 (**). The curves show mean value $\pm$ SEM. The bars presented in this figure represent the mean value whereas the symbols represent each individual adipocyte isolation.

Fig. 4. Adrenergic stimulation results in concentration-dependent heat production in inguinal white adipocytes.

a Heat flow, b accumulated heat, and c total accumulated heat for the first hour of the experiment for isolated mature inguinal white adipocytes (n = 10) from NMRI mice treated with Vehicle (black line and black bar) or 1 µM CL316243 (red line and red bar). d Concentration-dependent effects of CL316243 treatment in isolated mature inguinal white adipocytes (n = 3), the curve was fit using non-linear regression. Statistics was calculated using paired t-test on the underlying raw data. P value for vehicle vs. CL316243 is <0.0001 (****). The curves show mean value $\pm$ SEM. The bars presented in this figure represent the mean value whereas the symbols represent each individual adipocyte isolation.

Fig. 5. Oxygen consumption of mature inguinal white adipocytes isolated from 3-weeks-old NMRI mice directly correlated to heat production.

a Representative oxygen consumption traces of freshly isolated mature inguinal white adipocytes pre-treated with vehicle (black line) or CL316243 (red line) and measured in parallel. FCCP was titrated up to 40 µM, additions are indicated as arrows. b In another preparation of adipocytes, CL316243 was added to respiring cells during oxygen consumption recording, red indicating oxygen consumption after CL316243 addition. c Quantification of mature inguinal white adipocytes oxygen consumption measured as in a. The basal oxygen consumption of vehicle-treated cells (black) or oxygen consumption before CL316243 addition (red) was taken as 100%. FCCP-induced or CL316243-induced respiratory rate of the same population cells as well as respiratory rates of measured in parallel CL316243-treated cells in basal state and after FCCP were expressed relatively to this 100%. Data are represented as mean $\pm$ SEM, n = 3 for FCCP effect and n = 6 for CL316243 effect. Statistics was performed using two-way ANOVA followed by a Tukey’s multiple comparison test. P value for vehicle basal vs. CL316243 is 0.0399 (*), vehicle basal vs. FCCP is <0.0001 (****) and basal CL316243 vs. FCCP is 0.009 (###). d Representative recordings of oxygen concentration (blue line) and oxygen consumption rate (red line) of mature inguinal white adipocytes in Oroboros Oxygraph-2K in parallel to recordings of heat production in isothermal microcalorimeter. To mimic condition of open system in microcalorimeter, an oxygenation was constantly and continuously maintained in Oroboros chamber with exception of several first and last minutes of recordings when chamber was closed. The last period indicated as “WAT basal” corresponds to stable state of heat production (as shown on Fig. 2a). e Correlation between oxygen consumption rate (“WAT basal” as shown in d) and heat production (measured principally as shown in Fig. 2a simultaneously in 3–4 wells) in vehicle-treated cells. Linear curve fitting (solid line) gives the correlation coefficient equal 0.477. No extra heat production independent on oxygen was observed (heat equal –0.55 when oxygen equal 0). f Correlation between oxygen consumption rate and heat production in CL316243-treated cells. Linear curve fitting gives the correlation coefficient equal 0.456 (solid red line). Only very small extra heat production independent on oxygen was observed (heat equal 1.55 when oxygen equal 0). g Representative oxygen consumption trace of mature inguinal white adipocytes collected after 40 min measurement in isothermal microcalorimeter stimulated with CL316243 (1 µM). The bars presented in this figure represent the mean value whereas the symbols represent each individual adipocyte isolation.

Fig. 6. UCP1 is the primary driver of thermogenesis in brite adipocytes.

a, d Heat flow, b, e accumulated heat, and c, f accumulated heat expressed as a percentage increase over basal for the first hour of the experiment for isolated mature inguinal from WT mice (n = 4, solid line and solid bar) and UCP1 KO mice (n = 6, dashed line and empty bar) on the same genetic background treated with either vehicle (black) or 1 µM CL316243 (red). Statistics was calculated using paired t-test on the underlying raw data. P value for Vehicle vs. CL316243 is 0.0114 (*) in WT and 0.0172 (*) in UCP1 KO cells. The curves show mean value ± SEM. The bars presented in this figure represent the mean value whereas the symbols represent each individual adipocyte isolation.

Fig. 7. hMADS retain the capacity to produce heat in vitro.

a, d Heat flow, b, e accumulated heat, and c, f total accumulated heat for the first hour of the experiment for differentiated Human multipotent adipose-derived stem cells (hMADS) treated with either Vehicle (black line and black bar), CL316243 (n = 6, red line and red bar) and Noradrenaline (n = 5, green line and green bar) plated in separate wells. Statistics was calculated using t-test. P value for Vehicle vs. CL316243 is 0.0265 (*) and Vehicle vs. Noradrenaline is 0.0172 (*). The curves show mean value $\pm$ SEM. The bars presented in this figure represent the mean value whereas the symbols represent each individual well containing independent preparations of hMADS.