Thermoneutrality and Preclinical Obesity Models
The Need to Consider the Murine Thermoneutral Zone in Preclinical Obesity Studies
Mouse models are used to study obesity and to test drugs for humans that affect energy homeostasis and weight loss. These two species have different thermal biologies, yet their divergent strategies of energy expenditure in the aims of temperature maintenance are typically not considered.
Human ambience occurs around 20-22oC, a range wherein mice must expend significant amounts of energy to maintain a constant body temperature. This has profound effects on their physiology, endocrinology, and activity. Nevertheless, most preclinical obesity studies are conducted within the human ambient range, which may compromise translatability.
Ambient Temperature and the Thermoneutral Zone
Mice and humans maintain similar body temperatures (Tb) of about 36 to 38oC. However, mice are roughly 3000 times smaller by mass. Smaller animals have a larger surface area to volume ratio and lose heat more rapidly to their environment.
21oC is considered the lower boundary of a (clothed) human’s thermoneutral zone (TNZ). This is the range at which it can maintain a basal metabolic rate (BMR). In the TNZ, human Tb is largely supported by metabolic heat byproduct, and heat dissipation is actually a more imminent concern. In contrast, a vivarium-housed mouse at 21oC dedicates roughly one half of its total energy expenditure toward maintenance of Tb, and it exhibits 2.5X the metabolic rate of humans.
Mouse TNZ is achieved at higher temperatures, with most studies involving thermoneutrality being conducted at 30oC. In fact, mice demonstrate an interesting propensity: they adjust Tb from 36oC up to 38oC along the Ta range of 29oC to 34oC in order to keep energy expenditure, and thus BMR, stable. In part, this reflects the relative lack of heat dissipation mechanisms in rodents compared to humans.
Most vivaria maintain ambient temperature within the human TNZ, forcing study animals to engage in facultative thermogenesis. This has profound consequences on energy homeostasis and physiology. When housed at 30oC, mice consume less food, have lower heart rates and blood pressure, and reduce catecholamine and corticosteroid signaling.
What does this mean for mouse obesity models? Giles et. al. have shown that mice housed at 30oC and fed a high fat diet demonstrate enhanced weight gain over mice housed at 22oC, accompanied by profound exacerbation of NAFLD and associated markers of inflammation and liver dysfunction. Housing at thermoneutrality has even been shown to promote NAFLD induction in female mice, which are typically more resistant than males to developing this comorbidity from HFD.
Ambient Temperature and Preclinical Efficacy
Obesity models run at conventional, human-friendly ambient temperatures may blunt the severity or even the appearance of disease phenotypes that are needed for efficacy endpoints. They may also mask the effect of drugs that affect energy homeostasis for reversal of obesity. To date, unfortunately, the literature is sparse on proof-of-concept or preclinical efficacy data that put this theory to test.
Goldgof et. al. evaluated a discontinued weight loss drug, 2,4dinitrophenol (DNP), at human and mouse Ta’s. DNP is a proton ionophore, and acts as a mitochondrial uncoupler. At 22oC, systemic DNP-mediated heat generation substituted for UCP1 brown adipose tissue (BAT) thermogenesis, with no net effect on total energy expenditure or adiposity. In contrast, at 30oC the pharmacological increase in energy expenditure was noncompensatory. Therefore, the total energy expenditure was enhanced and adiposity was reduced compared to control mice.
Xiao et. al. evaluated the β3andrenergic agonist CL316243 at 22oC and 30oC. This type of agonism activates BAT through UCP1 induction, but also triggers lipolysis, beiging, and thermogenesis in white adipose tissue. Although CL316243 increased the energy expenditure of HFD-fed mice at both temperatures, reductions of body weight and adiposity were minimized at 22oC due to the animals exhibiting compensatory food intake. In contrast, at 30oC, food intake was less accentuated, and CL316243-treated mice became appreciably leaner. Furthermore, treatment at 30oC elicited more positive effects on insulin sensitivity that were not apparent at 22oC.
So what about using thermoneutral conditions to unveil the efficacy of drugs that act completely independently to axes involved in thermogenesis? This is ripe for investigation, as my searches identified no published studies. Surely, however, thermoneutrality represents an interesting new context in which to reevaluate therapeutic candidates that were abandoned after trials at conventional ambient temperature.
Mice engage thermogenic mechanisms at human ambient temperature which can be attenuated by housing at the murine thermoneutral zone. This strategy may prevent confounding of disease phenotypes through increased energy expenditure, and sensitize the effect of an obesity drug candidate. This is a demonstrably valuable strategy to enhance the development of NAFLD for research on the comorbidities of obesity.