Prolactin is named after its major role in lactation in mammals, even though it participates in multiple biological processes including reproduction, osmoregulation, immunoregulation, growth and energy metabolism (1–3), consistent with the wide distribution of prolactin receptors (PRLR) in the body (4). Prlr mRNA is expressed in numerous tissues involved in the control of energy balance, such as adipose tissue, pancreas, small intestine, liver and brain, where prolactin integrates endogenous or environmental signals to ensure metabolic homeostasis through modulation of food intake, lipid and glucose metabolism (3).

These metabolic actions of prolactin are fundamental during pregnancy and lactation, however, the association between elevated prolactin levels and increased body weight, food intake or adiposity in other situations such as prolactinoma occurrence or psychiatric treatments that target the dopamine type 2 receptors (DRD2), remains controversial. Studies in rats show that chronic hyperprolactinemia induced by regular injections of dopamine antagonists or exogenous prolactin, or ectopic pituitary transplants, is associated with increased food intake and body weight (5–7), whereas pharmacological suppression of prolactin secretion with the dopamine-agonist bromocriptine evokes the opposite outcome, being most effective in lactating rats and least in males (8). On the other hand, male mice with ectopic pituitaries show only a small increase in body weight and food intake, and a slight decline in fat mass (8). In the same line, hyperprolactinemic dopamine receptor D2 knockout (Drd2^-/- ) female mice have similar body weight to controls (9, 10), whereas body weight is remarkably increased in after 6 months of age in hyperprolactinemic mice with selective deletion of the dopamine receptor D2 in lactotropes (lacDrd2KO mice) (9).

Clinical data show an association of high very prolactin levels with increased prevalence of obesity and metabolic disorders such as dyslipidemia, glucose intolerance, and insulin resistance, and many of these disorders are improved by normalizing prolactin levels with dopamine agonists (8, 11–17). Moreover, genetic variants of the PRL locus (6p22.2 – p21.3) have been associated with alterations in body mass index in genome-wide association studies (18), and SNPs located near the PRLR gene correlate with an increased risk for gestational diabetes and obesity (19) strengthening the connection between prolactin and metabolic disorders. However, there is no strong evidence to indicate that high circulating concentrations of prolactin are a major factor in human obesity, and furthermore, prolactin levels below physiological range in humans and rodents associate with metabolic diseases implying also a protective effect of normal prolactin levels in metabolism (3, 20).

Therefore, the association of pathologically high prolactin levels and increased body weight is controversial but positive, suggesting that prolactin favors weight gain but additional factors may enhance or condition its effects.

High-Fat Diets (HFD) are well-known obesogenic stimuli, and in some circumstances, other challenges may potentiate or aggravate their metabolic effects, highlighting the multifactorial causation of obesity. In this context, we wished to determine the simultaneous effects of pathologically chronic hyperprolactinemia and HFD on food intake, adipose tissue accretion and expression of thermogenic and lipogenic markers, as well as glucose homeostasis in female mice.

PRLR signaling exerts crucial roles in the development and function of two major players of body energy balance: adipose tissue and the endocrine pancreas. Even though in transgenic mice lacking prolactin no profound metabolic phenotype was demonstrated (21), mice with Prlr deficiency showed a worsening in the induction of obesity or streptozotocin evoked diabetes (20, 22). Ample evidence has shown that this hormone participates in lipid metabolism, adipogenesis and adipocyte differentiation, as well as in pancreas development, beta islet proliferation, insulin synthesis and release (23), and the net effect of prolactin depends on its serum levels (3).

Adipose tissue is a metabolically dynamic organ, and is categorized in energy storing white adipose tissue (WAT), thermogenic brown adipose tissue (BAT) and thermogenic brown-like adipocytes named brite/beige adipocytes that are scattered within WAT (24). WAT, the main reservoir of energy is mainly categorized as subcutaneous (scWAT) and visceral (VAT) adipose tissue. Prlr mRNA has been documented in white adipocytes (25, 26), and hyperprolactinemic mouse models have increased adiposity as a result of either decreased lipolysis (9, 27) or increased lipogenesis in WAT (5). Nevertheless, opposite roles have also been described for prolactin or PRLR on adipogenesis regulation (20, 25, 28, 29).

On the other hand, BAT is a highly vascularized organ composed of adipocytes that contain small lipid droplets and a large number of mitochondria capable of oxidizing chemical energy to produce heat, providing thermogenic capacity to this tissue which depends mainly on the presence of the proton transporter uncoupling protein 1 (UCP1) (30). The role of prolactin in BAT physiology has not been extensively studied. Prlr knockout mice have smaller BAT depots and pre-adipocytes derived from this tissue are not capable of differentiating into mature adipocytes, a phenotype which can be reversed by the ectopic expression of Prlr (31).

Metabolic plasticity in response to hormonal stimuli or environmental cues has been identified as a hallmark feature of adipose tissue (32). Besides the well known plastic process of beiging, diet-induced obesity in mice evokes vascular remodeling and functional hypoxia leading to a “whitening” phenotype in BAT, characterized by mitochondrial dysfunction and loss, lipid droplet accumulation, and decreased expression of Ucp1 and Vascular endothelial growth factor, and tissue inflammation (33–36). BAT whitening is therefore associated to decreased thermogenic capacity and impaired energy balance, and the role of prolactin in this process has not been studied.

Finally, a reasonable body of evidence shows that prolactin signaling is implicated in the regulation of glucose homeostatic adaptations to pregnancy mainly through its impact in pancreatic islet cell physiology and glucose metabolism (37). In this sense, adequate levels of prolactin are beneficial for maintaining glucose homeostasis. However, pathologically high levels of prolactin may alter glucose tolerance, disrupt glucose signaling in pancreas, and induce insulin resistance (3).

We have recently described that lifelong hyperprolactinemia in lacDrd2KO female mice induced gradual metabolic alterations. Intriguingly, even though the elevated serum prolactin levels were observed since the first month of age, obesity was of late onset, beginning mildly in 6-month-old female mice and culminating in morbid obesity from 10 months of age onward (27, 38). The slow onset of the metabolic phenotype, led us to investigate whether exposure to a HFD might accelerate and intensify the late metabolic changes found in the hyperprolactinemic mouse model with special focus food intake, glucose homeostasis and adipose gene expression profiles, and the processes of whitening and beiging described in BAT and WAT, respectively.

Our results not only highlight the role of prolactin as a metabolic hormone, but also demonstrate that very high prolactin levels or exposure to a HFD in isolation may cause slight metabolic imbalance. However, the combination of these two challenges, or the combination of severe hyperprolactinemia with other metabolic challenges such as stress, anxiety, nutritional imbalance or endocrine disruptors, may lead to overweight or obesity.

Different experimental settings indicate that hyperprolactinemia is associated with weight gain in part as a result of increased food intake. For example, in female rats, drug-induced hyperprolactinemia, or 10-day prolactin administration, as well as ectopic pituitary transplants, increase food intake and body weight (5–7). Furthermore, previous evidence from our laboratory showed that hyperprolactinemic lacDrd2KO are overweight, but, despite early occurrence of hyperprolactinemia, obesity was gradual and of late onset beginning mildly in 6-month-old females and evolving to frank obesity at 10 months of age (9). These data are in accordance with clinical data, which show that not all patients with prolactinomas are obese. Because obesity and food intake behaviors are multifactorial in origin, we sought to combine high prolactin levels and a HFD in order to establish whether metabolic changes could be accelerated or potentiated by combining the metabolic challenges, and furthermore we wished to dissect the role of prolactin on BAT function in this situation. We therefore exposed the life-long hyperprolactinemic mouse model to CD or HFD for two months, starting at 3 months of age, and characterized the metabolic role of prolactin on brown and white adipose tissue depots and glucose homeostasis, in a context of normal or increased energy availability.

Food intake is regulated by an integrated system involving both homeostatic and hedonic brain circuits sensitive to peripheral factors (42). Homeostatic regulation of food intake comprises hypothalamic systems that drive food intake when energy stores are depleted, whereas hedonic or reward-based regulation is driven by dopaminergic pathways that may override homeostatic satiety signals during periods of relative energy abundance by increasing the desire to consume highly palatable, energy-dense food, when extra calories are unnecessary (43, 44). Our results show that alterations in food intake induced by prolactin may be associated with altered homeostatic and hedonic brain circuits. In this respect, the role of prolactin in controlling hedonic feeding has not been explored to date.

We detected effects for both diet and high prolactin levels on body weight progression. HFD increased body weight in both lacDrd2KO and control female mice, but importantly, this increase was evidenced much earlier in the hyperprolactinemic group, and the difference in body weight between genotypes increased significantly over time, clearly suggesting that chronic high prolactin levels aggravate body weight gain induced by HFD.

Increased food intake and body weight gain affected adipose tissue accretion. The identification of adipose tissue as a metabolic organ has achieved increasing importance in recent years as its dysfunction is considered one of the central events that triggers the development of obesity. In a context of positive energy balance favored by HFDs, the storage capacity of scWAT is limited, and ectopic fat accumulation can occur in other tissues such as liver, skeletal muscle and heart, or in VAT depots (45). Excessive ectopic lipid accumulation in VAT depots predisposes to the development of metabolic syndrome, defined by a cluster of interconnected factors that directly increase the risk of coronary heart disease, other forms of cardiovascular atherosclerotic diseases, and diabetes mellitus type 2 (46, 47). The role of prolactin in fat accumulation in the different adipose tissue depots has not been extensively studied, and our results dissect a differential impact of prolactin for each fat depot.

In accordance with other models of diet-induced obesity (48), mice of both genotypes exposed to a HFD for two months had significantly larger scWAT and visceral fat pads, the latter including gonadal and mesenteric adipose tissue. Interestingly, lacDrd2KO mice fed a control diet had heavier scWAT depots compared to their control counterparts, and exposure to HFD potentiated scWAT weight gain in this group, suggesting an impact of elevated prolactin levels specifically on scWAT weight.

To this respect, it has been described that scWAT is a metabolically plastic tissue, sensitive to the action of prolactin (49). Prlr knockout mice have impaired development of scWAT depots due to fewer adipocytes, indicating a role for prolactin on adipogenesis in this tissue (28). In this context, significant changes in the expression levels of genes related to metabolism were detected in scWAT of lacDrd2KO mice. Prlr mRNA levels were significantly reduced independently by genotype and diet. These results suggest that chronic hyperprolactinemia can lead to adipose tissue dysfunction and limit the effect of prolactin in this tissue; in contrast, prolactin induces its own receptor in most tissues (25). Results are consistent with previous evidence obtained from microarray studies in adipose tissue that demonstrated that Prlr is downregulated in diet-induced obese mice (50). Interestingly, HFD increased the expression levels of Ucp1 in scWAT in some samples in both genotypes, probably indicating heterogeneity of samples with regard to a beiging process. Previous evidence showed that a hypercaloric diet may induce the appearance of thermogenic adipocytes in WAT depots, particularly in scWAT, contributing to counteract the excess calories incorporated with the fat-enriched diet (51), but on the contrary, HFD may induce the whitening of beige depots (52). It has been described that the absence of prolactin stimulation in Prl knockout mice induces the appearance of beige cells with thermogenic capacity in perirenal adipose depots in response to HFD (49), but we did not observe changes in Ucp1 mRNA expression in scWAT depots related to genotype. Nevertheless, prolactin may impact on UCP1 independent mechanisms which contribute to the thermogenic action of white adipocytes, such as creatine-substrate cycling and Ca2+cycling (53). On the other hand, we found a strong inhibitory effect of prolactin and not HFD on Lpl and adiponectin expression levels in scWAT, which may suggest an additional mechanism participating specifically in prolactin-induced obesity. In concordance, it has been shown that elevated serum prolactin levels decrease mRNA expression of Lpl and adiponectin in adipocytes, in vivo and in vitro, and may contribute to the development of an obese phenotype (26, 54). LPL is the rate-limiting enzyme for the uptake and storage of triglyceride-derived fatty acids by adipose tissue (55). Reduced Lpl expression levels found in lacDrd2KO female mice regardless of the diet received, suggest that during chronic hyperprolactinemia scWAT depots may compensate adipocyte hypertrophy by limiting lipid uptake and storage capacity (56). On the other hand, adiponectin, a hormone synthesized and released into circulation by adipose tissue, regulates energy homeostasis playing an important role in the dialogue between adipose tissue and other metabolic tissues or organs (57). Decreased adiponectin levels in scWAT induced by high prolactin levels and not HFD, may be relevant to obesity and intolerance to glucose evidenced in lacDrd2KO mice, supporting the described role for low adiponectin levels in tissue resistance to insulin, and/or increased adiposity in obese patients, pigs and rodents (57–59).

Our results uncover a role for prolactin in BAT thermogenic gene expression, as well as an interaction with the effect of HFD in this adipose tissue. Plenty of evidence shows that obesity is associated with an increase in BAT weight, explained by a higher lipid content and a reduction in its thermogenic function (33, 60, 61). In this sense, heavier BAT depots do not necessarily indicate a greater thermogenic capacity but, instead, are associated with a defective tissue, as triglycerides are stored instead of being used for heat production (33).

Histological analysis of this tissue revealed increased size of lipid droplets within brown adipocytes, in response to either hyperprolactinemia or HFD independently, and, importantly, prolactin further increased lipid droplet size after exposure to an HFD. To be noted is that increased brown adipocyte size was evidenced in hyperprolactinemic non-obese CD fed mice. Therefore, the histomorphological alterations observed in BAT depots in the presence of prolactin suggest that the marginal BAT weight gain observed in lacDrd2KO mice could be associated to a lipid accumulation probably associated to a decrease in its thermogenic capacity (33), and that prolactin plays a major role.

During lactation, when prolactin levels are high, the expression levels of thermogenic genes in BAT decrease (62), indicating a negative association between prolactin and the thermogenic capacity of this tissue. Moreover, Prlr knockout mice fed a HFD for 16 weeks show resistance to weight gain compared to control mice partly due to increased expression levels of thermogenic markers in BAT (Ucp1, Pgc1a) and beiging in perirenal VAT (49), indicating improved thermogenesis in the absence of prolactin action. A negative association between high prolactin and the thermogenic capacity of BAT has also been proposed as a physiological adaptation during pregnancy and lactation, when glucose and triglycerides should not be burned to produce heat because they are needed for the pregnant and lactating mother, and glucose for the developing fetus. Furthermore, lactation performance is influenced by the capacity to dissipate body heat, and prolactin may participate in this process limiting thermogenesis (62).

Immunohistochemical and gene expression analysis of thermogenic markers of BAT, confirmed our histological observations that pointed to an alteration of the tissue in response to high prolactin levels. We studied UCP1 as well as cell death-inducing DFFA-like effector A (Cidea) and peroxisome proliferator activated receptor γ coactivator 1α (Pgc1a), two genes abundantly expressed in brown and beige adipocytes, which are critical regulators of BAT activity in response to environmental stimuli such as cold temperature and diet (63, 64). Significant decrements in the expression of Ucp1 and Pgc1a mRNA as well as UCP1 protein levels induced by prolactin but not by HFD, indicate that prolactin, independently of the diet or obesity, may predispose to the development of an obese phenotype by lowering the thermogenic capacity of BAT. According to previous evidence, expression levels of Ucp1 and Pgc1a are positively correlated because the activation of Pgc1α pathway induces mitochondrial biogenesis consequently increasing Ucp1 gene expression in BAT. Cidea was decreased by both prolactin and HFD, and the lowering impact of prolactin on BAT Cidea expression was significantly intensified when lacDrd2KO mice were exposed to a HFD, indicating a synergy of effects.

The observed reduction in the levels of expression of the typical markers of BAT, exclusively in hyperprolactinemic non obese and not in HFD fed mice point to an action of prolactin independently of food intake driven-obesity. These changes are concordant with quantitative analysis of histological sections which showcase the occurrence of BAT whitening, associated with an accumulation of large lipid droplets due to mitochondrial dysfunction and loss (33), and decreased expression levels of Ucp1, Pgc1a and Cidea. Collectively, these data suggest a negative action of prolactin on BAT thermogenic capacity, and point to a new avenue of prolactin action.

During BAT whitening there is a proneness to tissue inflammation and ER stress (34–36), increased autophagy (65, 66), and insulin resistance of the tissue (67, 68). Prolactin has been associated with inflammation in other tissues (69) but there are no data regarding an action of prolactin on BAT inflammation, ER stress or autophagy. In the present work, increased tendency in BAT mRNA levels of the proinflammatory cytokine Il1b suggest that prolactin may evoke inflammation of the tissue. Furthermore, a negative correlation of BAT UCP1 levels and NPY expression in the dorso medial hypothalamus has been described (70), and we previously showed that high prolactin levels increase Npy expression in the DMH (38) probably favoring food intake, and simultaneously decreasing BAT Ucp1, pointing to an indirect mechanism of action or prolactin on BAT function. Finally, the decrease in adiponectin expression levels in BAT in response to hyperprolactinemia and not HFD suggests a lower contribution to adiponectin plasma levels that could partly participate in the greater weight gain, given that significant negative associations have been described between adiponectin levels and body mass index (59, 71). Alternatively, obesity could be the cause of the low adiponectin expression levels.

An increase in pancreas weight was evidenced in lacDrd2KO female mice regardless of the type of diet received, suggesting a strong impact of elevated serum prolactin levels in the physiology of this organ. We have previously demonstrated that lacDrd2KO female mice fed a regular diet are hyperinsulinemic, and have increased pancreatic insulin content (27). This may be related to the increased tissue weight observed in the present experiments, and is in line with the role described for prolactin in pancreas development and proliferation (3). Glucose intolerance was observed in response to HFD or chronic hyperprolactinemia separately, and importantly, intolerance was exacerbated in HFD animals that were also hyperprolactinemic, suggesting an aggravating effect of prolactin on altered glucose homeostasis induced by HFD. To this respect, it has been described that prolactin administered in high doses alters the signaling mechanism through which glucose enters pancreatic β cells and therefore disrupts adequate insulin response to glucose which may finally lead to glucose intolerance (72) (27).

In conclusion, chronic high prolactin levels accelerate and aggravate the metabolic effects of exposure to a HFD, and have an independent impact on BAT. In normal weight hyperprolactinemic mice, exposure to a HFD enhanced body weight gain, glucose intolerance, and scWAT weight gain, to a greater degree than in control mice. In particular, the combination of metabolic insults (elevated prolactin and HFD) altered BAT induced marked tissue whitening and a pronounced decrease in the levels of expression of markers of thermogenic capacity. Furthermore, there was a synergy in the decrease of expression levels of Cidea in BAT in response to the combination of metabolic challenges, which further points to a mechanism involved in the worsening of the obese phenotype. But interestingly many metabolic alterations were mainly associated to high prolactin levels, and not to HFD or increased BW in our protocol, such as the decrease in mRNA levels of adiponectin and Lpl in scWAT, and of Ucp1, Pgc1a and adiponectin in BAT, pointing to an independent effect of prolactin.

Since the description of active and recruitable BAT in humans, increasing interest in its physiology and pathology has positioned this tissue in a new hierarchy as a possible target to tackle metabolic dysfunction. In this context, deciphering prolactin action, and the effect of prolactin antagonism on brown adipocytes may yield valuable tools in the effort to curtail obesity or accelerate weight loss.

Our results demonstrate that metabolic worsening can result from the combination of metabolic challenges. Therefore, even though prolactin is fundamental in orchestrating adaptive changes in metabolism during pregnancy and lactation, and physiological or slightly high levels of prolactin have a protective metabolic role, its pathological increase could be detrimental in an obesogenic environment that requires greater energy expenditure and not increased calorie intake. In particular, prolactin targets BAT inducing a decreased thermogenic capacity. These results should be considered in case of prolactinomas, or chronic psychiatric treatments that target the DRD2 and evoke chronic hyperprolactinemia. In this context, prolactin may represent an additional player in the development of obesity, which is multifactorial in origin, and therefore the interrelation of genetic, endocrine, behavioral and environmental factors deserves greater attention in the search of strategies to combat overweight.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Institutional animal care and use committee of the Institute of Biology and Experimental Medicine, Buenos Aires in accordance with the Animal Welfare Assurance for the Institute of Biology and Experimental Medicine, Office of Laboratory Animal Welfare, NIH, A#5072-01.

FL-V Conceived and designed research, performed experiments, analyzed data, interpreted results of experiments, prepared figures, drafted manuscript, edited and revised manuscript, approved final version of manuscript. CW performed experiments, analyzed data, interpreted results of experiments, edited and revised manuscript, approved final version of manuscript. ES Conceived and designed some experiments, analyzed data, edited and revised manuscript, approved final version of manuscript. AO performed experiments in vivo, approved final version of manuscript. JT performed immunohistochemical studies, approved final version of manuscript. DB-V Conceived and designed research, analyzed data, interpreted results of experiments, prepared figures, drafted manuscript, edited and revised manuscript, approved final version of manuscript. All authors contributed to the article and approved the submitted version.

This work was provided with financial support for the conduct of the research by Argentinean Agency for Promotion of Science and Technology (PICT 2016-526 and 2019-1619 to DB-V), National Research Council (CONICET, PIP 2021 #273) and Williams and Rene Baron Foundations to DB-V. FL-V and CW received a fellowship from the CONICET.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.