Our results present a multifaceted profile of how GHSR deletion modulates the effects of long-term (12-month) HFD in male and female rats. The main finding is that the deletion of GHSR protects against diet-induced weight gain and decreases food intake during HFD in male but not in female rats. The reduced food intake in our model was not related to changes in locomotor activity or anxiety-like behavior evaluated in the open field and novelty-suppressed feeding tests. All HFD groups, regardless of sex, accumulated more gonadal WAT, but in the males, WT rats had higher relative gonadal WAT weight (% of total body weight) than GHSR-KO. Using infrared thermography, we detected a higher body temperature of the interscapular region, where the BAT is mainly located, in GHSR-KO compared with WT rats in males but not females. Another novel finding of our study was that the HFD decreased brain FDG uptake in males of both genotypes, with a greater effect in WT than in GHSR-KO rats. In females, we observed the opposite, HFD attenuated FDG uptake more markedly, but also increased in certain brain areas in GHSR-KO rats. Furthermore, FDG uptake was increased in GHSR-KO male rats compared to WT, regardless of diet, an effect stronger in males than in females. Results of the RNA-sequencing in male rats showed upregulation of gene sets responsible for fat oxidation in the BAT of GHSR-KO and increased expression of connective tissue genes in WT, and these differences were more prominent in HFD. Furthermore, in the WAT during HFD, inflammation-related pathways were upregulated in WT compared to GHSR-KO. We also tested a therapeutically promising GHSR inverse agonist for the first time in both male and female mice. PF-5190457 is the only GHSR blocker which advanced to clinical trials and is well-tolerated in humans, and in our experiments icv. administration attenuated ghrelin-induced food intake interestingly only in male, but not female mice. The effect of GHSR inverse agonism on a more hedonic aspect of food intake, the HFD-induced binge-like eating, was not sex-dependent; icv. PF-5190457 reduced HFD intake in both sexes.
Our experiments demonstrate that diet-induced weight gain is attenuated in GHSR-KO male rats. The HFD in our model contained almost twice as many calories (5.24 vs. 2.89 kcal/g) as the regular diet, less carbohydrates (20% vs. 58%), and more fat (60% vs. 18%). The significantly reduced weight of GHSR-KO male rats in the HFD-fed group might result from decreased food intake compared to the WT male rats on the same diet. The initial characterization of our model24 used a regular diet for 34 weeks (vs. 48 weeks in the current study) in male rats only, and observed a small but significant reduction of body weight and food intake in the KO group. Of note, in the present study, we did not observe a decrease in body weight in rats on the regular diet, only on the HFD. In a previous 19-week study32 in male and female mice, the lack of GHSR reduced HFD intake and protected against HFD-induced weight gain compared to WT regardless of sex, whereas the lack of GHSR reduced weight in female but not in male mice fed a regular diet32. These findings indicate the complex interaction of sex and genetic modulation of the GHSR system in controlling food intake and body weight gain.
In the current study, GHSR deletion reduced normalized gonadal, but not inguinal, WAT weight (% of total body weight) regardless of diet. This finding may be explained by the lower expression of GHSR in the inguinal WAT than in gonadal WAT39. In rats, inguinal and gonadal fat pads are the main subcutaneous and visceral WAT depots, respectively40, suggesting that in the present study GHSR deletion reduced the ratio of visceral fat in the total body weight. The subcutaneous fat has a more beneficial metabolic profile compared to visceral fat in rodents and humans, by improving insulin action, longevity, protecting against ectopic fat deposition, lipotoxicity, and reducing tumorigenesis. The higher ratio of subcutaneous fat is also associated with lower triglyceride and higher high-density lipoprotein levels40. Deletion of GHSR may exert beneficial effects by reducing visceral (gonadal) WAT weight.
GHSR activity modulates the microstructure, gene-expression profile, and norepinephrine sensitivity of BAT, thereby affecting thermogenesis and energy expenditure. The number of mitochondria and the percentage of multilobular (brown) adipocytes is higher in GHSR-KO male mice41. GHSR expression in BAT increases with aging, leading to a thermogenesis reduction39. GHSR ablation attenuated the decline of thermogenesis, uncoupling protein-1 expression, and BAT’s norepinephrine content in male mice39. Consistent with this literature, we found that thermogenesis, measured by intrascapular temperature, was increased in GHSR-KO male rats, regardless of diet.
To gain a deeper understanding on the role of GHSR in BAT thermogenesis, we used RNA-seq to profile gene expression in the BAT in male rats. We found that GHSR gene deletion produced a BAT gene expression program characterized by the upregulation of skeletal muscle-related genes and other oxidative metabolism genes, such as tricarboxylic acid cycle and electron transfer genes, glycolytic genes, fat metabolism and β-oxidation genes, and mitochondrial genes. This effect of GHSR deletion became more pronounced during exposure to the HFD. Such a BAT gene expression profile has elements of the myogenic transcriptional signature that are dynamically regulated42–44. BAT and skeletal muscle, both of which contribute to the regulation of body temperature, have common cell lineages42,43. The skeletal muscle-like gene expression program in BAT and beige adipocytes is associated with heat generation45. Evidence shows that muscle-like actomyosin tensional response in BAT leads to calorie consumption that result in heat production and is induced by β-adrenergic activation45. Both BAT and skeletal muscle have high oxidative capacities and utilize glucose and fatty acids as substrates46,47. Glycolytic genes, fat metabolism, and β-oxidation genes were also increased in male rats by GHSR deletion. Increased glycolytic activity in beige fat in a myogenic state has also been associated with increased thermogenic capacity48.
WT male rats showed greater expression of extracellular matrix, integrins, and collagen genes. A late event in BAT gene expression response to HFD is the expression of connective tissue genes, including procollagen, procollagen cleavage, and remodeling genes, which takes place after the upregulation of the skeletal muscle-related gene expression program has subsided44. Thus, GHSR appears to direct the BAT gene expression program in male rats, both on the regular chow and on the HFD, possibly acting as a switch between a gene expression signature characterized by greater expression of the muscle-related gene expression program and one characterized by the expression of extracellular matrix, integrins, and collagen genes. The present results support GHSR as a therapeutic target that will implement the myogenic gene expression program in BAT, which has been proposed to be a therapeutic strategy for obesity and obesity-related metabolic disorders like type 2 diabetes49.
Gene expression analysis of the WAT of GHSR-KO male rats on the HFD vs. WT male rats on the same diet showed decreased expression of inflammation-related pathways in GHSR-KO male rats. As diet-induced obesity advances, angiogenesis can no longer keep up with the enlargement of adipose tissue. Pathological processes, such as hypoxia, may cause the dysregulation of adipokines, and adipocyte necrosis, leading to inflammation, fibrosis, and insulin resistance50. Confirming our findings, macrophage infiltration, as a marker of increased inflammation, was reduced in gonadal WAT of adipose tissue-specific GHSR-KO male mice51. Adipose tissue inflammation is reduced by interventions that improve fat oxidation and obesity52–54.
An additional finding of our study is the lipidomics analysis, based on an unbiased approach, to investigate the effect of HFD and genotype on the lipid profile of WAT and BAT. It is known that dietary fatty acid intake can cause changes in lipid composition. The HFD-fed rats in this study was high in saturated fatty acids, especially for triglyceride species55. We detected increased saturated and decreased unsaturated triglycerides in BAT and WAT due to HFD in both sexes. A recent study56 using an untargeted lipidomic approach for HFD-induced obesity in male rats showed a similar pattern in plasma, gonadal WAT, and liver. In our study, GHSR deletion affected the TG profile only in the WAT in males, leading to an increase in triglycerides with unsaturated fatty acids. This genotype-related difference was observed only with a regular diet, which can be explained by the robust decreasing effect of HFD on unsaturated triglycerides. We speculate that this effect of diet may have overshadowed the genotype effect.
The small animal PET highlighted differences in brain FDG uptake 50 weeks after initiation of a regular or high-fat diet in WT and GHSR-KO rats. HFD decreased brain FDG uptake, an observation consistent with early work in male Wistar rats57, male Sprague-Dawley rats58, and male C57BL/6J mice59. Yet, this work is novel not only because we investigated sex differences, but also for the length of HFD exposure, as previous studies measured FDG uptake only after 958,59, or 16 weeks57. The brain glucose uptake by glucose transporters is also affected by HFD, as GLUT-4 protein levels were decreased after 9 weeks of HFD in male C57BL/6J mice59. GHSR gene deletion diminished the negative effect of HFD on brain FDG uptake in male rats, and increased FDG uptake regardless of diet. GHSR may affect the glucose transporters on neurons (GLUT-3) and on astrocytes (GLUT-2)60, as Fuente-Martin et al. showed that both acute and chronic central ghrelin administration decreased hypothalamic GLUT-2 and GLUT-3 protein levels in vivo in male Wistar rats, and GHSR gene deletion abolished this effect in vitro60. In female rats, we observed the opposite, namely HFD markedly decreased brain FDG uptake in GHSR-KO, but not in WT rats. This result is consistent with a recent study that showed no changes in brain FDG uptake after 16 weeks of HFD in female, but a decrease in male Wistar rats57. In addition, the main effect of Genotype (GHSR-KO > WT) on brain FDG uptake was less prominent in females.
This Wistar GHSR-KO model was previously characterized in males24. In mice, GHSR deletion was protective against HFD-induced weight gain in both sexes32. Sexual dimorphism of the ghrelin system has been described; for example, Sprague-Dawley female rats61 had higher plasma ghrelin levels, lower LEAP-2 expression in the liver, and higher GHSR expression in the hippocampus, amygdala, and LH. Gonadal hormones may be responsible for this sexual dimorphism given that ovariectomy diminished, and estrogen replacement restored the differences in ghrelin levels between males and females61. In our Wistar rats, we did not find any difference in ghrelin or desacyl-ghrelin levels between females and males, regardless of genotype, but LEAP-2 plasma concentration was higher in males. Furthermore, estrogen decreases the sensitivity to ghrelin’s orexigenic effects, as males and ovariectomized females were more responsive to peripheral and central ghrelin administration than intact or estrogen-treated ovariectomized Long-Evans rats62.
The terminal hormone analysis established that chow-fed GHSR-KO male rats had higher peripheral levels of progesterone compared to chow-fed WT males, and HFD-fed GHSR-KO males. To date, the effects of ghrelin and GHSR on progesterone levels was only investigated in female rats63. Ghrelin administration reduced progesterone levels throughout the estrous cycle in rats64. The effect of obesity was examined in human males, in that there was a negative correlation between plasma progesterone levels and body weight, body mass index, and waist circumference65.
We amended our comprehensive characterization of GHSR-KO Wistar rats with central pharmacological interventions targeting GHSR in male and female C57BL/6J mice. PF-5190457 is a promising compound, the first GHSR blocker ever tested in clinical trials. This inverse agonist was found being safe and tolerable, and reduced food-seeking in humans30,36. Our novel finding was that its effect on ghrelin-induced food intake is sex-dependent, can be observed only in males, similar to what we described here in GHSR-KO rats and DIO. Neuronal activation in the arcuate nucleus of the hypothalamus is the key target of ghrelin for the induction of food intake. PF-5190457 also reduced this activation as measured by the number of cFos-IR cells. We tested PF-5190457 in a HFD-induced binge-like eating paradigm, which models how palatable foods induce hedonic food intake and hyperphagia26–28. Icv. PF-5190457 was effective in both sexes, and we could confirm that this effect is GHSR-dependent. This founding highlights the role of GHSR constitutive activity in binge eating, which involve the GHSRs on dopaminergic neurons in the reward system27,28.
Study limitations include that our endocrine outcome was a terminal hormone analysis at a single timepoint, before euthanasia; and we could not measure the binding affinity of the antibody in mice treated with the anti-ghrelin vaccine. Of note, we employed, herein, a global KO model and this approach holds limitations related to long-term adaptations that may occur in constitutive KO rodents66. Nonetheless, this approach holds important value, especially as a proof-of-concept related to the role of GHSR in a translationally valid model of obesity and its relevance for genetic mutations in ghrelin systems in humans, such as single nucleotide polymorphism67. Of note, albeit in a different context and phenotype, i.e., people with alcohol use disorder (for a review, see Farokhnia et. al68), recent human work suggests that GHSR inverse agonism leads to a reduction in food-seeking behaviors30.
In conclusion, we demonstrated a protective effect of GHSR deletion in diet-induced obesity in male, but not female, rats, which shows a dramatic sexual dimorphism. Our pharmacological studies in mice also revealed that the GHSR blockade has stronger effects in males than in females indicating that the pharmacological intervention of the GHSR in humans requires a precise assessment of the sex effects. Our results indicate an important role for GHSRs in the regulation of body weight, food intake, energy homeostasis, adipose tissue, and brain activity, and constitute a promising target for medication development for the treatment of obesity.