Pituitary gland agouti‐related peptide cells: Novel player controlling glucose metabolism

Abstract

The liver synthesizes and releases bile acids into the gut. Bile acids, either directly or indirectly, inhibit agouti‐related peptide (AGRP)‐B cells in the pars tuberalis of the pituitary gland. AGRP‐B cells are assumed to promote pancreatic insulin secretion and/or to improve insulin sensitivities in insulin sensitive organs, resulting in improved glucose tolerance.

Inter‐organ communication via neuronal systems is important for regulating systemic glucose metabolism. For example, the brain receives metabolic signals from the liver (via afferent splanchnic nerves) and relays them to the pancreas (via efferent vagal nerves), resulting in pancreatic β cell proliferation and glucose stimulated insulin secretion (GSIS) ¹ . Bile acids are another key player regulating glucose metabolism. Bile acids reportedly promote pancreatic insulin secretion via direct action on bile acid receptors expressed on pancreatic β cells and indirectly improve insulin sensitivity in adipose tissues and skeletal muscles via increased circulating fibroblast growth factor 21 ² . Interestingly, bile acids are also known to act directly on neurons (via receptors such as M2 muscarinic acetylcholine receptors) and to modulate brain functions (such as cognitive function) ³ . However, whether bile acids act on the brain to affect glucose metabolism remains unknown.

Liu et al. ⁴ have been tackling this important issue. In a previous study, they focused on taurocholate, a primary bile acid conjugated with taurine, and found that taurocholate treatment of ob/ob mice decreased brain Agouti‐related peptide (Agrp) messenger ribonucleic acid (mRNA) expression and improved glucose tolerance ⁴ . Moreover, additional AGRP knockout (KO) in ob/ob mice canceled the taurocholate‐mediated improvement of glucose tolerance. Their findings suggested a gut–brain inter‐organ communication in which bile acids affect brain AGRP cells to modulate systemic glucose metabolism. However, which portion(s) of the brain harboring the AGRP cells are involved in this inter‐organ communication, and whether the inter‐organ communication functions in organisms other than genetically obese mice, are important questions that have yet to be answered.

Recently, Liu et al. ⁵ demonstrated that cells in the pituitary gland express an AGRP splice variant crucial for modulating systemic glucose metabolism in non‐genetically obese mice ⁵ . They first searched a public genome database and recognized that there are three variants, originating in the mouse Agrp gene, which differ in the 5′‐untranslated regions (UTRs); AGRP‐A, AGRP‐B, and AGRP‐C. Hence, they tested whether and what variant expression(s) is affected by taurocholate. After 5 days of oral taurocholate treatment, mRNA expression levels of Agrp‐B, but not those of Agrp‐A and Agrp‐C, were reduced in brain samples containing the hypothalamus and pars tuberalis, the anterior region of the pituitary gland. These findings suggest that taurocholate inhibits AGRP‐B expressions in the hypothalamus and/or pars tuberalis. To identify the anatomical location of AGRP‐B cells, they generated Agrp‐B‐FLP FSF‐tdTomato mice, in which tdTomato is expressed under the 5′‐UTR of Agrp‐B. Interestingly, tdTomato was detected in the pars tuberalis, but not in the hypothalamus, indicating that AGRP‐B is expressed in the par tuberalis. In addition, in wild type mice, a 20 h fast did not increase Agrp‐B mRNA expressions in brain samples containing the hypothalamus and pars tuberalis, suggesting that fasting does not stimulate AGRP‐B activation. Moreover, in Agrp‐B‐FLP Ai193 mice, in which tdTomato is expressed under the 5′‐UTR of Agrp‐B, tdTomato and AGRP were co‐expressed in pars tuberalis cells, while neurons in the hypothalamic arcuate nucleus were tdTomato‐negative but AGRP‐immunoreactive. These results suggest that pars tuberalis cells express AGRP‐B, and that pituitary AGRP‐B cells are distinct from AGRP neurons in the hypothalamus.

To determine the developmental origin of AGRP‐B cells, the authors conducted immunostaining with pituitary cell markers. In Agrp‐B‐FLP Ai193 mice, tdTomato co‐localized with pituitary homeobox 1 (Ptx1), a transcription factor involved in the transcription of pituitary hormone genes, and chromogranin A (CGA), a molecule involved in secretory granule formation. These results suggest that AGRP‐B cells of the pars tuberalis are neuroendocrine cells of the pituitary gland.

Next, to examine whether suppression of AGRP‐B cells affects glucose tolerance, Liu et al. ⁵ performed experiments in two murine models; Agrp‐B KO mice and AGRP‐B silenced mice. First, they fed 2–3‐month‐old Agrp‐B KO mice and wild‐type mice a high‐fat diet (HFD) for 4 weeks and then performed glucose tolerance tests. In HFD‐fed Agrp‐B KO mice, the areas under the curve (AUCs) of blood glucose levels were smaller than those of wild‐type mice, indicating better glucose tolerance. Next, to silence AGRP‐B neurons using the designer receptor exclusively activated by designer drugs (DREADD) strategy, they generated Agrp‐B‐FLP Agrp‐IRES‐Cre RC:F‐PDi mice, in which inhibitory designer receptor hM4Di was specifically expressed in AGRP‐B cells. In these mice, treatment with a designer receptor agonist, Compound 21, was considered to silence AGRP‐B neurons. The authors fed 8–10‐month‐old mice the high‐fat diet for 7 weeks, followed by Compound 21 mixed in gavage for 5 days. Before (baseline) and after Compound 21 treatment, these mice underwent glucose tolerance tests. The blood glucose AUCs after Compound 21 treatment were smaller than those before treatment, suggesting that silencing AGRP‐B cells improved glucose tolerance. Moreover, serum insulin levels increased 15 min after glucose loading in Compound 21‐treated mice but not in baseline mice, suggesting enhanced GSIS with silencing of AGRP‐B neurons. The body weights were also reduced in Compound 21‐treated mice. Based on these results, the authors concluded that bile acids inhibit AGRP‐B cells, resulting in GSIS and glucose tolerance improvements (Figure 1). This line of research provides novel insights into the role of pituitary cells in systemic glucose metabolism. There are, however, further questions that need to be answered before these processes are fully understood.

Figure 1.

Proposed mechanism by which pituitary agouti‐related peptide (AGRP)‐B cells regulate glucose metabolism. The liver synthesizes and releases bile acids into the gut. Bile acids, either directly or indirectly, inhibit AGRP‐B cells in the pars tuberalis of the pituitary gland. AGRP‐B cells are assumed to promote pancreatic insulin secretion and/or to improve insulin sensitivities in insulin sensitive organs, resulting in improved glucose tolerance.

First, how AGRP‐B cells affect systemic glucose metabolism is unclear. Generally, improved glucose tolerance results from increased GSIS and/or improved insulin sensitivity. In this study, GSIS occurred in glucose‐loaded Compound 21‐treated mice, while loading of the same mice with an identical amount of glucose prior to Compound 21 treatment did not increase serum insulin levels at 15 min. This GSIS impairment in the baseline mice might be attributable to HFD pre‐loading. The effects of knockout of AGRP‐B or cell silencing of AGRP‐B neurons on glucose metabolism were examined only after long periods of HFD‐feeding. It is also important to unravel the physiological roles of AGRP‐B protein and expressing‐neurons in maintaining glucose homeostasis under normal‐diet conditions.

As for insulin secretion, it would be interesting to quantify serum insulin levels throughout the glucose tolerance test (GTT) periods and to evaluate insulin secretion by calculating the C_max and the AUC of serum insulin. Furthermore, whether the pancreatic β cell mass is affected by AGRP‐B knockout or AGRP‐B cell silencing is also important. These examinations may unravel the complex mechanisms by which pituitary AGRP‐B cells affect insulin secretion from pancreatic β cells. In addition to insulin secretion, whether AGRP‐B itself, and/or the cells producing it, modulate insulin sensitivity merits study. Given that body weight was reduced after Compound 21 treatment under high‐fat diet conditions, it is plausible that AGRP‐B cells affect systemic insulin sensitivity via body mass alterations. Related to this issue, the mechanism underlying reduced body weight by silencing AGRP‐B cells needs to be clarified. While AGRP synthesized in the hypothalamic arcuate nucleus is well known to induce body weight gain via increased food intake and decreased systemic energy expenditure, whether AGRP‐B synthesized in pituitary cells has similar effects is an issue of interest.

The next question is whether AGRP‐B cells in the pars tuberalis actually mediate the bile acid effects on improving glucose tolerance. Given that taurocholate treatment inhibits AGRP‐B cells and that their silencing increases glucose tolerance, whether taurocholate‐induced improvement of glucose tolerance is canceled in AGRP‐B‐silenced mice is important. If so, unraveling the molecular mechanism whereby taurocholate inhibits AGRP‐B cells merit investigation. In addition, there are several questions that must be answered to elucidate the role of AGRP‐B, and the neurons producing it, in regulating glucose metabolism. For instance, are AGRP‐B cells localized exclusively in the pars tuberalis, or are they also in other brain regions? What are the downstream targets of AGRP‐B neurons, and how do target cells affect glucose metabolism? Further studies are expected to confirm the role of AGRP‐B cells in the gut–brain inter‐organ communication mediated by bile acids.

Finally, considering therapeutic application, the pathophysiological significance of the gut–brain inter‐organ communication mediated by bile acids and AGRP‐B cells must also be understood. Given that AGRP‐B is upregulated under HFD‐fed conditions and AGRP‐B knockout and silencing the producing neurons prevent worsening of insulin secretion and insulin resistance, the pituitary AGRP‐B system may contribute to pathological conditions related to obesity. Further intensive studies are required to clarify these important issues.

In summary, pituitary gland AGRP‐B cells might be a novel player in systemic glucose metabolism control. Future research is anticipated to clarify the physiological and pathophysiological significance of AGRP‐B, as well as the pituitary neurons producing it, in complex brain‐orchestrated inter‐organ communication necessary for systemic glucose metabolism.

DISCLOSURE

The authors declare no conflict of interest.

Approval of the research protocol: N/A.

Informed consent: N/A.

Registry and the Registration No. of the study/trial: N/A.

Animal Studies: N/A.