Glucagon does not directly stimulate pituitary secretion of ACTH, GH or copeptin

may stimulate secretion of arginine-vasopressin (AVP)/copeptin, growth hormone (GH) and adreno-corticotrophic hormone (ACTH) from the pituitary gland. Nevertheless, the mechanisms and whether GCGR is present in human pituitary are unknown. In this study we found that intravenous administration of 0.2 mg glucagon to 14 healthy subjects was not associated with increases in plasma concentrations of copeptin, GH, ACTH or cortisol over a 120-min period. GCGR immunoreactivity was present in the anterior pituitary but not in cells containing GH or ACTH. Collectively, glucagon may not directly stimulate secretion of GH, ACTH or AVP/ copeptin in humans but may instead be involved in yet unidentified pituitary functions.


Introduction
Glucagon is a 29 amino acid peptide hormone secreted from pancreatic alpha cells and a key regulator of hepatic glucose production.The effects of glucagon are mediated by binding to the glucagon receptor (GCGR) [1,2].Although known for more than 100 years, knowledge of glucagon function and GCGR expression in developing and adult brain as well as the pituitary gland is very limited.However, considering the current focus on glucagon-based therapies as a possible treatment option for obesity and related conditions this area warrants further attention [3].
Glucagon has been reported to stimulate secretion of adrenocorticotrophic hormone (ACTH), growth hormone (GH) and argininevasopressin (AVP)/copeptin [4][5][6][7][8][9].For glucagon to directly stimulate secretion of ACTH, GH and AVP/copeptin GCGR must be expressed on somato-and corticotrophs in the anterior pituitary as well as on magnocellular hypothalamic AVP neurons or their axons projecting to the posterior pituitary.GCGR has been identified in rat pituitary using radiolabeled glucagon [10] but whether the human pituitary has GCGR expression is not known.
Here we investigated localization and function of the GCGR in the pituitary gland using immunohistochemistry, fluorescence in situ hybridization and measurement of pituitary-related hormones in plasma following intravenous administration of 0.2 mg glucagon to 14 healthy subjects.

Subjects, glucagon administration and plasma samples
Plasma samples from 14 healthy subjects (7 men + 7 women) were obtained from the GLUSENTIC Trial (NCT04907721) [11].Subjects were tested in the morning following an overnight fast.A catheter was placed in the antecubital vein in both arms in which glucagon was administered in one arm and venous blood samples were drawn from the other.A bolus injection of 0.2 mg glucagon (Novo Nordisk GlucaGen Hypokit) was intravenously administered at time 0. Just before glucagon administration a baseline blood sample (time 0) was drawn.Blood was obtained at times 10, 30, 60 and 120 minutes following glucagon administration.
The trial was approved by the scientific-ethical committee of the Capital Region of Denmark (H-20023717) and carried out according to the Declaration of Helsinki.Prior to inclusion subjects were informed about potential risks and discomfort with participation.Written consent was obtained from each subject.

Pituitary tissue samples and immunohistochemistry
Pituitary samples from four human fetuses (135-200 mm crownrump length) corresponding to 16-21 weeks post conception were obtained from the Human Embryonic/Fetal Biobank, Department of Cellular and Molecular Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark, previously described in detail [12] (Ethical approval: Danish Regional Committee on Health Research Ethics KF-V.100.1735/90& KF-11 2006 -4838).One adult human pituitary sample (approved for research use by local ethical committee), five rat (E15.5 to P0) and five adult mouse pituitary samples were also included.Mice were treated according to principles of Danish Law on Animal Experiments (LBK NR474, May 15, 2014) and Dyreforsoegstilsynet (license number: 2021-15-0201-00929) and rat samples were retrieved from existing biobank.Tissue was fixed in either 10% neutral buffered formalin, Bouin or Stefanini's fixatives.Preparation and immunostaining of sections for bright field light microscopy as well as single-and double labeling immunofluorescence was performed as previously described [12][13][14].Details of the primary and secondary antibodies including information regarding dilutions, heat induced epitope retrieval and suppliers are listed in Table 1.

Fluorescence in situ hybridization
For detection of Gcgr mRNA, a plasmid (MC203290) containing the mGcgr cDNA was obtained from Origen.After expansion, the full coding sequence (nt.115 -1575, BC057988) was amplified by PCR using Pfu polymerase and the following primers mGcgr Fw: TGTA-GAATTCATGCCCCTCACCCAGCTC and mGcgr Rv: TTCA-GAATTCAGGTGGGGCTGTCAGCCAA.The obtained PCR-fragment was inserted in the EcoRI site of pBluscriptKS+, and the resulting plasmid mGcgrKS+(1.3)was sequence verified.mGcgrKS+(1.3)linearized with either XhoI or XbaI was used as template for the antisense and sense probe, respectively, and RNA labelling with Digoxigenin-11-UTP were made using T7-polymerase (antisense) and T3-polymerase (sense).
In situ hybridization was performed on two adult mouse pituitary specimens.Mouse tissue was obtained and fixated as described for immunohistochemistry.For hybridization antisense and sense probe were diluted 1:4000 and 1:1000, respectively.Following overnight hybridization, sections were incubated with a sheep anti-digoxigenin antibody (Roche Cat# 11207733910, RRID:AB_514500, diluted 1:50) followed by washing and incubation with Alexa 488-conjugated tyramide (Thermo Fisher Cat# B40953, diluted 1:100).After washing slides were mounted in glycerol DAPI solution.
Deconvolution is stated in associated figure legends.

Statistical analysis
Statistics were performed in GraphPad Prism version 9.4.1 for Windows.To evaluate changes in plasma concentration of the measured hormones following glucagon administration Friedmans test was performed.If significant, this was followed by Dunn's multiple comparisons test versus baseline.A p-value <.05 was considered statistically significant.Data are presented as medians with interquartile ranges (IQR).
No significant changes in plasma concentration of GH (p=.091) or ACTH (p=.204) were observed between any of the included time points following glucagon administration (Fig. 1C, E).Plasma copeptin had a decreasing tendency from baseline until end point evaluation (p=.060) (Fig. 1F).

GCGR in rodent and human pituitary
GCGR immunoreactivity was found in anterior pituitary from fetal and newborn rat, adult mouse and fetal and adult human (Fig. 2A, B, C and Fig. 3D, E).In situ hybridization on adult mouse pituitary supported these findings, revealing Gcgr mRNA expression in anterior pituitary (Fig. 3A, B, C).Double immunostaining of an adult human anterior pituitary neither showed overlap between GCGR and GH nor GCGR and ACTH (Fig. 2D, E).A subpopulation of anterior pituitary cells positive for GCGR had a stellate appearance and therefore double immunostaining for S-100 (marker of folliculostellate cells) and GCGR was performed.However, no overlap between S-100 and GCGR was found in adult human anterior pituitary (not shown).

Discussion
Our study demonstrates that intravenous administration of 0.2 mg glucagon does not lead to a significant increase in plasma copeptin, GH, ACTH or cortisol over a 120-min period in healthy young subjects.
Surprisingly, median plasma cortisol was significantly lower at the end of the glucagon test compared to baseline.However, a similar decrease in plasma cortisol was observed by Arvat et.al between 08.00 and 10.00 h following both intravenous glucagon (1 mg) and saline administration [16].Since plasma cortisol concentration has been reported to drop by approximately 30 nmol/L every hour from morning until noon [17,18] the observed decrease in plasma cortisol concentration is believed to be due to circadian variation rather than a derived inhibitory effect of glucagon.
Plasma glucagon and glucose concentrations were also monitored.Glucagon stimulates glycogenolysis and gluconeogenesis [19].An Fig. 3. Glucagon receptor localization in adult mouse pituitary gland.Immunofluorescence microscopy of adult mouse pituitary following in situ hybridization (A-C) and immunohistochemistry (D,E).A shows glucagon receptor mRNA (Gcgr, green staining) in anterior pituitary following hybridization with the antisense probe (GCGR, green; DAPI, red).B, serving as a control for A, shows no glucagon receptor mRNA (Gcgr, green staining) following hybridization with the sense probe (GCGR, green; DAPI, red).The dashed line marks the posterior pituitary.C is a subsection of anterior pituitary presented in A. Note, GCGR protein is also distributed in cells of the anterior pituitary in D (GCGR, red; DAPI, blue).E is a higher magnification of D representing a deconvoluted Z-stack of images.E shows GCGR (red dots) in the membrane of anterior pituitary cells (GCGR, red; DAPI, blue).Abbreviations: AP, anterior pituitary; GCGR, glucagon receptor; PP, posterior pituitary.Scale bars: C, 50 μm; D, 20 μm; E 7 μm.

I. Stangerup et al.
evanescence effect of glucagon on blood glucose and the ability for glucagon to, at least to a certain degree, lower blood glucose has previously been reported as well as greatly debated for decades [3,[19][20][21].In vitro studies have shown that glucagon is able to act as an insulin secretagogue [21].These mechanisms may explain the initial increase of plasma glucose followed by a tendency towards lower concentrations compared to baseline.
The glucagon stimulation test (GST) has been reported as an alternative to the insulin test to address GH and ACTH reserve.When performing GST, GH should reach plasma concentrations of > 3 µg/L in adults to rule out GH deficiency when BMI is < 25 kg/m2 whereas the cut-off value for plasma cortisol for secondary adrenal insufficiency is debated and to a greater degree assay dependent [4,[22][23][24].Of note, all included healthy subjects failed to reach plasma concentrations of GH above 3 ug/L.This study design differs from a regular GST in both glucagon dosage, administration form and time for end point evaluation.This could explain some of the observed differences in hormone response following glucagon administration.
A lower glucagon dosage was used in this study compared to a regular GST (0.2 mg vs. 1-1.5 mg).Despite the lower dosage, supraphysiological concentrations and a 100-fold increase in plasma glucagon were still observed in all subjects 10 minutes following glucagon administration.Few older studies have also reported that the effects of glucagon on GH and ACTH pituitary secretion are exclusively seen following intramuscular or subcutaneous and not intravenous administration [16,25].However, higher plasma concentrations are observed upon intravenous administration [20].Hence, if stimulatory effects of glucagon on pituitary secretion of copeptin, GH and ACTH were mediated by glucagon directly acting on magnocellular AVP neurons, somato-and corticotrophs an intravenous dosage of 0.2 mg would be expected to be sufficient.
The peak in copeptin, GH, ACTH and cortisol following GST has been described for most people to occur between 120 and 180 minutes following glucagon administration [6,7,25].It is therefore possible that an increase in plasma pituitary hormones is missed due to early end point evaluation which together with a small sample size are limiting factors of this study.However, the half-life of intravenous administered glucagon is short (4-8 minutes) [26][27][28] and plasma glucagon was close to baseline after 30 minutes.Furthermore, a direct effect on the anterior pituitary would be expected to cause hormone secretion within minutes [29].The described delayed pituitary hormone response compared to the glucagon peak therefore favors the assumption that any stimulatory effect of glucagon on AVP/copeptin, GH and ACTH secretion is mediated through indirect rather than direct mechanisms.Supporting these hypotheses, GCGR immunoreactivity in anterior pituitary did not localize with GH or ACTH in an adult human pituitary.
Presence of GCGR in anterior pituitary was consistent across species as well as in both fetal and adult brain and compliant with Gcgr mRNA expression in mouse anterior pituitary.However, only one adult human pituitary was investigated and mRNA expression data were limited to mouse specimens.Hence, further validation of presented findings is still needed.
Agranular folliculostellate cells are thought to serve as local regulators of the granular pituitary cells through not yet fully understood mechanisms [30].Despite a stellate appearance of a subpopulation of GCGR immune-positive cells there was no overlap of GCGR with S-100, a marker of folliculostellate cells.Thus, the GCGR positive cells in anterior pituitary still warrant further investigation and could reveal new avenues in glucagon research.

Declaration of Competing Interest
Associate Prof. Nicolai J. Wewer Albrechtsen has received funding from and served on scientific advisory panels and/or speakers' bureaus for Boehringer Ingelheim, MSD/MERCK, Regeneron, Roche, Novo Nordisk and Mercodia.Remaining authors have no conflict of interest to declare.

Fig. 1 .
Fig.1.Changes in plasma concentrations of glucagon, glucose, ACTH, cortisol, GH and copeptin following glucagon administration.Plasma concentrations of glucagon (A), glucose (B), ACTH (C), cortisol (D), GH (E) and copeptin (F) were analyzed 10, 30, 60 and 120 minutes following intravenous administration of 0.2 mg glucagon to 14 healthy human subjects (7 males and 7 females).Baseline plasma samples were obtained at timepoint 0 where after glucagon was immediately administered.Plasma glucagon (A) and glucose (B) rapidly increased but were no different from baseline within 60 minutes following glucagon administration.Plasma cortisol (D) decreased throughout the 120 minutes whereas no changes in plasma ACTH (C), GH (E) or copeptin (F) were observed following glucagon administration.Data are presented as medians with interquartile ranges.* p< .05 vs. 0 (baseline) using Dunn's multiple comparisons test.Abbreviations: ACTH, adrenocorticotropic hormone; GH, growth hormone.

Fig. 2 .
Fig. 2. Glucagon receptor localization in rat and human pituitary gland.Bright-field (A-C) and double-immunofluorescence microscopy (D,E) of newborn rat (A) and adult human pituitary (B-E).Rat tissue was a section containing the entire head of the animal keeping the pituitary in situ in the scull.(A) and (B) show glucagon receptor (GCGR, brown staining) immunopositive cells in anterior pituitary and no specific reaction in posterior pituitary (A,B).Note, GCGR staining in pars intermedia of newborn rat pituitary (A).C is a subsection of anterior pituitary cells in B. GCGR immunoreactivity does not overlap with ACTH (D) or GH (E) immunopositive cells in anterior pituitary (GCGR, red; ACTH, green; GH, green).Abbreviations: ACTH, adrenocorticotropic hormone; AP, anterior pituitary; GCGR, glucagon receptor; GH, growth hormone; PI, pars intermedia; PP, posterior pituitary; RP, Rathke's pouch.Scale bars: A, 500 μm; B, 2000 μm; C, 50 μm; D-E, 20 μm.

Table 1
Primary and secondary antibodies used for immunohistochemistry.