Glucagon exerts effects on the mammalian heart. These effects include alterations in the force of contraction, beating rate, and changes in the cardiac conduction system axis. The cardiac effects of glucagon vary according to species, region, age, and concomitant disease. Depending on the species and region studied, the contractile effects of glucagon can be robust, modest, or even absent. Glucagon is detected in the mammalian heart and might act with an autocrine or paracrine effect on the cardiac glucagon receptors. The glucagon levels in the blood and glucagon receptor levels in the heart can change with disease or simultaneous drug application.
Left Atrium | Right Atrium | Ventricle | Remarks | |
---|---|---|---|---|
Cat | [3,5,25][3][5][25] PCE [25] PIE |
[5,25,26][5][25][26] In vivo: PIE [5,27][5][27] PIE: papillary muscle [27] PIE: perfused heart [28] No PIE in heart failure [25,29][25][29] PIE in heart failure |
||
Dog | [3,4][3][4] PIE | [3,4,30,31,32,33,34,35,][3]36,[4]37,[38,39,30][31][32][33 |
Compound: | Affinity at GR |
Organism Cells |
Half Life | |
---|---|---|---|---|
Antagonist HM15136 |
[95] Human GR: EC50 = 92 pM | CHO, Mice | In mice: 136 h | |
Antagonist: nonpeptide:2-(4-pyridyl)-5-(4-chlorophenyl)-3-(5-bromo-2-propyloxy-phenyl)pyrrole (L-168,049) | Human GR [96] IC50 = 3.7 nM | ][34][35][36][37][38][39]40[40] PCE [31,36][31][36] PIE |
[3,4,34,35,39,40][3][4][34][35][39][40] PIE [4] failing dog ventricle |
[40] Coronary perfusion enhanced |
Embryon-ic chick heart | [41 | |||
] | ||||
[ | ||||
54 | ||||
] | ||||
[ | ||||
72 | ||||
] | ||||
No PCE | ||||
Species | Right Atrium | Left Atrium | Ventricle | ||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Cat | [28] AC stimulation in normal heart [28] No stimulation of AC in failing hearts |
[74] AC stimulation [74] PDE not inhibited |
|||||||||||||||||||||||||||||
Chicken embryonic ventricular cardio-myocytes | [75] Calcium transients increased | Mice | |||||||||||||||||||||||||||||
] | PCE | ||||||||||||||||||||||||||||||
Dog | [31] | ||||||||||||||||||||||||||||||
AC not stimulated | |||||||||||||||||||||||||||||||
Antagonist: nonpeptide: N-[3-cyano-6-(1,1-dimethylpropyl)-4,5,6,7-tetrahydro-1-benzothien-2-yl]-2-ethylbutanamide (SC203972) | [ | [96] Human GR IC50 = 181 nM31] AC not stimulated |
[76] cAMP increase |
Mice | Frog | Embryonic chick heart | [42] PIE | ||||||||||||||||||||||||
[ | 31 | ] | AC stimulated | [41] AC stimulated | [41] Glucagon binding increased with age | ||||||||||||||||||||||||||
Antagonist, peptide: desHis1-Pro4-glucagon |
[97] Human GR IC50 = 1 nM |
Mice | persistent biological effects |
Guinea pig | [3,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57][3][43][ | Frog heart44 | |||||||||||||||||||||||||
Antagonist, peptide, des-His1-[Glu9]glucagon, GR specific, | [77]][45][46][47][48][49][50][ PDE inhibition | [98]51][52][53][54][55][56][57] PIE | Rat GR[43,[44,45,7747,58,59][43]][44][45][47][58][59] PCE | LTCC stimulation[ | |||||||||||||||||||||||||||
HEK | 44 | , | 45][44][45] No inotropic effect | ||||||||||||||||||||||||||||
Human cardiac tissue (isolated) | [50 | Guinea pig ventricle] No inotropic effect | [45 | ||||||||||||||||||||||||||||
LY2490921 GR unspecific | ,77][45][77] cAMP not increased | [23,50 | [59 | [98] Rat GR: 1.3 µM, [98] Human GLP-1-R: 1.2 µM][23][,76][59]50] No inotropic effect |
[76] AC not stimulated[50,60,61][50][60][61] No inotropic effect, [60] APD shortened, in vivo: PIE [49,62][49 |
[42][62] PIE in left ventricular papillary muscle strips from failing hearts | |||||||||||||||||||||||||
, | 77 | HEK | ][42][77] PDE inhibition | Human patient or healthy volunteer In vivo |
[38,51,52,53,63,64][38][51][52][53][63][64] PCE | [38,3851,][5152,][5253,]63,64][[ | |||||||||||||||||||||||||
Human atrium | 53 | ] | [ | ||||||||||||||||||||||||||||
Antagonist, peptide desHis1Glu9(Lys30PAL)-glucagon |
[99] GR: 170 pM | 78] AC stimulation and inhibition | [63][64] PIE (1–5 mg i.v.) | [38,52,63,64][38][52][63][64] Vascular peripheral resistance decreased, [38,51,52,63][38][51][52][63] nausea [51,63][51][63] vomiting, [64] flushing, [64] palpitations, [64] diarrhoea, and [64] hyperglycemia [52] Coronary flow increased, [52] oxygen consumption increased, [52 |
] Blood glucose increased | ||||||||||||||||||||||||||
Mice, HEK293 | persistent biological effects | ||||||||||||||||||||||||||||||
Monkey | Human | ||||||||||||||||||||||||||||||
Antibody REMD2.29 | [20] | [ | 42] PIE | ||||||||||||||||||||||||||||
AC stimulation in normal adult heart, | [74] No stimulation of AC in adult failing hearts, [74] AC stimulation human fetal heart |
[ | [100] 30 pM79] AC stimulation, [79] PDE not inhibited |
Mouse adult |
|||||||||||||||||||||||||||
Rabbit | [76] AC not stimulated in membranes[18] PCE, [54] No PCE |
[50 | [76] AC not stimulated in membranes, [80] AC stimulated in nuclei |
||||||||||||||||||||||||||||
Glucagon: physiological agonist | [95] Human GR: 800 pM | ] | No PIE | ||||||||||||||||||||||||||||
[ | 98] Rat GR: 400 pM [98] Rat GLP-1R: 4.9 nM |
[95] 5 min | Mouse fetal and neo-natal | Monkey | [81] PDE inhibition [42[65] PCE late-term fetal mouse heart |
]Neonatal mouse cardiomyocyte [54,65][54][65] PIE, [55 | |||||||||||||||||||||||||
Sacubitril: inhibitor of glucagon degradation | AC not stimulated | [101] Inhibition of glucagon metabolism: about 1 nM [42] AC not stimulated |
sacubitrilat,[56,57][ | 42]55][56][57] PCE | |||||||||||||||||||||||||||
AC not stimulated | Rabbit | [66] No PCE | [3,66][3][66] No effect | ||||||||||||||||||||||||||||
Mouse | |||||||||||||||||||||||||||||||
Anti-sense for GR | [77] PDE inhibition | ||||||||||||||||||||||||||||||
[ | 102] Mice, [103] patients T2DM |
Rat, adult | [3] No inotropic effect [67, |
Neonatal rat heart68] | [72][67][68] PIE | [23,44,67,68,69,70][23][44][67][68][69][70] PCE [3] No inotropic effect |
No AC stimulation[3,44,[45,44][68,45][70,71][3]68][70][71 | ] | |||||||||||||||||||||||
PIE | |||||||||||||||||||||||||||||||
Antagonist: Exenadin 9–39 | [98] IC50: Human GLP-1-R: 17 nM | [ | 44 | ,68][44][68 | ] Relaxation shortened | ||||||||||||||||||||||||||
Rat, fetal | Rat heart | [42,45,82,83][42][45][54, | |||||||||||||||||||||||||||||
Exenadin4: GLP-1-receptor agonist | [98[82][83] AC stimulation | ] EC50: Human GLP-1-R: 30 pM[45,[457284]][84] LTCC stimulation
AC: activity of cardiac adenylyl cyclase, PDE: phosphodiesterase, cAMP: 3′,5′ cyclic adenosine monophosphate, and LTCC: L-type calcium ion channel.
GR: glucagon receptor, GLP-1: glucagon-like-protein-1, min: minutes, h = hours, EC50 = half maximally stimulatory concentration, IC50 = half maximally inhibitory concentration. pM: pico (10−12) mole per liter, nM: nano (10−9) mole per liter, and µM: micro (10−6) mole per liter. HEK: human embryonic kidney 293 cells, CHO: Chinese hamster ovary cells, and T2DM: type 2 diabetes mellitus.
However, the published data would be consistent with the assumption that, at least in the non-failing human heart, the positive inotropic effect of glucagon is mediated, in all likelihood, by two receptors. At low glucagon concentrations, glucagon would stimulate the glucagon receptors in human atrial and ventricular preparations. At higher glucagon concentrations, glucagon would stimulate the GLP-1 receptors. When glucagon stimulates the glucagon receptor, as well as when glucagon stimulates the GLP-1 receptor, an increase in the cAMP concentrations in cardiomyocytes would follow, and this would increase the force of contraction in any case (Figure 1).
Interestingly, as heart failure (detected clinically, patients aged from 18 years to 62 years) increased, the positive inotropic effect was weaker and was even absent in papillary muscle strips from patients with end-stage heart failure, and, in their samples, glucagon failed to increase the activity of AC [60].
As animal models, to better understand the role of the glucagon receptors, mice with a generalized knockout of the glucagon receptor or heart-specific knockdown of the glucagon receptor have been generated [18]; they have a cardiovascular phenotype [18] that will be discussed below. However, these mice would also be helpful in the future to study the inotropic and chronotropic effects of glucagon. One would predict that glucagon should be unable to increase the force of contraction in the atrium and/or ventricle of mice with a knockout of the glucagon receptor.
3. Glucagon Receptor RegulationIn principle, the expression of the glucagon receptor can be regulated, and thereby the function of glucagon can be changed in the heart. In isolated, cultivated adult rat hepatocytes after 24 h, the glucagon receptor (at the mRNA and protein level) is down-regulated in the presence of glucose and upregulated in the presence of cAMP-increasing agents, acting via the stimulation of adenylyl cyclase (AC) through forskolin, the activation of the glucagon receptor through glucagon, or the inhibition of cAMP degradation (using a phosphodiesterase (PDE) inhibitor, 3–isobutyl-1-methyl-xanthine) and the stimulation of the β-adrenoceptor (by isoprenaline, [104]). If we assume that the glucagon receptor stimulates AC, then the measurement of AC is a surrogate parameter of glucagon-receptor–protein expression. Under this limitation, one can try to interpret older data generated before cloning the glucagon receptor was described.
Experimental hypothyroidism reduces the efficacy (but not the potency) of glucagon to stimulate AC (Table 2) in the hearts of rats, conceivably via a reduction in the density of the glucagon receptors [105]. Chronic β-adrenergic stimulation with a parenteral injection of living rats with isoprenaline likewise reduces the efficacy (but not potency) of glucagon to increase the activity of cardiac AC [106]. In the hearts of obese or hypertensive rats, the efficacy of glucagon to stimulate the activity of AC is reduced. In the dog heart, glucagon is more effective in stimulating the activity of AC in the ventricle than the atrium [81]. In the monkey atrium, glucagon fails to activate AC [81]. In human atrial or ventricular preparations, 10 μM of glucagon increases the activity of AC [107]. The problem is that very high concentrations of glucagon are required to detect an increase in AC activity.
Nevertheless, artifacts should be ruled out. Hence, there is a clear need to also measure in protein levels the expressional changes alluded to above for the glucagon receptor, for example, in hypothyroidism or after prolonged β-adrenergic stimulation, to confirm or refute this earlier work.
Like other heptahelical receptors, the glucagon receptor exhibits homologous desensitization due to receptor downregulation. At least this could be concluded when rats were treated with injections of 0.5 milli gram (mg) of glucagon/150–200 g (g, body weight) over 10 days [108]. The density of the signal (the binding of the membranes from the rat liver to the radioactive glucagon) declined [108]. However, PCR or better Western blot data in the heart would also be interesting under these conditions to confirm this reduced glucagon receptor expression over time. In a guinea pig left atrium, 1 μM of glucagon increased the force and cAMP levels, and both these parameters were desensitized by a 15 min pre-treatment with 2 μM of glucagon [43]. Similarly, the desensitization of glucagon in cardiac contractility is well known [3,44,45][3][44][45]. In mice in vivo, glucagon induced tachycardia [65]. In the liver, the desensitization of AC by glucagon has been convincingly demonstrated [109]. Desensitization for all G-protein-coupled receptors involves isoforms of the GTP-binding protein receptor kinase (GRK), protein kinase C, and a cAMP-dependent protein kinase (PKA, [110], Figure 1).
Glucagon can lead to the phosphorylation of the glucagon receptor in Chinese hamster ovary cells transfected with this receptor [111]. One would predict this to occur in the human heart, but this has not yet been reported. For receptor regulation, it is essential to know that glucagon receptors can be ubiquitylated [112,113,114][112][113][114]. This ubiquitination is involved in the internalization of the glucagon receptor upon agonist occupation. In addition, ubiquitination seems to drive biased agonism; it can lead to the stimulation of the non-canonical pathway comprising β-arrestin and MAP kinases [113]. The expression of the glucagon receptor in a transfected cell line could be reduced by antidiabetic drugs, namely thiazolidinediones [115]. One could speculate that this mechanism might be active in the heart. This is consistent with the information available for the promoter of the glucagon receptor, which contains an element involved in the mechanism of action of thiazolidinediones [116]. In the rat glucagon receptor gene, a glucose response element reduces the transcription of the glucagon receptor and, thus, its expression [117]. The promoter region of the human glucagon receptor has been studied [116]. This is relevant for the downregulation of the glucagon receptor. There is evidence for a cAMP-mediated downregulation of promoter transcription in cell culture studies [116]. This would, for instance, explain why glucagon can reduce the expression of the glucagon receptor by raising the cAMP level. This is a protective mechanism against deleterious increases in the cellular (in our case, cardiac) cAMP levels by auto-inhibiting the action of glucagon. Similarly, glucose autoinhibition exists. In liver cells in culture, high glucose concentrations reduce the efficacy of the glucagon receptors to increase glucose concentrations further [100]. Whether this holds true in the heart remains to be studied.
4. Glucagon Receptor Agonists and AntagonistsAs already mentioned above, glucagon has about 140 times less affinity for glucagon-like peptide receptors (about 130 nM) than specific agonists at these receptors (review: [9]). GLP-1-receptor expression as mRNA is three times more abundant in the human atrium than the human ventricle or cardiomyocytes from the left ventricle [94].
Here, species differences occur again, making it challenging to translate animal data directly to patients. In rat cardiomyocytes, glucagon-like peptide-1 increased cAMP levels but reduced contractility. This was accompanied by a reduced pH in the cardiomyocytes and a subsequent desensitization of the myofilament to calcium [118] (review: [119]). In contrast, the stimulation of GLP-1 receptors increased the force of contraction in human atrial preparations [94]. Peptides similar to glucagon can be used as agonists and antagonists at the glucagon receptors [119,120][119][120]. Taking a pharmacokinetic approach, non-peptide antagonists at the glucagon receptor have been developed that are per-orally available and might play a role in treating diabetes [12,119][12][119]. Some authors have claimed that glucagon fails to pass through the blood–brain barrier [121]. Hence, glucagon infused to treat the heart should not have direct side effects in the brain. However, this view has been challenged in recent years; some have claimed that glucagon can pass into the brain and exert physiological effects in the brain [122].
There has been some success in devising drugs comprising glucagon, a linker, and triiodothyronine. This bifunctional agonist was helpful in a mouse model of diabetes for preserving cardiovascular function [123]. Peptide fragments of glucagon or mutated glucagon stimulate the activity of AC in a cell culture system, usually with a potency and effectivity less than those of native glucagon. As such, these derivatives of glucagon are partial agonists, and thus, they are also antagonists at the glucagon receptor [82]. Another classification of glucagon receptor antagonists is to divide them based on their chemistry into small organic molecules (not protease-sensitive peptides), antibodies, or antisense ribonucleic acid (RNA) [124]. Theoretically, antisense RNA could also be encapsulated in a virus. Antagonists were initially designed to reduce the blood glucose levels (mice and monkeys: [125]).
Peptide analogs of glucagon have been radioactively labeled and used as tracers to study the glucagon receptor occupation in the living body [126]. Like insulin, glucagon can precipitate in solution or stick to the vessel wall. Hence, some have recommended the addition of bovine serum albumin and an acidic pH for its storage [127]. Others have developed a mutated glucagon called dasiglucagon, which contains seven amino acid mutations compared to glucagon and does not cause fibrils to form in aqueous solutions [113]. |