List of content
  1. Introduction
  2. The incretin GLP-1 and its elevation during systematic inflammation
  3. GLP-1R agonists as diabetes therapeutic agents
  4. The anti-inflammation features of GLP-1 and its based diabetes drugs
  5. GLP-1 based drugs in lung injury studiesV.I. Studies in chronic obstructive lung diseases and chronic lung inflammationV.II. Studies in acute lung injuryV.III. Combined use of MSC and GLP-1 based drugs
  6. Summary and perspective
  7. References
List of Abbreviations: ALI, acute lung injury; ANP, natriuretic peptide; ARDS, acute respiratory distress syndrome; BALF, brochoalveolar lavage fluid; BMI, body mass index; COPD, chromic obstructive lung disease; CRP, C-reactive protein; DPP-4, dipeptidyl peptidase 4; DPP-4i, DPP-4 inhibitor; EMA, European Medicines Agency; eNOS, endothelial NO synthase; FDA, FGF10, fibroblast growth factor 10; FOXA2, forkhead box A2; Food and Drug administration; GIP, gastric inhibitory polypeptide; GLP-1, glucagon-like peptide-1; GLP-1R, GLP-1 receptor; GLP-1RA, GLP-1R agonist; GLP-2, glucagon-like peptide-1; GWAS, genome-wide association study; HBA1c, Hemoglobin A1c; hMSCs, human MSCs; hs-CRP, high-sensitivity CRP; ICAM, intercellular adhesion molecule-1; IEL, intraepithelial lymphocyte; IFN-γ, interferon γ; IL-1α; IL-β, IL-1RA, interleukin 1 receptor antagonist; interleukin β; IL-4, interleukin 4; IL-6, interleukin 6; IL-8, interleukin 8; IL-13, interleukin 13; IL-33, interleukin 33; ILC2s, group 2 innate lymphoid cells; i.p, intraperitoneal; KGF, keratinocyte growth factor; KO, knockout; LPL, lipoprotein lipase; LPS, lipopolysaccharide; MCP-1, monocyte chemoattractant protein-1; MPGF, major pro-glucagon fragment; MSC, mesenchymal stem cells; NEP24.11, neutral endopeptidase 24.11; NF-κB, nuclear factor-kappa B; NO, Nitric Oxide; NLRP3, NLR family pyrin domain containing 3; PBMC, peripheral blood mono-nuclear cells (also known as MNC, mononuclear cells); PC1/3, pre-hormone convertase 1/3; PC2, pre-hormone convertase 2; PGF-2α, 8-iso-prostaglandin F2a; PKA, protein kinase A; PKG, protein kinase G; PMN, polymorphonuclear neutrophil; PPAR, peroxisome proliferator-activated receptor; sGC, soluble guanylyl cyclase; SGLT-2, sodium-glucose cotransporter-2; SP-1, surfactant protein A; STZ, streptozotocin; SVF, sromal vascular fraction; T1D and T2D, type 1 and type 2 diabetes; TG, triglyceride; Th2, T helper 2; TNF-α, tumor necrosis factor α; Tregs regulatory T cells; TxNIP, thioredoxin-interacting protein; TTF-1, thyroid transcription factor-1; VCAM-1, vascular cell adhesion molecule-1.
Introduction
Glucagon-like peptide-1 (GLP-1) is an incretin hormone produced in endocrine L cells, located mainly in distal ileum and the colon (1-4). Postprandial GLP-1 secretion leads to reduced plasma glucose level by mechanisms including the stimulation of insulin secretion, the inhibition of glucagon release, as well as the delay of gastric emptying (5). Furthermore, plasma GLP-1 elevation or GLP-1-based drug administration may directly reduce food intake, involving its function in the brain, mediated by GLP-1 receptor (GLP-1R), which is known to be expressed in brain hypothalamus and elsewhere (6-8). Various GLP-1 based drugs (also known as GLP-1R agonists, GLP-1RAs) have been developed and approved by Food and Drug administration (FDA), European Medicines Agency (EMA), or other authorities for diabetes treatment since 2005 (9). They are now widely utilized in treating type 2 diabetes (T2D) without side effects of weight gain and hypoglycemia, compared with various sources of commercial insulin (10). GLP-1 based drugs, such as liraglutide (commercially known as Victoza®) and semaglutide (Ozempic®) were also approved by FDA for chronic weight management in patients with obesity, overweight, or a weight related comorbid condition.
Studies on native GLP-1 and the use of these drugs in animal models or in treating patients with T2D have also uncovered their potent anti-inflammatory functions (11, 12). Since inflammatory responses also play important roles in the development and progression of diseases other than T2D, repurposing GLP-1 based drugs has been attracting researcher’s attention in various fields.
In this review, we have briefly summarized the discovery of GLP-1 as an incretin hormone and the development of GLP-1 based diabetes drugs. We then discussed studies leading to the recognition of the anti-inflammatory and immune regulatory functions of GLP-1 and its based drugs. The focus of this review, however, is on literature discussions of the discovery and functional assessment of potential therapeutic effects of GLP-1 based drugs in chronic and acute lung injuries. For more detailed discussions on utilization or potential utilization of GLP-1 based drugs in the treatment of diabetes, cardiovascular diseases, and neurodegenerative brain disorders, please see excellent review articles elsewhere (1, 13-18).
The incretin GLP-1 and its elevation during inflammation
GLP-1 was recognized as the 2nd incretin hormone back to 1983 (19, 20). Ebert and colleagues have observed that in the rat model, incretin activity was still preserved after gastric inhibitory polypeptide (GIP, the 1st incretin identified in 1960s, also known as glucose-dependent insulinotropic polypeptide) was removed from gut extracts by immune-adsorption (20). Following the isolation of the proglucagon gene (GCG /Gcg ) cDNA from fish, hamster, rat, mice, and humans, it was evident that in addition to encoding glucagon, a counter-regulatory hormone of insulin;GCG/Gcg cDNAs also encode two additional polypeptides defined as glucagon-like peptide 1 (GLP-1) and glucagon-like peptide 2 (GLP-2) (19, 21-27).
Gcg (GCG in humans) is abundantly expressed in pancreatic α-cells, intestinal endocrine L cells, and certain neuronal cells in the brain (3, 22, 28). Post-translational processing of the pro-hormone proglucagon occurs in tissue-specific manners via prohormone convertases (PC) known as PC1/3 and PC2. As shown in Figure 1A, in pancreatic α-cells, which mainly expresses PC2, the pro-hormone proglucagon is cleaved to produce the active hormone glucagon and other products including major proglucagon fragment (MPGF). In the brain and the intestinal endocrine L cells, expression of PC3 (also known as PC1) leads to the catalysis of pro-glucagon into GLP-1 and GLP-2, as well as glicentin and oxyntomodulin (23, 25, 26, 29-33).
Full length GLP-1 consists of 37 amino acid residues, and it becomes biologically active after it is truncated at the N -terminus to form GLP-17-37 (Figure 1B) or GLP-17-36 amide (not shown), with the latter to be more abundant in the circulation after meals (34, 35). As mentioned above, GLP-1 is the 2nd incretin hormone recognized to date (36-39) while GIP being the first one (20, 40, 41). Incretins are defined as gut produced hormones that can stimulate insulin secretion in glucose-concentration dependent manner. The inhibitory effect of GLP-1 on glucagon secretion, however, was not shared by GIP (42). Instead, a study showed that GIP might stimulate glucagon secretion from pancreatic islet α-cells (42). Native GLP-1 (both GLP-17-37 and GLP-17-36 amide) can be cleaved by the ubiquitously expressed enzymes dipeptidyl peptidase 4 (DPP-4) to produce GLP-19-37 or GLP-19-36amide, while further cleavage by neutral endopeptidase 24.11 (NEP 24.11) leads to the production of GLP-128-36 and GLP-132-36(43-48). Although certain biological functions of GLP-19-36amide, GLP-128-36amide and GLP-132-36amide have been described in pre-clinical investigations by our team and others (46-49), those are generally considered as inactive “degradation” products of GLP-1. Half-life of GLP-1 is relatively short, around 5-6 min in human plasma (34). For mechanisms underlying GLP-1 secretion, please see review articles by our team and by others elsewhere (50-53).
In humans, circulating GLP-1 level starts to ramp up only a few minutes after nutrient intake. It reaches the peak around 1 hour (54). Among the nutrient components, glucose was shown to be the strongest stimulus of GLP-1 secretion followed by sucrose, starch, triglycerides (TG), and certain amino acids (50, 51). Animal model studies have demonstrated that systemic inflammation induced by endotoxin (lipopolysaccharide, LPS) can also stimulate GLP-1 secretion in mice (55-57). Kahles and colleagues observed that among the inflammatory stimuli including endotoxin, interleukin 6 (IL-6), and IL-1β, it appears that IL-6 was sufficient and necessary to directly stimulate GLP-1 production and release, as in IL-6 knockout mice, endotoxin induced GLP-1 secretion was blunted (57). Kahles and colleagues have also reported that in an intensive care unit (ICU) cohort, GLP-1 plasma levels correlated with inflammation markers and the disease severity (57). Consequently, they suggested that GLP-1 serves as a link between the immune system and the gut (57). Indeed, metabolic illness and inflammatory diseases share certain common therapeutic targets (58). Individuals underwent cardiac surgery or autologous stem cell transplantation had up to 2-fold higher levels of circulating GLP-1, reported by Lebherz and colleagues, as well as Ebbesen and colleagues (59, 60). Patients with severe burn injury produced 3-fold more plasma GLP-1, while patients who died from severe burn injury had 5-fold higher GLP-1 levels than those who survived (61). In addition, patients that suffered from sepsis combined with T2D displayed an enhanced activation of endogenous GLP-1 system compared to non-diabetic patients (62). Thus, in both animal models and in human subjects, systematic inflammation can cause plasma GLP-1 elevation. Further investigations are required to determine whether plasma GLP-1 level can be developed as a biomarker for diagnosis and prognosis of inflammatory responses and inflammatory diseases. Patho-physiologically, elevated GLP-1 level during systematic inflammation may serve as a self-defence mechanism.
GLP-1R agonists as diabetes drugs
Although GIP was discovered more than a decade earlier compared with GLP-1, for various reasons, it has not yet been developed as a therapeutic agent. In 2005, the first GLP-1 based drug, exenatide (with the commercial name Byetta®), was approved by FDA for T2D treatment. Since then, ten additional GLP-1R agonists (GLP-1RAs) have been approved for T2D treatment. Table 1 lists those GLP-1RAs, as well as four DPP-4 inhibitors (DPP-4i) and DPP-4i based compound drugs.
As shown in Figure 1C, there are six GLP-1RAs currently approved by FDA as treatment options for T2D, including exenatide, liraglutide, lixisenatide, dulaglutide, albiglutide and semaglutide. Exenatide was developed based on studies on a peptide isolated from the saliva of the Gila monster, known as exendin-4. Exendin-4 contains 39 amino acid residues with a half-life longer than 10 min and sharing 53% amino acid sequence homology with human GLP-1 (63, 64). As a synthetic version of exendin-4, exenatide is resistant to DPP-4-induced degradation which contributes to a longer half-life about 2.4 h (65, 66). Lixisenatide, another derivative of exendin-4 with a half-life of about 3 h, was approved by FDA in 2016 (67, 68). Liraglutide (Victoza®) was approved by FDA in 2010, which is a modified human GLP-1, sharing 97% sequence identity with native GLP-1. The non-covalent binding with albumin further enhanced its stability. It is the first long-acting compound of GLP-1RAs, with a much longer half-life of 13 h (69-71). The other two long-acting GLP-1RAs, albiglutide and dulaglutide, consist of a dimer of human GLP-1 molecules that are fused to recombinant human albumin and a modified human immunoglobulin G4 heavy chain, respectively. Therefore, they have further extended half-life of about 5 days (10). Semaglutide (Ozempic) is the most recent approved long-acting GLP-1RAs for T2D in 2017, with a half life of 7 days (72, 73). Specifically, an equipotent once-daily oral administration form of semaglutide was approved in 2019 and greatly improved medication compliance (74).
As shown in Figure 1B, native GLP-1 can be cleaved by DPP-4, which is a ubiquitously expressed peptidase. DPP-4 can also inactivate GIP. Thus, DPP-4 inhibition can prevent degradation of both native GLP-1 and GIP. DPP-4i can specifically inhibit the enzymatic degradation activity of DPP4 by over 80%, leading to a doubling of active GLP-1 level (75). Sitagliptin (Januvia), developed by Merck & Co, was the first DPP-4i approved by the FDA as a T2D drug in 2006, followed by saxagliptin, linagliptin and alogliptin (76-78). DPP-4i can be administered orally and formulated either as a single-ingredient product or in combination with other diabetes medicines, including metformin (Table 1).
The anti-inflammation features of GLP-1 and its based diabetes drugs
Systemic inflammation is usually characterized by elevated pro-inflammatory cytokines and imbalanced immune cells in the circulation. As the first FDA approved GLP-1-based diabetes drug, the anti-inflammatory features of exenatide have been intensively investigated in patients with T2D. As early as 2007, Viswanathan and colleagues have demonstrated that in subjects with T2D, exenatide had two “non-metabolic actions”: the effect on attenuating plasma C-reactive protein (CRP) levels and the effect on lowering systolic blood pressure (11). A few years later, Kim and colleagues showed in mice that cardiomyocyte GLP-1R activation promoted the translocation of the rap guanine nucleotide exchange factor Epac2 to the membrane, leading to atrial natriuretic peptide (ANP) elevation, which lowers blood pressure (79). Interestingly, they have also located GLP-1R expression in mouse cardiac atria (79). In 2011, Wu and colleagues showed that in patients with T2D, 16-week exenatide treatment had not only reduced body mass index (BMI) and improved hemoglobin A1c (HbA1c) and glucose profile; but also decreased circulating levels of inflammatory markers including high-sensitivity C-reactive protein (hs-CRP) and monocyte chemoattractant protein-1 (MCP-1) (80). Furthermore, the level of oxidative stress marker 8-iso -prostaglandin F2a (PGF2α), was also reduced following exenatide treatment (80). The protein and mRNA levels of a battery of pro-inflammatory cytokines including tumor necrosis factor alpha (TNF-α), IL-1β and IL-6 in peripheral blood mono-nuclear cells (PBMC or MNC) were also shown to be suppressed by exenatide treatmentin subjects with T2D (81, 82). Moreover, both investigations observed the anti-inflammatory effect of exenatide in the absence of body weight loss in patients with T2D with 12-week exenatide treatment (81, 82). Thus, the anti-inflammatory effect of exenatide may not always be secondary to its body weight lowering effect (81-83). The anti-inflammatory effect of liraglutide was also demonstrated very recently by Zobel and colleagues in subjects with T2D (84). In this clinical trial, subjects with T2D were on 26-week liraglutide treatment. Zobel and colleagues observed discrete modulatory effect of liraglutide on the expression of inflammatory genes in PBMCs. Importantly, such modulatory effect was not observed in the in vitro settings with direct liraglutide treatment in the human monocytic cell line THP-1 (84). Furthermore, Zobel and colleagues reported that GLP-1R expression cannot be detected in the THP-1 cell line or in PMBCs (84).
The anti-inflammatory effect of GLP-1 based diabetes drugs was also investigated in various animal models. Although GLP-1 based diabetes drugs showed no improvement in patients with type 1 diabetes (T1D), Sherry and colleagues demonstrated that exenatide could facilitate the reversal of T1D in NOD mice treated with the “therapeutic” anti-CD3 monoclonal antibody (85). Mechanistically, the facilitation is likely involving the increase of anti-inflammatory subsets of T lymphocytes, such as T helper 2 (Th2) and regulatory T cells (Tregs) in mice (85-87). More recent studies have further demonstrated the T lymphocyte regulatory function of liraglutide and dulaglutide, as well as the DPP-4i sitagliptin (88-90). The DPP-4i linagliptin was also shown to attenuate obesity-related insulin resistance and inflammation by modulating M1/M2 macrophage polarization, reported by Zhuge and colleagues (91). In a Wistar rat model with intraperitoneal (i.p .) LPS challenge, exenatide treatment was shown to attenuate neutropenia, associated with decreased levels of a battery of pro-inflammatory cytokines, including IL-1α, IL-1β, IL-6, TNFα, and IFNγ (92). Utilizing the pro-adipocytic 3T3-L1 and RAW264.7 macrophage cellular models, several studies have shown that the DPP-4i anagliptin or liraglutide can inhibit nuclear factor-kappa B (NF-κB) pathway and secretion of a battery of pro-inflammatory cytokines (93-95). Although a few studies have indicated the expression of GLP-1R in rodent immune cells (87, 90), as mentioned above, a more recent human study by Zobel and colleagues showed that the repressive effect of liraglutide on expression of inflammatory genes in PBMCs was not observed in thein vitro settings with direct liraglutide treatment (84). In addition, Zobel and colleagues cannot detect GLP-1R in the THP-1 cell line or in primary MNCs (84). Figure 2 summarizes our current knowledge on the anti-inflammatory features of GLP-1 based diabetes drugs. It remains to be determined whether those anti-inflammatory and immune cell modulatory effects of GLP-1RAs are mediated directly by GLP-1R that are expressed in majority of immune cells, or by a small portion of immune cells, or by yet to be further explored mechanism.
GLP-1 based drugs in lung injury studies
During the past decade, there are substantial controversies in literature regarding GLP-1R expression in a few extra-pancreatic organs, including that in the liver, adipose tissues and immune cells (9, 96-98). Nevertheless, in vivo effects of GLP-1 based drugs in the liver and other extra-pancreatic organs are clear and substantial (96-101). The controversies could be partially due to the lack of reliable anti-GLP-1R antibodies (96, 97). It remains to be determined whether those extra-pancreatic functions of GLP-1 and its based drugs are mediated by certain brain-peripheral tissue axis or by a small portion of GLP-1R-producing cells scattered within each of those organs. However, we have learned for more than 25 years that lung is an extra-pancreatic organ, which exhibits the highest GLP-1R expression (102, 103). Hence, great efforts have been made in clinical trials and in various lung injury models, seeking for the possibility to repurpose GLP-1-based drugs in chronic and acute lung injury treatment. Here we will present our literature review on clinical investigations as well as studies with chronic lung injury and acute lung injury animal models. We will then summarize a few more recent studies on “therapeutic effect” of combined use of human mesenchymal stem cells (hMSCs) and GLP-1 based drugs in mouse acute lung injury models.