Pneumonia Induced Rise in Glucagon Promotes Endothelial Damage and Thrombogenicity

Meet the First Author, see p 1032

Community-acquired pneumonia is a leading cause of hospitalization globally¹ and significantly associated with an increased risk of cardiovascular events, like myocardial infarction and stroke, within the first 30 days after respiratory infections.² The pathophysiological mechanisms connecting community-acquired pneumonia to cardiovascular events remain poorly understood, highlighting the need for further research to develop effective preventive strategies.

In this study, we examined pathological endothelial alterations in an experimental model of pneumococcal pneumonia. Wild-type male adult C57BL/6J mice (17 weeks old) were inoculated intranasally with 5×10⁵ Streptococcus pneumoniae (S pneumoniae) serotype 4 (TIGR4), while controls received a sham infection with PBS (Figure Ai). Antibiotic treatment began 24 hours postinfection (p.i.) with subcutaneous ampicillin, followed by oral amoxicillin-clavulanic acid until the end of the study.

Figure.

Glucagon increase caused by pneumococcal pneumonia promotes endothelial damage and thrombogenicity. Ai, C57BL/6J mice were randomly assigned to control or pneumonia groups and subjected to carotid injury (CI). Aii, Hematoxylin and eosin-stained lung images. Aiii, Lung injury scores per time point (n=6^(t1)). Aiiii, Evans blue-stained injured carotid arteries 3 days post-CI. Bi, Plasma proteomics analyzed using RStudio (n=5^(t1)). Bii, ELISA-measured glucagon levels in mice 48 hours and 10 days postinfection (p.i.; n=6, Control+CI n=8^(t1)). Ci, Real-time oxygen consumption rate (OCR) in human aortic endothelial cells (HAECs) stimulated with glucagon for 6 hours. Ciii, ROS production after glucagon stimulation for 3 hours (n=6^(ANOVA2)). Ciii, Scratch-wound healing assay with representative HAEC migration images. Di and Dii, ICAM-1 (intercellular adhesion molecule-1; n=6^(ANOVA2)) and E-selectin (n=6^(KW4)) expression in HAECs measured by flow cytometry as percentage of CD54⁺ or CD62E⁺ singlets. Diii, Quantification of THP-1 cell adhesion to HAECs under flow (n=8^(ANOVA2)). Diiii, Quantification of CD42b⁺ coverage on HAEC monolayer (n=6, 100 nM glucagon, n=5^(KW4)). Ei, Blood glucagon concentration measured by ELISA 48 hours p.i. in mice treated with liraglutide (n=6^(t1)). Eii, Evans blue-stained injured carotid arteries 3 days post-CI (pneumonia group results are the same as in Aiiii). Fi, Thrombus formation in carotid arteries of control (n=6), pneumonia (n=6), and liraglutide (n=4^(KW4)) groups at 3 time points. Fii, Time-course of thrombus area. Data were analyzed with GraphPad Prism 9 (mean±SEM). Normality was checked via the Shapiro-Wilk test. Normally distributed data were analyzed with unpaired 2-tailed Student t tests^(t1) or 1-way ANOVA^(ANOVA2) with Bonferroni post hoc tests. Nonparametric data were analyzed with Mann-Whitney U test U^(MW3) or Kruskal-Wallis test^(KW4) with Dunn post hoc tests. FCCP indicates carbonyl cyanide-p-trifluoromethoxyphenylhydrazone; ROS, reactive oxygen species; and THP-1, Tohoku Hospital Pediatrics-1, a human leukemia monocytic cell line.

Pneumonia severity and the treatment efficacy were assessed through histological lung injury scores. Lung injury was evident 48 hours p.i. but resolved by day 10 p.i., indicating effective treatment (Figure Aii and Aiii). To explore the impact of pneumonia on vascular regeneration, we induced carotid artery injury using bipolar microforceps 7 days p.i. De-endothelialization was quantified via Evans blue staining 3 days postinjury (10 days p.i.). Our results demonstrated significantly impaired wound healing in infected mice compared with controls (Figure Aiiii).

Murine plasma proteome analysis identified significantly elevated glucagon levels in infected mice (Figure Bi). ELISA analysis confirmed a rise in circulating glucagon levels at 10 days p.i., but not during the acute phase of pneumonia at 48 hours p.i. (Figure Bii).

To explore a potential causal relationship between elevated glucagon levels postpneumonia and impaired vascular regeneration, we examined the impact of glucagon on endothelial functional properties. In vitro assays revealed that glucagon disrupted mitochondrial function (Figure Ci), increased reactive oxygen species (ROS) production (Figure Cii), and compromised the migratory ability of human aortic endothelial cells (Figure Ciii). Through these effects, glucagon may exacerbate endothelial damage and hinder recovery.

Glucagon heightened inflammatory responses, as indicated by increased expression of adhesion molecules such as ICAM-1 (intercellular adhesion molecule–1) and E-selectin (Figure Di and Dii), along with increased monocyte adhesion to human aortic endothelial cells (Figure Diii). Moreover, glucagon enhanced endothelial thrombogenicity, as evidenced by enhanced platelet-endothelial interactions (Figure Diiii).

To counteract these effects, we administered the GLP-1R (glucagon-like peptide-1 receptor) agonist liraglutide to mice with pneumonia, which reduced postinfection blood glucagon levels (Figure Ei) and significantly improved vascular regeneration (Figure Eii).

Finally, we observed accelerated thrombus formation and vessel closure at the lesion site in infected mice by intravital microscopy (Figure Fi and Fii). These prothrombotic effects of pneumonia were diminished in infected mice treated with liraglutide (Figure Fi and Fii).

Overall, the study examined the consequences of bacterial pneumonia on endothelial inflammation, thrombogenicity, and regeneration after injury, yielding several important findings: (1) pneumococcal pneumonia impairs vascular regenerative potential after injury; (2) pneumonia increases endothelial inflammation and thrombogenicity following vascular injury; and (3) elevated glucagon levels after pneumonia may mediate, at least partly, these pathological endothelial alterations.

Vascular endothelial damage is a hallmark of respiratory infections. While proper endothelial repair mitigates pathological alterations of the vascular wall, impaired regeneration promotes vascular thrombogenicity,³ raising the risk of cardiovascular complications like myocardial infarction and ischemic stroke.

We identified elevated glucagon levels in the circulation of mice as a potential factor impairing endothelial healing. Our in vitro studies showed that glucagon stimulation impairs endothelial cell physiology, including impaired migratory capacity and mitochondrial bioenergetic function. Interestingly, the highest glucagon levels were observed during the recovery phase from the acute infection 10 days p.i., suggesting a delayed response rather than a link to acute metabolic demand. Early glucagon suppression by the GLP-1R agonist liraglutide restored endothelial healing and prevented increased thrombogenicity in mice with pneumococcal pneumonia. However, beside its glucagon-lowering effect, liraglutide may also directly protect endothelial cells.

In conclusion, our findings suggest a novel mechanism linking elevated blood glucagon levels after pneumococcal pneumonia to impaired endothelial regeneration and increased thrombogenicity. Modulating glucagon release may provide a novel approach to counteract the increased atherothrombotic cardiovascular risk associated with pneumonia.

ARTICLE INFORMATION

Acknowledgments

The authors thank Ulrike Behrendt, Denise Barthel, and Minoo Moobed for excellent assistance.

Sources of Funding

This study was funded by the German Federal Ministry of Education and Research (SYMPATH and 01ZX1906B) and by grants from the German Heart Research Foundation (F/01/22) to A. Haghikia.

Disclosures

None.

Data Availability

Data, methods, and materials are available upon reasonable request.