COVID-19 generates hyaluronan fragments that directly induce endothelial barrier dysfunction
Kimberly A. Queisser, Rebecca A. Mellema, Elizabeth A. Middleton, Irina Portier, Bhanu Kanth Manne, Frederik Denorme, Ellen J. Beswick, Matthew T. Rondina, Robert A. Campbell, Aaron C. Petrey
Kimberly A. Queisser, Rebecca A. Mellema, Elizabeth A. Middleton, Irina Portier, Bhanu Kanth Manne, Frederik Denorme, Ellen J. Beswick, Matthew T. Rondina, Robert A. Campbell, Aaron C. Petrey
View: Text | PDF
Research Article COVID-19 Vascular biology

COVID-19 generates hyaluronan fragments that directly induce endothelial barrier dysfunction

Abstract

Vascular injury has emerged as a complication contributing to morbidity in coronavirus disease 2019 (COVID-19). The glycosaminoglycan hyaluronan (HA) is a major component of the glycocalyx, a protective layer of glycoconjugates that lines the vascular lumen and regulates key endothelial cell functions. During critical illness, as in the case of sepsis, enzymes degrade the glycocalyx, releasing fragments with pathologic activities into circulation and thereby exacerbating disease. Here, we analyzed levels of circulating glycosaminoglycans in 46 patients with COVID-19 ranging from moderate to severe clinical severity and measured activities of corresponding degradative enzymes. This report provides evidence that the glycocalyx becomes significantly damaged in patients with COVID-19 and corresponds with severity of disease. Circulating HA fragments and hyaluronidase, 2 signatures of glycocalyx injury, strongly associate with sequential organ failure assessment scores and with increased inflammatory cytokine levels in patients with COVID-19. Pulmonary microvascular endothelial cells exposed to COVID-19 milieu show dysregulated HA biosynthesis and degradation, leading to production of pathological HA fragments that are released into circulation. Finally, we show that HA fragments present at high levels in COVID-19 patient plasma can directly induce endothelial barrier dysfunction in a ROCK- and CD44-dependent manner, indicating a role for HA in the vascular pathology of COVID-19.

Authors

Kimberly A. Queisser, Rebecca A. Mellema, Elizabeth A. Middleton, Irina Portier, Bhanu Kanth Manne, Frederik Denorme, Ellen J. Beswick, Matthew T. Rondina, Robert A. Campbell, Aaron C. Petrey

×

Figure 6

HA fragments present in COVID-19 plasma promote endothelial dysfunction.

Options: View larger image (or click on image) Download as PowerPoint
HA fragments present in COVID-19 plasma promote endothelial dysfunction....
(A) LMVEC were seeded on permeable supports (3 μm pore size) placed into a 24-well plate and grown to confluence. Cells were treated with or without HA purified from patient plasma samples (750 ng HA) or 1.5% patient plasma for 16 hours at 37°C to induce endothelial barrier disruption. The upper chamber was replaced with FITC-conjugated dextran (1 mg/mL, 40 kDa) in PBS, and medium from the lower chamber was measured after 1 hour. (B) LMVECs were grown to confluence in a transwell chamber and coincubated in the presence or absence of the ROCK inhibitor Y27632 (10 μM), in the presence or absence of biosynthetic 4 kDa HA (1000 ng), or in the presence or absence of HA purified from patient plasma for 16 hours at 37°C to induce endothelial barrier disruption, and barrier function was measured as above. (C) LMVEC were grown on coverslips until confluency in the presence or absence of Y27632 prior to treatment with either HA purified from healthy donor plasma, COVID-19 patient plasma, or LMW-HA (4 kDa). Cells were fixed, permeabilized, and stained with an antibody against VE-cadherin. Scale bar: 20 μm. (D) VE-Cadherin immunostaining was quantified by using 3 independent experiments and normalized to the number of endothelial cells per field using ImageJ. (E) LMVECs were treated for 48 hours in the presence or absence of siRNA (10 nM) or a CD44-HA blocking antibody (KM114, 1 μg) prior to treatment with HA and measurement of barrier function. In some experiments, HA was digested with Streptomyces hyaluronidase (HA’ase) as a specificity control. Data are reported as mean ± SEM; n = 5 independent experiments of at least 4 patients each, 1-way ANOVA followed by Tukey’s multiple comparison tests. Different alphabetical superscripts are significantly different from each other; P < 0.05.