MMP-2 and MMP-9 levels in plasma are altered and associated with mortality in COVID-19 patients

Abstract

Respiratory symptoms are one of COVID-19 manifestations, and the metalloproteinases (MMPs) have essential roles in the lung physiology. We sought to characterize the plasmatic levels of matrix metalloproteinase-2 and 9 (MMP-2 and MMP-9) in patients with severe COVID-19 and to investigate an association between plasma MMP-2 and MMP-9 levels and clinical outcomes and mortality. MMP-2 and MMP-9 levels in plasma from patients with COVID-19 treated in the ICU (COVID-19 group) and Control patients were measured with the zymography. The study groups were matched for age, sex, hypertension, diabetes, BMI, and obesity profile. MMP-2 levels were lower and MMP-9 levels were higher in a COVID-19 group (p < 0.0001) compared to Controls. MMP-9 levels in COVID-19 patients were not affected by comorbidity such as hypertension or obesity. MMP-2 levels were affected by hypertension (p < 0.05), but unaffected by obesity status. Notably, hypertensive COVID-19 patients had higher MMP-2 levels compared to the non-hypertensive COVID-19 group, albeit still lower than Controls (p < 0.05). No association between MMP-2 and MMP-9 plasmatic levels and corticosteroid treatment or acute kidney injury was found in COVID-19 patients. The survival analysis showed that COVID-19 mortality was associated with increased MMP-2 and MMP-9 levels. Age, hypertension, BMI, and MMP-2 and MMP-9 were better predictors of mortality during hospitalization than SAPS3 and SOFA scores at hospital admission. In conclusion, a significant association between MMP-2 and MMP-9 levels and COVID-19 was found. Notably, MMP-2 and MMP-9 levels predicted the risk of in-hospital death suggesting possible pathophysiologic and prognostic roles.

Keywords: COVID-19 pathophysiology; MMP-2; MMP-9; Metalloproteinases; SARS-COV2 infection.

Graphical abstract

Fig. 1

Plasmatic levels of metalloproteinase-2 (MMP-2, (A)) and metalloproteinase-9 (MMP-9, (B) in control (n = 29) and COVID-19 (n = 53) subjects’ group. Data are presented as mean ± SEM, normalized to control group, * ** *p < 0.0001 determined by Mann-Whitney test. AU = arbitrary unit.

Fig. 2

MMP-2 plasma levels in normotensive Control group (Control non-HPT, n = 15) and Hypertensive Control group (HPT, n = 14), normotensive COVID-19 patients (COVID-19 non HPT, n = 24) and hypertensive COVID-19 patients (COVID-19 HPT, n = 29) (A); MMP-2 plasma levels in Control group with normal BMI (Control non-obese, n = 16) and Control group with obesity (Control obese, n = 13) and COVID-19 group with normal BMI (COVID-19 non obese, n = 26) and COVID-19 group with obesity (COVID-19 obese, n = 27) (B). Data are presented as mean ± SEM, normalized to control group *p< 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001,determined by Kruskal-Wallis test and Dunn’s post hoc test. AU = arbitrary unit. No differences were observed in the MMP-2 levels for intra group comparisons between non-HPT vs HPT or non-obese vs obese.

Fig. 3

MMP-9 plasma levels in normotensive Control group (Control non-HPT, n = 15) and Hypertensive Control group (Control HPT, n = 14), normotensive COVID-19 patients (COVID-19 non-HPT, n = 24) and hypertensive COVID-19 patients (COVID-19 HPT, n = 29) (A); MMP-2 plasma levels in Control group with normal BMI (control non-obese, n = 16) and Control group with obesity (Control obese, n = 13) and COVID-19 group with normal BMI (COVID-19 non obese,n = 26) and COVID-19 group with obesity (COVID-19 obese, n = 27) (B). Data are presented as mean ± SEM, normalized to control group ***p < 0.001, ****p < 0.0001 determined by Kruskal-Wallis test and Dunn’s post hoc test. AU = arbitrary unit. No differences were observed for the MMP-9 levels for intra group comparisons between non-HPT vs HPT or non-obese vs obese.

Fig. 4

MMP-2 plasma levels in and COVID-19 patients that received inhibitor of renin-angiotensin system (iRAAS) treatment before hospitalization (COVID-19 on-iRAAS, n = 24) and COVID-19 patients without iRAAS treatment before hospitalization (COVID-19 off-iRAAS, n = 29)(A); MMP-2 plasma levels in COVID-19 patients that received corticosteroid treatment (COVID-19 on-corticosteroid, n = 47) and COVID-19 patients that without corticosteroid treatment (COVID-19 off-corticosteroid, n = 6) in the first day of hospitalization (B); MMP-2 plasma levels in and COVID-19 patients that had Acute Kidney Injury (AKI) during hospitalisation (without AKI, n = 27) and COVID-19 patients with AKI during hospitalization (with AKI, n = 26) (C); MMP-9 plasma levels in COVID-19 patients that received iRAAS treatment before hospitalization (COVID-19 on-iRAAS, n = 24) and COVID-19 patients without iRAAS treatment before hospitalization (COVID-19 off-iRAAS, n = 29) (D); MMP-9 plasma levels in COVID-19 patients that received corticosteroid treatment (COVID-19 on-corticosteroid, n = 47) and COVID-19 patients that without corticosteroid treatment (COVID-19 off-corticosteroid, n = 6) in the first day of hospitalization (E); MMP-2 plasma levels in and COVID-19 patients that had Acute Kidney Injury (AKI) during hospitalisation (without AKI, n = 27) and COVID-19 patients with AKI during hospitalization (with AKI, n = 26) (F). Data are presented as mean ± SEM, normalized to group not on medication or who had no AKI, *p < 0.05 determined by Mann-Whitney test. AU = arbitrary unit, ns = non significant.

Fig. 5

Relationship between MMP-2 plasmatic levels (A), MMP-9 plasmatic levels (B), and death in survivor (n = 30) and non-survivor (n = 23) groups on the first day after admission in ICU care. Data are presented as mean ± SEM, normalized to survivor group, *p < 0.05, ***p < 0.001, determined by Mann-Whitney test.

Fig. 6

Mortality prediction showed on the ROC curve including SAPS3 and SOFA scores ability to predict death on COVID-19 severe patients analyzed by multiple logistic regression AUC= 0.651, p = 0.062 (A); Mortality prediction showed on the ROC curve including MMP-2, MMP-9 associated with BMI, age, AKI and hypertension ability to predict death on COVID-19 severe patients analyzed by multiple logistic regression AUC= 0.864, p < 0.0001 (B).