Coronavirus disease 2019 is associated with low circulating plasma levels of angiotensin 1 and angiotensin 1,7

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) binds to human host cells using angiotensin-converting enzyme 2 (ACE2).1 It had been hypothesized that decreased ACE2 activity due to SARS-CoV-2 binding and internalization may lead to elevated levels of its substrate angiotensin II (Ang2) and low levels of its product angiotensin 1,7 (Ang1,7).2 We recently demonstrated that circulating levels of Ang2 and aldosterone are not increased in coronavirus disease 2019 (COVID-19).3 However, Ang2 can be metabolized by other enzymes, thus not eliminating the possibility that renin–angiotensin–aldosterone system (RAAS) dysfunction could result in a functional deficiency of Ang1,7. Ang1,7 opposes the vasoconstrictive, proinflammatory, and pro-oxidative properties of Ang2.2 In animal models of acute respiratory distress syndrome (ARDS), Ang1,7 administration was shown to reduce acute lung injury.4 Clinical trials have started testing a synthetic Ang1,7 peptide (TXA127) in patients with COVID-19. Here, we present early data on circulating concentrations of Ang1,7 and angiotensin 1 (Ang1) in symptomatic SARS-CoV-2-infected patients in whom we had previously reported Ang2 levels. Adults with suspected COVID-19 presenting to the Emergency Department (ED) at the University of Cincinnati Medical Center and requiring a clinical blood draw were prospectively enrolled. The sample collection was performed under an institutional review board-approved waiver of informed consent. Samples were collected into prechilled tubes with protease inhibitor present to prevent degradation of angiotensin peptides. Only patients with a positive reverse transcription-polymerase chain reaction test for SARS-CoV-2 on nasopharyngeal swabs were included. Patients taking ACE inhibitors (ACEi) or angiotensin receptor blockers (ARBs) were excluded. As a control group, deidentified samples from healthy normotensive outpatient adults were employed. Plasma concentration of Ang1 (Enzo Life Sciences) and Ang1,7 (Novus Biologicals) was measured using Enzyme-Linked Immunosorbent Assay (ELISA) with duplicate measures to ensure test reliability with averages reported. Plasma concentrations were reported as median with interquartile range (IQR) and compared between groups using the Mann–Whitney U test. A total of 43 COVID-19-positive patients were initially enrolled, 13 of whom were excluded due to taking an ACEi or ARB. Three patients did not have aliquots to run in duplicate and were, hence, excluded. Thus, our final study population consisted of 27 patients with COVID-19 and 14 healthy normotensive controls. There were no significant differences with respect to sex (p = .165) or age (p = .226) between healthy controls and patients with COVID-19. The most common comorbidities were diabetes (30.7%), hypertension (30.7%), hyperlipidemia (25.9%), and obesity (22.2%). The median blood pressure at ED presentation was 137 (IQR: 118–148)/73 (IQR: 69–82) mmHg. Seventeen (63.9%) patients were hospitalized and seven (25.9%) required intensive care unit (ICU) admission. The plasma levels of Ang1 and Ang1,7 are shown in Figure 1. Compared to healthy controls, patients with COVID-19 had significantly lower levels of Ang1 (146.8 [IQR: 9.6–219.7] vs. 704.6 pg/ml [IQR: 526.3–1038.0]; p < .001) and Ang1,7 (168.1 [IQR: 132.1–267.9] vs. 318.1 pg/ml [IQR: 226.7–370.6]; p = .003) (Figures 1A and 1D). No statistically significant difference was found for either Ang1 (p = .399) or Ang1,7 (p = .272) between patients with COVID-19 discharged versus admitted from the ED (Figures 1B and 1E). However, patients with COVID-19 requiring an ICU during the course of illness displayed significantly lower Ang1,7 on admission compared to those not requiring an ICU (124.3 [IQR: 66.7–168.3] vs. 196.5 pg/ml [IQR: 142.0–288.8]; p = .013) (Figure 1C). A similar trend towards low circulating Ang1 was also observed in patients with COVID-19 needing an ICU (30.3 [IQR: 3.90–140.3] vs. 164.2 pg/ml [IQR: 112.5–449.3]; p = .054), although not statistically significant (Figure 1F). In this study, we present findings of low levels of Ang1,7 in patients with COVID-19 as compared to healthy controls. Moreover, we observed lower levels of Ang1, which suggests that the RAAS system may be relatively suppressed in patients with COVID-19. This may be due to low plasma renin activity or decreased angiotensinogen levels, which should be measured in future studies. As ACE2 activity may be attenuated by SARS-CoV-2 binding, the Ang1 may, therefore, be mostly metabolized via ACE, thus generating an adequate amount of Ang2, as previously observed,3 along with the low Ang1,7 levels presented in this report. Further investigations measuring ACE2 levels and ACE2 activity are needed to fully interpret these findings. Interestingly, Ang1,7 at admission was found to be even lower in patients needing ICU admission. Although it is unclear if this is due to a higher rate of viral attenuation of ACE2 or a nonspecific characteristic of ARDS or critical illness, these data lend credence to the ongoing trials employing synthetic Ang1,7 in COVID-19. Given that low levels of Ang1,7 are observed at admission suggests Ang 1,7 supplementation early in the disease course may be a reasonable therapeutic strategy. On the contrary, due to low circulating Ang1 and normal Ang2, recombinant ACE2 supplementation alone may not be an effective strategy for overcoming an Ang 1,7 deficit in COVID-19. Other factors should be considered when interpreting these findings. First, Ang1,7 is difficult to measure due to rapid degradation, as well as the potential for variable sensitivity and specificity when measured via ELISA. We encourage confirmation studies utilizing other biochemical techniques. Second, it is possible that circulating levels may not reflect tissue levels, including in the lung. While samples from bronchoalveolar lavage would be needed to confirm these findings, this approach is limited by the lack of a routine clinical indication for this procedure in patients with COVID-19. Third, the physiology of the RAAS is complex and may be impacted by many variables, including the state of the illness itself and underlying comorbidities.5 Future analyses with larger samples and more diverse control groups should control for these potential confounding variables. The authors declare that there are no conflict of interests. This study was supported by the University of Cincinnati College of Medicine Special Coronavirus 2019 Research Pilot Grant Program. Further material support was provided by CinCor Pharma. The data underlying this article will be shared on a reasonable request to the corresponding author.

Topics & Concepts