Alpha-1-antitrypsin antagonizes COVID-19: a review of the epidemiology, molecular mechanisms, and clinical evidence

There is burgeoning evidence that alpha-1-antitrypsin (AAT), the most abundant endogenous serine protease inhibitor (serpin), inhibits SARS-CoV-2 infection and mitigates many of the pathogenic mechanisms of COVID-19. Herein, we review the literature and mechanisms by which AAT antagonizes SARS-CoV-2 and COVID-19.

Upon binding of the spike protein of SARS-CoV-2 to its receptor angiotensin converting enzyme 2 (ACE2) on the cell membranes, the spike protein is sequentially cleaved by the cell surface serine proteases furin at the S1/S2 site and transmembrane protease 2 (TMPRSS2) at the S2’ site [1,2]. Moreover, neutrophil elastase, the canonical target of AAT, also cleaves near the S1/S2 site for the SARS-CoV-2 variant with the D614G mutation in the spike protein [3]. These modifications of the spike protein enable fusion of the viral envelope with the cell plasma membrane.

SARS-CoV-2, the cause of COVID-19, is the third beta-coronavirus and second sarbecovirus to emerge in humans in this century. It is likely that additional coronaviruses will jump the species barrier, potentially leading to new pandemics. Hence, despite the availability of efficacious COVID-19 vaccines, future SARS-related strains and other coronaviruses are likely to remain problematic because: (i) vaccine-refusal within a segment of the population, estimated to occur in 25–30% of the U.S. population per 2021 Gallup Poll and Monmouth University Polling Institute as well as the 4% vaccination rate in resource-poor countries, seedbeds for development of new SARS-CoV-2 variants; (ii) breakthrough cases even in those who have been vaccinated, especially in immunocompromised individuals [4–9]; (iii) fatigue in public health measures; (iv) anti-viral agents Paxlovid (nirmatrelvir + ritonavir) and molnupiravir were largely studied in the context of earlier SARS-CoV-2 strains with more recent occurrence of breakthrough symptoms and viral replication [10,11]; and (v) evolution of new SARS-CoV-2 strains such as the now dominant omicron strains (e.g. BQ.1, BQ.1.1, and XBB.1.5), which are much more transmissible, more immune evasive, less responsive to vaccines/boosters with protection that declines more rapidly, and/or increased or complete resistance to existing monoclonal antibodies [12–21]. Even as recently as September 2022, when the number of patients hospitalized for COVID-19 were declining, 500–1000 individuals still die each day (182 000–365 000 per year) from COVID-19 in the U.S. In contrast, 12 000–52 000 people die from influenza each year in the U.S. for the last 10 years before COVID-19.

Pneumonia/acute lung injury with refractory hypoxemia and septic shock due to a hyperinflammatory response — due largely to increases in inflammatory cells recruited and activated by interleukin-1 (IL-1), IL-6, IL-8, IL-17, and tumor necrosis factor (TNF) — are the most common causes of COVID-19 deaths [22–26]. Severe COVID-19 is also associated with CD4⁺ and CD8⁺ T cell lymphopenia and decreased interferon-gamma (IFNγ)-positive CD4⁺ T cells [27]. The current treatment for severe COVID-19 is largely supportive care combined with the glucocorticoid dexamethasone, which reduced mortality by 2.9% in those receiving supplemental oxygen only and by 12.1% in those requiring mechanical ventilation [28]. Remdesivir is also used for hospitalized patients [29] although its efficacy is controversial [30,31]. Anakinra (IL-1 receptor antagonist, IL-1Ra) has been evaluated for treatment of severe COVID-19 with no or marginal success [32–36].

Antagonists to IL-6 and IL-6 receptor (IL-6R) have shown either negative results or benefit in some patients recalcitrant to dexamethasone [37]. Several concerns with the use of antagonists to IL-6 or IL-6R are that IL-6 helps induce antibody production and activates T cells against infectious agents [22,38]. The FDA has also issued a Black Box warning for the anti-inflammatory Janus kinase 1 inhibitors (baricitinib, tofacitinib, upadacitinib) due to risks of serious cardiac events and cancer.

AAT is the third most abundant protein in plasma and the most prevalent serpin in the body. As an acute phase protein, AAT levels increase three- to five-fold with systemic infection or inflammation, underscoring its important homeostatic function [39,40]. Persons with a normal SERPINA1 gene are designated as protease inhibitor MM (PiMM). Two of the most common abnormal AAT alleles are Z-AAT (rs28929474) and S-AAT (rs17580), with the PiZZ, PiSZ, and some PiMZ genotypes resulting in severe or relative AAT deficiency. Some AAT alleles (F and I) encode a dysfunctional AAT protein but with normal circulating levels.

It has been estimated that approximately 100 000 individuals in the U.S. have severe AAT deficiency, with <10% currently identified [41]. However, this assessment is likely to be underestimated based on the calculation that if there are 13 million subjects with chronic obstructive pulmonary disease (COPD) in the U.S. — and that AAT deficiency (mostly PiZZ) is a well-established risk factor for COPD and PiZZ comprises 1–3% of all COPD cases [42] with both lower and higher prevalence reported based on the cohort defined [43] — then 130 000–390 000 with known COPD are predicted to be AAT deficient (PiZZ). The number of AAT deficient individuals in the U.S. population is likely to be even more as some patients with AAT deficiency have not developed COPD (e.g. children), COPD has yet to be detected in them, or have undiagnosed AAT deficiency-associated liver disease. Even in the absence of frank AAT deficiency, the AAT response to a systemic infection may be inadequate as has been shown for hospitalized COVID-19 patients [44,45] and among millions with certain heterozygous AAT mutations (PiMZ) [39]. Furthermore, oxidation of methionine 351 and/or 358 of even normal AAT may cause loss of its serpin activity [46]. Oxidation may also induce AAT polymerization [47,48]. Because COVID-19 is associated with increased oxidative stress [49], AAT may be rendered ineffective even with increased levels. Indeed, the concept of ‘relative AAT deficiency’ has been raised wherein external stressors such as oxidative stress, smoke exposure, high glucose levels, and bacterial proteases inactivates normal AAT [50].

COVID-19 cases were increased in parts of Italy with greater prevalence of AAT deficiency [51]. AAT-deficient subjects were also found to be 8.8-fold more likely to have symptomatic COVID-19 than the general population [52]. Shapira et al. [53] found a significant direct correlation between the frequency of the PiZ and PiS alleles with COVID-19 death rates in 67 countries. Yoshikura [54] also reported a robust correlation between the Pi*Z variant and the number of COVID-19 cases (correlation coefficient (CC) = 0.8584) and deaths (CC = 0.8713) in 68 countries. Both Europeans and Latinos have relatively higher rates of PiZZ, PiSZ, and PiMZ, estimated to be ∼12%, and are also the predominant ethnic groups in countries having the highest COVID-19 cases and deaths [55–58]. Furthermore, evidence implicates AAT deficiency itself — and not its pulmonary sequelae of emphysema or bronchiectasis – as a major risk factor for SARS-CoV-2 infection [57]. Hospitalized patients with severe COVID-19 had significantly lower plasma AAT levels than those admitted for non-COVID-19 pneumonia [59]. In a multi-center study of over 2000 patients with COVID-19, possessing the Pi*Z allele and/or plasma AAT level <116 mg/dl were highly associated with severe COVID-19 compared with non-severe disease [60]. In 105 subjects with AAT deficiency and COVID-19 infections in 10 European countries, poor outcome was more likely with PiZZ genotype than PiSZ; interestingly, only 14 subjects were receiving regular AAT augmentation and they showed a trend toward better outcome [61]. Murgia et al. [62] analyzed the effect of combined AAT anomalies (PiMZ, PiSZ, and PiMS) and air pollution on COVID-19 using the AAT*air pollution global risk score. They found that the ranking of the AAT*air pollution global risk score paralleled with the ranking of countries stratified by incidence of COVID-19 deaths [62].

McElvaney et al. [63] found the IL-6:AAT ratio is markedly elevated in critically ill COVID-19 patients compared with controls and that this ratio directly correlated with prolonged hospital stay and mortality. Since this study did not specifically target AAT-deficient subjects, it suggests that relatively normal or even modestly elevated AAT levels in the face of a severe infection may be wholly inadequate. Similarly, a study of 137 hospitalized patients with COVID-19 (of whom 56 were critically ill) found that the AAT:IL-6 ratio (reverse ratio from aforementioned study) was lower in non-survivors and those with thrombosis [64]. In COVID-19-associated acute respiratory distress syndrome, manifested by increased neutrophil elastase activity, there is an excess of protease:anti-protease ratio [65]. Interestingly, IL-6 induced AAT and tocilizumab (an IL-6R antagonist) down-regulated AAT in COVID-19 [65].

In contrast with the aforementioned studies, a community-based cohort study from the U.K. showed there was no increase in either SARS-CoV-2 infection or death rates in the presence of S- or Z-AAT alleles [66]. Similarly, in 61 Swedish patients hospitalized with COVID-19, only two were found to be PiMZ and none for PiMS, PiSZ, PiSS, or PiZZ [67]. While it is difficult to reconcile these incongruent findings between AAT anomalies and hospitalized COVID-19 cases, one plausible explanation is that some with known AAT deficiency took greater precautions as evinced by the finding that the AlphaNet (AAT-deficient) population had lower case rate but greater hospitalization rate for COVID-19 during the survey period from January 2020 to February 2022 [68]. In a large genome-wide association study (GWAS) meta-analysis of over 125 000 COVID-19 cases and 2.5 million controls, SERPINA1 (AAT) gene polymorphism was not identified to be a major associative factor for severe COVID-19 [69]. However, the benefits of GWAS are well known but the limitations often overlooked. GWAS studies use power in numbers for complex and common diseases that have minor allele frequencies (MAF), typically >5% for common variants with small effects across large numbers of unrelated individuals; in contrast, genes with rare variants such as SERPINA1 with MAF <5% typically associated in related individuals with large effects would be missed. The variants in GWAS themselves are not always the causal variant but may be signal for a nearby functional variant. Additional limitations of GWAS include the epidemiological design and the methodology (SNP arrays with imputation vs. whole genome sequencing) where SNP arrays may not be designed to capture genes with rare variants [70,71]. In summary, uncommon and rare genetic variants — as illustrated by SERPINA1 genotypes that cause AAT deficiency — are not well captured by GWAS arrays [72,73]. Indeed, although one study found that the Z allele of SERPINA1 gene was associated with COPD with an odds ratio of 1.17–1.74 [74], seven other GWAS or exome-wide studies did not find an association between COPD and the Z allele [75-81]; yet, the Z-AAT allele is strongly linked clinically to COPD [82]. In other words, if AAT deficiency had not been fortuitously identified in 1963, it is unlikely that GWAS or exome-wide studies would have identified SERPINA1 variants (causing AAT deficiency) as a significant risk/associative factor for COPD.

It was reported 30 years ago, in the context of HIV infection, that an engineered AAT (α1-PDX) inhibits furin, the serine protease that cleaves the spike protein of SARS-CoV-2 before cleavage by TMPRSS2 [83,84]. Five years before the COVID-19 pandemic, AAT was reported to inhibit TMPRSS2, also required to activate the hepatitis C virus [85]. Following binding to their cell surface receptors, the coronaviruses capable of causing fatal disease — with SARS-CoV and SARS-CoV-2 binding ACE2 and MERS-CoV binding DPP4 — all require TMPRSS2 to process the viral spike protein to gain cellular entry.

Because AAT is a potent serpin, it has the potential to inhibit both TMPRSS2 and SARS-CoV-2 infection. Indeed, in HEK293T cells engineered to overexpress TMPRSS2, physiologic concentrations of AAT potently inhibited TMPRSS2 activity [86,87]. Ritzmann et al. [88] showed in both two-dimensional submerged airway epithelial cells and three-dimensional airway epithelial organoid cultures that physiologic concentrations of AAT (1 and 5 mg/ml) significantly inhibited TMPRSS2 activity and decreased intracellular burden of SARS-CoV-2 [88]. Wettstein et al. [89] also demonstrated that AAT inhibited SARS-CoV-2 infection of human airway epithelial cells (hAEc). Oguntuyo et al. [90] reported that sera from SARS-CoV-2 naïve individuals inhibited cellular entry of SARS-CoV-2 and identified AAT as the molecule responsible. AAT strongly inhibits elastase, which also cleaves the spike protein at the S1/S2 site of a SARS-CoV-2 variant [3]. Singh and co-workers [91] generated variants of SerpinB3 — through mutagenesis of its reactive center loop — that were more potent in inhibiting TMPRSS2 than AAT. Rosendal et al. [92] used a three-dimensional lung cell model using primary cells from multiple donors and found that more resistant donors to in vitro SARS-CoV-2 infection expressed higher transcript levels of SERPINA1 (AAT gene), SERPINE1, and SERPINE2 in airway cells (basal, ciliated, club, and goblet cells) [92].

Negatively charged polysaccharides are capable of augmenting the interaction between a serpin and its cognate serine protease [87,93]. Heparin has the highest negatively charge density of any biologic molecules. We found that AAT inhibition of TMPRSS2 activity and of the intracellular burden of human coronavirus 229E (HCoV-229E) in primary hAEc were augmented significantly by enoxaparin (a low molecular mass heparin) [87]. Detailed in silico modeling demonstrated: (i) suboptimal electrostatic complementarity between AAT and TMPRSS2 at both the buried interface and solvent-exposed interface rim in a molecular charge model (Figure 1A); (ii) the reactive center loop of AAT (Figure 1B, magenta) adopts an inhibitory-competent conformation compared with the crystal structure of TMPRSS2 (Figure 1B, cyan) bound to an exogenous (nafamostat) or endogenous (HAI-2) inhibitors of TMPRSS2; and (iii) negatively charged heparin (stick figures in Figure 1A–C) bridges adjacent electropositive patches at the TMPRSS2–AAT interface, neutralizing otherwise repulsive forces [87].

In addition to its anticoagulant effect and its ability to augment AAT inhibition of TMPRSS2, heparin may antagonize COVID-19 by other means (Table 1). Since AAT and enoxaparin individually and synergistically antagonize several of the pathogenic mechanisms of COVID-19, we posit that they will be more effective than agents that only inhibit TMPRSS2 such as camostat [94].

In addition to inhibiting TMPRSS2, preventing the processing of the spike protein and cellular entry of SARS-CoV-2, AAT has a panoply of other activities that could antagonize both the intracellular virus and the multiple pathogenic mechanisms associated with severe COVID-19 (Figure 2 and Table 1) [23,95]. Such host-directed activities are less likely to be affected by SARS-CoV-2 mutations.

AAT augments host immunity via several mechanisms against various pathogens including influenza [96], HIV [97–101], Pseudomonas aeruginosa [102], Mycobacterium intracellulare [103], and Streptococcus pneumoniae [104]. AAT reduced the burden of cell-associated M. intracellulare through induction of autophagy [103]. Since autophagy has been implicated in controlling MERS-CoV [105], AAT-induced autophagy is a plausible mechanism by which SARS-CoV-2 is killed.

AAT inhibits production of pro-inflammatory cytokines [63,103,106–113], induces expression of the anti-inflammatory IL-1Ra [112], and skews T cell differentiation toward the T regulatory phenotype [114]. These anti-inflammatory effects of AAT — augmented by an increase in sialylation of AAT [110] — could mitigate the injurious hyperinflammatory response typically seen seven to 10 days after the symptom onset of severe COVID-19. In contrast with the anti-inflammatory effects of glucocorticoids, AAT induces a more selective cytokine shift toward resolution; e.g. suppression of pro-inflammatory cytokines plus induction of endogenous IL-1Ra, whereas glucocorticoid causes a more global immunosuppression including inhibition of IL-1Ra expression [112]. Interestingly, when AAT and glucocorticoid are combined in macrophages, lower glucocorticoid concentrations used allowed up-regulation of IL-1Ra, whereas higher glucocorticoid levels antagonized the effects of AAT [112]. Vitamin D deficiency may also lead to decreased release of AAT from T cells, causing decreased release of the anti-inflammatory cytokine IL-10 [115,116]. Tumpara et al. [117] showed that recombinant AAT (made in Chinese hamster ovary cells) was more potent than plasma-derived AAT in inhibiting pro-inflammatory cytokine production from peripheral blood mononuclear cells stimulated with both lipopolysaccharide and spike proteins.

Recently, we reported the novel finding that AAT binds to the cytoplasmic glucocorticoid receptor (GR) in macrophages, with the AAT–GR complex having gene regulatory, anti-inflammatory, and host-defense properties [118]. Perhaps relevant to COVID-19, we found that the AAT–GR complex inhibited IL-8 production from stimulated macrophages. Since IL-8 is a potent chemokine for neutrophils and neutrophilia is a poor prognostic marker for COVID-19 [119], AAT antagonism of IL-8 production may be another mechanism against COVID-19.

AAT has been shown to protect the lungs from acute lung injury [120,121]. Since an imbalance of protease and anti-protease is implicated in the pathogenesis of COVID-19 acute lung injury, absolute or relative deficiency of functional AAT is a plausible risk factor for severe COVID-19 [122,123]. In 39 critically ill patients with COVID-19, the median AAT level was modestly increased (by 1.5-fold the upper limits of normal), but there was a disproportionately increase in blood matrix metalloproteinase-12 (MMP-12, produced mainly by macrophages) and neutrophil elastase by ≥10-fold and four- to seven-fold, respectively [124]. Furthermore, the peak MMP-12 and neutrophil elastase levels were higher in those who subsequently died. When AAT phenotype was analyzed temporally, there was an increase in sialylation of the M0 and M1 AAT glycoforms in all those who died and in 59% of those who survived, suggesting that a shift in sialylation was a counterregulatory response to increase anti-elastase activity in these critically ill COVID-19 patients [124]. Interestingly, the increased protease:anti-protease ratio seen with COVID-19 may be due to cleavage of AAT by proteases, as evinced by increased AAT fragments C-36 (cleavage product of neutrophil elastase) and C-42 (cleavage product of MMP-12) (Figure 2) [125]. Additionally, cathepsin B may play a role in the hyperinflammatory response of COVID-19 [126] and AAT inhibits neutrophil elastase induction of cathepsin B in vivo and in vitro (Figure 2) [127]. Elastase also induces MMP-2, the plasma levels of which correlated with in-hospital death in patients with severe COVID-19 [128]. Through AAT inhibition of neutrophil elastase, subsequent inhibition of MMP-2 production is another mechanism by which AAT antagonizes injury to the lung extracellular matrix (Figure 2) [127,129]. AAT also has been shown to have anti-apoptotic function in epithelial and endothelial cells [130,131] (Figure 2), which may not be necessarily dependent on its anti-elastase activity as has been shown for protection against hepatocellular death [132]. In assessing lung function 12 months after recovery from COVID-19, those with PiMS and PiMZ genotypes had lower AAT levels (compared with PiMM), especially in those with persistently impaired lung function [133].

In 31 patients with SARS-CoV-2 infection, plasma levels of exosome-associated neutrophil elastase were elevated compared with control subjects and positively correlated with signs of endothelial damage and circulating AAT levels [134]. Whereas AAT was able to inhibit purified neutrophil elastase in macrovascular endothelial cells, exosome-associated neutrophil elastase was resistant to AAT inhibition.

ADAM17 is a cell surface protease that causes shedding of ACE2, the cell surface receptor for SARS-CoV-2 [135] (Figure 3A). While this cleavage would theoretically reduce initial binding SARS-CoV-2 to the cell membrane, reduction of membrane-bound ACE2 (mACE2) may increase lung injury and edema because ACE2 has been shown to protect the lungs from injury — by reducing both bradykinin production and neutrophil infiltration as well as catalyzing angiotensin II to the anti-inflammatory angiotensin-(1–7) and angiotensin-(1–9); both metabolic products of angiotensin II also protect lung epithelial cells from death [136–138]. ADAM17, aka TNF converting enzyme, also cleaves membrane-bound TNF to its soluble form, perpetuating the inflammatory response including the induction of MMP-2 (Figure 3A).

Both membrane-bound and soluble IL-6R have pro-inflammatory function but the former occurs mostly in macrophages, whereas the latter occurs in all cell types (Figure 3B,C). Upon binding of IL-6 to membrane-bound IL-6R, the IL-6–IL-6R complex engages gp130, a cell surface molecule that transmits downstream intracellular signaling, a process known as cis-IL-6 signaling (Figure 3B,C, left-hand side). Soluble IL-6R binding to IL-6 can also bind to gp130, a process known as trans-IL-6 signaling (Figure 3C, right-hand side). During inflammatory states, ADAM17 is activated and cleaves membrane forms of IL-6 and IL-6R to their soluble forms [139–141], resulting in more widespread IL-6-mediated inflammation (Figure 3C). ADAM17 also induces shedding of fractalkine, a membrane-bound chemokine that facilitates intercellular adhesion and mainly found in neurons and endothelial cells [142]; thus, ADAM17 may reduce intercellular adhesion between endothelial cells.

AAT inhibits activity of ADAM17 [106,132,143]. Given that ADAM17 may augment inflammation and other mechanisms that promote lung injury, AAT inhibition of ADAM17 is likely to be anti-inflammatory and to protect the lungs from injury. While AAT inhibition of ADAM17 could also preserve ACE2 on the cell surface, such preservation of the receptor for SARS-CoV-2 may paradoxically decrease viral infectivity to other cells by decreasing the egress of newly synthesized viral particles [138,144].

Elevated absolute neutrophil number, increased neutrophil percentage, and high neutrophil:lymphocyte ratio in the blood predict progression to severe COVID-19 [119]. One study showed that neutrophil:lymphocyte ratio ≥4.5 and ≥6.5 were associated with significantly increased severity and mortality, respectively [145]. There is also increasing evidence that aberrant formation of neutrophil extracellular traps (NETs) — comprised of neutrophil-derived decondensed chromatin decorated with elastase, cathepsin G, and histones — play a pathogenic role in the immunothrombosis, mucous secretions, and cytokine production seen with severe COVID-19 [146-152]. In hospitalized patients with COVID-19, increased NETs formation was found in the plasma, tracheal secretions, and lung tissues [152]. In post-mortem lung tissues of individuals who died from COVID-19, NETs were found in the airways, interstitium, and arterioles [151]. Since elastase plays a key role in NETs formation by degrading specific histones and promoting chromatin decondensation [153], AAT has the potential to inhibit pathogenic NETs formation [154]. This notion is supported by an ex vivo pre-COVID-19 study showing that AAT is able to change the shape and adherence of NETs [155]. However, in the bronchoalveolar lavage fluid of COVID-19 patients, it was found that AAT–elastase complex was not found (due to elastase binding to acute phase proteins, histones, and C3) and that exogenous AAT did bind to NETs but did not qualitatively inhibit NETs formation [156].

AAT is presently administered to patients with confirmed AAT deficiency and airflow limitation [82]. A patient survey of AAT-deficient patients showed that AAT replacement reduced the number and severity of lung infections [157]. In 19 AAT-deficient emphysema patients in whom AAT was abruptly discontinued due to insurance reimbursement issues, the number of pulmonary exacerbations — historically often due to viral or bacterial infections — dramatically rose [158].

A 43-year-old woman with cystic fibrosis, who developed severe respiratory deterioration with marked increase in pro-inflammatory cytokines in the airways and the blood following acquisition of COVID-19, was administered intravenous AAT at 120 mg/kg/week for four consecutive weeks [159]. Following each AAT dose, both plasma and airway IL-1β, IL-6, and IL-8 levels decreased and airway neutrophil elastase level decreased, accompanied by clinical and radiographic improvements [159].

An open-label study administered AAT to nine patients requiring hospitalization for mild or moderate COVID-19 — four received inhaled AAT at 100 mg/day for 7 days while five received inhaled AAT at the same dose plus intravenous AAT at 60 mg/kg on Days 1, 3, and 5 [88]. Two of the patients also received remdesivir and two received dexamethasone. In all nine patients, the oxygenation improved, serum C-reactive protein decreased, and all were discharged in good functional status. While there was no control group in this study, an analysis of a contemporaneous observational cohort (CORSAAR) of 23 COVID-19 patients matched for age and disease severity revealed three COVID-19 deaths and a delayed decrease in C-reactive protein in the historical ‘control’ group that did not receive exogenous AAT [88]. Three Israeli patients who were critically ill with COVID-19 (all mechanically ventilated) were administered three doses of intravenous AAT (1–2 days apart) early in the hospitalization (initiated Day 2–7 of hospitalization) and all improved objectively (as measured by the SOFA score and PaO₂/FiO₂ ratio) within a day of administering the AAT; a fourth patient who received only one dose of AAT at ∼Day 21 of hospitalization succumbed [160]. In a phase 2, multi-center, randomized, double-blind, placebo-controlled trial of 36 patients with moderate to severe COVID-19-associated acute respiratory distress syndrome who received either weekly placebo (n = 11), weekly intravenous AAT at 120 mg per kg (n = 13), or AAT once followed by weekly placebo (n = 12), AAT treatment (in the two latter groups) resulted in significant decrease in IL-6 compared with the placebo group at 1 week [161]. However, AAT administration (n = 25) also reduced soluble receptor for TNF [161], which may increase free TNF. While there was no difference in mortality or ventilator-free days, the AAT-treated patients had a trend toward decreased time on the ventilator P = 0.44 [161]. Strassmair and Stangl [162] have proposed the idea that delivery of mesenchymal stem cells that express AAT, with their ability to ‘home-in’ to the lungs, might be an effective mechanism to deliver AAT.

Several large studies demonstrate robust epidemiologic association between AAT deficiency and increased COVID-19 infection, hospitalization, severe disease, and mortality. Enoxaparin augments AAT inhibition of TMPRSS2 activity and thus may synergize with AAT in inhibiting SARS-CoV-2 entry of cells. Because both AAT and enoxaparin have biological properties that antagonize several pathogenic mechanisms of COVID-19, this pharmaceutical combination is promising for those critically ill from recalcitrant COVID-19. Thus, future translational investigations and clinical trials should be undertaken with combined AAT and enoxaparin in patients who are severely ill with COVID-19.

• While there are effective vaccines against the prevention of severe COVID-19, treatment for patients who are critically ill with COVID-19 remains suboptimal.

• Based on the emerging evidence that AAT inhibits SARS-CoV-2 infection and many of the pathogenic mechanisms of COVID-19, a review of this topic is timely.

• Several large studies demonstrate robust epidemiologic association between AAT deficiency and increased COVID-19 infection, hospitalization, severe disease, and mortality. Enoxaparin, a low molecular mass heparin, augments AAT inhibition of TMPRSS2 activity and thus may synergize with AAT in inhibiting SARS-CoV-2 entry of cells. Because both AAT and enoxaparin have biological properties that antagonize several pathogenic mechanisms of COVID-19, this pharmaceutical combination is promising for those critically ill from recalcitrant COVID-19. Thus, future translational investigations and clinical trials should be undertaken with combined AAT and enoxaparin in patients who are severely ill with COVID-19.

The authors declare that there are no competing interests associated with the manuscript.

This work is funded by Veterans Affairs Merit Award, 1 I01 BX005651-01A2. We are also grateful to Nina C. Khoo, M.D. for her generous and kind support.

AAT

alpha-1-antitrypsin

ACE2

angiotensin converting enzyme 2

COPD

chronic obstructive pulmonary disease

GR

glucocorticoid receptor

GWAS

genome-wide association study

hAEc

human airway epithelial cells

IL-1

interleukin-1

IL-6R

IL-6 receptor

MAF

minor allele frequencies

MMP-12

metalloproteinase-12

PiMM

protease inhibitor MM

TMPRSS2

transmembrane protease 2

TNF

tumor necrosis factor