Glutamine is a neutral, conditionally essential amino acid that is indispensable during periods of rapid growth or in pathological states such as trauma, sepsis, and infection ^[1][2][3][4][12,13,14,15]. It is the most abundant amino acid in the blood and is used as a nitrogen and carbon source to synthesize amino acids, lipids, and nucleic acids. Glutamine enters cells via multiple membrane transporters that make up the solute carrier (SLC) superfamily. At least fourteen glutamine transporters have been identified at the molecular level that belong to four distinct families: SLC1, SLC6, SLC7, and SLC38 ^[5][6][7][24,25,26]. These glutamine transporters often share specificity with other neutral or cationic amino acids and have distinct transport modes. Some glutamine transporters are obligatory exchangers, while others function as active transporters in one direction. While most glutamine transporters import glutamine, some mediate the efflux of glutamine, which is coupled to the influx of other amino acids allowing for regulation of the intracellular pools of glutamine and amino acids. Once transported into the cell glutamine is used for the biosynthesis of glucosamines, nucleotides, and asparagine and may activate mammalian target of rapamycin complex 1 (mTORC1) in the cytoplasm. However, a majority of glutamine is metabolized to glutamate and ammonia (NH₃) by the mitochondrial enzyme glutaminase (GLS) ^[8][27]. There are two isoforms of GLS, GLS1 and GLS2, but GLS1 is preferentially expressed by immune and vascular cells ^[9][10][11][28,29,30]. Interestingly, a recent report found that GLS1 also triggers mitochondrial fusion in a non-enzymatic manner, demonstrating that this enzyme regulates both mitochondrial metabolism and dynamics ^[12][31]. The GLS1 product NH₃ induces the expression of heme oxygenase-1 (HO-1) and stimulates autophagy, which collectively allow cells to survive harmful stimuli, including inflammatory stress ^[13][14][15][32,33,34]. Furthermore, glutamate is exported from the mitochondria to the cytosol, where it is utilized to generate the tripeptide (glutamate, glycine, cysteine) antioxidant glutathione (GSH) and various non-essential amino acids such as alanine, aspartate, proline, ornithine, serine, cysteine, glycine, and arginine via the concerted action of several enzymes. While aspartate is utilized to generate nucleotides, arginine is metabolized to the critical signaling gas nitric oxide (NO) by NO synthase (NOS), see ^[16][22]. The release of NO by ECs plays a key role in maintaining vascular homeostasis. NO regulates blood flow and pressure by blocking arterial tone. It also exerts a potent antithrombotic effect by inhibiting platelet adhesion and aggregation. In addition, NO prevents neointima thickening and vascular occlusion by retarding SMC proliferation, migration, and extracellular matrix deposition. Moreover, NO suppresses inflammation by inhibiting the expression of adhesion receptors, the synthesis of proinflammatory cytokines and chemokines, and the recruitment, infiltration, and activation of leukocytes within blood vessels. Cytosolic glutamate can also be converted back to glutamine by glutamine synthetase (GS) at the expense of NH₃- and adenosine triphosphate (ATP) ^[17][35]. Alternatively, mitochondrial glutamate is converted into alpha-ketoglutarate (αKG) by glutamate dehydrogenase 1 (GLUD1) or by the aminotransferase glutamic-pyruvic transaminase 2 (GPT2) and glutamic-oxaloacetic acid transaminase 2 (GOT2). Mitochondrial αKG participates in the tricarboxylic acid (TCA) cycle, supporting oxidative phosphorylation and the generation of electron donors (nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FADH₂)) and ATP. However, under hypoxic conditions, αKG supports the reductive carboxylation pathway yielding citrate, which is used for fatty acid synthesis. In addition, αKG is transported to the cytosol, where it activates -mTORC1, promotes collagen synthesis via the activation of prolyl-4-hydroxlase, and serves as an important cofactor for nuclear enzymes involved in the epigenetic modification of histones and DNA. Thus, by replenishing TCA intermediates, glutamine works as an anaplerotic substrate, a privileged role that the amino acid plays in several types of normal and neoplastic cells.

2. Vascular Disease in COVID-19

COVID-19 is accompanied by a significant risk of ischemia-related vascular disease. Studies during the early phase of the pandemic found that the coordinated activation of inflammatory and thrombotic responses, thrombo-inflammation, is a primary cause of morbidity and mortality in patients with COVID-19 ^[18][98]. Indeed, laboratory studies detect the presence of a procoagulant state with elevated levels of circulating D-dimer and fibrinogen, a mild prolongation of prothrombin time in the plasma, and minimal thrombocytopenia in many patients hospitalized with COVID-19 ^{[19][20][21][22]}[99,100,101,102]. Moreover, cross-sectional studies demonstrate that the rate of thromboembolic events such as thrombosis and pulmonary embolism are exceedingly high in critically ill patients with COVID-19 ^[23][24][25][103,104,105]. Post-mortem examinations of lung tissue in critically ill patients reveal a high frequency of platelet-fibrin thrombi in the small arteries and capillaries, suggesting a hypercoagulable condition that precipitates both venous and arterial thrombosis ^{[25][26][27][28][29]}[105,106,107,108,109]. The elevated rate of arterial thrombosis in COVID-19 possibly plays a role in the higher frequency of myocardial infarction, ischemic stroke, and acute limb ischemia in this patient cohort ^{[24][30][31][32]}[104,110,111,112]. Autopsy findings also indicate a high prevalence of alveolar microthrombi in patients with COVID-19. These occluding microthrombi extend beyond the lungs to the heart, kidney, and liver, suggesting the widespread presence of thrombotic microangiopathy in COVID-19 ^[33][113]. The etiology of thrombosis in patients with COVID-19 is multifactorial involving complement activation and cytokine release, coagulation abnormalities, platelet hyperactivity and apoptosis, the formation of neutrophil extracellular traps, and endothelial dysfunction ^[33][34][5,113]. However, endothelial malfunction is a central feature in COVID-19 thrombotic complications. In particular, the endothelium undergoes a prothrombotic transformation involving the loss of the glycocalyx and cytoprotective signaling, and the generation of thrombotic effectors to promote fibrin formation, platelet adhesion, and complement activation ^[35][115]. Clinical signs of endothelial inflammation and death in COVID-19 are common and found in multiple organs ^[18][36][37][98,116,117]. Autopsy studies on lung specimens from SARS-CoV-2-infected patients detected severe EC damage with evidence of apoptosis and loss of tight junctions ^[38][118]. In addition, biomarkers of endothelial dysfunction are elevated in patients with COVID-19. Plasma von Willebrand factor (vWF) is increased in COVID-19 patients with high concentrations associated with severe disease ^[39][40][41][119,120,121]. Moreover, circulating levels of P-selectin, E-selectin, soluble intercellular adhesion molecule-1 (ICAM-1), thrombomodulin, angiopoietin-2, and plasminogen activator inhibitor-1 (PAI-1) are augmented in COVID-19 ^[18][42][43][98,122,123]. The release of vWF, selectins, and ICAM-1 following EC activation binds platelets, neutrophils, and monocytes to initiate thrombosis, while the shedding of thrombomodulin by pro-inflammatory cytokines would further promote the procoagulant and inflammatory milieu within the vasculature of COVID-19 patients ^[18][98]. In addition, the release of PAI-1 by ECs would reduce fibrinolysis by inhibiting the tissue plasminogen activator, thereby stimulating fibrin deposition in the microvasculature. Circulating ECs, which reflect cells that are detached from the damaged blood vessel, are also higher in critically ill patients and may contribute to the enhanced vascular permeability in patients with COVID-19 ^[44][124]. A recent meta-analysis disclosed that several biomarkers of endothelial dysfunction are significantly associated with increased composite poor outcomes in COVID-19 patients ^[45][125]. COVID-19 also negatively impacts vascular tone. Endothelium-dependent microvascular reactivity in the skin is severely restricted in critically ill COVID-19 patients ^[46][126]. In addition, systemic endothelium-dependent vasodilator responses are markedly decreased in patients with severe or mild to moderate COVID-19 relative to healthy individuals, and this is paralleled by an increase in circulating inflammatory cytokines and chemokines ^[47][127]. Moreover, endothelium-dependent vasodilation remains impaired shortly after acute COVID-19 but improves following long-term recovery from the disease ^[48][49][128,129]. Together, these studies suggest that NO bioavailability is compromised in COVID-19 patients. This is consistent with reports showing a decrease in serum NO metabolites in COVID-19 patients, most likely due to the consumption of NO by reactive oxygen species ^[50][51][130,131]. In addition, the degradation of the endothelial glycocalyx in COVID-19 may also contribute to the decline in NO synthesis since it is critically involved in flow-mediated NO production ^[52][53][54][132,133,134]. The reduction in endothelial NO may further aggravate ischemic injury in COVID-19 by promoting blood vessel spasm and eliminating a key brake on platelet activation and aggregation. Moreover, the loss of NO-mediated vasodilation is further amplified by an angiotensin converting enzyme (ACE)/ACE2 imbalance in COVID-19 that favors angiotensin II-mediated vasoconstriction ^[55][135]. In fact, spontaneous and severe coronary vasospasm has been reported in patients with COVID-19 ^[56][57][58][136,137,138].

3. Glutamine Deficiency in COVID-19

Metabolomic studies have consistently identified a decline in circulating levels of glutamine in multiple patient populations with COVID-19 that is correlated with disease severity ^{[59][60][61][62][63][64][65][66][67][68]}[23,139,140,141,142,143,144,145,146,147]. COVID-19 patients have a significantly reduced glutamine to glutamate ratio, indicating the increased utilization of glutamine. Consistent with this notion, reductions in glutamine are accompanied by an increase in plasma GLS activity ^[69][70][71][148,149,150]. Furthermore, the decrease in glutamine concentration in severe COVID-19 patients is positively correlated with lactate dehydrogenase, C-reactive protein, and pCO₂ and positively correlated with pO₂, disclosing formerly unknown consequences of low glutamine in severe COVID-19 pathologies ^[72][151]. Moreover, meta-analysis showed that elevated glutamine is related to a decreased risk of COVID-19 infection and severe COVID-19, whereas raised glutamate levels are associated with increased risk of infection and serious disease ^[73][152]. Intriguingly, an advanced bioinformatic platform reported that glutamine was the top candidate of over 26,000 FDA-approved drugs tested for reversing coronavirus associated alterations in gene expression ^[74][153].

4. Strategies Targeting Glutamine in COVID-19

There is an emerging realization that amino acids play an essential role in preserving immune function and vascular health ^{[16][75][76][77][78][79][80]}[21,22,154,155,156,157,158]. Considerable evidence indicates that the bioavailability of glutamine is critically depleted in COVID-19 patients, perhaps secondary to a rise in GLS activity that directs the conversion of glutamine to glutamate. The resulting decline in glutamine will compromise EC function and NO production by limiting cellular arginine levels. The loss of glutamine also stimulates vascular inflammation, vascular cell thrombogenicity, platelet activation and aggregation, and reduces the viability of ECs by restricting NO synthesis and the induction of HO-1. In addition, glutamine deficiency causes a broad dysfunction of the immune system that worsens the gravity of COVID-19. The consumption of glutamine in COVID-19 will also increase oxidative stress by reducing the synthesis of GSH and the expression of HO-1. Collectively, these actions related to glutamine insufficiency will promote vasospasm, thrombosis, and NETosis, resulting in vascular occlusion and organ failure. Crucially, comorbidity-associated glutamine deficiency is a predisposition to severe COVID-19 ^[81][159]. Multiple approaches may be used to mitigate the loss of glutamine in COVID-19. Dietary supplementation of glutamine has proven effective in correcting deficiencies in circulating glutamine in cardiometabolic disease and hemolytic disorders and is now prescribed in patients with sickle cell disease ^[82][83][84][160,161,162]. However, glutamine generates ATP and precursors for the synthesis of macromolecules required for virus assembly, and restoration of this amino acid may augment SARS-CoV-2 replication and infection. In fact, host cell glutamine metabolism has been identified as a potential treatment against COVID-19 ^[85][86][163,164]. Nevertheless, preliminary clinical studies support the use of glutamine in COVID-19 patients. An initial small, retrospective, cross-sectional study found that the oral ingestion of glutamine powder (30 g/day) with meals shortens hospital stays and lessens the need for intensive care in patients with COVID-19 ^[87][165]. In addition, a single-blind randomized clinical trial reported that intravenous glutamine administration (0.4 g/kg/day) lowers the inflammatory response in COVID-19 patients admitted to the intensive care unit but does not improve short-term mortality in this patient cohort ^[88][166]. More recently, a larger case–control study noted that the short-term meal-time consumption of glutamine (30 g/day) reduces serum markers of inflammatory and oxidative stress and increases appetite in hospitalized COVID-19 patients ^[89][167]. A major limitation of current studies is that it is not known whether glutamine administration elevates glutamine availability in COVID-19 patients. Due to the extensive metabolism of glutamine by the splanchnic circulation, high doses of glutamine may be needed to fully restore plasma glutamine in these patients ^[90][168]. While glutamine is usually administered using the free form, the use of more stable dipeptide formulations, including L-glycyl-L-glutamine, L-arginyl-L-glutamine, and L-alanyl-L-glutamine, should be considered since they possess a more favorable pharmacokinetic profile ^[90][91][168,169]. The use of L-arginyl-L-glutamine in COVID-19 is especially appealing since deficits in both glutamine and arginine have been detected in this disease ^[59][92][93][23,170,171]. While a plethora of studies have confirmed the safety of dietary glutamine supplementation, care should be implemented when using this amino acid in critically ill patients ^[94][95][172,173]. Clearly, larger, multi-center dose-escalation studies employing various preparations of glutamine are needed to clarify the safety, utility, and dosing requirements for this amino acid in patients with COVID-19. The prophylactic use of glutamine in individuals with a high risk for poor outcomes following SARS-CoV-2 infection should also be considered. The administration of specific metabolites of glutamine with known beneficial actions should also be considered. In this respect, the thoughtful delivery of NH₃ may be advantageous given its ability preserve to EC viability and stimulate HO-1 expression. In addition, NH₃ and its chloride salt inhibits the replication of several viruses, including the influenza virus, reovirus, papilloma virus, and the infectious pancreatic necrosis virus ^{[96][97][98][99]}[174,175,176,177], and may be efficacious against the SARS-CoV-2 virus ^[100][178]. Aside from NH₃, the use of another gas such as NO holds promise in managing COVID-19 ^[101][102][179,180]. Both inhaled NO and NO donors have a wide range of antiviral activity and serve as a first line of defense against foreign invaders. In addition, NO plays a critical role in maintaining vascular integrity and endothelial function, controlling the response of immune cells and platelets, and has pulmonary protective properties. Furthermore, the glutamine metabolite αKG may be of benefit to patients with COVID-19. αKG has immuno-augmenting properties and a recent report found that ingestion of αKG inhibits SARS-CoV-2 replication and reduces inflammation, thrombosis, and apoptotic cell death in the lungs of in infected animals ^[103][181]. Moreover, αKG restores EC function in cells deprived of glutamine, and oral administration of αKG enhances NO synthesis by ECs in diet-induced obese rodents ^{[7][104][105][106]}[26,84,85,182]. Thus, the delivery of NH₃, NO, and αKG provides an attractive means for combatting COVID-19. Alternatively, downstream effectors of glutamine may be employed to treat COVID-19. In this respect, HO-1 represents an attractive target for COVID-19 since it is strongly induced by glutamine and possesses potent antiviral and vasoprotective properties ^{[13][107][108][109]}[32,183,184,185]. HO-1 or its reaction products CO and biliverdin/bilirubin suppress many viral infections, including influenza A, hepatitis B, human immunodeficiency virus, Ebola, dengue, and Zika, and there is a strong possibility that they may control SARS-CoV-2 infections by restoring IFNγ production and/or blocking viral proteases and RNA polymerases ^{[110][111][112]}[186,187,188]. In addition, HO-1 and its end products confer anti-thrombotic effects by lessening endothelial injury, the expressing of adhesion receptors, and inflammatory responses, diminishing procoagulant molecules, such as vWF, PAI-1, and tissue factors ^{[113][114][115][116][117][118][119][120][121][122]}[93,94,95,96,97,189,190,191,192,193]. Interestingly, an increase in HO-1 expression is observed in critically ill COVID-19 patients, and this may serve as an adaptive mechanism to counteract increased heme levels driving coagulation and thrombosis in these patients ^[123][124][194,195]. Several approaches may be taken to target HO-1 in COVID-19. Numerous inducers of HO-1 have been identified and shown to be effective in preclinical studies. Heme and its derivates are potent inducers of HO-1 that elicit beneficial effects in animal models of infection, inflammation, and cardiovascular disease ^{[110][113][114][115][116][117]}[93,94,95,96,97,186]. However, the induction of HO-1 by heme analogues in patients is transient and does not persist for more than one week ^[125][196]. Moreover, heme is a danger-associated molecular pattern molecule that may exacerbate inflammation in SARS-CoV-2-infected patients ^[126][197]. More promising, the synthetic triterpenoid bardoxolone methyl is a strong inducer of HO-1 that has generated some positive results in patients with kidney disease ^{[127][128][129]}[198,199,200]. In addition, dimethylfumarate is another activator of HO-1 that is approved for use in multiple sclerosis and, importantly, shows efficacy in an animal model of vascular occlusion ^[130][201]. The direct application of HO-1-derived products affords another avenue for ameliorating COVID-19-mediated complications. While the inhalation of CO has been successfully employed in a myriad of preclinical studies ^{[113][114][115][116][131]}[93,94,95,96,204], its translation to the clinic remains problematic, owing to safety concerns and suboptimal dosing regimens ^[132][205]. The difficulties associated with CO inhalation protocols led to the development and use of CO-releasing molecules that liberate controlled amounts of CO in response in specific stimuli ^[133][134][206,207]. In addition, the use of organic-based click-and-release prodrugs that employ a chemical reaction to generate CO provides another mode for the application of this gas ^[135][208]. Furthermore, saturated solutions of CO impart a simple vehicle for gas delivery that has been demonstrated to prevent vascular occlusion in injured rat arteries and in a murine model of sickle cell disease ^[136][137][209,210]. In a similar manner, saturated solutions of biliverdin may also be used to attenuate COVID-19-related issues ^{[138][139][140]}[211,212,213]. However, the limited solubility and stability of bilirubin in aqueous solutions hampers its direct administration. Instead, a pegylated form of bilirubin that self-assembles into hydrophilic nanoparticles may be used, as this formulation was shown to be non-toxic and effective in a mouse model of ulcerative colitis ^[141][214]. One potential drawback with the use of bile pigments is their rapid metabolism by the liver which necessitates frequent dosing. However, the rise in circulating bilirubin following its administration may be extended by blocking its metabolism by uridine diphosphate-glucuronosyltransferase 1A1 (UGTA1A). Indeed, atazanavir and canagliflozin are dual HO-1 inducers and UGTA1A inhibitors that raise plasma bilirubin concentrations and improve vascular function in diabetic patients, offering proof of principle for this strategy ^{[120][142][143][144][145][146]}[191,202,215,216,217,218].