Autophagy as an emerging target for COVID-19: lessons from an old friend, chloroquine

ABSTRACT

During the last week of December 2019, Wuhan (China) was confronted with the first case of respiratory tract disease 2019 (coronavirus disease 2019, COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Due to the rapid outbreak of the transmission (~3.64 million positive cases and high mortality as of 5 May 2020), the world is looking for immediate and better therapeutic options. Still, much information is not known, including origin of the disease, complete genomic characterization, mechanism of transmission dynamics, extent of spread, possible genetic predisposition, clinical and biological diagnosis, complete details of disease-induced pathogenicity, and possible therapeutic options. Although several known drugs are already under clinical evaluation with many in repositioning strategies, much attention has been paid to the aminoquinoline derivates, chloroquine (CQ) and hydroxychloroquine (HCQ). These molecules are known regulators of endosomes/lysosomes, which are subcellular organelles central to autophagy processes. By elevating the pH of acidic endosomes/lysosomes, CQ/HCQ inhibit the autophagic process. In this short perspective, we discuss the roles of CQ/HCQ in the treatment of COVID-19 patients and propose new ways of possible treatment for SARS-CoV-2 infection based on the molecules that selectivity target autophagy.Abbreviation: ACE2: angiotensin I converting enzyme 2; CoV: coronavirus; CQ: chloroquine; ER: endoplasmic reticulum; HCQ: hydroxychloroquine; MERS-CoV: Middle East respiratory syndrome coronavirus; SARS-CoV: severe acute respiratory syndrome coronavirus; SARS-CoV-2: severe acute respiratory syndrome coronavirus 2

Coronaviridae Orthocoronavirinae Betacoronavirus Sarbecovirus (coronavirus [CoV]) causes infections (mild to severe) in both mammalian and avian species and comes under the category of positive-sense single-stranded RNA viruses [1]. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), also referred to as 2019-novel coronavirus (2019-nCoV), causes coronavirus disease 2019 (COVID-19). Among the CoV family, four viruses (229E, OC43, NL63, and HKU1) and two viruses (severe acute respiratory syndrome coronavirus [SARS-CoV] and Middle East respiratory syndrome-related coronavirus [MERS-CoV] [2,3]) cause the cold, and severe respiratory diseases, respectively. Although the SARS-CoV-2 surface morphology is known (Figure 1), the complex structure and specific modifications are not documented. It is hypothesized that, like other recent coronaviruses (SARS-CoV and MERS-CoV) [2,3], SARS-CoV-2 has spread to humans from wild animals (i.e., it is a zoonotic disease). Bats have been identified as a source for the SARS-CoV (bat-SL-CoVZC45, MG772933.1), which has a very high nucleotide similarity (96%) with SARS-CoV-2 [4]. In addition, pangolins are also suspected for the zoonotic transmission of this virus into humans [5,6]; however, speculation of zoonotic transmission of SARS-CoV-2 requires further scientific validation. SARS-CoV-2 was first isolated from the bronchoalveolar-lavage fluids of Chinese adult patients [7]. Later, a positive case was confirmed in the US with a history of recent travel to China [8]. The clinical symptoms, especially signs of a severe pneumonia-like fever, cough and chest/breathing discomfort, have been observed, primarily 7 d after infection, but with symptoms appearing over a span of 2–14 d. SARS-CoV-2 infection in severe cases leads to pneumonia, pulmonary edema, lung damage, and failure of other vital organs, such as liver, kidney, and heart [9–11]. Patients who are suffering from severe SARS-CoV-2 infection need supportive mechanical ventilator management in intensive care units, with a high risk of other complications resulting from the infection.

Figure 1.

Coronavirus (SARS-CoV-2) structure, replication mechanism and possible actions of CQ/HCQ. SARS-CoV-2 is a positive-sense single-stranded RNA virus. The nuclear material is encapsulated by proteins in the form of a helical nucleocapsid, which is surrounded by a lipid envelope. (A) The virus contains different structural proteins, such as open reading frame polyproteins, spike glycoprotein (S; 1273 amino acids [aa]), an envelope protein (E; 75 aa), membrane protein (M; 222 aa), and nucleocapsid protein (N; 419 aa) [7]. G, glycosylation site. (B) SARS-CoV-2 entry and infection of the lungs. (C) SARS-CoV-2 enters the host cells (preferably lung epithelial) via ACE2 receptors. After fusion with the cell and lysosomal membrane, and following uncoating of the viral nucleocapsid in the cytoplasm, replication of the virus begins. Subsequently, the single-stranded RNA is replicated into complete virions by using the host cell nuclear, ER, and Golgi machinery. Virus release is mediated by exocytosis. Chloroquine inhibits autophagy by increasing the lysosomal pH. By creating a basic pH in the lysosomes, chloroquine inhibits the enzymatic activity of lysosomal enzymes and prevents the fusion of viral-lysosomal membranes. ACE2, angiotensin I converting enzyme 2; CQ, chloroquine; HCQ, hydroxychloroquine; ER, endoplasmic reticulum; HE, hemagglutinin-esterase; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; TMPRSS2, transmembrane serine protease 2

The immune system in severely ill COVID-19 patients is deeply compromised [12]. COVID-19 patients with progressive disease experience a so-called cytokine storm (IL1B/IL-1β, IL2, IL7, CXCL8/IL8, IL9, IL10, IL17A, CSF2/GM-CSF [colony stimulating factor 2], CSF3/G-CSF, IFNG/IFN-γ, TNF/TNF-α, CXCL10/IP10 [C-X-C motif chemokine ligand 10], CCL2/MCP1 [C-C motif chemokine ligand 2], CCL3/MIP1A, CCL4/MIP1B) syndrome [13], which leads to severe lung damage and the development of acute respiratory distress syndrome [10,14–16].

Alarmingly, the whole world is witnessing the COVID-19 pandemic with 3,640,000 positive cases and a > 5% (255,000) death rate as of 5 May 2020. However, the death rate depends on multivariable conditions, such as comorbidity, age, state of the immune system, and others. As with other RNA viruses, the nuclear material of SARS-CoV-2 is covered by a protective lipid layer, which helps the virus to get absorbed by the cells of the ocular, nasal, and buccal mucosa. Virus spike proteins are responsible for viral entry into the host cells. CoVs, including SARS-CoV-2, bind to ACE2 (angiotensin I converting enzyme 2) on the target cell surface, which acts as an entry receptor for the virus [17,18] (Figure 1). It should be noted that ACE2 expression is not restricted to human airway epithelial cells; rather, it is also highly expressed in kidney, heart and small intestine, hence making these organs vulnerable to viral attack [19,20]. Furthermore, SARS-CoV-2 uses serine protease, TMPRSS2 (transmembrane serine protease 2) for the priming of spike (S) protein and fusion of viral and host cellular membranes [21–23]. Unpublished data also suggest that SARS-CoV-2 uses BSG/CD147/EMMPRIN (basigin [OK blood group]) for the invasion of host cells, although precise mechanisms are not known [24]. Previous studies have identified sialic acid-containing glycoproteins and gangliosides in the respiratory tract as additional receptors that could bind to the S protein of human coronavirus [25]. The architecture of this receptor binding site is also conserved for hemagglutinin-fusion-esterase glycoproteins of influenza virus and hemagglutinin-esterase protein of coronavirus.

A growing number of clinical trials based on FDA-approved drugs and biologics are currently underway worldwide [26–28]. A vast literature has already documented the usefulness of such therapeutic strategies including antivirals, receptor-binding domain/RBD-ACE2 blockers, S protein cleavage inhibitors, fusion protein core blockers, host/viral protease inhibitors, selective and nonselective immunosuppressors, monoclonal/polyclonal antibodies that block pro-inflammatory signals, convalescent plasma therapy and others [29–32]. Nonetheless, novel therapies, including antiviral drugs, safer anti-inflammatory compounds and protective vaccines, are eagerly awaited.

CoVs, including SARS-CoV, MERS-CoV, and SARS-CoV-2, depend on lysosomal proteases for the infection of the host. In fact, it has been proposed that trypsin-like proteases in the lysosomes of respiratory tract epithelial cells act as a major determinant for the respiratory tropism of SARS-CoV [33,34]. These lysosomal proteases cleave the surface S proteins of coronaviruses and promote fusion of viral and lysosomal membranes. Recent electron microscopy studies revealed that SARS-CoV-2 particles are found floating extracellularly and are also seen within membrane-bound vesicles in the cytosol (Figure 1) [7]. In addition, virus replication in the host cell starts at the endoplasmic reticulum-Golgi intermediate compartment [35], which is linked to autophagosome biogenesis, where the viral genome has vital interactions with other proteins to assemble a complete virion [35]. Considering these events and their close link with decisive elements of autophagy, it seems logical to focus our attention on the autophagic pathway as a possible option to control the SARS-CoV-2 infection.

Protocols based on chloroquine (CQ) and its less toxic derivative hydroxychloroquine (HCQ) (Figure 2) have gained consideration for use in the treatment of SARS-CoV-2 infection [36,37]. CQ and HCQ are known regulators of endosomes/lysosomes , which are crucial, and absolutely required in autophagy [38]. CQ and HCQ are available, respectively, in unique phosphate and sulfate forms. HCQ is more soluble and less toxic than CQ [39] and both have a long half-life (40–60 d) [40]. The mechanism of CQ function is well-established. Being a diprotic weak base (pKa 8.5), the unprotonated form of CQ easily diffuses across the membranes of cells and organelles. However, once protonated (either mono- or di-protonated), as occurs for example within the endosome or lysosome, CQ loses the capacity to diffuse across a membrane [41] (Figure 2(B)). Once accumulated in lysosomes, CQ interferes with lysosome-mediated autophagy function by increasing the endosomal/lysosomal pH (Figure 2(B)) (reviewed elsewhere [38,40,42,43]). Thus, by inhibiting lysosome-mediated proteolysis, CQ alters the formation of autolysosomes. In addition, several studies have reported that CQ causes inhibition of endosomal toll-like receptors [44–46], the innate receptors that sense pathogens, and initiate signaling events to mount inflammatory responses.

Figure 2.

Chemical structures of (A) chloroquine and hydroxychloroquine and (B) unprotonated and (di)protonated chloroquine (see the description in the text)

CQ and HCQ have long been used to treat malaria (HCQ was first synthesized in 1946). CQ and HCQ exert their anti-malarial activity by inhibiting hemozoin (polymerized crystalline heme) formation in the parasite food vacuoles, which are the main sources for the conversion of host hemoglobin into amino acid residues [47]. Being an acidic environment, the food vacuole is a favorable site for CQ and HCQ entrapment, where they selectively bind to heme and prevent hemozoin formation. If the parasite is not converting the heme to hemozoin, the free heme is available for the oxidation and release of superoxide radicals [48,49]. This oxidative stress damages the food vacuoles and, as a consequence, the parasite. Also, monobasic forms of CQ and HCQ display inhibitory activity on the heme polymerase enzyme, which helps in the polymerization of heme [48].

During the current COVID-19 pandemic, CQ and HCQ have been administered as first-line drugs in several studies [36,37]. Although these clinical trials were done rapidly, in small cohorts of patients, often with no control groups, and were poorly controlled, a few [36] have described some positive effects of CQ and HCQ in SARS-CoV-2-infected patients [26,37,50–55]. A recent initial clinical trial in 100 COVID-19 patients has suggested the efficacy of CQ, which mitigates SARS-CoV-2-induced pneumonia [37]. Another retrospective study involving 568 critically ill COVID-19 patients indicated that patients who received HCQ along with standard antiviral drugs and antibiotics had a significantly lower mortality rate [56]. Furthermore, a multi-center observational study implied the usefulness of CQ (500 mg/once daily) without any significant toxicity for the treatment of COVID-19 [57]. Though other pre-clinical and clinical studies and a meta-analysis did not find clinical benefits of HCQ for COVID-19 [58–60], the short-term use of CQ/HCQ as an adjunct therapy in COVID-19 infection still seems advisable, particularly during early phase, to block virus replication, shedding and community spread. A total of 148 clinical trials (by mid April 2020) have been planned on CQ/HCQ either alone or in combination for the treatment of COVID-19.

CQ/HCQ has been commonly used as a prominent tool to monitor autophagy via lysosomal acidification inhibition. For in vitro experiments, a dose range of 10–100 μM has been used for the inhibition of autophagy in various cells [61]. For in vivo experiments in mice, a dose of 30–70 mg/kg body weight (not more than 3 doses/week) has been used [62]. However, in the treatment of systemic lupus erythematosus and other autoimmune diseases, and depending on the severity of the disease, patients have received 200–400 mg of CQ/HCQ either by a single daily dose or two divided doses [63]. For the treatment of COVID-19, different clinical trials have explored CQ/HCQ at a dose range of 400–600 mg/d (not more than 10 g/14 d schedule) [36,64]. However, in view of well-known secondary effects, CQ/HCQ have to be used under strict medical vigilance. HCQ analogs with less toxicity are under development. Though inconclusive, based on data from the open-label trials that used combination of HCQ and azithromycin in COVID-19 patients (Zithromax) [65–67], randomized clinical trials are underway that combine these two molecules. Of note, azithromycin also acts by preventing lysosomal acidification [68]. It will also be of interest to explore other molecules, including peptides that directly affect endo-lysosomes [38].

Being an autophagy inhibitor, CQ/HCQ might also block the entry (endocytosis), uncoating and exit (exocytosis) process of SARS-CoV-2. During endocytosis and exocytosis, cells produce danger signals that eventually activate the bystander innate immune cells (Figure 1) [52,69]. Moreover, inhibition of autophagy also leads to the reduction of recycled materials, which may serve as a nucleation material for SARS-CoV-2. Nevertheless, these propositions need to be strictly validated in dedicated experiments. It appears that all malaria-infected populations are partly immune to SARS-CoV-2.

In addition to its effect on lysosomal pH, CQ exhibits broad-spectrum antiviral and immunomodulatory properties [70,71], which might work in a synergistic manner to benefit COVID-19 patients. CQ inhibits glycosylation of the ACE2 receptor, which is essential for the SARS-CoV-2 S protein to bind and infect the host cells [72]. Computational studies have further determined that CQ and HCQ can interfere with S protein binding of gangliosides [73]. Of note, inflammatory cytokines are responsible for lung damage in SARS-CoV-2 infection. Interestingly, CQ inhibits inflammatory cytokines at the transcription level, independent of the lysosome [74]. CQ/HCQ inhibits the activation of monocytes, macrophages [75], and T cells, and thereby reduces the release of pro-inflammatory cytokines (IL1B, IL6, TNF, IFNG, and others) [76]. However, a recent in vitro study performed in peripheral blood mononuclear cells has revealed that HCQ has minimal immunomodulatory activity [77]; but it must be noted that this study was performed under non-stimulatory conditions and might not mimic inflammatory conditions observed in SARS-CoV-2 infection or other autoimmune and inflammatory diseases.

The therapeutic use of CQ/HCQ is strictly regulated due to their cardiovascular side effects. HCQ, although less toxic than CQ, still displays some well-known undesirable properties. The main concern is with regard to cardiomyopathy and heart failure, and the development of retinopathy due to HCQ affinity to melanin-containing cells, a secondary effect noticed in patients with lupus disease [78–81]. Therefore, it is recommended that the daily dose of CQ/HCQ in autoimmune patients should not exceed 5 mg/kg/d. Cardiac adverse effects were also noticed in high-dose CQ (12 g for 10 d)-treated severely ill COVID-19 patients or those receiving a combination of HCQ and azithromycin [82,83].

Thus, irrespective of conflicting results in COVID-19 patients [64], many countries are using CQ/HCQ, either alone or in combination, as a first-choice treatment. Based on the current evidence, we speculate that the therapeutic outcome of CQ/HCQ in COVID-19 patients depends on many factors, such as pharmacokinetics (bioavailability) [84], pharmacodynamics, severity of the disease, comorbidity, ethnicity, age and others. Heterogeneity in the treated COVID-19 patients with respect to severity, genetic background, and pre-existing comorbidities are the major confounding factors that contributed to conflicting results on the therapeutic use of CQ/HCQ [85]. In addition, the dose of the drug also matters: many times, CQ with ≥ 500 mg/d (but not more than 600 mg/d) along with azithromycin have shown greater beneficial effects than lower doses [55,86]. This notion is also supported by mathematical modeling [84].

Considering the present situation with COVID-19, efforts must be focused on fast-tracking drug development and using the rich knowledge gained during previous SARS-CoV and MERS-CoV outbreaks. Several FDA- or EMA-approved therapeutic molecules are available for treating patients with COVID-19. Based on the current knowledge with CQ/HCQ, molecules that target autophagy pathways might be useful to treat COVID-19 patients. However, randomized controlled clinical trials are urgently needed to validate their therapeutic benefits and safety either as monotherapy or combination therapy with antiviral molecules. This view is also supported by a recent observational study on the use of HCQ in consecutive hospitalized COVID-19 patients [87]. Although the recent randomized clinical studies are not supportive of HCQ therapy for COVID-19 (both therapeutic and postexposure prophylaxis) [88,89], the other questions such as pre-exposure prophylaxis use of CQ/HCQ [90], the effect of CQ/HCQ on the degree of disease severity with age, virus load and virus transmission remain for future research. Despite differing results on CQ/HCQ, it is apparent that tailor-made changes on the pharmacophores of aminoquinolines would be an advantage for treating other viral diseases. 

Funding Statement

This work was supported by the National Institutes of Health [GM131919]; European Union Horizon H2020 programme [H2020-SC1-2019-874653-INDIGO]; Laboratoire d’Excellence en Recherche sur le Médicament et l’Innovation Thérapeutique (FR) [ANR-10-LABX-0034].

Disclosure statement

The authors declare no conflict of interest. All declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.