1. Introduction
The novel coronavirus 2019, now officially termed as SARS-CoV-2, causes the coronavirus disease 2019 (COVID-19) by infecting the respiratory system [
1]. The disease was first detected and reported in Wuhan, China, in December 2019 [
2], and has now spread to over 150 countries on all continents except Antarctica. According to the World Health Organization (WHO) Coronavirus Disease Dashboard, the global tally of coronavirus cases has approached 13.5 million, with a death toll of over 580,000 [
3] at the time of writing (15 July 2020). The United States alone has counted over 3.4 million infections and over 131,000 deaths, contributing more than a quarter of both overall infections and deaths globally (John Hopkins University Coronavirus Resource Center) [
4]. The WHO characterized the COVID-19 as a pandemic on 11 March 2020 [
5]. The fatality rate of the disease is particularly high among patients who are older and who have underlying health issues, such as cancer, diabetes, and compromised lung function or lung disease.
The United Nations (UN) reported that 35 million people worldwide suffer from substance use disorders (SUDs) [
6]. In the U.S., the number of individuals experiencing SUDs is 20.3 million [
7]. A large number of individuals with SUDs have underlying health conditions, particularly cardiovascular and lung diseases and hepatitis C or HIV-1 infections. Together with complicated socioeconomic issues, these populations are particularly vulnerable to COVID-19 [
8]. Despite the fact that researchers and clinicians around the world have collected and disseminated tremendous amounts of data on COVID-19, we have very little knowledge of the interactions and comorbidity of COVID-19 and SUD. In this review, we will analyze relevant COVID-19 and SUD literatures, and highlight the susceptibilities and vulnerabilities for individuals with SUD.
SARS-CoV-2 is known to attack the respiratory tract and could lead to severe lung damage or pulmonary fibrosis [
2]. Smokers of tobacco and marijuana, and possibly people who vape [
9], are susceptible to chronic obstructive pulmonary disease (COPD) [
10,
11], which could cause severe complications of COVID-19 and lead to a higher fatality rate [
12,
13]. Other substances of abuse, such as opioids and methamphetamine, function increasingly through the brain and immune system to indirectly affect the respiratory system. Opioid use disorder (OUD) tends to slow breath rate and decrease blood oxygen content (hypoxemia). An extended period of hypoxemia is one of the major causes of overdose fatality. COVID-19, which shrinks lung capacity, could heighten the condition caused by opioid overdose [
14,
15]. Methamphetamine may contribute to pulmonary hypertension and edema through cardiomyopathy and restricted blood circulation [
16]. Many substances of abuse also impair the bidirectional interactions between the brain and immune responses, resulting in an increase in the infection rates among individuals with SUDs. Some drugs exert proinflammatory effects in the central nervous system (CNS), leading to neuroinflammation. The buildup of proinflammatory cytokines and chemokines in the CNS may exaggerate the already excessive inflammatory response in the peripheral tissues of COVID-19 patients.
In addition to pathological risks that patients with SUDs are facing, highly risky behaviors can put them into even greater jeopardy in the pandemic. Suicide mortality associated with SUDs is significantly higher compared to the general population across all categories, including age, gender, income, and education, and the relative risk of suicide is more prominent in women. People with multiple alcohol, drug, and tobacco use disorders are at a particularly heightened risk of suicide [
17]. In a position paper, the International Society of Addiction Medicine (ISAM) Practice and Policy Interest Group noted that people with SUDs suffer from serious health complications, including chronical infections, weakened immune systems, various respiratory, cardiovascular, and metabolic disorders, and a range of psychiatric comorbidities. While they are stigmatized and marginalized with limited access to healthcare, the difference in perceived danger and risk-taking behaviors may put people with SUDs at a higher rate of mortality [
18]. Due to the lack of research on SUDs and COVID-19, the group also put together recommendations for health service providers and policymakers regarding the comorbidity of COVID-19 infection and SUDs [
18].
2. SARS-CoV-2 Infection and COVID-19 Pathogenesis
Coronaviruses are named for their crown-like spikes protruding from the surface of the virion, and can be classified into four genera, known as α, β, γ, and δ. There are seven known coronaviruses that can infect humans. Four of them, α-coronaviruses 2229E and NL63, and β-coronaviruses OC63 and HKU1, infect people on a regular basis and cause common cold symptoms. The other three, all belonging to the β-coronavirus subfamily, are believed to have originated from bats and evolved through jumping animal species to become new human coronaviruses [
19,
20]. These include the SARS-CoV, the coronavirus that caused severe acute respiratory syndrome (SARS) outbreak in 2003, the MERS-CoV, the coronavirus that caused Middle East Respiratory Syndrome (MERS) outbreak in 2012, and the SARS-CoV-2, the novel coronavirus that is causing the current COVID-19 pandemic railing the whole world.
Like all other coronaviruses, SARS-CoV-2 contains a single-strand (ss) RNA genome. The sequence of the 29,903 nucleotides (nt) long SARS-CoV-2 genome was first reported by Chinese scientists and made publicly available on GenBank with accession number MN908947 [
21]. The genome organization of SARS-CoV-2 is similar to that of other representative β-coronaviruses. As illustrated in
Figure 1a, it is comprised of a 5′-untranslated region (UTR), and at least ten open reading frames (ORFs) that encode non-structural and structural viral proteins, followed by the polyA tail at the 3′-end. The first ORF is the 21,291-nt replicase gene,
ORF1ab, encoding 16 non-structural proteins (NSP1-16). The subsequent ORFs encode four structural proteins, spike (S), envelope (E), membrane (M), and nucleocapsid (N), as well as several accessory proteins that do not participate in viral replication and transcription. Structural proteins are important in maintaining viral structural and genomic stability. An illustration of SARS-CoV-2 viral structure is shown in
Figure 1b. The S protein is responsible for infection and transmission. Non-structural proteins, such as the RNA-dependent RNA polymerase (NSP12), have functions in viral genome transcription and replication. The SARS-CoV-2 genome was found to have 79.6% sequence similarity to SARS-CoV, and shares 96% identity at the whole-genome level to a bat coronavirus (Bat CoV RaTG13) detected in
Rhinolophus affinis [
20].
Similar to SARS-CoV, SARS-CoV-2 recognizes the angiotensin converting enzyme 2 (ACE2) receptor by its S protein and utilizes it for cell entry [
20,
22]. The heavily glycosylated S protein triggers virus cell entry by fusing the receptor binding domain (RBD) on the S1 subunit to the host ACE2 receptor, engaging the transition of S2 subunit to a stable post-fusion conformation [
23]. Cryo-electron microscopy (EM) structures of the pre-fusion [
23] and post-fusion structures [
24] of the S protein have been reported. The SARS-CoV-2 S protein has been shown to have a much higher binding affinity to the ACE2 than the SARS-CoV S protein [
23,
25]. The S protein contains 22 N-linked glycans, and the complex glycosylation is likely to play a role in shielding and camouflaging for immune evasion of the virus [
26,
27]. The S protein is activated by type II transmembrane serine protease (TMPRSS2), a host protease co-expressed with ACE2 on the cell surface [
24,
28]. In cells not expressing TMPRSS2, other proteases, such as cathepsin B/L, may activate the S protein and facilitate viral entry [
29].
Upon cell entry, SARS-CoV-2 has a similar life cycle and pathogenesis as other β-coronaviruses, including SARS-CoV and MERS-CoV [
30]. Upon ACE2 receptor binding, the virus fuses its membrane with the host cell plasma membrane, releasing its genomic RNA into the cytoplasm. Since the viral RNA is similar to the human messenger RNA (mRNA), it triggers the host ribosome to start translating the viral RNA and producing viral proteins. The viral replicase ORF is translated into two overlapping polyproteins, PP1a (NSP1-11) and PP1ab (NSP1-16), which require extensive processing. NSP5, the 33.8-kDa main viral protease (M
pro), also referred to as the 3-chymotrypsin-like protease (3CL
pro), performs the function by autolytic cleavage of the protease itself, and then subsequently digests the polyproteins into 16 non-structural proteins. NSP12, known as the RNA-dependent RNA polymerase (RdRp), together with NSP7 and NSP8, carries out the critical process of the viral RNA synthesis, and is central to the viral replication and transcription cycle. The N-terminal non-structural protein, NSP1, has been shown to bind to the 40S small ribosomal subunit, shutting down all host cell protein production by blocking the mRNA entry tunnel. NSP1 binding to ribosomes and blocking host cell translation effectively inhibits type-I interferon (IFN-I)-induced innate immune response by turning off the retinoic acid-inducible gene (RIG)-I antiviral sensor [
31]. The inhibition of the IFN-I-induced innate immunity allows the assembly of viral particles inside the host cell. The newly produced structural proteins, S, M, and E, are inserted into the endoplasmic reticulum (ER) or Golgi membrane, while the N protein associates with the newly synthesized viral RNA to stabilize the genome. The viral particles are assembled into the ER-Golgi intermediate compartment (ERGIC), fuse with the plasma membrane, and bud off the host cell. The released virions will further infect more cells. The functions of other NSPs are not fully understood. A comparative structural genomics study revealed a possible functional intra-viral and human-virus interaction network of NSPs [
32]. Recurrent mutations in the SARS-CoV-2 genome have been identified in some NSPs and the S protein, suggesting ongoing adaptations of the coronavirus through transmission [
33]. Particularly, the D614G mutation in the S protein makes it more stable, and the virus becomes more infectious and transmissible [
34,
35]. This mutated virus is the dominant form in Europe and North America since March 2020 [
36].
3. Vulnerability of Substance Use Disorders (SUDs) in COVID-19
Underlying medical conditions can put individuals at increased risk for severe illness from COVID-19. The comorbid conditions include COPD, cardiovascular diseases, other chronical respiratory diseases, diabetes, obesity, and cancer. According to a large-scale study with 72,314 cases conducted by the Chinese CDC, case-fatality and mortality rates are significantly increased in patients with comorbid conditions comparing to those with no underlying conditions (
Table 1) [
12]. In a study in New York City, the epicenter of the COVID-19 pandemic in the U.S., comorbid conditions are highly associated with hospitalization and severity of the illness (
Table 2) [
37].
A nationwide case-control study in Korea also confirmed that diabetes, hypertension, and chronic respiratory disease, among others, were associated with severe COVID-19 [
38]. Individuals with SUDs commonly experience respiratory and cardiovascular disorders, including hypertension and COPD, and have undermined immune systems, making them particularly vulnerable in COVID-19. A significant portion of individuals with SUDs have underlying medical conditions, and are more likely to be marginalized. According to a recent study in British Columbia, Canada, among 19,005 individuals who had one or more non-fatal overdose events between 2015 and 2017, 10,649 (56.0%) had a record of receiving social assistance, and 5716 (30.0%) had no fixed address record. These individuals with a history of overdose are more likely to have at least three known chronical conditions associated with COVID-19 severity, including chronical pulmonary disease, diabetes, and coronary heart disease, with adjusted odds ratios (ORs) to be 2.01, 1.24 and 2.08, respectively, with reference to people without an overdose [
39]. During the COVID-19 pandemic, risks of abusing substances and addictive behaviors are also increasing. The stress and social isolation associated with the response to COVID-19 increases the risk of alcohol abuse and misuse, which is known to suppress immune systems and cause emotional dysregulation [
40]. A study in China showed that relapses of alcohol and smoking abuse were prominent (18.7% and 25.3%, respectively), and 32.1% of regular drinkers and 19.7% of regular smokers increased alcohol and cigarette consumption [
41].
SARS-CoV-2 can attack and damage human organs through two major events: direct viral attacks against target organs and abnormal immune responses and inflammation [
42]. Initial evidence focuses on the damage to the respiratory system and the lung, and is correlated with clinical symptoms of the patients [
2,
12,
20,
21]. The identified viral entry receptor, ACE2, is abundantly expressed in the epithelial cells along the respiratory tract and the lung alveoli [
29,
43,
44]. High level expression of ACE2 receptors is also reported in organs and tissues outside the respiratory system, including the heart, kidney, and intestine [
45]. Therefore, these organs are the potential targets of and could be damaged by SARS-CoV-2. As mentioned earlier, smoking of tobacco and marijuana directly impairs respiratory system. Other substances of abuse can cause cardiovascular diseases, which amplify respiratory and pulmonary complications. We will discuss these complications in the next section.
More severely, ACE2 is abundantly expressed in vascular endothelial cells [
45]. Several clinical cases have been reported to indicate direct involvement of vascular endothelial cells in COVID-19 pathology at different organs, suggesting that the damages to the lung, heart, kidney, liver, small intestine, and bowel, are actually caused by endotheliitis (endothelialitis). Direct viral infection of endothelial cells induces inflammation and inflammatory cell death at the endothelium (
Figure 2) [
46]. Comparing lung tissues from deceased patients of acute respiratory distress syndrome (ARDS) associated with influenza and COVID-19, the lungs from COVID-19 patients displayed distinctive vascular impairments of the pulmonary vessels. Most significantly, viruses were found inside the endothelial cells of the lung tissues from COVID-19 patients, which disrupted cell membrane, caused prevalent thrombosis with microangiopathy, and induced elevated intussusceptive angiogenesis [
47]. A greater number of ACE2+ endothelial cells were found in COVID-19 patients, correlating to changes in endothelial morphology, including disruption of endothelial cell junctions, cell swelling, and detachment from the basal membrane. The presence of the SARS-CoV-2 virus inside the endothelial cells, together with the induced inflammation, may directly contribute to the endothelial injury [
46,
47]. Although there are no reported cases yet, one potential target of SARS-CoV-2 infection is the brain microvascular endothelial cell (BMVEC). BMVECs line up the microcapillary beds and form the blood-brain barrier (BBB) together with brain astrocytes and pericytes. The BBB prevents pathogens and toxins from trespassing into the brain side. Tight-junction (TJ) protein complexes, composed of occludin, claudins, junctional adhesion molecules (JAMs) and membrane-directed scaffolding protein zonulae occludentes (ZO), form a physical inter-endothelial barrier that strictly controls migration of molecular and cellular contents from the circulation into the brain (
Figure 2) [
48,
49]. High expression of efflux pumps and stereospecific solute transporters at the endothelium additionally limits molecules from crossing the barrier [
50,
51,
52]. The BBB plays an essential role in protecting the brain from pathogen invasion. Viral infection of BMVECs could result in endothelial dysfunction, leakage, and even rupture, and is detrimental to the BBB integrity. The damaged BBB allows the virus to migrate into the brain side, and infect neuronal tissues [
53]. Another possible route of CNS invasion could be through invading the peripheral nerve terminals and then entering the CNS via trans-synaptic transfer [
54]. The first case of meningitis associated with SARS-CoV-2 has been reported, in which viral RNA was detected in the cerebrospinal fluid (CSF) of the patient [
55]. Clinical evidence from Wuhan, China showed that more than 1/3 of COVID-19 patients had neurological symptoms, and CNS involvement was linked to the prognosis and severity of the disease [
56,
57]. Substances of abuse have been shown to severely compromise the endothelial barrier at the BBB, leading to increased BBB permeability and possibly intensified brain damage in COVID-19 (
Figure 2).
The other aspect of COVID-19 pathology involves abnormal immune responses, which could exaggerate into a cytokine storm [
58,
59]. SARS-CoV-2 dramatically promotes host cell kinase activities, including casein kinase II (CK2) and p38, and stimulates production of diverse cytokines [
60]. Evidence has shown that COVID-19 elevates proinflammatory cytokines and chemokines, including tumor necrosis factor (TNF)-α, interleukin (IL)-1β and IL-6, granulocyte-colony stimulating factor (G-CSF), interferon γ (IFN-γ)-induced protein-10 (IP-10), monocyte chemoattractant protein-1 (MCP-1), and macrophage inflammatory proteins-1α (MIP-1α) [
61,
62,
63]. Although there has been no reported evidence, it is possible that pattern recognition receptors, such as toll-like receptors (TLR3, TLR7, and TLR8), RIG-I, and nucleotide-binding oligomerization domain (NOD)-like receptors (NLRP1, NLRP3, and NLRP12), are also activated by COVID-19 through innate immunity [
64]. It has been reported that IL-6 was significantly increased in severe COVID-19 cases, and its level was closely correlated with the severity of patients [
65]. Human bronchial epithelial cells infected with SARS-CoV-2 showed elevated expression of type I and type III IFNs and IL-6 [
44]. Furthermore, type I interferon, IFN-α, stimulates the expression of ACE2, the molecular target of SARS-CoV-2, in primary human nasal epithelial cells [
43]. Type III interferon, IFN-λ, has been shown to disrupt the lung epithelial barrier by direct inhibition of lung epithelial proliferation and repair, contributing to COVID-19 pathogenesis in the lower airways [
66]. The upregulation of IL-6 and other proinflammatory cytokines was also observed in SARS cases [
67] and influenza infection [
68]. Substances of abuse can induce high level of expression of proinflammatory cytokines and chemokines in the CNS and cause neuroinflammation, which can worsen inflammatory responses in COVID-19. Details will be discussed in the next section.
The bidirectional communication between the brain and the immune system plays a critical role in COVID-19 pathogenesis. It has been well established that brain-immune interactions are widespread and significant. For instance, the immune system produces hormones and neurotransmitters [
69,
70,
71], while the anterior pituitary cells in CNS produce proinflammatory cytokines, such as IL-6 [
72]. Microglial cells are immune effectors in the CNS, which produce and secrete cytokines and neurotrophic or neuron survival factors upon inflammation and injury [
73]. For pathogen infections, the innate immunity provides the first line of defense through recognition of pathogen-associated molecular patterns (PAMPs), initiating nonspecific cellular and humoral responses and rapidly activating nonspecific neural responses, including systemic hormonal responses through the hypothalamic-pituitary-adrenal (HPA) axis (
Figure 3) [
74]. Consisting of the hypothalamus of the brain, and endocrine organs, the pituitary and cortex of the adrenal glands, the HPA axis is responsible for systematic inflammation control. The paraventricular nucleus (PVN) of the hypothalamus plays the main governing role of the HPA axis, releasing a wide range of neuropeptides, including the corticotrophin-releasing hormone (CRH) and arginin-vasopressin (AVP). These neuropeptides reach the anterior pituitary (AP) to activate corticotrope cells to secrete adrenocorticotropic hormone (ACTH). ACTH subsequently enables the synthesis and secretion of glucocorticoids in the zona fasciculata of the adrenal cortex through melanocortin type 2 receptors [
75]. The physiological feedback loop involves releasing immune mediators and cytokines, such as IL-1, IL-6, and tumor-necrosis factors (TNFs), by the innate immune system to activate neural responses, which amplify local inflammation to contain and eliminate pathogen invasions. Upon pathogen clearance, the brain responds by activating the HPA axis and releasing anti-inflammatory molecules, glucocorticoids, from the adrenal cortex. The release of this final product of the HPA axis sends a signal to the immune system to terminate the inflammatory responses, completing the hormonal negative-feedback loop. Glucocorticoids also negatively regulate the HPA axis itself, restoring host homeostasis, including CNS and cardiovascular system, as well as metabolic balances. The interplay between the nervous system and the immune system plays a critical role in forming a cohesive and integrated early host response for pathogen clearance through an optimized innate inflammatory response. Impairment of the HPA axis by various substances of abuse could render the host highly susceptible to inflammation and even increased mortality from septic shock from exposure to infectious and proinflammatory triggers, including COVID-19. Inappropriate and excessive CNS responses could predispose the host to extreme inflammation, including cytokine storm that has been observed clinically in influenza [
68,
76], SARS [
77], and COVID-19 [
59]. Excessive activation of the HPA axis and the release of glucocorticoids by several substances of abuse will suppress the activities of various immune cells, including macrophages, dendritic cells, and T cells, and will inhibit activities of NK cells, B cells, and T cells (
Figure 3) [
74]. The immunosuppression reduces antibody production, cytotoxicity, and T cell-mediated immune responses, and is linked to higher incidences of pathogen infections, slowed recovery, and severe disease progression in COVID-19.