GD is a classic example of a lysosomal storage disease and has a worldwide incidence of approximately 1/40,000 to 1/60,000 [158]. GD is classified into Types 1, 2, and 3, which are all caused by pathogenic variants in GBA1 (in human)/Gba1 (in mouse) that lead to the functional disruption of the encoded lysosomal enzyme, acid β-glucosidase (β-D-glucosyl-N-acylsphingosine glucohydrolase, EC 4.2.1.25; GCase) and to excess tissue accumulation of glucosylceramides (GC) [23]. Type 1 GD mainly affects the liver, spleen, lung, bone, and kidney [77,159]. A significant proportion of patients with Type 1 GD also have CNS manifestations characterized by the mild brain inflammation [66,67,160,161,162]. Type 2 and Type 3 GD are the neuronopathic forms of GD (nGD), which are characterized by severe and chronic brain inflammation that leads to the loss of neurons and early death (e.g., <2 years in patients with Type 2 nGD and 10–40 years in patients with Type 3 nGD) [162,163,164,165,166,167,168,169,170,171,172,173,174,175]. The main neurological symptoms of Type 2 and 3 nGDs are characterized by selective degeneration of the cerebellar dentate nucleus and the dentato-rubro-thalamic pathway, generalized epilepsy and seizures, horizontal saccadic eye movements, ataxia, spasticity, oculomotor abnormalities, hypertonia of the neck muscles, extreme arching of the neck, bulbar signs, limb rigidity, occasional choreoathetoid movements, and progressive dementia [176,177,178,179,180,181,182,183,184,185,186,187,188,189]. Enzyme replacement therapy (e.g., imiglucerase, velaglucerase, or taliglucerase) and substrate reduction therapy (e.g., eliglustat and miglustat) are available to treat many of the visceral aspects of GD but have no impact on the CNS disease and are of limited benefit in the management of immune inflammation and disease complications in multiple organs such as the bones, lungs, and lymph nodes [70,190,191,192]. The development of alternative therapies, i.e., gene, substrate reduction, and enzyme replacement therapies, has been hampered by limitations in understanding disease pathogenesis, inability of therapies to fully cross the blood-brain barrier, and toxicity concerns due to procedural risks [159,193,194,195,196].

The complement system comprises a group of liquid and cell membrane-associated proteins, which are mainly produced in the liver. However, certain brain cells, such as microglial cells, astrocytes, and neurons, are also involved in the direct synthesis of complement proteins [197,198,199]. Complement activation is a complex process, which largely occurs by the classical, alternative, and lectin pathways [200]. The classical pathway is activated by ligation of the IgG and/or IgM immune complexes (ICs) to their corresponding receptors and/or C1q on the cell surface. The alternative pathway is activated by binding of spontaneously activated C3 protein (C3b fragment) to host and non-host cell surfaces [201,202]. The lectin pathway is activated by the binding of the mannan-binding lectin (MBL) to mannose-containing carbohydrates or related ficolins to certain carbohydrates or acetylated structures [198,203,204]. Each of the complement activation pathways follows a series of reactions generating common key components known as C3 and C5 [205]. The downstream cleavage of C3 by the C3 convertases causes the formation of C3a and C3b [205]. Similarly, the downstream cleavage of C5 by the C5 convertases causes the formation of C5a and C5b [205]. C3a binds the C3aR receptor, and C5a binds C5aR1 and C5aR2 receptors [205]. C3b is a major opsonin that induces the tagging and phagocytic uptake of pathogens, and C5b initiates the terminal complement pathway, resulting in the formation of the membrane attack complex (MAC) composed of C5b, C6, C7, C8 and multiple C9 molecules [35,205,206,207,208]. Several activated components of the complement system are essential for controlling cellular and metabolic functions in both visceral organs and the CNS [209,210,211,212,213]. However, the abnormal activation and production of complement components such as C3, C3b, iC3b, CR3, C5, C5a, and MAC/C5b-9 can contribute to visceral and CNS tissue damage in several diseases including, but not limited to, allergic diseases, cardiovascular diseases, age-related macular degeneration, systemic lupus erythematosus, traumatic brain injury, stroke, neuromyelitis optica spectrum disorders, and neurodegenerative diseases such as amyotrophic lateral sclerosis, Alzheimer’s disease and Huntington disease [214,215,216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,232].

Complement activation has been observed in H5N1 and H1N1 influenza virus infections [233]. The sera and lung tissues of the MERS-CoV-infected, hDPP4-transgenic (hDPP4-Tg) mouse model showed complement activation and increased production of C5a and C5b-9, as well as increased viral replication [234,235]. Furthermore, targeting C5aR with anti-C5aR antibody treatment in the MERS-CoV-infected hDPP4-Tg mouse model decreased viral replication and damage to lung and spleen tissue [234,235]. The lung of the mouse-adapted SARS-CoV animal model has also demonstrated complement activation and higher virus replication [31]. To investigate whether SARS-CoV triggered complement activation in the mouse model, C3^-/- and background matched control wildtype (WT) mice were infected with SARS-CoV-2 and their lung tissues analyzed for immune inflammation. The data showed that C3^-/- mice infected with SARS-CoV-2 were protected against the development of respiratory dysfunction and lung damage when compared to WT mice [31].

SARS-CoV-2 infected primary human airway epithelial cells also caused complement activation and increased production of C3a [236]. C3aR and C5aR targeting in cellular models also showed protection from lung tissue damage [236]. Lung and skin tissues and sera from patients with severe COVID-19 have shown increased complement activation and massive generation of C5a, C5b-C9, and C5aR1 in blood and pulmonary myeloid cells [209]. Studies have shown that SARS-CoV-2 induced hyperinflammation is linked to the development of acute respiratory distress syndrome (ARDS), systemic clotting, and a variety of cutaneous manifestations in patients with COVID-19 [80,81,82,237,238,239,240,241]. Strikingly, administration of C3 (AMY101) or C5 (eculizumab) targeting drugs have shown marked reduction in the SARS-CoV-2-induced development of severe disease symptoms and lung damage in COVID-19 patients [80,81,82]. These data suggest that complement activation and the resultant activation of the C5a–C5aR1 axis is critical for propagating the disease complications in patients with COVID-19.

Investigations into the mechanisms by which complement activation occurs in COVID-19 revealed (1) the receptor-binding domain of the spike protein of SARS-CoV-2 binds to specific IgG/IgM antibodies, (2) heparan sulfate binds to the spike protein of SARS-CoV-2, and (3) recognition of the spike and nucleocapsid proteins of SARS-CoV-2 by lectin pathway components subsequently triggers complement activation by the classical, alternative, and lectin pathways [32,35,242,243,244]. Mouse IgGs (IgG1, IgG2a/c, IgG2b, and IgG3) and their matching receptors (e.g., FcγRI, FcγRIIb, FcγRIII and FcγRIV) differ from the human IgGs (e.g., IgG1, IgG2, IgG3, and IgG 4) and their receptors (e.g., FcγRI, FcγRIIa, FcγRIIc, FcγRIIIa, FcγRIIb, and FcγRIIIb) [245]. Human or mouse FcγRI binds only to the monomeric IgG, but all the other FcγRs in the mouse (e.g., FcγRIIb, FcγRIII and FcγRIV) and human (e.g., FcγRIIa, FcγRIIb, FcγRIIIa and FcγRIIIb) can bind to IgG-ICs [245,246]. SARS-CoV-2 infection can result in the sustained development of memory B cells and long-lived bone marrow plasma B cells [247,248,249] and the production of IgG, IgM, and IgA antibodies to S, the RBD of S, and N proteins. The SARS-CoV-2-specific IgA/IgM antibodies are present only for a short period of time, whereas the SARS-CoV-2-specific IgG antibodies persist for long periods of time, thereby providing protection against SARS-CoV-2 in patients with COVID-19 [250,251].

The crosslinking of IgG2a/IgG2b ICs to the FcγRIII/FcγRIV in mice and IgG1 IC crosslinking to the activating FcγR in humans cause optimal complement activation and C5a formation [23,252,253,254,255,256,257,258]. Analysis of IgG and their receptors revealed higher expression of the activating FcγR receptors [259] and increased production of SARS-CoV-2-specific IgG antibodies (IgG1 > IgG2 > IgG3) in patients with COVID-19 [23,251,260]. These findings suggest that the IgG isotypes developing in COVID-19 contribute to complement activation. We and others have reported higher levels of complement activation products (e.g., C1q, C3, C4b C5a, C3aR, C5aR1, and C-type lectins) in mouse models and patients with GD. Studies on the Gba1 mutant mouse model, specimens from Type 1 GD patients, and induced pluripotent stem cell (iPSC)-derived Mɸs and lymph nodes from Type 2 and type 3 nGD patients have shown that activation of complement and the C5a–C5aR1 axis leads to tissue inflammation and organ failure in GD [126]. Further, we and others have shown that genetic deficiency of C5aR1 or pharmaceutical targeting to block this receptor in mouse models of GD and in human cell models resulted in marked reduction of immune inflammation and tissue damage (e.g., lung, liver, and spleen) [79,258]. Increased production of autoantibodies specific to GC, IgG2a/c and IgG2b are seen in the Gba1^9V/- mutant mouse model of GD. Also, high levels of GC-specific IgG1 autoantibodies and lower levels of IgG2 and IgG3 autoantibodies have been observed in GD patients [258]. Similarly, experimental mouse models of GD and GD patients have shown elevated levels of mouse/human specific FcγRIII and FcγRIV [258,261,262]. C1q-binding to IgG-IC drives a series of events resulting in the proteolytic cleavage of C4, C2, and C3 and eventually of C5 into C5a and C5b. Further, C5a can be generated locally in immune cells through an FcγR-dependent mechanism involving LAT phosphorylation and production of a C5-cleaving protease [215,252,253,254,255,256,257,263,264,265]. We have observed that massive GC accumulation drives the production of GC-specific IgG2a/c and IgG2b-ICs in Gba1^9V/- mice, as well as IgG1 and IgG3 ICs in GD patients to initiate complement activation through the classical pathway [258]. These findings suggest common involvement of complement activation and C5a–C5aR1 axis-mediated tissue inflammation in GD and COVID-19. Therefore, targeting complement at the level of the C5a–C5aR1 axis could control SARS-CoV-2-induced tissue inflammation in COVID-19 in addition to GD patients with COVID-19.

Several studies have shown that in vivo and ex vivo stimulation of immune cells with immune complexes (e.g., IgG and IgM ICs) and complement activation products (e.g., C3, C3a, C3b, and C5a) can cause excess secretion of lysosomal enzymes. Sphingolipids are ubiquitous components of the plasma membrane of eukaryotic cells and are essential for controlling cell proliferation, survival, and death [4,307]. However, prolonged overabundance of sphingolipids is detrimental, as seen in several lysosomal storage disorders such as GD, Fabry disease, Tay-Sachs disease, Sandhoff disease, Krabbe disease, and Niemann–Pick disease [308]. GC is an important sphingolipid, representing the backbone of more than 450 structurally different glycosphingolipids. Activation of GCS, the enzyme that places a glucosyl moiety onto ceramide, is the first pathway-committed step in the production of more complex GSLs, such as lactosylceramide and gangliosides [309].

Increased formation of GSLs such as GM2-ganglioside and lactosylceramide, have been observed upon infection by Zika virus and hepatitis C virus, respectively [310,311]. Human cytomegalovirus infection also causes an excess formation of ceramide and GM2-ganglioside [312]. Dengue virus causes the increased production of ceramide and sphingomyelin [313]. Influenza virus infection has also demonstrated increased synthesis of GC and sphingomyelin [314,315]. Conversely, suppression of the biosynthesis of cellular GSLs results in the inhibition of maturation of influenza virus particles in vitro [316,317]. It was recently suggested that GSLs support SARS-CoV-2 replication and infection [318,319]. Elevated levels of 3-ketosphinganine (16-0, 18-0, 18-1, and 20-0), sphinganine, sphingosine, dihydrosphingosine, ceramides, dihydroceramides, sphinganine-1-phosphate (S1P), GA1, and GM3 have been observed in SARS-CoV-2 infection in vitro and in vivo in experimental mouse and human cells models as well as in the sera/plasma of COVID-19 patients [1,2,3,4]. Furthermore, drugs targeting GCS inhibited the replication of SARS-CoV-2 [78].

Pathogenic variants in GBA1/Gba1 and the resultant deficiency of acid-β glucosidase drive massive increases of GCs: 18-0, 20-0, 22-1, 22- 0, 24-1, and 24-0 in tissues (e.g., lung, liver, spleen, and brain) from mouse models and human patients with type 1, 2 and 3 GD [66,67,121,161,165,166,167,258,320,321,322,323,324,325,326]. Genetic deficiency or pharmaceutical targeting of C5aR1 in mouse models and human cellular models of GD (e.g., macrophages, and dendritic cells), showed marked reduction in the generation of GCS and GCs (e.g.,16-0, 18-0, 20-0, 22-1, 22- 0, 24-1, and 24-0) [79,258]. Importantly, studies have shown that a majority of GD patients treated with substrate reduction and/or enzyme replacement therapies and experienced SARS-CoV-2 infection only developed mild symptoms of COVID-19 and survived. These findings suggest that the C5a–C5aR1-induced excess sphingolipid production propagates the disease in COVID-19 and GD. Thus, targeting the C5a–C5aR1 axis-mediated overproduction of GC could control SARS-CoV-2 replication and prevent development of severe disease in patients with COVID-19 and GD patients with COVID-19.