Molecular mechanisms of SARS-CoV-2 entry: implications for biomedical strategies

SUMMARY

The mechanisms by which viruses enter host cells are crucial for their ability to infect and cause disease, serving as major targets for both host immune responses and therapeutic strategies. The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) entry process is primarily driven by the binding of the viral spike (S) protein to the angiotensin-converting enzyme 2 (ACE2) receptor, in conjunction with the activity of endosomal cathepsin L and the serine protease transmembrane protease serine 2 (TMPRSS2). Nevertheless, recent scientific advances have expanded our understanding of SARS-CoV-2 entry mechanisms, uncovering alternative receptors and novel cofactors that may enhance viral tropism and adaptability. Given the critical role of the SARS-CoV-2 S protein in mediating host cell entry, it has become a primary target for prevention and therapeutic strategies. However, the continuous spread of SARS-CoV-2 has led to the emergence of S protein variants that may potentially confer a fitness advantage or modify key aspects of SARS-CoV-2 biology, such as transmissibility, infectivity, antigenicity, and/or pathogenicity, posing significant challenges to the efficacy of current interventions. In this review, we provide an updated and comprehensive overview of the latest advances in SARS-CoV-2 entry pathways and molecular mechanisms, exploring their implications for antiviral drug discovery, vaccine design, and the development of other biomedical strategies while addressing the challenges posed by the ongoing evolution of the virus.

INTRODUCTION

Coronaviruses (CoVs) are a diverse group of enveloped, single-stranded, positive-sense RNA viruses responsible for both established and emerging diseases in many avian and mammal species (1, 2). In humans, CoVs have been historically associated with mild respiratory illnesses in healthy individuals and more severe diseases in immunocompromised individuals, children, and the elderly (human coronaviruses [hCoV]-HKU1, -NL63, -229E, and -OC43) (1, 3). However, the emergence of highly pathogenic CoVs, such as severe acute respiratory syndrome coronavirus 1 (SARS-CoV) in 2002, Middle East respiratory syndrome coronavirus (MERS-CoV) in 2012, and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in 2019, has shifted this paradigm (4–8).

As of January 31, 2025, the global circulation of SARS-CoV-2, the etiological agent of the ongoing coronavirus disease 2019 (COVID-19) pandemic, has led to over 7,777 million confirmed cases and more than 7 million fatalities worldwide (World Health Organization). Furthermore, recent data suggest that nearly 15 million deaths globally may have been directly or indirectly linked to the virus during the first 2 years of the COVID-19 pandemic, highlighting the devastating global impact of the crisis (9).

Despite extensive past and ongoing vaccination efforts (10), COVID-19 cases continue to surge worldwide, with new waves of infection constantly being registered (11, 12). This persistent rise is driven by several factors, including the emergence of new viral variants, enhanced rates of transmission, waning immunity after vaccination or natural infections, different immunization rates, and the limitations of current vaccines to prevent transmission (13–15). These challenges underscore the need for continued vigilance and the development of more effective strategies to control the pandemic and mitigate its long-term impact.

The SARS-CoV-2 virion is mainly composed of four structural proteins: the spike (S), membrane (M), envelope (E), and nucleocapsid (N) proteins (16). Among them, the S protein is essential for viral entry into host cells, mediating both the binding to the angiotensin-converting enzyme 2 (ACE2) receptor and the subsequent fusion with the host cell membranes (17). It is a type I transmembrane glycoprotein that assembles into homotrimers on the viral envelope, with each virion containing an average of 15–30 S trimers, which confer the characteristic crown-like appearance of the virus (17, 18).

Comprising 1,273 amino acids, the S protein is initially synthesized as a single polypeptide and subsequently cleaved into two subunits: the surface-exposed S1 subunit, which binds to the ACE2 receptor and determines virus cell tropism, and the transmembrane S2 subunit, which anchors the protein to the viral envelope and facilitates membrane fusion between host cells and viral particles. After cleavage, S1 and S2 remain non-covalently associated with the mature S protein (2, 17, 18). The S1 subunit consists of an N-terminal domain (NTD) and a receptor-binding domain (RBD), which is responsible for binding to the ACE2 receptor on host cells (Fig. 1). Notably, there is a 72% similarity between the amino acid sequences of the RBDs of SARS-CoV-2 and SARS-CoV, with the RBD of SARS-CoV-2 showing greater affinity for ACE2 (19). The S2 subunit is the transmembrane component of the spike protein. Its N-terminal region includes an S2′ internal cleavage site, followed by the fusion peptide (FP), a sequence of 20–25 hydrophobic amino acids that anchors the viral spike protein into the host cell membrane (17, 20). In addition, the S2 subunit includes two heptad repeat (HR) domains, HR1 and HR2, which are highly conserved among CoVs and play a crucial role in the fusogenic mechanism of the virus, a transmembrane domain, and a C-terminal cytoplasmic tail (17, 21). At the junction of the S1 and S2 subunits (S1/S2), the S protein harbors a polybasic cleavage site (PRRAR) that is recognized and cleaved by the protease furin. This feature is present in MERS-CoV and hCoV-OC43 but is absent in SARS-CoV and other SARS-related CoVs, which may partially explain the higher infectivity of SARS-CoV-2 compared to SARS-CoV (2, 19, 22–25).

Fig 1.

Domain organization, conformational dynamics, and mutation hotspots of the SARS-CoV-2 spike protein. (A) The spike protein is divided into distinct domains, each represented by a unique color for clarity. Key structural and functional regions are annotated, including the S1 and S2 subunits, S1/S2 and S2′ protease cleavage sites, N-terminal domain (NTD), receptor-binding domain (RBD), C-terminal domain (CTD), fusion peptide (FP), heptad repeats (HR1 and HR2), transmembrane domain (TM), and cytoplasmic tail (CT). Representative glycan icons are also shown according to their positions within the sequence. Protease cleavage sites are marked with scissors, highlighting their importance in viral entry and fusion processes. S1/S2 is a polybasic cleavage site recognized and cleaved by the host protease furin, whereas S2' internal cleavage site is cleaved during viral entry by TMPRSS2 on the plasma membrane or by cathepsin L in the endosomal lumen following CME (17). (B) Cryo-electron microscopy structures of full-length S trimers from the D614G variant in the conformation closed or receptor-inaccessible, with three RBDs down (PDB: 7KRQ), and open or receptor-accessible, with one RBD up (PDB: 7KRR). (C) Conformational transition of the SARS-CoV-2 Wuhan-Hu-1 S2 subunit from the prefusion (PDB: 6XR8) to the post-fusion (PDB: 6XRA) state. Following the release of the S1 subunit, the S2 subunit undergoes dramatic conformational changes, transitioning into the post-fusion state. This process begins with the unfolding of HR1, which drives the insertion of the FP into the target membrane. Subsequent refolding of HR2 brings the FP and the transmembrane domain into proximity, ultimately facilitating membrane fusion (17). (D) Zoom into the S1 trimer from the D614G variant in the open conformation to show spatial mapping of mutation hotspots, including residues T372, L452, S477, E484, F486, N501, and D614 (colored in green), which positively impact viral infectivity or transmissibility. Note that this early variant already has T372A and D614G mutations.

SARS-CoV-2 entry relies on the sequential cleavage of the S protein at the S1/S2 and S2ʹ cleavage sites, steps crucial for membrane fusion. The initial cleavage at S1/S2 occurs during viral maturation within the infected cell, priming the S protein for subsequent stages of viral entry (17, 26). However, this cleavage also destabilizes the S protein, as the S1 and S2 subunits remain weakly associated. These weak interactions can be easily disrupted, causing the S2 trimer to prematurely adopt a postfusion conformation, which results in a non-functional S protein (27, 28). Rather than eliminating the furin-cleavage site, SARS-CoV-2 rapidly acquired the globally prevailing D614G mutation in early 2020. This mutation stabilizes the S protein and reduces the shedding of S1 from viral-membrane-bound S2, a change associated with increased infectivity and viral load in patients with COVID-19 (29–33).

Approximately 40% of the S protein surface is shielded by host-derived glycans, with each trimer displaying 22 putative N-glycosylation sites and 17 potential O-glycosylation sites (34–36). Glycosylation of the S protein has been described as an important modulator of viral infection, influencing receptor interactions and the host immune response (36). Importantly, glycosylation is also a major factor in immune evasion, as it can mask critical epitopes that would otherwise be targeted by both innate and adaptive immune responses (37).

Given the critical role of the SARS-CoV-2 S protein in mediating host cell entry, it has become a primary target for prevention and therapeutic strategies. Efforts to block SARS-CoV-2 entry have included approaches such as blocking binding with the ACE2 receptor, SARS-CoV-2 entry inhibitors, antibody-based therapies targeting the S protein, host protease inhibitors, and vaccines designed to neutralize conserved epitopes in the S protein (38, 39). However, the continuous spread of SARS-CoV-2 has led to the emergence of S protein variants that may potentially confer a fitness advantage or modify key aspects of SARS-CoV-2 biology, such as transmissibility, infectivity, antigenicity, and/or pathogenicity (40, 41), posing significant challenges to the efficacy of current interventions.

The objective of this review is to present a comprehensive overview of the molecular mechanisms underlying SARS-CoV-2 entry, focusing on the role of the S protein and its interactions with host cellular factors (Table 1). Furthermore, we explore the implications of these processes for the design of biomedical strategies, including antiviral therapies and vaccine development, while addressing the challenges posed by the ongoing evolution of the virus.

TABLE 1.

Host factors involved in SARS-CoV-2 entry into human cells^a

Host factor	Function in SARS-CoV-2 entry	Mechanism of interaction with the S protein	Cell type/tissue	Candidate drugs	References
ACE2	Main receptor	Binds to the RBD	Ciliated cells (nasal epithelium), type II alveolar cells, epithelial cells (small intestine, testicles, kidneys, cardiac muscle, colon, thyroid), Langerhans islets, brain, and macrophages	Anti-hACE2 mAbs: 4G8C11 and 3E8	(42, 43)
				Small-molecule inhibitors: IFITM, MLN-4760, SB27001 and its derivatives	(44–46)
				Repurposed drug: dalbavancin, curcumin, glycyrrhizin	(47–50)
				GB-2	(51, 52)
Furin	Protease (primes the S protein during viral maturation)	Cleaves the S protein at the S1/S2 cleavage site	Transmembrane protein of the Golgi apparatus from the infected cell	MI-1851	(26)
				CMK + naphthofluorescein	(53)
				Permethrin	(54)
TMPRSS2	Protease	Cleaves the internal S2′ site	Type II pneumocytes, ileal absorptive enterocytes, and nasal ciliated cells (co-express TMPRSS2 and ACE2)	Active site inhibitors: Nafamostat, camostat, and GBPA (FOY251)	(55, 56)
				Serine protease inhibitors: Aprotinin	(57)
				Peptidomimetic inhibitors: MM3122, MM3123, MI-432, MI-1900, and N-0385	(26, 58, 59)
				Dual-inhibition (TMPRSS2/CTSL): Compound 221-148 and Omicsynin B4	(60, 61)
TMPRSS4	Protease	Cleaves the internal S2′ site	Small intestinal enterocytes	Not currently available
TMPRSS11D (HAT)	Antiviral protease	Cleavage at R682	Ciliated bronchial epithelial cells and the extracellular milieu	Exogenous HAT	(62)
NSPs: CatG and NE	Antiviral proteases	Cleavage at multiple sites	Secreted by neutrophils	CatG	(63)
Cathepsin L	Protease	Cleavage of the internal S2′ site	TMPRSS2-deficient cells	Bafilomycin A1	(64, 65)
				Peptide-based inhibitors: P9, P9R, and 8P9R	(66–68)
				Omeprazole	(66)
				Fluoxetine	(69–71)
				Teicoplanin	(72–74)
				CTSL inhibitors: K777, MDL-28170, and E64-d	(75, 76)
				Dual-inhibition (TMPRSS2/CTSL): Compound 221-148 and Omicsynin B4	(60, 61)
				Calpain inhibitors II/XII (blocks CTSL and 3CLpro)	(66, 72)
TMEM106B	Receptor (in vitro)	Binds to the RBD through its C-terminal LD loops	ACE2-negative airway cells (H255 cells), human intestinal epithelial cells (HIEC-6) and brain-derived cells: glioma-derived U-87 cells, patient-derived glioblastoma cells, and induced pluripotent stem cell (iPSC)-derived astrocytes	Not currently available
TfR	Receptor	Binds to the RBD	Lung cells (nasal epithelial, trachea, lung)	sTfR and transferrin Anti-TfR antibodies Synthetic peptides: SL8 and QK8	(77)
AXL	Receptor	Binds to the NTD	Pulmonary and bronchial epithelial cells, type I and II alveolar epithelial cells, fibroblasts, and immune cells	Recombinant sAXL	(78)
				AXL inhibitors: gilteritinib and nintedanib	(79)
				AXL-related inhibitors: imatinib, ruxolitinib, tofacitinib, baricitinib	(80–83)
				Combinatory therapies: baricitinib + remdesivir, baricitinib + LPV/R, baricitinib + corticosteroid	(82, 84, 85)
TIM-1 (KIM-1)	Receptor	Binds to the PS through its IgV domain	Lung and kidney epithelial cells, endothelial cells, and brain microvascular endothelial cells	Recombinant TIM-1 proteins	(86)
				miR-142	(87)
ASGR1	Receptor	Binds to the RBD (higher affinity) and the NTD through its CRD	Hepatocytes, liver cell lines, ACE2-deficient cells, and specific immune cells	Not currently available
KREMEN1	Receptor	Binds to the RBD, NTD, and S2 through its CUB domain	ACE2-negative cells and specific immune cells	Not currently available
CD147 (basigin)	Receptor	No direct RBD binding; mediates Arf6-dependent viral endocytosis	Pulmonary and tubular epithelial cells, T cells, fibroblasts, megakaryocytes, platelets, and cardiac pericytes (non-infective, only activation)	Exogenous CD147	(88)
				Meplazumab	(88–90)
				Niclosamide	(91)
CD209 (DC-SIGN)	Receptor and viral attachment factor (trans-infection of susceptible cells)	Binds to the RBD through its CRD	Dendritic cells and tissue-resident macrophages, including alveolar and dermal macrophages	Glycomimetic CD209 antagonists	(92)
CD209L (L-SIGN)	Receptor and heterodimer with ACE2	Binds to the RBD through its CRD	Lung and kidney epithelial cells (particularly in type II alveolar cells) and endothelial cells of the lungs, liver, and renal arterioles	sCD209L	(93)
				Man84 and dimeric Man84	(94)
NRP1	Co-receptor	Binds to the CendR at the C-terminal end of the S1 subunit	Endothelial cells, NRP1 +olfactory epithelial cells, cerebral cortex, and astrocytes	Peptide-based inhibitors: ATWLPPR, Tuftsin, and EG00229, and its derivative EG01377	(95–99)
				Small-molecule antagonists	(100)
				Phenothiazines	(101)
HSPG/Hp	Co-receptor	Binds to the S1/S2 site and the RBD	Ubiquitously expressed	Unfractionated Hp and LMWH	(102, 103)
				Modified Hp: Roneparstat, pentosan polysulfate, mucopolysaccharide polysulfate, and sulodexide	(104–106)
				HS-mimetic: Pixatimod	(107)
				LF and bovine LF	(108, 109)
				Tilorone	(110, 111)
				Mitoxantrone and pixantrone	(112, 113)
				Brilacidin	(114)
SR-B1	Co-receptor, binds to HDL-bound virus	Binds HDL-bound virus and facilitates its interaction with the S1 subunit for viral entry	Lung (alveolar), colon, retina, small intestine, and testis epithelial ACE2+ cells	SR‐BI antagonist: ITX5601	(115)
Vimentin	Co-receptor	Binds to a separate motif in the RBD, distinct from the ACE2-binding site, promoting viral attachment	Epithelial and endothelial cells	rhRod	(116)
				Anti-VIM mAbs	(117, 118)
				ALD-R49	(119)
				Withaferin A (WFA)	(117)
Sialic acids	Host factor that enhances viral attachment	Binds to the RBD and NTD	Lung and oral epithelial cells	Sialic acid derivatives	(120)
B0AT1	Host factor that may enhance viral entry by forming a supercomplex with ACE2	Stabilizes ACE2 dimerization, forming a supercomplex that allows simultaneous interaction with two S proteins for enhanced viral entry	Intestine and kidney epithelial cells	Nimesulide	(121, 122)
GRP78	Host factor that enhances viral entry by complexing with the S protein and ACE2 and is a potential alternative receptor	Binds to the RBD through its SBD	ACE2+ cells, monocytes, and macrophages (ACE2-independent SARS-CoV-2 entry)	hMAb159	(123)
				S protein-derived cyclic peptides	(124)
				YUM70	(125)
				Repurposed drugs: BOLD-100 and Oleandrin	(126, 127)
Integrin α5β1	Potential alternative receptor	Binds to the S2 subunit	Pulmonary epithelial cells, vascular endothelial cells, and T cells	ATN-161	(128, 129)
Integrin αvβ3	Potential alternative receptor	Binds to the RGD motif	ACE2-negative endothelial cells and ACE2+ +human cardiac myocytes	Anti-αvβ3 mAbs αvβ3 antagonist: Cilengitide	(130)
DPP4/CD26	Potential alternative receptor	Binds to the RBD through its α/β-hydrolase domain	ACE2-negative cells, lung parenchyma, vascular endothelium, fibroblasts of human bronchi, and lymphocytes	Repurposed DPP4 inhibitors: Sitagliptin and Linagliptin	(131)
				Small-molecule inhibitors: N-0385	(132)
				Spironolactone + DPP4 inhibitors	(133)
TLR4	Potential alternative receptor, facilitating CME	Binds to the S1 and S2 subunits through its extracellular domain (in silico)	Epithelial cells, renal proximal and distal tubular epithelial cells and NK cells (activation, no productive infection)	TLR4 antagonists: Eritoran and Resatorvid (TAK-242)	(134, 135)
				Repurposed drugs: Glycyrrhizin and Nifuroxazide	(134, 136)
TLR1	Alternative receptor (ACE2-independent entry in the late infection stage)	Binds to the S1 and S2 subunits through its extracellular domain (in silico)	Myeloid cells (abortive replication)	TLR1 inhibitor: Cu-CPT22	(137)
CD4	Potential co-receptor (facilitates viral entry following ACE2-dependent entry)	Binds to the RBD through its NTD	CD4+ T cells	sCD4 and RPA-T4	(138)
NTG-IIA (MYH9)	Potential co-receptor (facilitates CME following ACE2-dependent entry)	Binds to the S1 (NTD) and S2 through the C-end domain of the non-muscle myosin heavy chain	Lung cells	Not currently available

^a This table summarizes key host factors implicated in the viral entry process, detailing their specific roles, mechanisms of interaction with the S protein or primary receptors, and the cell types or tissues where they facilitate viral entry. Additionally, the table includes potential candidate therapeutic agents that target these host factors.

SARS-COV-2 ENTRY MECHANISMS

SARS-CoV-2 S protein and ACE2 interaction

ACE2 is the main receptor for SARS-CoV-2 and plays a central role in the viral infection process. It is predominantly expressed on the apical membrane of polarized cells and is found across over 72 tissues and 180 cell types (139). In the respiratory system, ACE2 expression is more abundant in the nasal epithelium, particularly in ciliated cells, and progressively decreases along the respiratory tract, reaching its lowest levels in the lower airway, where it is mainly localized in type II alveolar cells (140). This distribution pattern of ACE2 expression levels correlates with the viral infection gradient, as nasal ciliated cells are the primary targets for viral replication during the early stages of infection (140, 141).

Beyond the respiratory tract, ACE2 is also highly expressed in several other tissues, including the small intestine, testicles, kidneys, cardiac muscle, colon, and thyroid (142). This widespread distribution contributes to explaining the systemic effects of SARS-CoV-2 infection, including gastrointestinal, renal, and cardiovascular complications (143, 144). ACE2 is also present in the pancreatic islets of Langerhans and the brain, raising the possibility of direct viral infection in these tissues. However, the association between SARS-CoV-2 and conditions such as type 2 diabetes or neurological symptoms remains unclear, especially since these manifestations are also observed in mild or long-COVID cases without evidence of active infection (145–149).

The primary pathway of SARS-CoV-2 infection begins with the adsorption of the S protein to the ACE2 receptor on the surface of the target cell. This interaction induces extensive conformational changes in the S1 and S2 subunits, bringing the viral envelope into close proximity to the host cell membrane and exposing the internal S2′ cleavage site (Fig. 1). Cleavage at this site is mainly facilitated either by transmembrane serine protease 2 (TMPRSS2) on the plasma membrane or by soluble cathepsin L within the endosomal lumen following clathrin-mediated endocytosis (CME). This cleavage event leads to the externalization of the FP, which integrates into the host cell membrane, initiating fusion. As a result, SARS-CoV-2 enters the host cell through one of two primary mechanisms: direct fusion of the viral envelope with the host cell membrane or fusion with endosomal membranes following receptor-mediated endocytosis (17, 150) (Fig. 2). These events ultimately lead to the release of the viral genome into the cytoplasm, initiating the replication cycle of SARS-CoV-2 (17, 18).

Fig 2.

Mechanisms of SARS-CoV-2 cellular entry. This figure illustrates the multiple pathways utilized by SARS-CoV-2 to enter host cells, highlighting direct membrane fusion and endosomal entry mechanisms, as well as ACE2-dependent and ACE2-independent pathways. It also details the key host factors involved, including co-receptors and their primary entry pathways, along with the different alternative receptors identified.

Importantly, in addition to the classical membrane-bound ACE2-mediated viral entry, recent studies have highlighted an alternative entry mechanism involving soluble ACE2 (sACE2), generated by Membrane-type 1 Matrix Metalloproteinase (MT1-MMP)-mediated cleavage of membrane-bound ACE2. This truncated sACE2 retains spike-binding capacity and facilitates viral entry into otherwise non-permissive cells. In addition, sACE2 can form complexes with vasopressin and engage non-canonical receptors, potentially broadening viral tropism and contributing to systemic spread and tissue damage (151–153).

A deeper understanding of the detailed mechanisms underlying viral entry is essential for developing targeted therapeutic strategies to block viral infection and prevent COVID-19 transmission.

Role of proteases in viral entry

TMPRSS2

TMPRSS2 is a type II transmembrane serine protease (TTSP) that contains an N-terminal cytoplasmic tail, a transmembrane domain, and three extracellular domains: a low-density lipoprotein receptor class A domain (LDLa), a scavenger receptor cysteine-rich domain (SRCR), and a C-terminal serine protease domain. The latter domain contains a catalytic triad (H296, D345, and S441) that mediates substrate cleavage (154). TMPRSS2 is predominantly expressed on the plasma membrane of epithelial cells across various tissues, including the gastrointestinal, respiratory, and urogenital tracts. Notably, co-expression of ACE2 and TMPRSS2 mRNAs was observed in type II pneumocytes, ileal absorptive enterocytes, and nasal ciliated cells (140, 154, 155).

Although the precise physiological function of TMPRSS2 in the host remains unclear, its involvement in facilitating viral entry is well established. TMPRSS2 has been shown to proteolytically activate the glycoproteins of a broad range of respiratory viruses, including the F protein of human metapneumovirus (HMPV), the hemagglutinin (HA) protein of influenza A virus (IAV), and the S protein of several human CoVs, such as SARS-CoV, MERS-CoV, hCoV-229E, hCoV-OC43, hCoV-HKU1, and SARS-CoV-2 (20, 156–161).

While both SARS-CoV and SARS-CoV-2 utilize TMPRSS2, SARS-CoV-2 shows greater dependence on this protease compared to SARS-CoV, likely due to the presence of the furin cleavage site (S1/S2) in the SARS-CoV-2 S protein (17). Some studies suggest that the sequence between the S1 and S2 subunits of SARS-CoV is not a suitable substrate for TMPRSS2, whereas it is efficiently cleaved by cathepsins. Replacing the SARS-CoV-2 S1/S2 site with the corresponding SARS-CoV sequence impaired the infection of TMPRSS2⁺ human airway cells, underscoring the critical role of this multibasic site in TMPRSS2-mediated entry (162).

TMPRSS2 expression determines the pathway through which SARS-CoV-2 enters host cells (163). In the absence of TMPRSS2, the complete virus-ACE2 complex is internalized through CME, and the virus-host membrane fusion takes place in endosomes. By contrast, the presence of TMPRSS2 on the plasma membrane enables direct fusion of the viral and host cell membranes at the surface, facilitating rapid viral entry and bypassing the need for endosomal trafficking (17, 154, 163) (Fig. 2). TMPRSS2-mediated viral entry not only accelerates viral fusion and replication, but also elicits early antiviral responses and decreases viral sensitivity to specific host restriction factors, leading to increased rates of cytopathology, apoptosis, and necrosis (154, 164). In addition, differential TMPRSS2 usage among SARS-CoV-2 variants has been described, highlighting its role in shaping the host-pathogen interaction (165).

Cathepsins

Although TMPRSS2-mediated viral entry enhances entry efficiency and infectiousness of SARS-CoV-2, cleavage of SARS-CoV-2 S protein at the S2′ site can also be mediated by cathepsins, especially the lysosomal cysteine protease cathepsin L (CTSL) (17, 166).

Cathepsins are low-pH-dependent proteases with endopeptidase and exopeptidase activities involved in protein degradation in the late endosomes and lysosomes. They are classified into aspartic (cathepsins D and E), serine (cathepsin G), and cysteine proteases (cathepsins B, C, K, L, S, and V) based on their substrate specificity (17, 154). Among cysteine proteases, cathepsins B (CTSB) and CTSL have been extensively studied for their role in viral entry (167–169).

During SARS-CoV-2 infection, the cathepsin-mediated pathway becomes critical in cells with low or absent TMPRSS2 expression, as CTSL activity provides an alternative route for viral entry through the endocytic pathway (17). Notably, recent data suggest that the Omicron subvariants are progressively shifting from relying predominantly on TMPRSS2-dependent entry, as observed in earlier variants, to preferential entry through the endosomal pathway (165, 170). Since the viral entry pathway may significantly influence the effectiveness of antivirals (39), this shift should be continuously monitored for its potential impact on antiviral drug development and to ensure the efficacy of current and future therapeutic strategies targeting SARS-CoV-2.

Other proteases

Although TMPRSS2 is the most extensively studied TTSP in SARS-CoV-2 entry, other proteases, including TMPRSS4, TMPRSS11A, TMPRSS13, and matriptase (ST14), also contribute to the viral infection process (17, 171, 172). TMPRSS4 plays a complementary role in priming the S protein, particularly in small intestinal enterocytes (173, 174). Exogenous expression of TMPRSS13 and TMPRSS11D enhanced SARS-CoV-2 entry and replication in vitro (172). Interestingly, TMPRSS11D, also known as human airway trypsin-like protease (HAT), has been recently identified as an antiviral protease (62). The inactive precursor of HAT is expressed on the plasma membrane of ciliated bronchial epithelial cells in the airways. Once processed, the mature HAT enzyme is released into the extracellular space (175). Mature HAT is involved in local defense and pathologic processes, such as asthma, chronic bronchitis, pulmonary fibrosis, and influenza virus infection (175, 176). Recent studies demonstrated that mature HAT cleaves the SARS-CoV-2 S protein at R682, inhibiting its attachment to host cells and membrane fusion. Importantly, the exogenous addition of HAT notably reduced the infectivity of ancestral SARS-CoV-2 in vivo (62). Interestingly, mutations near the R682 cleavage site, such as P681R in the Delta variant and P681H in the Omicron variants (including subvariants BQ.1.1 and XBB.1.5), make the virus resistant to the antiviral effect of HAT (62).

Neutrophil elastase (NE), a key member of the neutrophil serine proteases (NSPs), is another significant protease involved in SARS-CoV-2 entry. NE was initially observed to cleave the ACE2 ectodomain from the plasma membrane of respiratory epithelial cells. This cleavage reduces the binding efficiency of SARS-CoV-2 to these cells, which may mitigate the severity of the infection (177). Interestingly, it has been described that individuals with cystic fibrosis, a condition characterized by elevated NE levels in the airway, experience lower rates and reduced severity of SARS-CoV-2 infection compared to the general population, highlighting a potential protective role of NE in these patients (177, 178).

Additional studies have revealed that NSPs, including cathepsin G (CatG) and NE, further contribute to the innate defense against SARS-CoV-2 infection by proteolytically inactivating the S protein and reducing lung inflammation (63). These NSPs, secreted by neutrophils recruited to the lungs during COVID-19 pneumonia, can cleave the S protein at multiple sites, preventing its binding with ACE2 in vitro. CatG treatment significantly inhibits viral replication in mice lungs, whereas NE and/or CatG deficiency exacerbates lung pathology (63).

Recent studies have highlighted the role of members from the MMP and a disintegrin and metalloproteinase (ADAM) families in facilitating SARS-CoV-2 entry. Notably, the entry of the Omicron BA.1 variant into primary nasal epithelial cells was found to be mediated by proteases from the MMP/ADAM families, suggesting that these proteases may play a significant role in driving the tropism shift of the Omicron variant, enabling its enhanced adaptation to nasal epithelial cells (179, 180).

ADDITIONAL HOST ENTRY FACTORS

The SARS-CoV-2 S protein interacts directly or indirectly with a range of host cell surface molecules, including C-type lectins, glycan receptors, and phosphatidylserine receptors, which have been identified as co-receptors, attachment factors, or even alternative viral receptors capable of mediating ACE2-independent entry (Table 1) (17, 34, 181–185). These host-encoded proviral factors present promising targets for the development of broad-spectrum antiviral therapies (186, 187). Reflecting the complexity of SARS-CoV-2 interaction with host cells, a recent study identified 293 host proteins that bind to the S protein in human lung cells. This was achieved by co-immunoprecipitating Calu-3 lysates incubated with spike S1-S2 proteins using protein A/G-agarose beads, followed by analysis with liquid chromatography-tandem mass spectrometry (LC-MS/MS). Among these, 42 proteins were membrane-associated proteins, with nine expressed at the plasma membrane (77).

Alternative SARS-CoV-2 receptors

SARS-CoV-2 has been detected in a wide range of organs in COVID-19 patients, even in those with low or absent ACE2 expression, suggesting the presence of ACE2-independent alternative receptors (188–192). For a protein to be considered a receptor, it must bind to the SARS-CoV-2 S protein and mediate effective viral infection. Here, we will outline several alternative SARS-CoV-2 receptors, grouped by functional classes, that play a role in the virus's ability to infect cells using ACE2-independent pathways.

Receptors involved in endocytic pathways

Transmembrane protein 106B (TMEM106B)

TMEM106B has emerged as an alternative receptor mediating ACE2-independent SARS-CoV-2 entry into ACE2-negative cells (182). This type II transmembrane protein is localized in late endosomes and lysosomes. It comprises a cytosolic NTD, a transmembrane helix, and a glycosylated C-terminal luminal domain (LD), which can be cleaved by lysosomal proteases (193). TMEM106B forms homodimers and heterodimers with its homolog, transmembrane protein 106C (TMEM106C), and is highly expressed in the brain, heart, thyroid, adrenal glands, and testis (194–196).

The ability of SARS-CoV-2 to infect ACE2-negative airway cells was previously associated with the E484D mutation in the S protein, but the underlying mechanism remained elusive (197). Baggen et al. (2023) revealed that TMEM106B acts as a SARS-CoV-2 receptor and that the E484D mutation increases the affinity of the S protein for TMEM106B, facilitating the infection of ACE2-negative cells. Cryogenic electron microscopy (cryo-EM) analysis demonstrated that the LD of TMEM106B directly interacts with the RBD of SARS-CoV-2 S protein, specifically residues 443–495 (182). Recent research has identified two critical loops of the SARS-CoV-2 S protein that mediate recognition of TMEM106B. Residues 471–491 of the RBD serve as an anchoring loop, while residues 444–451 stabilize the interaction. These binding dynamics vary among Omicron subvariants. For instance, while BA.2.86 demonstrates an enhanced ability to infect cells with low ACE2 expression, EG.5.1 exhibits reduced binding efficiency, likely due to diminished electrostatic complementarity within the large anchoring loop (198).

Two mechanisms have been proposed for TMEM106B facilitation of SARS-CoV-2 entry in vitro. The virus can either bind to small amounts of TMEM106B present on the plasma membrane or co-internalize it with the assistance of heparan sulfate proteoglycans (HSPGs) to facilitate initial attachment. Inside endosomes, TMEM106B stabilizes the conformational changes in the S protein that lead to S2′ site exposure, enabling viral envelope fusion with the endosomal membrane. This function mimics the role of ACE2, though with a significantly lower binding affinity (182, 199).

TMEM106B-mediated SARS-CoV-2 infection has been observed in intestinal and brain-derived cells with low or undetectable ACE2 expression, potentially contributing to the multi-organ pathology associated with COVID-19 (182). In vitro experiments have shown that monoclonal antibodies targeting TMEM106B effectively blocked this entry mechanism, presenting a promising alternative therapeutic strategy (182). However, Yan et al. (2024) suggest that while TMEM106B-mediated SARS-CoV-2 entry is robust in vitro, it has limited relevance for in vivo infections in hosts with endogenous ACE2 expression (199).

Human transferrin receptor (TfR)

TfR is a ubiquitously expressed type II transmembrane protein located on plasma and endosomal membranes. It forms a homodimer, with each monomer consisting of a small cytoplasmic NTD, a transmembrane helix, and a large extracellular domain. The extracellular domain includes a protease-like domain, an αIII-2 helical domain that facilitates dimerization, and a membrane-distal apical domain (200). TfR primarily facilitates iron uptake by forming a complex with diferric transferrin (Tf), which is internalized via CME. In addition, TfR is exploited by various viruses to enhance their cellular entry (201, 202).

Recent research identified TfR as an alternative receptor enabling ACE2-independent SARS-CoV-2 infection (77). This is facilitated by a strong interaction between TfR and the S protein. ELISA and surface plasmon resonance (SPR) assays demonstrated a direct, high-affinity interaction between the human TfR and the RBD of the S protein, a specificity not observed with TfR from mice or Syrian hamsters. Structural modeling localized the key interaction region to the αIII-2 helical domain (residues 524–537) of TfR (77).

Asialoglycoprotein receptor 1 (ASGR1)

ASGR1 is a C-type lectin transmembrane receptor predominantly expressed in hepatocyte membranes that regulates plasma glycoprotein concentration by mediating the endocytosis of desialylated glycoproteins (203). A genome-wide CRISPR-Cas9 screening identified ASGR1 as an alternative receptor facilitating ACE2-independent SARS-CoV-2 entry, particularly in hepatocytes, which express minimal ACE2 levels (204, 205). ASGR1-mediated entry was further confirmed in ACE2-negative Huh-7 cells and liver cell lines, where ASGR1 knockout or inhibition by monoclonal antibodies or siRNA significantly reduced viral infection (204).

ASGR1 interacts with the SARS-CoV-2 S protein through its carbohydrate recognition domain (CRD), efficiently binding the RBD and, to a lesser extent, the NTD (204, 206). In vivo studies using transduced mice expressing ASGR1 demonstrated higher viral titers in the lungs compared to controls, emphasizing its role in viral pathogenesis (204). Furthermore, single-cell RNA sequencing of nasopharyngeal and pharyngeal samples from COVID-19 patients revealed ASGR1 expression in epithelial and immune cells, with viral presence frequently associated with these populations (206).

Receptors involved in the immune response

Cluster of differentiation 147 (CD147)

CD147, commonly known as basigin, is a transmembrane glycoprotein that belongs to the immunoglobulin superfamily (207). It is widely expressed in epithelial and immune cells and plays important biological roles, including acting as a negative regulator of T-cell activation and forming a complex with cyclophilin A to facilitate HIV-1 entry in host cells (208, 209).

CD147 has been proposed to be an alternative receptor for SARS-CoV and SARS-CoV-2 infections, facilitating ACE2-independent viral entry (88, 210). Blocking CD147, either through knockout or with Meplazumab, a monoclonal anti-CD147 antibody, significantly inhibits SARS-CoV-2 infection in vitro. Interestingly, human CD147 expression allows viral entry into non-susceptible cells, including T cells that naturally lack ACE2 (88). In vivo studies in human CD147 knock-in NSG mice showed increased susceptibility to SARS-CoV-2 compared to wild-type NSG mice, emphasizing its critical role in viral entry (211).

CD147 is upregulated in COVID-19 patient samples, particularly on the basolateral side of renal tubular epithelial cells, where it mediates SARS-CoV-2 entry into renal tubules. This leads to the activation of CD147 binding partners, such as cyclophilins and integrins, which contribute to inflammation and renal tubular damage (212, 213). Furthermore, CD147 has also been linked to SARS-CoV-2-induced pulmonary fibrosis, a significant post-COVID condition (89), as well as to internalization in megakaryocytes and platelets, which may exacerbate inflammation and increase platelet reactivity in COVID-19 patients (214, 215). CD147 has also been associated with SARS-CoV-2-induced cardiac pericytes dysfunction, where the S protein stimulates the phosphorylation of extracellular signal-regulated kinase 1/2 (ERK1/2) in pericytes through the CD147 receptor, potentially contributing to microvascular damage in COVID-19 (216).

However, the role of CD147 in SARS-CoV-2 infection has been disputed based on the lack of a direct interaction between CD147 and the RBD of the S protein (217). This suggests that additional proteins may be involved in mediating the interaction between CD147 and the virus.

T-cell immunoglobulin mucin domain 1 (TIM-1)

TIM-1 is a phosphatidylserine receptor (PSR) and type I transmembrane protein characterized by an external mucin domain and a variable immunoglobulin-like (IgV) domain (212). It facilitates CME by binding phosphatidylserine (PS) present on viral envelopes through its IgV-domain, a mechanism exploited by viruses such as hepatitis A and C, HIV-1, Ebola, and Dengue viruses (218–220).

In the context of SARS-CoV-2 infection, TIM-1 is thought to function by binding to virion-associated PS, rather than directly interacting with the spike protein (183). Interestingly, TIM-1 is highly expressed in lung and kidney epithelial cells in COVID-19 patients, and it has been postulated to function as an alternative receptor for SARS-CoV-2 in ACE2-negative cells (212, 221–223). This is particularly relevant given the frequent occurrence of kidney damage in COVID-19 patients, even in those without pre-existing kidney disease (212).

C-type lectins DC-SIGN (CD209) and L-SIGN (CD209L)

CD209 and CD209L are C-type lectins that have been suggested as receptors for SARS-CoV and SARS-CoV-2, facilitating ACE2-independent viral entry (93, 224, 225). These lectins are known to recognize glycans linked to viral glycoproteins, enabling viral attachment to the target cell, cell-mediated transmission of the virus, or direct infection (17, 225).

CD209 is primarily expressed in dendritic cells (DCs) and tissue-resident macrophages, including alveolar and dermal macrophages (226–228). It binds N-oligosaccharides linked to the SARS-CoV S protein, mediating viral attachment to DCs. Although these cells are not productively infected, they facilitate trans-infection by transmitting the virus to susceptible target cells via a synapse-like structure (225). CD209 plays a similar role in cell-mediated transmission of other viruses such as HIV-1, Ebola virus, and cytomegalovirus (229–231). However, in SARS-CoV-2 infection, apart from facilitating trans-infection, CD209 binds to the RBD of the S protein through its CRD, allowing direct infection (93).

CD209L is highly expressed in lung and kidney epithelial cells, particularly in type II alveolar cells, and endothelial cells of the lungs, liver, and renal arterioles (93, 224). It also binds to the RBD from the SARS-CoV-2 S protein, mediating ACE2-independent viral entry (93). CD209L has an N-oligosaccharide linked at N92, which influences its binding efficiency with SARS-CoV-2 S protein, as removal of the N-glycosylation at this site enhances RBD interaction (93, 224). Since CD209L also interacts with ACE2, heterodimerization of these receptors has been suggested as a mechanism for viral entry in cells co-expressing both proteins (93). In human endothelial cells, permissive to SARS-CoV-2 infection, the inhibition of CD209L and administration of soluble CD209L restricts viral entry (93).

Both CD209 and CD209L act as alternative receptors for SARS-CoV-2 in disease-relevant cell types, including those within the vascular system. Consequently, they represent promising targets for antiviral therapies, particularly in tissues where ACE2 expression is limited (93).

Receptors involved in apoptosis

Tyrosine-protein kinase receptor UFO (AXL)

AXL is a glycoprotein broadly expressed across human tissues, with particularly high levels in the lungs, often exceeding ACE2 expression. Particularly, it is highly expressed in type I and II alveolar epithelial cells, fibroblasts, and immune cells (232). AXL belongs to the tyrosine-adenosine mutant (TAM) receptor family and serves as a PSR that recognizes cells displaying externalized PS during apoptosis or cellular stress (233). Notably, PSRs also mediate the entry of a wide range of additional enveloped viruses, including Zika, Ebola, influenza, and Dengue viruses (219, 233, 234).

AXL has been identified as an alternative receptor for SARS-CoV-2. Unlike ACE2, different studies suggest that this alternative receptor binds specifically to the NTD of the SARS-CoV-2 S protein (78). The residues His61 and Glu85 of AXL play critical roles in this interaction, as their mutation disrupts viral attachment (235, 236). However, additional studies were unable to detect direct interactions between AXL and the purified SARS-CoV-2 S protein, indicating that PSRs do not interact directly with the S protein but rather bind to PS in the virion membrane to mediate virus uptake (183).

Experimental studies demonstrate that overexpression of AXL enhances SARS-CoV-2 entry into pulmonary and bronchial epithelial cells lacking ACE2 and TMPRSS2 expression, while AXL knockout significantly reduces viral infection (78). Recent studies confirm that AXL is indispensable for ACE2-independent viral entry in pulmonary epithelial cells, especially for variants like Delta, which contain variations in the spike NTD favouring AXL interaction (235).

Kringle containing transmembrane protein 1 (KREMEN1)

KREMEN1, a high-affinity receptor for dickkopf homolog 1 (DKK1) and a Wnt-signaling antagonist (237, 238), has also been described as an alternative receptor for SARS-CoV-2 (206). It interacts with the RBD, NTD, and S2 regions of the S protein through its CUB domain, displaying higher affinity for the RBD (204, 206).

In vitro studies demonstrated that KREMEN1 supports SARS-CoV-2 entry in ACE2-deficient cells (206). Transduced mice expressing KREMEN1 showed significantly increased viral titers in the lungs compared to controls, reinforcing its role in mediating ACE2-independent infection (204). Clinical samples from COVID-19 patients also highlighted a higher prevalence of KREMEN1 expression in both epithelial and immune cells. SARS-CoV-2-positive cells in these samples often correlated with KREMEN1 expression, underscoring its contribution to viral tropism and entry in ACE2-limited contexts (206).

Co-receptors

Among all the proteins identified for their ability to bind the SARS-CoV-2 S protein, the majority enhance ACE2-dependent viral entry. These co-receptors either form heterodimers with ACE2 or mediate viral attachment to the host cell, facilitating the interaction between SARS-CoV-2 S protein and ACE2. Here, we will list some of these co-receptors and their role in SARS-CoV-2 entry.

Neuropilin-1 (NRP1)

NRP1 is a co-receptor for SARS-CoV-2 infection (239). It is a type 1 transmembrane glycoprotein with multiple functions in the central nervous system (CNS) and endothelial cells, where it mediates cell adhesion and acts as a receptor for vascular endothelial growth factor (VEGF165), regulating angiogenesis (240, 241). The extracellular domains of NRP1, a1/a2 and b1/b2, mediate cell adhesion. While class 3 semaphorins, a type of neuronal chemorepellants, bind to both domains, VEGF165 specifically interacts with the b1/b2 domain. This domain binds to the C-end Rule motif (CendR), a pentabasic sequence at the C-terminal end of VEGF165 (242). NRP1 also interacts through its b1 domain with the polybasic sequence (RRAR) at the C-terminal end of the S1 subunit from the SARS-CoV-2 S protein, termed as CendR peptide, which is only exposed after the S1/S2 site is cleaved by furin (150, 243). This interaction stabilizes the C-terminal end of the S1 subunit (residues 640–685), inducing the release of the N-terminal portion of the S2 subunit (residues 686–700). As a result, S1 dissociates from S2, and the S2′ site is exposed to the host protease TMPRSS2, enhancing viral entry (243). Notably, a monoclonal antibody targeting NRP1 has been shown to block this effect, reducing TMPRSS2-mediated viral entry (239).

The Omicron variant contains mutations near the CendR motif of the S1 subunit, which, according to in silico molecular docking analysis, may strengthen the interaction between the S1 C-terminal end and NRP1, potentially contributing to the increased infectivity of this variant (244).

NRP1 is highly expressed in olfactory epithelial cells, where it facilitates SARS-CoV-2 infection despite their relatively low expression of ACE2 (239, 245). In rhesus macaques, SARS-CoV-2 infection leads to significant overexpression of NRP1 in the cerebral cortex, accompanied by an upregulation of inflammatory response genes. These changes create a positive feedback loop, amplifying viral infection within the CNS, exacerbating the hyperinflammatory state, and potentially leading to viral encephalitis (246). NRP1 also mediates SARS-CoV-2 entry in astrocytes, leading to reactive astrogliosis characterized by increased type I interferon production, inflammation, and downregulation of transporters for water, ions, and neurotransmitters. These changes create a toxic environment that harms neighboring neurons, leading to their dysfunction and death (247). On the other hand, the binding of SARS-CoV-2 to NRP1 prevents the binding of VEGF-A, disrupting the VEGF-A/NRP1 signaling pathway. This is related to impaired pain signaling and analgesia, which may contribute to asymptomatic COVID-19 and increased virus transmission (248).

NRP1 has another isoform, NRP2, which can also bind to SARS-CoV-2 S protein (95). NRP2 expression has recently been associated with increased viral proliferation in inflamed tissues. Proinflammatory cytokines, including tumor necrosis factor α (TNFα) and interleukin-1 beta (IL-1β), induced NRP2 upregulation, with no similar effect observed for NRP1 (249). This suggests that NRP2 may facilitate SARS-CoV-2 infection in coordination with NRP1 and ACE2 under inflammatory conditions.

Heparan sulfate (HS) and heparin (Hp)

ACE2-mediated entry of SARS-CoV and SARS-CoV-2 relies on cell surface heparan sulfate proteoglycans (HSPGs) as an assisting co-receptor (250). HSPGs are expressed on the plasma membrane of most mammalian cells and in the extracellular matrix (251). HS and Hp are glycosaminoglycans (GAGs), negatively charged linear polysaccharides formed by repeating disaccharide units of N-acetylglucosamine and D-glucuronic or L-iduronic acid in HS and Hp, respectively (252).

SARS‐CoV‐2 entry is inhibited by knocking out genes involved in HS biosynthesis or incubating cells with an HS mimetic. Hp and HS directly bind to the S protein, facilitating viral adsorption to the plasma membrane and increasing accessibility to host cell surface proteases (181, 250, 253). Three HS binding sites have been identified in the SARS-CoV-2 S protein: the furin-cleavage site (residues 681 –686) and two motifs within the RBD, “YRLFRKS” (residues 453–459), and “SKPSKRS” (residues 810–816). Having three different sites of possible attachment to HS/Hp increases the possibility of the S protein binding to GAGs on the host cell surface (253). The S protein from Delta and Omicron variants showed higher affinity to Hp than the WT S protein (102).

Scavenger receptor B type 1 (SR-B1)

SR-B1 is the major receptor for high-density lipoprotein (HDL) and plays a key role in facilitating ACE2-dependent entry of SARS-CoV-2 by enhancing viral attachment to tissues where SR-B1 and ACE2 are co-expressed, such as the lung, colon, retina, small intestine, and testis (115). The S1 subunit of the SARS-CoV-2 S protein binds to cholesterol and HDL, which are recognized by the central domain of SR-B1, facilitating the S-ACE2 interaction and subsequent viral internalization (115, 254). Lower HDL levels upregulate SCARB1 (the gene encoding SR-B1), potentially increasing susceptibility to SARS-CoV-2 infection. Conversely, knockdown of SCARB1 or the use of SR-B1 antagonists significantly reduces viral infection in vitro (115, 174). Blocking the cholesterol-binding site of the S1 subunit with a monoclonal antibody also inhibits HDL-enhanced SARS-CoV-2 infection, which may explain the association of low HDL levels with increased COVID-19 severity (115, 255).

SR-B1 has previously been identified as a co-receptor of hepatitis C virus (HCV) in hepatocytes (256) and a facilitator of Dengue virus infection, particularly through interactions with ApoA1 (257). These observations highlight its role in multiple viral infections, making SR-B1 a potential target for therapeutic interventions. Apart from being a co-receptor, SR-B1 modulates immune responses by regulating lymphocyte proliferation and macrophage activity, typically suppressing the release of pro‐inflammatory cytokines. The depletion and exhaustion of SR‐B1 during SARS‐CoV‐2 infection contribute to the development of dyslipidemia and may exacerbate the cytokine storm and hyperinflammatory state observed in severe COVID-19 cases (254).

Vimentin (VIM)

VIM has been identified as a potential co-receptor for SARS-CoV-2 through LC/MS-MS analysis of proteins with S protein-binding ability (258). It is a type III intermediate filament protein ubiquitously expressed on the surface of human cells, with the highest levels of expression found in ovarian and adipose tissue, followed by endothelial cells, fibroblasts, and monocytes (196, 259). VIM is also secreted into the extracellular space by endothelial cells and macrophages (260, 261). Its role as a receptor or co-receptor has been documented for various viruses, including Dengue virus, influenza A, and SARS-CoV (262–264).

Extracellular VIM binds to the S protein of both pseudotyped and infectious SARS-CoV-2, facilitating viral adsorption to ACE2 and promoting infection (117). Studies show that VIM and ACE2 co-expression significantly increases SARS-CoV-2 entry, while shRNA-mediated knockdown of surface VIM reduces infection in endothelial cells (117). Moreover, treatment of ACE2/A549 cells with purified VIM or coculture with VIM +cells increases ACE2-dependent viral entry, highlighting VIM’s role in supporting infection (258). Furthermore, surface VIM expression is upregulated following SARS-CoV-2 infection (117).

VIM specifically binds to the RBD from SARS-CoV-2 S protein through its extracellular rod II domain, a lectin-like domain, without competing with ACE2 for RBD binding (117, 258). The CR3022 antibody, which targets a unique RBD epitope without interfering with ACE2 binding, inhibits VIM-S protein interaction and SARS-CoV-2 entry into endothelial cells. This suggests a crucial role for VIM in facilitating endothelial cell infection, potentially contributing to vascular complications associated with COVID-19 (258). In addition, inhibiting VIM significantly reduces SARS-CoV-2-induced cytopathic effects and enhances cell survival, emphasizing the importance of further exploring VIM’s role in the later stages of SARS-CoV-2 infection (117).

Sialic acids

Sialic acids, a family of nine-carbon monosaccharides derived from neuraminic acid, are usually located at the terminal positions of glycans on glycolipids and glycoproteins in the plasma membrane (265, 266). Sialic acids are abundantly expressed on epithelial cells, especially in the lungs and the oral cavity, where they frequently function as receptors or attachment factors for respiratory viruses, including hCoVs (266–268). While the NTD of SARS-CoV-2 S protein exhibits weak binding to sialic acids, the RBD has been shown to recognize sialic acid-containing oligosaccharides, particularly monosialylated gangliosides such as GD1a and GM3, with a binding affinity comparable to that for HS (269–271). Enzymatic cleavage of the sialic acid residues confirms their critical role in viral attachment (270). In addition, neuraminidase treatment or inhibition of sialic acid biosynthesis significantly reduces SARS-CoV-2 infection in ACE2+ cells (269). The virus binds to sialic acid moieties on the plasma membrane, which act as anchors, increasing its residence time on the cell surface and promoting ACE2-mediated entry (269, 272). This dual-receptor mechanism may partly explain the rapid spread of SARS-CoV-2 compared to SARS-CoV and MERS-CoV. Notably, the Beta variant (501Y.V2-1) demonstrates a higher affinity for 9-O-acetylated sialic acids through its NTD, suggesting that SARS-CoV-2 may evolve to exploit glycan receptors for dual functionality through both the RBD and NTD, thereby enhancing its binding versatility (271).

Sodium-dependent neutral amino acid transporter (B⁰AT1)

The collectrin-like domain of ACE2 mediates its homodimerization at the plasma membrane, where it forms a supercomplex with B⁰AT1. This supercomplex enhances viral binding and entry by stabilizing the interaction between ACE2 and the SARS-CoV-2 S protein (273). Within the ACE2-B⁰AT1 supercomplex, each ACE2 receptor binds the RBD of different S proteins, while B⁰AT1 stabilizes the structure. This configuration allows two different S proteins to interact with ACE2 simultaneously, promoting efficient viral entry (121, 273, 274). B⁰AT1 mediates the absorption of neutral amino acids in the intestine and their reabsorption in the kidney (275). It is primarily expressed in the gastrointestinal tract and kidney, where its localization to the plasma membrane depends on chaperones such as ACE2 (in the intestine) or collectrin (in the kidney) (274, 275). Although B0AT1 is absent in the lung, a lung-specific homolog may form a similar complex with ACE2, potentially facilitating SARS-CoV-2 infection in respiratory tissues (17, 276). The structural arrangement of the ACE2-B⁰AT1 supercomplex, particularly near other co-receptors like NRP1, highlights its role in optimizing viral entry into host cells (243). Targeting this supercomplex presents a promising therapeutic strategy, particularly in tissues where ACE2 and B0AT1 are co-expressed.

Glucose-regulated protein 78 (GRP78)

GRP78, also known as immunoglobulin heavy chain binding protein (BiP), is a chaperone from the heat shock protein 70 (HSP70) family (277). It is a type II transmembrane protein with an ATP-binding domain (ABD) at its N-terminal and substrate-binding domain (SBD) at its C-terminal (278). GRP78 is expressed in the lumen of the endoplasmic reticulum (ER), where it is involved in the unfolded protein response (UPR). By refolding misfolded proteins or directing them to degradation pathways, GRP78 helps maintain protein homeostasis within the cell. During ER stress, GRP78 is overexpressed and translocated to the plasma membrane, where it facilitates the entry of various pathogens, including viruses, bacteria, and fungi (278).

GRP78 has been proposed as an accessory host factor for SARS-CoV-2 entry. Its SBD directly binds to the RBD of the SARS-CoV-2 S protein, with the highest affinity observed for the Omicron variant (124, 279). This interaction enables GRP78 to form a complex with the S protein and ACE2 on the plasma membrane, as demonstrated in vitro (123).

Knockdown of GRP78 and treatment of lung epithelial cells with a humanized monoclonal antibody anti-GRP78 (hMAb159) significantly reduced ACE2 expression and SARS-CoV-2 infection in vitro. These findings suggest that GRP78 is a promising antiviral target for SARS-CoV-2 and other viruses that exploit it for entry (123).

Recent studies have proposed GRP78 as an alternative receptor capable of mediating ACE2-independent SARS-CoV-2 entry, particularly into monocytes and macrophages, which constitutively express GRP78 on their plasma membrane (280). This pathway contributes to immune dysregulation and hyperinflammation, which are hallmarks of severe COVID-19. Targeting GRP78 may therefore not only limit viral entry but also mitigate hyperinflammation associated with the disease, offering a dual therapeutic opportunity (280).

Integrins

Integrins, particularly α₅β₁ and α_vβ₃, have been implicated in SARS-CoV-2 entry. However, their classification as alternative viral receptors remains debated due to the absence of evidence of productive viral replication. Their involvement in SARS-CoV-2 infection was initially predicted through in silico analyses, highlighting the interaction between the conserved RGD motif (403-Arg-Gly-Asp-405), near the ACE2-binding site in the RBD of the S protein and integrins (281). Among the seven heterodimeric integrins capable of binding RGD, α₅β₁ and α_vβ₃ were confirmed as key mediators of viral interaction in vitro (130, 282).

Integrin α₅β₁, the primary receptor for fibronectin, supports SARS-CoV-2 adhesion and facilitates ACE2-independent entry in pulmonary epithelial cells (283, 284). In vascular endothelial cells, binding of the SARS-CoV-2 S protein to integrin α₅β₁ activates NF-κB, leading to increased production of pro-inflammatory cytokines, upregulation of ACE2, enhanced coagulation factor production, and increased endothelial permeability (282). These effects contribute to vascular damage and inflammation associated with COVID-19, presenting integrin α₅β₁ as a promising target for treating vascular inflammation in SARS-CoV-2-infected patients (282).

In T cells, integrin α₅β₁ mediates viral entry, suppressing proliferation and promoting activation and pro-inflammatory cytokine secretion, which could exacerbate T-cell dysregulation and increase disease severity (285). Recent studies also show that while α₅β₁ does not directly mediate viral entry in endothelial cells, it promotes S protein-mediated cell-cell fusion in cooperation with ACE2. This process involves a direct interaction between α₅β₁ and the S2 subunit of the S protein, rather than the RGD motif (286). This could be explained because the S2 subunit has two potential integrin-binding motifs: LDS (Leu441-Asp442-Ser443) and LDI (Leu585-Asp586-Ile587) (285).

Similarly, integrin α_vβ₃ is ubiquitously expressed across the host, especially in endothelial cells, and facilitates ACE2-independent CME of SARS-CoV-2. Following its entry, SARS-CoV-2 remodels cell phenotype and promotes angiogenesis, changes that align with the observed cardiovascular complications associated with COVID-19. However, α_vβ₃-mediated entry does not lead to productive viral replication across SARS-CoV-2 variants (130, 287).

Collectively, these findings highlight integrins as modulators of SARS-CoV-2 infection. Their roles in facilitating entry, promoting inflammation, and contributing to tissue-specific pathophysiology make them promising targets for therapeutic intervention, particularly in mitigating vascular and immune dysregulation during COVID-19.

Dipeptidyl peptidase 4 (DPP4)

DPP4, also known as adenosine deaminase complexing protein 2 (CD26), was initially proposed as a potential alternative receptor for SARS-CoV-2 due to its ability to bind the MERS-CoV S protein and computational docking analyses with SARS-CoV-2 (288, 289).

DPP4 is a type II transmembrane glycoprotein composed of an N-terminal cytoplasmic tail, a transmembrane domain, and an extracellular domain that includes an α/β-hydrolase domain (290). DPP4 is widely distributed in the lung parenchyma, vascular endothelium, fibroblasts of human bronchi, and lymphocytes, suggesting a potential role in SARS-CoV-2 pathogenesis across multiple tissues (291, 292). While data support DPP4 as a potential alternative receptor for SARS-CoV-2 in ACE2-negative cells, where it interacts with the RBD of the S protein through its α/β-hydrolase domain, the experimental evidence supporting productive infection via this pathway remains limited (131, 293).

Toll-like receptors (TLRs)

TLR4 and TLR1 have emerged as potential contributors to SARS-CoV-2 infection and pathogenesis. These receptors may facilitate viral entry and exacerbate hyperinflammatory responses, which are key features of severe COVID-19 (294).

TLR4, a well-characterized pattern recognition receptor (PPR), has been shown to bind the SARS-CoV-2 S protein with a higher affinity than ACE2 (295). This interaction facilitates viral CME, increases ACE2 expression on cell surfaces to facilitate viral entry, and activates TLR4 signaling pathways (296). This ultimately leads to the production of proinflammatory cytokines and chemokines, driving a cytokine storm, the primary cause of mortality in patients with severe COVID-19 (134).

In the lungs, TLR4 activation is particularly detrimental. Destruction of type II alveolar cells after infection reduces surfactant production, increases air-tissue surface tension, and exposes epithelial TLR4-binding sites. The proposed model suggests that SARS-CoV-2 may bind TLR4 to upregulate ACE2 via interferon-stimulated genes (ISGs), facilitating viral entry and triggering MyD88-dependent proinflammatory signaling (134). TLR4 activity also contributes to systemic inflammation in tissues such as the heart, kidneys, skin, and gastrointestinal tract, where it is expressed (297).

Recent research identified TLR1 as a crucial factor responsible for ACE2-independent viral entry into myeloid cells and subsequent inflammatory response triggered by interactions of TLR1 with E and M proteins (137). They advance our understanding of COVID-19 progression by proposing a model where, during the early stage of infection, SARS-CoV-2 infects ACE2+ respiratory tract epithelial cells by an ACE2-dependent mechanism, leading to strong viral replication but scarce inflammatory response due to inhibition of the NF-κB signaling pathway by ORF6. Conversely, during the late stage of infection, the virus uses TLR1-mediated entry into the recruited ACE2⁻ immune cells, which leads to abortive replication as it blocks subgenomic RNA transcription, preventing the translation of structural proteins, including ORF6. However, nsp14 is directly translated from genomic RNA, promoting IKK phosphorylation and resulting in hyperactivation of the NF-κB signaling, which ultimately enhances proinflammatory cytokine production by immune cells, leading to hyperinflammation (137).

Cluster of differentiation 4 (CD4)

Lymphopenia is a hallmark of COVID-19-associated immune dysregulation, strongly linked to disease severity (298). However, the mechanisms by which SARS-CoV-2 induces immune dysfunction remain incompletely understood. Recent findings support the direct infection of CD4+ T helper cells as a potential contributor to immune impairment in severe cases.

SARS-CoV-2 infects human CD4+ T helper cells, but not CD8+ T cells, with viral RNA detected in blood and bronchoalveolar lavage samples from patients with severe COVID-19 (299). The RBD from SARS-CoV-2 S protein directly binds to the NTD of CD4, facilitating SARS-CoV-2 entry into CD4+ T cells in vitro and in vivo. Blocking this interaction with soluble CD4 (sCD4) or CD4 monoclonal antibodies like RPA-T4 significantly reduces viral loads in these cells, confirming the critical role of CD4 for viral entry (138).

Although ACE2 expression is low in CD4+ T cells, SARS-CoV-2 infection in these cells is dependent on ACE2 and TMPRSS2. This was demonstrated by the significant reduction in infection upon ACE2 blockade with polyclonal antibodies and TMPRSS2 inhibition using camostat mesylate (138). These findings indicate that CD4 acts as a co-receptor, supporting ACE2-dependent viral entry through a cooperative mechanism. In addition, they observed that SARS-CoV-2 infection impairs CD4+ T-cell function, induces cell death, and promotes elevated IL-10 expression, which is linked to viral persistence and heightened disease severity. These findings suggest the dual role of CD4-mediated viral entry, as it facilitates viral replication while impairing T helper cell functionality, contributing to immune dysfunction and increased disease severity (138).

The diverse mechanisms of SARS-CoV-2 entry into host cells highlight the complexity of its interactions with various cellular factors, including canonical receptors, co-receptors, and alternative receptors (Fig. 2). These interactions enable the virus to infect a wide range of cell types, even in tissues with minimal ACE2 expression, contributing to its broad tissue tropism and pathogenesis.

EVOLUTION OF THE SPIKE PROTEIN AND ITS IMPACT ON VIRAL ENTRY

Since 2019, the natural selection of variations, particularly in the S protein, has driven the emergence of a diverse subset of SARS-CoV-2 variants. As of October 2021, the European Centre for Disease Prevention and Control (ECDC) classified SARS-CoV-2 variants into three categories based on their potential transmissibility, virulence, and capacity to cause severe disease: variants of concern (VOCs), variants of interest (VOIs), and variants under monitoring (VUMs). Previously identified VOCs include Alpha (B.1.1.7), Beta (B.1.351), Gamma (P.1), Delta (B.1.617.2), and the more recent Omicron (B.1.1.529). Notable VOIs include Mu (B.1.621) and Lambda (C.37) (300, 301). As of January 31, 2025, no SARS-CoV-2 variants meet the VOC criteria, with Omicron BA.2.86 and KP.3 representing the only VOIs, and Omicron XEC classified as VUM.

The first major evolution of the virus occurred with the emergence of the Alpha VOC in December 2020, after nearly a year of global circulation (302–304). This VOC was followed by the lineages Beta (305) and Gamma (306), each found to harbor mutations which have been shown to impair antibody recognition and neutralization (13). These lineages were rapidly replaced by Kappa (B.1.617.1) and Delta (307, 308), which produced a sharp increase in infection rates worldwide. The Omicron VOC was first described in November 2021, being identified as one of the most highly modified and transmissible variants among the VOCs (309). It rapidly gained significant attention as it triggered new waves of global infections and rapidly evolved into a series of “first-generation” sublineages, such as BA.2, BA.4, and BA.5, followed by “second-generation” descendants such as BQ.1, XBB.1.5, EG.5, and KP (310–314). Currently, “second-generation” Omicron sublineages are globally dominant, exhibiting enhanced fitness characteristics (315–319). All these variants accumulated multiple amino acid substitutions, particularly in the S protein, which conferred a selective advantage in the context of viral evolution and population-level immunity (13, 313, 314, 320).

As previously mentioned, the S protein plays a crucial role in viral pathogenesis, as viral entry into host cells is facilitated by its binding to the human ACE2 receptor (1, 321, 322). Due to its critical importance in viral transmissibility and immune evasion, the S protein is under strong selective pressure, driving its continuous and rapid evolution (41, 323, 324). These evolutionary changes can significantly affect the structural stability of the S protein and its capacity to bind to human receptors (13, 41, 324–327). In particular, certain substitutions in the S protein may enhance viral entry by enhancing the cleavage of the S1 and S2 subunits, facilitating greater membrane fusion and more efficient cell entry, ultimately increasing the transmissibility of SARS-CoV-2 variants (13, 41, 328–330).

Across all SARS-CoV-2 variants compared to the original Wuhan strain, over 190 unique mutations have been identified in the S protein. Approximately 60%–70% of these mutations positively impact infectivity or transmissibility, including T372A, L452R, S477N, E484K, F486P, N501Y, and D614G. Conversely, 30%–40% have either neutral effects or may reduce transmissibility, such as K417N, G446S, L455F, L455S, F456L, E484A, F486V, G496S, Q498R, and Y505H (Fig. 3) (323). In addition, these mutations may compromise the efficacy of neutralizing antibodies, potentially reducing the effectiveness of existing SARS-CoV-2 vaccines (331–334).

Fig 3.

Impact of amino acid substitution on SARS-CoV-2 viral entry in VOCs and selected sublineages. Each colored (dark gray, blue, and red) box indicates the presence of specific mutations in each variant lineage, determined by a frequency of 75% or higher in the analyzed sequences (data sourced from Outbreak. Info as of December 2, 2024). Asterisks (*) mark mutations for which experimental data remain inconclusive, though preliminary studies suggest potential functional effects. Blue boxes highlight an enhancement in ACE-2-binding affinity and/or viral entry efficiency, while red boxes indicate a reduction in these parameters. The figure categorizes mutations based on the functional subunits of the S protein: S1, which is responsible for host receptor binding, and S2, which facilitates membrane fusion. It also identifies structural domains within the S1 subunit, including the NTD, RBD, and CTD.

Certain changes in the RBD are particularly significant, as they can enhance ACE2 binding and viral entry. Some mutations improve receptor affinity directly by altering molecular interactions, while others stabilize the RBD in a favorable conformation for receptor accessibility (335). This region is also crucial for generating neutralizing antibodies, making it a hotspot for immune-evading mutations (336).

Another critical region is the furin cleavage site located between the S1 and S2 subunits, which contains a unique amino acid sequence associated with increased infectivity (38). Specific changes in this site, including those found in the Alpha (P681H) and Delta (P681R) variants, increase the basicity of the furin cleavage site, enhancing its susceptibility to cleavage and thereby facilitating more efficient host cell entry by priming the S protein (329, 337). By contrast, SARS-CoV-2 variants without this cleavage site show reduced infectivity (24).

In contrast to the S1 domain, the S2 domain displays fewer mutations, particularly between Omicron sublineages, indicating a more conserved mutational landscape (338). Mutations such as N764K, D796Y, Q954H, and N969K are shared among Omicron variants, while specific changes like N856K and L981F are specific to BA.1 (339).

The most significant mutations in the S protein that influence viral entry and transmissibility include the following.

D614G

The first major adaptive change identified in the S protein was D614G, which increased ACE2 binding affinity and enhanced the efficiency of furin-mediated cleavage (340, 341). This amino acid substitution quickly became the globally dominant variant through positive selection and is now a fixed polymorphism in nearly all circulating SARS-CoV-2 variants (337, 342, 343). Notably, a correlation between ACE2 expression and the prevalence of the D614G mutant has been proposed. Specifically, populations with higher ACE2 expression, such as those in Asia, may have favored the persistence of wild-type strains, while the positive selection of the D614G mutation in other regions could be linked to the lower ACE2 expression levels observed in African, North American, and European populations (340, 342). This amino acid substitution induces a structural change in the S protein, disrupting the stabilizing bond between D614 in the S1 subunit and T859 in the S2 subunit. As a result, the S protein undergoes a significant conformational shift, markedly increasing the proportion of S proteins in the open conformation (from 18% to 58%), thereby enhancing their interaction with the ACE2 receptor (30, 340, 344).

H655Y

Phylogenetic analysis of global SARS-CoV-2 genome sequences revealed the early emergence of the H655Y mutation in New York in March 2020. This mutation, which has been detected in several subsequent lineages, including Gamma and Omicron, suggests its potential contribution to viral adaptation and evolution, as it enhances both viral replication and spike protein cleavage (345). Interestingly, this change, which is present in the S1 domain outside of the RBD, has been linked to a reduction in neutralizing activity when targeted by certain monoclonal antibodies (346). Moreover, this change has also been naturally selected in cats and mice, suggesting that it may have a beneficial effect on expanding the viral host range and increasing susceptibility (347, 348). Previous studies suggested that the combination of H655Y, N679K, and P681H mutations may promote more efficient S1/S2 cleavage by furin, thereby facilitating viral entry into host cells and enhancing both replication capacity and infectivity (349, 350).

K417T/N, S477N, E484K/A, and N501Y

Changes such as N501Y, S477N, and E484K, located within the RBD, likely contributed to increased ACE2-binding affinity in different SARS-CoV-2 variants (351). Among these, the N501Y substitution emerged independently in multiple SARS-CoV-2 VOCs, including Alpha, Beta, and Gamma (302, 305, 306). This change is associated with increased ACE2 receptor-binding affinity and enhanced viral fitness (352). The E484K mutation (E484A in Omicron) further enhances the RBD affinity for the ACE2 receptor by inducing local structural rearrangements (353–355). These changes improve the hydrogen bonding between Q493 of the S protein and H34 of ACE2. By contrast, the K417T mutation (K417N in Omicron) disrupts the salt bridge with D30 of the ACE2 receptor, decreasing both binding affinity and infectivity (323). However, this loss of binding affinity was compensated by the accompanying E484K and N501Y mutations, resulting in improved affinity for ACE2 (323, 351, 356). Furthermore, it has been described that the K417T/N mutation has progressively facilitated immune escape throughout the pandemic, enhancing the ability of SARS-CoV-2 to evade immunity and promote viral transmission (323).

G446S, G496S, and Y505H

The changes G446S, G496S, and Y505H, located within the RBD of some Omicron sublineages, significantly alter interactions with the ACE2 receptor, resulting in reduced binding affinity (327, 357, 358). Structurally, the G446S mutation disrupts the hydrogen bond with Q42 of ACE2, while the G496S mutation eliminates the hydrogen bond with K353. In addition, the Y505H substitution reduces interaction sites by removing its contact with E37 of ACE2 (327). Despite these changes weakening the RBD-ACE2 interaction, these mutations play a significant role in antibody evasion, allowing the virus to partially evade immune responses (323).

F486V/S/P

Among the Omicron sublineages, polymorphisms such as F486V are uniquely found in BA.4, BA.5, and BQ.1 variants. This change enables the virus to escape neutralizing antibodies by reducing ACE2-binding affinity (359). Other polymorphisms at the same position, including F486S and F486P, have been detected in subvariants like XBB.1.5, EG.1, JN.1, and KP.2.3. These mutations are associated with decreased antibody recognition and ACE2 binding, with F486V being particularly implicated in immune escape and potential changes in viral infectivity (360). Despite its prevalence, the impact of the F486P mutation is still unclear. While some studies suggest that this mutation enhances viral fitness by strengthening its binding to the ACE2 receptor, others report a reduction in binding affinity (311, 361, 362).

S477N

The S477N mutation in the spike protein, found in all Omicron sublineages, plays a dual role (363). It introduces two new hydrogen bonds with ACE2, strengthening ACE2 binding while conferring resistance to different monoclonal antibodies. This mutation may compensate for the reduced binding affinity caused by other immune-evading mutations, such as K417N, G446S, E484A, G496S, and Y505H (354, 357, 363, 364). While some studies report lower binding affinity, others suggest an increase in ACE2 interaction, emphasizing its complex and context-dependent effects (361, 365).

L455F and F456L

L455F and F456L mutations, also known as FLip mutations, have been reported in several XBB sublineages, including XBB.1.5.70, HK.3, JG.3, and JD.1.1, and are associated with enhanced fitness and immune evasion of SARS-CoV-2 (362, 366, 367). It has been described that, while the F456L mutation slightly reduces the binding between the RBD and ACE2, the L455F mutation causes a more pronounced reduction in this affinity. However, when combined, these mutations reposition H34 on ACE2, forming new hydrogen bonds with the RBD and increasing binding affinity through an epistatic effect (362). Furthermore, the L455S mutation, found in the JN.1 strain, a descendant of BA.2.86, which has outcompeted all earlier variants to become the dominant circulating variant, significantly promotes immune escape at the expense of reducing receptor-binding affinity (319, 367–371).

L452R

The L452R mutation is present in Omicron sublineages such as BA4, BA.5, and BQ.1 variants. This substitution may drive evolutionary progression by altering electrostatic interactions, potentially enhancing binding through the altered configuration of L452, which influences interactions with the negatively charged ACE2 surface (38, 372–374). Consequently, L452R enhances the fusogenicity and binding affinity of SARS-CoV-2, contributing to increasing the viral infectivity by enhancing the cleavage of the spike protein (359, 375, 376). Interestingly, this mutation was found to potentially enhance the immune escape ability of the virus from convalescent plasma and monoclonal antibodies (354, 377, 378).

K417N, N440K, and T478K

K417N, N440K, and T478K mutations are located in the S1 domain and are shared with ancestral VOCs. These polymorphisms have been previously linked to enhanced ACE2 binding and/or immune evasion (348, 379). The T478K mutation, in particular, plays a pivotal role in stabilizing and modifying the RBD loop (residues 473–490), enhancing its interaction with the ACE2 receptor. This modification increases viral binding affinity, which may contribute to enhanced infectivity and fitness (380).

T372A

T372A disrupts the N-X-T/S motif, preventing N-linked glycosylation at residue N370. This change enhances the ability of SARS-CoV-2 to interact with host cell surface glycans, thereby strengthening its binding affinity for ACE2. As a result, this substitution likely served as an early adaptation in the pandemic, significantly increasing ACE2 affinity (323, 381, 382).

Overall, the changes observed throughout the evolution of SARS-CoV-2, particularly those found in VOCs, primarily serve to enhance viral fitness compared to the wild-type strain. Structural analysis of the Omicron S trimer bound to ACE2 has shown that Omicron exhibits a 6- to 9-fold increase in binding affinity for ACE2 compared to earlier variants. This heightened affinity is mainly attributed to the additional mutations in the RBD, which improve molecular interactions and facilitate viral entry into host cells. These adaptations likely contribute to the enhanced transmissibility of Omicron and its ability to evade immune defenses (41, 383). Collectively, these changes highlight the remarkable evolutionary flexibility of SARS-CoV-2 to optimize transmission across diverse genetic backgrounds.

IMPLICATIONS FOR ANTIVIRAL DRUG DEVELOPMENT

A deep understanding of the molecular mechanisms underlying SARS-CoV-2 entry is key for the development of new antiviral drugs aimed at preventing infection and mitigating COVID-19 progression. This section explores several complementary strategies to disrupt viral entry pathways, including therapies that target ACE2, S-priming proteases, alternative receptors, co-receptors, and other host factors essential for viral entry (Table 1). These approaches hold significant promise for future therapeutic applications, particularly as direct S protein-targeting therapies face potential limitations in efficacy against rapidly emerging variants (384).

ACE2-based therapies

The rapid evolution of SARS-CoV-2 and the emergence of VOCs have posed significant challenges to therapies targeting the S protein (385). As a result, directly targeting ACE2 with anti-human ACE2 monoclonal antibodies (mAbs) has emerged as a promising therapeutic alternative. For instance, 4G8C11 effectively blocks SARS-CoV-2 entry into ACE2⁺ human cells without interfering with the carboxypeptidase activity of ACE2 (42). Another notable anti-ACE2 mAb, 3E8, demonstrates broad-spectrum efficacy in vitro and in animal models against multiple VOCs, including Alpha, Beta, and Gamma, and related hCoVs like SARS-CoV and hCoV-NL63 (43). When delivered via mRNA platforms, 3E8 has also shown potential against Omicron variants in lung-targeted animal models (386).

An additional approach involves the use of interferon-inducible transmembrane (IFITM) proteins, which disrupt the cholesterol concentration in endosomes, preventing the entry of enveloped viruses such as SARS-CoV, SARS-CoV-2, MERS-CoV, and hCoV-NL63 (44, 166, 387). Small molecules, such as MLN-4760, SB27001, and its derivatives (SB27012, SB27047, and SB27041) interfere with ACE2 conformational changes, blocking its interaction with the RBDs from various SARS-CoV-2 VOCs (45, 46). Although it has been shown that these inhibitors only partially reduce SARS-CoV-2 infection in vitro, they hold potential for use in combination with other therapeutic strategies (72).

Another promising anti-COVID-19 drug candidate is Dalbavancin, a Food and Drug Administration (FDA)-approved lipoglycopeptide antibiotic that directly binds to ACE2 with high affinity, blocking its interaction with the SARS-CoV-2 S protein. Dalbavancin has been shown to significantly reduce viral replication and tissue damage in both mouse and rhesus macaque models, providing long-lasting protection due to its extended plasma half-life (47). Furthermore, a curcumin-based film spray, which disrupts the interaction between the SARS-CoV-2 S protein and ACE2 (48), has demonstrated inhibitory effects on both SARS-CoV-2 and influenza virus infections in vitro (49). Interestingly, the natural compound glycyrrhizin demonstrates antiviral activity against SARS-CoV-2 infection in vitro by targeting ACE2, reducing its binding with the SARS-CoV-2 S protein, and downregulating ACE2 expression (50).

A comprehensive search of ongoing and completed clinical studies, listed onhttps://clinicaltrials.gov/https://clinicaltrials.gov/, reveals that 45 trials are investigating therapeutic strategies targeting ACE2 in the context of COVID-19. Of these, 20 studies have been completed, 6 have been terminated, and 3 are actively recruiting participants.

Protease-based therapies

Targeting proteases has emerged as a promising strategy for combating SARS-CoV-2 and other respiratory viruses (388). Active site inhibitors of TMPRSS2, such as nafamostat, camostat mesylate, and its metabolite 4-(4-guanidinobenzoyloxy) phenylacetic acid (GBPA) form covalent acyl-enzyme complexes with the catalytic serine (Ser441) of TMPRSS2, preventing S protein cleavage and effectively blocking SARS-CoV-2 entry (55, 56, 72). Despite their potential, TMPRSS2 inhibitors exhibited limited antiviral activity against the Omicron variant (389). A newer class of reversible TMPRSS2 peptidomimetic inhibitors, including MM3122, MM3123, MI-432, MI-1900, and N-0385, has demonstrated superior selectivity and antiviral potency as compared with nafamostat and camostat (26, 72). Among them, N-0385 has shown potent inhibition of different SARS-CoV-2 VOCs, including Alpha, Beta, Gamma, Delta, and Omicron (58, 390). MM3122, which also targets the proteases matriptase and hepsin, effectively inhibits SARS-CoV-2 replication in human lung epithelial cells, being also effective against the EG.5.1 variant. When administered, both prophylactically and therapeutically, it significantly reduces lung pathology and suppresses proinflammatory cytokine production following SARS-CoV-2 infection (59).

SARS-CoV-2 S priming can also be inhibited by furin inhibitors, such as MI-1851, which efficiently reduces SARS-CoV-2 replication in vitro. Combining MI-1851 with various TMPRSS2 inhibitors, such as MI-432 or MI-1900, further enhances the antiviral activity against SARS-CoV-2 (26). Treatment with decanoyl-RVKR-chloromethylketone (CMK), a peptidomimetic furin inhibitor, in combination with naphthofluorescein, a non-competitive small-molecule inhibitor of furin that also decreases viral RNA transcription, prevents the production of mature viral particles, inhibiting viral entry and syncytium formation (53, 66). A study identified permethrin as a selective non-competitive inhibitor that binds to a novel allosteric pocket of furin (54).

Inhibition of SARS-CoV-2 CME, either by preventing CTSL activation or directly inhibiting its protease activity, offers a promising therapeutic strategy. CTSL activation requires an acidic environment, typically found in late endosomes and lysosomes, making endosomal acidification inhibitors potential candidates to block viral entry (391). Hydroxychloroquine, for instance, was initially proposed for its ability to increase endosomal pH, partially restricting SARS-CoV-2 entry into TMPRSS2⁺ cells (392). However, clinical trials have shown it to be ineffective in treating patients hospitalized with COVID-19 (393). Bafilomycin A1 (Baf-A1), an inhibitor of the vacuolar-ATPase proton pump (v-ATPase), in combination with camostat, effectively prevents SARS-CoV-2 infection in TMPRSS2⁺ cells (64). Moreover, Baf-A1 has shown effectiveness against Beta and Delta variants in vivo, making it a promising candidate for COVID-19 treatment (65). Peptide-based inhibitors such as P9, derived from mouse β-defensin-4, and its derivatives (P9R and 8P9R) also target endosomal acidification by preventing proton transport into the endosome (67). 8P9R demonstrates dual efficacy by blocking CME and TMPRSS2-mediated entry mechanisms through viral particle aggregation (66), significantly inhibiting SARS-CoV and SARS-CoV-2 infection in mouse and hamster models (67, 68). Chimeric peptides that combine 8P9R with antibody fragments or sACE2 offer a potential anti-inflammatory therapeutic approach (67). Several FDA-approved drugs have also been repurposed to inhibit CTSL activation. Omeprazole, a v-ATPase inhibitor, reduces CTSL-mediated processing of the S protein in vitro, but its clinical potential requires further investigation (66). Similarly, fluoxetine, a widely used antidepressant and functional inhibitor of acid sphingomyelinase (FIASMA), prevents endosomal acidification by inhibiting the v-ATPase, while FIASMA inhibition disrupts cholesterol distribution (69, 394, 395). This leads to cholesterol accumulation in endosomes and a reduction of its content in other cellular membranes, not only limiting viral envelope formation but also impairing SARS-CoV-2 propagation (70). Furthermore, fluoxetine, whose therapeutic potential for COVID-19 has been explored in clinical trials, modulates inflammatory responses by inhibiting IL-6 and NF-κB signaling, potentially mitigating the progression of cytokine storms (71). Numerous CTSL inhibitors are also undergoing preclinical and clinical trials (66, 72). For instance, K777, a dipeptide-vinyl sulfone compound, is an irreversible CTSL inhibitor that effectively prevents the entry of SARS-CoV-2 VOCs across various host cell models (75). MDL-28170, a reversible CTSL and CTSB inhibitor, is a promising candidate for broad-spectrum antiviral therapies, demonstrating efficacy against both SARS-CoV-2 and Ebola virus infections in vitro and in vivo (75). E64d, a cysteine protease inhibitor, effectively prevents viral entry, including the Omicron variant, in TMPRSS2⁻ cells (76, 396). Similarly, teicoplanin, an antibiotic commonly used against Gram-positive bacteria, inhibits CTSL activity and restricts SARS-CoV-2 entry in vitro (73). These CTSL inhibitors could be simultaneously administered with TMPRSS2 inhibitors to block both viral entry mechanisms (60). Recently, compounds 212–148 and Omicsynin B4 have been identified for their dual inhibition of TMPRSS2 and CTSL, preventing infection of various SARS-CoV-2 VOCs in vitro (60, 61). Another promising category, calpain inhibitors II/XII, targets both CTSL and the viral 3CLpro protease, achieving potent antiviral effects at lower doses and further minimizing the risk of resistance development (66, 72).

A comprehensive search of ongoing and completed clinical studies, listed onhttps://clinicaltrials.gov/https://clinicaltrials.gov/, reveals that 237 trials are investigating the use of protease-specific inhibitors as a treatment for COVID-19. Of these, 112 studies have been completed, 33 have been terminated, 7 are active but not recruiting, and 17 are actively recruiting participants.

Therapies targeting alternative receptors

The interaction between SARS-CoV-2 S protein and TfR disrupts cellular iron transport, causing extracellular iron accumulation, which increases susceptibility to ferroptosis and may exacerbate multi-organ damage. Notably, this iron dysregulation has been associated with COVID-19 progression (77). Potential therapies targeting this mechanism include soluble TfR, transferrin, anti-TfR antibodies, and synthetic peptides (SL8 and QK8), which have been shown to successfully inhibit viral entry in vitro. In addition, the administration of anti-TfR antibodies in a monkey model reduced viral replication and pneumonia, suggesting that TfR antagonists could be a promising therapeutic approach against COVID-19 (77).

The AXL and TIM-1 receptors have also emerged as potential therapeutic targets to block alternative SARS-CoV-2 entry pathways. AXL, in particular, is a candidate for drug repurposing with available kinase inhibitors such as gilteritinib, which strongly inhibits AXL-S protein interaction and significantly reduces infection across different SARS-CoV-2 VOCs, including Alpha, Beta, Delta, and Omicron (79). Beyond their direct antiviral effects, these AXL inhibitors have anti-inflammatory and antifibrotic properties, which may help mitigate severe COVID-19 complications (79). Other AXL-related inhibitors, including imatinib mesylate, ruxolitinib, baricitinib, and tofacitinib, have also shown promise in drug repurposing efforts (80–83, 397, 398). Given the multitargeted nature of these kinase inhibitors, optimizing target specificity remains important (79). These inhibitors can also be used in combinatorial therapies with other antiviral agents or mAbs. For instance, baricitinib has shown synergistic inhibition of SARS-CoV-2 in combination with remdesivir, lopinavir/ritonavir (LPV/R), and corticosteroids (82, 84, 85, 399). On the other hand, TIM-1 inhibitors, such as recombinant TIM-1 proteins, have shown the ability to block entry of various SARS-CoV-2 VOCs and may be useful in preventing kidney epithelial cell infection, potentially reducing kidney damage in COVID-19 patients (86, 212). In addition, miR-142, a potential broad-spectrum agent against envelope viruses that use TIM-1, regulates TIM-1 transcription in endothelial cells, offering the potential to mitigate complications such as stroke and systemic inflammation associated with COVID-19 (87).

The CD147 receptor not only mediates SARS-CoV-2 infection, but also promotes post-COVID-19 pulmonary fibrosis by contributing to inflammation, fibroblast activation, and extracellular matrix remodeling (89). In this context, meplazumab, a humanized anti-CD147 mAb, has been shown to mitigate fibroblast activation and decrease susceptibility to pulmonary fibrosis in animal models (88, 89). Furthermore, this mAb effectively inhibits viral infection of megakaryocytes and prevents the cytokine storm induced by different SARS-CoV-2 VOCs, including Alpha, Beta, Gamma, and Delta (214, 400). Notably, its administration in clinical trials has demonstrated accelerated recovery in COVID-19 patients (90). Niclosamide, an FDA-approved anti-helminthic drug, has emerged as a promising repurposed inhibitor of CD147. It inhibits syncytia formation, SARS-CoV-2 replication, CD147 glycosylation, and CD147 expression by reducing the translocation of the RNA-binding protein Human-antigen R (HuR) to the cytoplasm, which regulates CD147 post-transcriptional regulation by binding to its 3′-UTR (91, 401).

As previously mentioned, CD209 is another alternative receptor for SARS-CoV-2 and plays a role in mediating attachment to dendritic cells and cell-mediated viral transmission in infections like HIV-1, Ebola, and cytomegalovirus (93, 229–231). Targeting CD209 provides a promising, broad-spectrum, host-directed strategy to mitigate excessive innate immune responses and prevent disease progression. A novel class of potent glycomimetic CD209 antagonists effectively inhibits the binding of SARS-CoV-2 S protein to CD209 and prevents the cell-mediated trans-infection of ACE2⁺ cells (92). Inhibitors targeting CD209L, a variant found in airway epithelial cells and endothelial cells, could be used to prevent impaired angiogenesis and vascular injury during viral infection (93). Among these inhibitors, Man84, a mannose-based glycomimetic and its dimeric form, is a CD209L selective ligand that effectively inhibits SARS-CoV-2 infection, particularly in the respiratory tract (94).

Therapies targeting co-receptors and other host entry factors

Peptide-based inhibitors such as ATWLPPR (A7R), Tuftsin, and EG00229 mimic the binding mechanism of SARS-CoV-2 S protein or VEGF-A165 to the b1/b2 domain of NRP1, demonstrating potential in blocking viral entry (96–98). Previous studies have shown that incubation with EG00229 reduces the efficiency of SARS-CoV-2 infection in vitro (95). In addition, two small-molecule antagonists have been identified as promising candidates to block SARS-CoV-2 entry as they produce a stronger inhibition of CendR binding to the b1 domain of NRP1 (100). Furthermore, phenothiazines, including alimemazine, have recently been recognized for their ability to interfere with the S protein-NRP-1 interaction, inhibiting the entry of SARS-CoV-2 in vitro (72, 101).

Hp is another potential therapeutic target to treat COVID-19, which provides benefits beyond its anticoagulant properties. It neutralizes SARS-CoV-2 infection by binding the S protein HS-binding sites and has anti-inflammatory effects by interacting with chemokines and DAMPs released during viral infection (103, 402). Unfractionated Hp is more effective than low-molecular-weight heparin (LMWH) at inhibiting S protein interactions with target cells (102, 103). Modified Hp, such as Roneparstat, a heparinase inhibitor, exhibits broad-spectrum antiviral activity against SARS-CoV-2, HTLV-1, and HIV-1, while reducing cytokine release in vitro (104). Other promising repurposed drugs for treating COVID-19 include mucopolysaccharide polysulfate, an antithrombotic heparinoid effective against wild-type and Delta SARS-CoV-2 variants, and pixatimod (PG545), a synthetic HS-mimetic in clinical trials for cancer which inhibits different SARS-CoV-2 VOCs in vitro, including Alpha, Beta, Delta, and Omicron and provides prophylactic protection in animal models (105, 107). Pentosan polysulfate, a semi-synthetic Hp-like glycosaminoglycan used to treat interstitial cystitis, exhibits stronger SARS-CoV-2 inhibition than Hp in vitro, despite weaker anticoagulant effects (106). In addition, sulodexide, a mixture of heparin and dermatan sulfate, has shown clinical benefit in improving the outcomes of COVID-19 patients (403).

Lactoferrin (LF), known for its broad-spectrum antiviral activity, binds to HSPGs on the cell surface, disrupting the interaction between SARS-CoV-2 and HSPGs and making it a promising treatment for SARS-CoV-2 and other common hCoVs (108). In addition, small-molecule inhibitors such as Tilorone, a pan-antiviral agent, and Raloxifene also inhibit SARS-CoV-2 infection through the suppression of HSPG-dependent endocytosis, and their combination shows synergistic antiviral effects (110, 166). Similarly, mitoxantrone, a repurposed FDA-approved drug, and its derivative, pixantrone, interfere with the interaction between HSPGs and the S protein, effectively inhibiting HSPG-dependent viral entry of SARS-CoV-2 in vitro and reducing syncytium formation (112, 166). However, in a mouse model, pixantrone shows only modest inhibitory activity against viral entry and replication (113). Brilacidin, with dual antiviral activity, inactivates SARS-CoV-2 and targets HSPGs on the host cell surface, preventing the infection of various hCoVs, including hCoV-229E, hCoV-OC43, and hCoV-NL63 (114).

VIM is a key co-receptor for SARS-CoV-2 infection, with recombinant rod domain of vimentin (rhRod) and anti-VIM mAbs identified as potent inhibitors of SARS-CoV-2 entry by binding to the S protein (116–118). Anti-inflammatory agents targeting VIM, such as Withaferin A (WFA), disrupt viral entry by aggregating vimentin filaments (117). In addition, ALD-R491 inhibits the interaction between the S protein and ACE2, reducing viral CME and egress, which leads to reduced disease symptoms and lung damage in mouse models (119).

As previously mentioned, treatment with hMAb159, a humanized anti-GRP78 mAb, significantly reduces ACE2 expression and SARS-CoV-2 infection in lung epithelial cells (123). Studies of cyclic peptides derived from the SARS-CoV-2 S protein have shown strong binding to GRP78, revealing them as novel potential antiviral agents (124). YUM70, a small-molecule inhibitor of GRP78, effectively blocks viral entry of various SARS-CoV-2 variants in vitro and improves survival in transgenic mice while reducing lung damage (125). Other small-molecule inhibitors repurposed as COVID-19 therapeutic candidates include oleandrin (OLN), which suppresses GRP78 expression and blocks SARS-CoV-2 infection, and BOLD-100, which efficiently suppresses stress-induced GRP78 upregulation in the ER in clinical cancer trials (126, 280, 404). These findings reinforce GRP78 as a versatile and promising antiviral target, whose inhibition may additionally help to reduce the hyperinflammation associated with COVID-19 (123, 280).

Fibronectin-derived peptides, such as ATN-161, have antiviral potential by inhibiting α₅β₁ integrin interaction with the SARS-CoV-2 S protein, reducing viral infection in vitro and in transgenic mice (66, 128, 129). Similarly, cilengitide, an α_vβ₃ integrin antagonist, blocks viral attachment and protects endothelial integrity, preventing barrier permeability caused by SARS-CoV-2 infection (130, 287).

DPP4 inhibitors like Sitagliptin and linagliptin, commonly used for type 2 diabetes treatment, reduce SARS-CoV-2 infectivity by disrupting the DPP4-RBD interaction and blocking ACE2 glycosylation, weakening the binding of the S protein to the receptor (131, 405). N-0385, previously mentioned as a small-molecule inhibitor of TMPRSS2, also functions as a DPP4 inhibitor and modulates inflammatory pathways (58, 132). This dual mechanism positions N-0385 as a promising drug candidate for reducing SARS-CoV-2 infection and the associated inflammatory damage. When combined with Spironolactone, which antagonizes TMPRSS-2 and ADAM17, these DPP4 inhibitors may improve clinical outcomes by decreasing SARS-CoV-2 entry and providing enhanced anti-inflammatory, antiproliferative, and antifibrotic effects (133).

Targeting TLR4 offers a strategy to mitigate the severe inflammatory responses and respiratory complications associated with COVID-19 (294, 406). Pulmonary surfactants like Beractant (bovine-derived) and Colfosceril palmitate (synthetic) are being tested in clinical trials (NCT04384731) for their efficacy in treating severe respiratory disease (134). TLR4 antagonists such as Eritoran have shown potential in reducing systemic inflammation and myocardial damage in animal models (134). In addition, Resatorvid (TAK-242), a small-molecule inhibitor that disrupts TLR4 adaptor interactions, effectively reduced NF-κB activation, cytokine production, and lung permeability in animal studies (134). There are also additional FDA-approved and natural compounds that target TLR4 and offer additional therapeutic potential. Among them, Nifuroxazide, an oral antibiotic which reduces TLR4 content and modulates downstream inflammatory signaling, and glycyrrhizin, which also antagonizes TLR4 while having anti-inflammatory effects, stand out as potential candidates for mitigating the inflammatory responses associated with SARS-CoV-2 infection (134, 136).

VACCINE DESIGN CONSIDERATIONS

The development and implementation of vaccines have been critical in controlling the COVID-19 pandemic, significantly reducing the incidence of cases, hospitalizations, and deaths globally (10, 407). Among past and current vaccine candidates, the S protein of SARS-CoV-2 has been the primary focus due to its central role in viral entry and its ability to elicit neutralizing antibodies (nAbs) (Table 2) (10, 407). However, the rapid evolution of SARS-CoV-2 and the generation of viral mutations that may reduce the neutralizing capacity of antibodies elicited by current vaccines have highlighted the challenges in developing long-lasting and broadly effective vaccines.

TABLE 2.

Overview of major vaccines developed targeting the SARS-CoV-2 S protein^a

Type of vaccine	Name	Manufacturer	Route of administration	Target	Virus variant	Countries with approval	References
RNA	Comirnaty (BNT162b2) and Cominarty Bivalent	Pfizer/BioNTech	IM	S protein	Original/Omicron XBB1.5, BA.1, BA.4-5, JN.1, KP.2	>160	(408–411)
	Spikevax (mRNA-1273) and Spikevax Bivalent	Moderna Biotech	IM	S protein	Original/Omicron XBB1.5, BA.1, BA.4-5, JN.1, KP.2	>90	(412–414)
	Arcturus (ARCT-154)	Arcturus Therapeutics	IM	S protein	Original	Limited to Japan	(415–417)
	AWcorna (ARCoV)	Biotechnology/Suzhou Abogen Biosciences	IM	RBD	Original	2	(418–421)
	TAK-919 (with Moderna technology)	Takeda	IM	S protein	Original	Limited to Japan	(422)
Vector (non-replicating)	Vaxzevria (AZD1222/ChAdOx1 nCoV-19)/Covishield	AstraZeneca/Oxford University	IM, IN	S protein	Original	>150	(423–426)
	Sputnik V (Gam-COVID-Vac) and Sputnik Light	Gamaleya Institute	IM, IN	S protein	Original	>70	(427–432)
	Jcovden (Ad26.COV2.S)	Janssen (Johnson & Johnson)	IM	S protein	Original	>110	(433–437)
	Convidecia (Ad5-nCoV) and Convidecia Air	CanSino Biologics Inc./Beijing Institute of Biotechnology	IM, IN	S protein	Original	>10	(438–441)
	iNCOVACC (BBV154)	Bharat Biotech International Limited	IN	S protein	Original	Limited to India	(442–444)
Protein subunit	Nuvaxovid (NVX-CoV2373)/Covovax	Novavax Serum Institute of India	IM	S protein	Original, Omicron XBB1.5, JN.1	>40	(445–448)
	Corbevax	Biological E.	IM	RBD	Original	2	(449–451)
	BIMERVAX (PHH-1V)	HIPRA Human Health	IM	RBD	Alpha and Beta variants	>40	(452–454)
	EpiVacCorona	State Research Center of Virology and Biotechnology	IM	Peptides from the S protein	Original	4	(455–457)
	Abdala	Centro de Ingeniería Genética y Biotecnología	IM	RBD	Original	6	(458–460)
	VidPrevtyn Beta	Sanofi/GSK	IM	S protein	Beta	>30	(461–463)
	MVC-COV1901	Medigen	IM	S protein	Original	4	(464–467)
	Zifivax (ZF2001)	Anhui Zhifei Longcom	IM	RBD	Original	4	(468–471)
	Soberana-02 and Soberana-Plus (FINLAY-FR-1A)	Instituto Finlay de Vacunas	IM	RBD	Original	4	(472–476)
	V-01	Livzon Mabpharm Inc	IM	RBD	Original	Limited to China	(477–480)
	SKYCovione (GBP510)	SK Bioscience/GlaxoSmithKline	IM	RBD	Original	Limited to Korea	(481–483)
	Razi Cov Pars (RCP)	Razi Vaccine/Serum Research Institute	IN	S protein	Original	Limited to Iran	(484–487)
	SpikoGen	Vaxine Pty Ltd/CinnaGen Co.	IM	ECD	Original	Limited to Iran	(488–490)
	IndoVac (CoV2-IB 0322)	PT Bio Farma	IM	RBD	Original	Limited to Indonesia	(491, 492)
	TAK-019 (Novavax formulation)	Takeda Pharmaceutics	IM	S protein	Original	Limited to Japan	(493–495)
DNA	ZyCoV-D	Cadila Healthcare	ID	S protein	Original	Limited to India	(496–498)
VLP	Covifenz	Medicago	IM	S protein	Original	Limited to Canada	(499–501)

^a The table includes information on the vaccine type, name, manufacturer, route of administration (intranasal (IN), intramuscular (IM), intradermal (ID)), specific target, virus variants addressed, the number of countries where approval has been granted (https://covid19.trackvaccines.org/), and supporting references.

Vaccines targeting the S protein typically include either the full-length glycoprotein or specific domains (Table 2) (10, 407, 502). In this context, the full-length S protein can be presented in its native form or with structural modifications to optimize immunogenicity, production, and stability. For instance, mRNA vaccines such as Pfizer-BioNTech (Comirnaty) and Moderna (Spikevax) incorporate proline substitutions (e.g., K986P and V987P) to stabilize the spike protein in its prefusion conformation (412, 503). This modification enhances antigen presentation to the immune system and improves production efficiency and storage stability. In addition, some configurations, such as those used in Novavax (NVX-CoV2373), include the elimination of the furin cleavage site to inhibit cleavage processing (445).

Vaccines targeting specific domains of the S protein provide a more focused approach to inducing strong immune responses while reducing off-target effects. In these approaches, the RBD is a key target due to its high immunogenicity and critical role in viral attachment (504, 505). Innovative strategies to enhance RBD-based vaccine efficacy include conjugation to virus-like particles, trimerization using the T4 fibritin foldon domain to mimic the native structure of the S protein, or the use of ferritin-based designs, such as RBD-NTD-ferritin nanoparticles (506–508).

While the majority of neutralizing activity in convalescent and vaccine-elicited sera is directed against the RBD, other neutralizing antibodies recognize regions within the S1 and S2 subunits (504, 509–511). In the S1 subunit, the NTD is a key target, as some non-RBD-targeting nAbs can neutralize the virus by destabilizing the S protein structure (510, 512, 513). Meanwhile, in the S2 subunit, antibodies targeting conserved regions such as the FP can effectively block the conformational changes required for viral entry or inhibit membrane fusion (510, 514). These conserved epitopes in the S2 subunit are particularly attractive for the development of universal vaccines, as they remain less prone to mutations across variants (338, 515, 516).

Despite its critical role as a vaccine target, the S protein presents several challenges that complicate vaccine design and implementation. One significant issue, as described earlier, is that the S protein is heavily glycosylated, with several glycosylation sites that can shield about 40% of the protein surface, camouflaging important epitopes from humoral and cellular components (34–36). This structural feature limits the accessibility of key neutralizing regions, potentially affecting vaccine efficacy (517). However, most of these glycosylation sites are highly conserved, suggesting they play crucial structural or functional roles within the spike protein (518). This conservation could potentially reduce the likelihood of vaccine escape by future SARS-CoV-2 variants, though further research is required to fully understand its implications.

Another challenge comes from the conformational dynamics of this protein. The S protein alternates between “closed” and “open” states, with the RBD only exposed in the open state (519). This conformational flexibility allows the virus to balance receptor binding and immune evasion by concealing key epitopes in the closed state (40, 520). To solve this challenge, a number of vaccine candidates rely on the incorporation of key mutations in the S protein to stabilize the prefusion state to decrease the transition to the postfusion conformation. Key mutations incorporated into the S protein to stabilize its prefusion conformation include the K986P/V987P substitutions. However, since this approach does not entirely overcome the challenge of achieving broad protection against diverse variants, alternative stabilization strategies and innovative vaccine designs are being explored (521–523).

Despite their demonstrated success in reducing severe disease, hospitalizations, and deaths during the COVID-19 pandemic, current vaccines face certain challenges that highlight the need for continued innovation. While the primary goal of vaccination is to prevent serious illness rather than infection, the effectiveness of current COVID-19 vaccines in preventing mild or asymptomatic infections is limited, particularly with the emergence of immune-evasive variants like Omicron (524–526). Furthermore, immune responses in certain individuals, including those with immune deficiencies or undergoing immunosuppressive treatments, may not be robust enough to confer adequate protection, underscoring the need for vaccines that provide broader and more durable immunity (527–530).

Next-generation vaccines must not only offer improved protection against variants but also ensure strong mucosal immunity to combat respiratory transmission. The development of vaccines with enhanced immunogenicity, such as those delivered intranasally or through alternative platforms, may hold the key to boosting local immune responses and improving vaccine accessibility, especially for high-risk populations (531).

In the longer term, the development of broadly effective multivalent vaccines, which incorporate antigens from multiple variants, as well as pan-coronavirus vaccines, will be critical in overcoming the limitations of current vaccines (532–534).

A comprehensive search of ongoing and completed clinical studies, listed onhttps://clinicaltrials.gov/https://clinicaltrials.gov/, reveals that 1,363 trials are investigating the use of vaccines against COVID-19. Of these, 658 studies have been completed, 52 have been terminated, 111 are active but not recruiting, and 99 are actively recruiting participants.

NEW DIRECTIONS IN BIOMEDICAL STRATEGIES AND THERAPEUTIC INTERVENTIONS

To address the challenges associated with the fast evolution of the virus, researchers worldwide are continuously working on new treatments and vaccines to counteract the changing variations (535). Among the approaches already in use or under development, the use of mAbs, which demonstrated effectiveness in both therapeutic and preventive applications, and combination therapies, which integrate antiviral drugs with treatments such as mAbs, COVID-19 convalescent plasma (CCP), and intravenous immunoglobulin (IVIG), are promising strategies to mitigate severe outcomes and address the challenges posed by viral evolution (536–541). This section will explore these innovative approaches, with a particular focus on those targeting viral entry mechanisms, reviewing potential treatments currently in clinical trials, and highlighting their ability to improve treatment efficacy.

mAbs

Throughout the COVID-19 pandemic, mAbs have been rapidly developed and tested for their safety and efficacy in clinical trials for both prevention and treatment of the disease due to their high specificity and reliability (536, 542, 543). Most therapeutic mAbs currently in use are of the IgG immunoglobulin G (IgG) subtype, which is preferred for its ease of production and long circulating half-life (544). These neutralizing antiviral antibodies, which can directly neutralize the virus independently of other immune components or cells, have been employed against various viral infections, including Ebola virus and respiratory syncytial virus (545, 546). In the context of SARS-CoV-2, the first mAb therapies became available within the first 10 months of the COVID-19 pandemic, but their efficacy has been affected by the emergence of new variants, making many treatments less successful (547, 548). This type of treatment is characterized by its ability to block the attachment of SARS-CoV-2 to the human ACE2 receptor by binding to the receptor RBD of the S protein, thereby inhibiting viral entry (549). Unlike vaccines, which stimulate humoral and adaptive immunity, mAbs can serve as a crucial tool for patients with weakened immune responses (550, 551). Interestingly, a study identified over 4,000 SARS-CoV-2 S protein-specific memory B cells, among which 453 were found to produce neutralizing antibodies. Among these, the most effective neutralizing antibodies targeted the RBD, demonstrating therapeutic efficacy at lower doses, which made them both more cost-effective and easier to produce sustainably (552). As a result, many different mAbs were developed and received emergency use approvals for treating high-risk patients with mild to moderate COVID-19, leading to reduced viral loads, fewer hospitalizations, and improved symptoms (536, 543, 553, 554).

While combination antibody therapies targeting non-overlapping epitopes have been developed to address the loss of efficacy caused by the emergence of escape mutants, this approach has proven insufficient against the rapid emergence of new variants and subvariants, which frequently accumulate mutations and deletions in antibody-binding domains (552, 554–556). As a result, mAbs therapy is not currently recommended for the treatment of acute COVID-19 or as post-exposure prophylaxis. The mAbs currently approved by the European Medicines Agency (EMA), in monotherapy or in combination, which are mainly used for the treatment of the early stage of infection, individuals with risk factors for severe disease, and for pre-exposure prophylaxis in subjects where vaccination is not recommended, are tixagevimab-cilgavimab, regdanvimab, casirivimab-imdevimab, and sotrovimab (554).

A comprehensive search of ongoing and completed clinical studies, listed onhttps://clinicaltrials.gov/https://clinicaltrials.gov/, reveals that 150 trials are investigating the use of mAbs for COVID-19 treatment. Of these, 68 studies have been completed, 22 have been terminated, 2 are active but not recruiting, and 10 are actively recruiting participants.

Combination therapies

Throughout the COVID-19 pandemic, numerous combination therapies have emerged, with some of them aimed at blocking the binding of SARS-CoV-2 to the receptor, thereby inhibiting viral entry into human cells (535, 557). These therapies typically combine antivirals with immunomodulators, targeting multiple disease mechanisms to improve treatment outcomes. Based on data fromhttps://clinicaltrials.gov/https://clinicaltrials.gov/, there are currently 299 clinical studies investigating combination therapies against COVID-19. Of these, 112 studies have been completed, 21 have been terminated, 10 are active but not recruiting, and 32 are actively recruiting participants. Below, we outline some of the key trials involving combination therapies with a focus on viral entry that have either been completed or are currently underway:

IBIO123

IBIO123, developed by Immune Biosolutions, is a mixture of mAbs developed to neutralize SARS-CoV-2 by targeting its S protein and preventing viral entry into human cells. This therapy consists of three fully human recombinant monoclonal IgGs: IBIO-1 (63%), IBIO-2 (5%), and IBIO-3 (32%). Formulated for inhalation, IBIO123 is delivered directly to the lungs, where it acts locally to neutralize the virus (558). IBIO-1 and IBIO-3 bind non-competitively to the RBD, while BIO-2 binds to a highly conserved epitope in the S2 subunit (559–562). Results from phases 1 and 2 clinical trials (NCT05639166) have shown a reduction of respiratory symptoms with no major safety concerns. Therefore, IBIO123 offers promising benefits in terms of symptom resolution, making it a good candidate treatment for symptomatic COVID-19 patients (558).

Ronapreve (casirivimab-imdevimab)

Ronapreve, also known as REGN-COV and developed by Roche and Regeneron Pharmaceuticals, is based on the combination of two non-competitive mAbs (casirivimab and imdevimab). These mAbs bind to the non-overlapping region of the RBD of the SARS-CoV-2 S protein, preventing the virus from entering human cells by inhibiting the binding of the virus to the ACE2 receptor (542). Ronapreve has demonstrated efficacy in reducing viral load, alleviating symptoms, and significantly decreasing the risk of COVID-19-related hospitalization or mortality (563–565). A phase 3 clinical trial (NCT04852978) was recently completed, examining the safety and immune response of ronapreve combined with Moderna’s mRNA-1237 vaccine in individuals with chronic conditions. In addition, a phase 4 clinical trial (NCT05502081), completed last year, evaluated the effectiveness and safety of this combination compared with standard antiviral therapies such as remdesivir and favipiravir in hospitalized COVID-19 patients. The results of these latter trials, however, are still pending publication.

Evusheld (tixagevimab-cilgavimab)

Evusheld, also known as AZD7442, is a combination of two human mAbs, tixagevimab and cigalvimab, designed to be administered synergistically in the pre-exposure prophylaxis and treatment of COVID-19 (566). Both antibodies target the SARS-CoV-2 S protein by binding to non-overlapping RBD epitopes, inhibiting viral entry by blocking the binding to ACE2 (567, 568). A recently completed phase 3 clinical trial (NCT05780437) evaluated the safety and efficacy of this combination therapy in hospitalized patients, leading to its authorization for pre-exposure prophylaxis in adults and adolescents. This authorization was supported by its demonstrated efficacy against Omicron subvariants BA.1.1, BA.2, BA.4, and BA.5, along with its ability to alleviate symptoms in immunocompromised populations (550, 569). Currently, an active phase 4 clinical trial (NCT05982704) is evaluating the neutralizing activity and efficacy of Evusheld and regdanvimab against SARS-CoV-2 in patients with moderate to severe COVID-19.

Nirmatrelvir/ritonavir and rosuvastatin

Nirmatrelvir is an antiviral medication that acts as an orally active 3C-like protease inhibitor, ritonavir is an antiretroviral drug, and rosuvastatin is a statin medication that is involved in the indirect blocking of viral entry (570, 571). This combination therapy has recently completed a phase 1 clinical trial (NCT05898672) focused on the evaluation of the pharmacokinetics and safety of nirmatrelvir/ritonavir when co-administered with rosuvastatin in healthy individuals. This effect has been observed in studies that associate the use of statins with a reduced severity of COVID-19 infections. Preliminary results are not currently specified, but the design and objectives are detailed.

BAM/ETE (bamlanivimab/LY3819253 plus etesevimab/LY3832479)

BAM/ETE is a combination therapy comprising two potent neutralizing mAbs designed to combat SARS-CoV-2. Both antibodies bind to distinct and non-overlapping regions of the RBD in the SARS-CoV-2 spike protein, thus blocking the binding of the virus to ACE2. This dual binding mechanism prevents viral entry into host cells and promotes virus neutralization (572, 573). A phase 2 clinical trial was recently completed to assess the efficacy and safety of BAM/ETE in individuals recently diagnosed with mild to moderate COVID-19 (NCT04634409, NCT04427501). Results from phase 2 demonstrated a significant reduction in viral load, while preliminary phase 3 findings showed a 70% reduction in hospitalization risk with no reported deaths (574).

Remdesivir and baricitinib

Remdesivir is an antiviral drug that inhibits the RNA-dependent RNA polymerase enzyme, which is essential for viral replication (575). Baricitinib, on the other hand, is an oral immunomodulator that acts as a Janus kinase (JAK) inhibitor. It is used in moderate to severe COVID-19 patients to manage excessive immune responses, such as cytokine storms, and to reduce pulmonary dysfunction and the need for supplemental oxygen. In addition, baricitinib inhibits the signaling pathways involved in CME, a mechanism that SARS-CoV-2 may exploit for viral entry into host cells (85, 576, 577).

This combination therapy is being evaluated in a phase two clinical trial (NCT04321993) to assess its safety and effectiveness in hospitalized patients with moderate to severe COVID-19. The therapy is being tested alongside other treatments, such as tocilizumab. Preliminary results suggest that this combination reduces recovery time and improves clinical outcomes in patients (82).

CONCLUSIONS AND PERSPECTIVES

Understanding the molecular mechanisms underlying the multistep entry processes of SARS-CoV-2 into host cells has been pivotal in advancing the development of safe and effective biomedical strategies to combat the COVID-19 pandemic. While the S protein-ACE2 interaction, together with the roles of TMPRSS2 and endosomal cathepsins, represents the basis of SARS-CoV-2 entry mechanisms, recent discoveries of alternative receptors and cofactors have provided new insights into viral tropism, adaptability, and immune evasion strategies (Table 1).

In this review, we provide a comprehensive and multidimensional synthesis of molecular data on SARS-CoV-2 entry, focusing on the interactions of the viral S protein with host cell factors such as ACE2, TMPRSS2, and other alternative receptors, as a means toward the development of effective therapeutic and preventive measures. These interactions represent ways by which the virus can infect numerous types of tissues, adapt to host responses, and evade immune defense. Furthermore, the emergence of major mutations in the S protein has significantly enhanced transmissibility, immune evasion, and pathogenicity in circulating variants, presenting ongoing challenges to public health and therapeutic efficacy.

This review highlights the critical role of the S protein in mediating host-virus interactions, the complex interplay of viral variants and host proteases, and the increasingly important role of alternative receptors and co-receptors in ACE2-independent viral entry (Table 1). These findings directly inform antiviral strategy development, including virus entry inhibitors, neutralizing antibodies, and next-generation vaccines (Table 2). Future research must delve deeper into the molecular and structural basis of SARS-CoV-2 entry, particularly regarding emerging variants, with a special focus on elucidating the role of alternative receptors and their contribution to tissue tropism and pathogenesis. In addition, the dynamics of S protein evolution require continuous monitoring to anticipate possible adaptations that may have an impact on therapeutic and preventive measures.

Therapeutic innovation should focus on multi-targeted approaches combining entry inhibitors, host protease modulators, and immune-boosting strategies. The development of a pan-coronavirus vaccine and multivalent formulations that take into account the molecular characteristics shared by all CoVs could offer robust defenses against future zoonotic outbreaks by providing enhanced long-term protection. In addition, the integration of structural biology, computational modeling, and high-throughput screening technologies will be critical to accelerate the discovery of effective antiviral agents.

Ultimately, a comprehensive, multidisciplinary approach is essential to advance these research priorities, building resilient public health strategies and ensuring preparedness for the ongoing evolution of SARS-CoV-2 and other future emerging infectious diseases.

Biographies

Irene Santamaría-Castro obtained her degree in Biochemistry at the University of Málaga, where she developed a strong interest in Microbiology, leading her to join the Department of Microbiology as an intern in 2023/2024. She is currently pursuing a Master’s in Cellular and Molecular Biology, with a specialization in Microbiology at the same institution. Her research centers on the innate immune response to SARS-CoV-2 infection and viral immunopathology, conducted at the Immune Response to Viral Infections Lab (IBIMA-Plataforma BIONAND/Universidad de Málaga), supported by the Spanish government. Her passion for virology was sparked during the COVID-19 pandemic, although her fascination with viruses began earlier, inspired by her mother’s work on fish virology. For her Bachelor's thesis, she developed tools for in vitro SARS-CoV-2 studies, and in her Master’s thesis, she aims to identify new interferon-stimulated genes with antiviral activity against this virus.

Rocío Leiva-Rebollo is a Postdoctoral Researcher at Universidad de Málaga / IBIMA Plataforma BIONAND, where she is part of the research group focused on the immune response to viral infections, including SARS-CoV-2, influenza, and HIV. She obtained her Ph.D. in Cellular and Molecular Biology from the University of Málaga in 2019, where she began her research into antiviral immunity. She has been active in the field of viral immunology since 2016. In 2024, she completed a postdoctoral fellowship at the Icahn School of Medicine at Mount Sinai (New York, USA), specializing in respiratory viruses. Her current research aims to identify and characterize interferon-stimulated genes responsible for blocking viral infection, with a particular focus on HIV. Her work contributes to understanding host-pathogen interactions and the identification of future antiviral strategies.

Sonia Marín-Wong is a researcher at the Biomedical Research Institute of Malaga (IBIMA Plataforma BIONAND). She earned her B.Sc. in Biochemistry from the University of Malaga and completed her M.Sc. at the Autonomous University of Barcelona. During her academic journey, she developed a strong interest in virology and its impact on human health, which has guided her current research. Her work focuses on investigating the antiviral effects of interferons during SARS-CoV-2 infection and their role in modulating disease severity.

Jose Manuel Jimenez-Guardeño is a Group Leader at the University of Málaga / IBIMA Plataforma BIONAND, where he co-leads the Immune Response to Viral Infections lab. Over the past 15 years, he has been researching highly pathogenic human viruses, including SARS-CoV, SARS-CoV-2, HIV-1, and influenza virus, aiming to unravel the cellular and molecular mechanisms of antiviral immunity. After completing his BSc in Biology at the Universidad de Málaga and a MSc in Molecular Biomedicine at the Universidad Autónoma de Madrid (Spain), he completed his PhD at the National Center for Biotechnology (CNB-CSIC, Spain) under the supervision of Prof. Luis Enjuanes, where he focuses on investigating the molecular basis of SARS-CoV pathogenesis. Next, he joined the lab of Prof. Michael Malim at King’s College London (UK) to investigate the antiviral immune response to different human pandemic viruses, including HIV-1, SARS-CoV-2, and the influenza virus.

Ana Maria Ortega-Prieto is a Group Leader at IBIMA Plataforma BIONAND, where she co-leads the Immune Response to Viral Infections lab. After completing a BSc in Biology (Universidad de Málaga) and an MSc in Molecular and Cellular Biology (Universidad Autónoma de Madrid), she joined the CBMSO-CSIC (Spain) to carry out her PhD thesis on antiviral drug strategies. Here, she focused on investigating the characterization of compounds with antiviral activity against HCV. In 2015, she joined the laboratory of Dr. Marcus Dorner at Imperial College London (UK) to investigate the development of advanced in vitro and in vivo model systems to study different viral infections, including HBV, HCV, HIV, and Zika virus. She later continued her research at King’s College London (UK) in the laboratory of Prof. Michael Malim, where she investigated virus-host interactions and immune responses to SARS-CoV-2, HIV, and the influenza virus.