Single-Domain Antibodies as Therapeutics for Respiratory RNA Virus Infections

Over the years, infectious diseases with high morbidity and mortality disrupted human healthcare systems and devastated economies globally. Respiratory viruses, especially emerging or re-emerging RNA viruses, including influenza and human coronavirus, are the main pathogens of acute respiratory diseases that cause epidemics or even global pandemics. Importantly, due to the rapid mutation of viruses, there are few effective drugs and vaccines for the treatment and prevention of these RNA virus infections. Of note, a class of antibodies derived from camelid and shark, named nanobody or single-domain antibody (sdAb), was characterized by smaller size, lower production costs, more accessible binding epitopes, and inhalable properties, which have advantages in the treatment of respiratory diseases compared to conventional antibodies. Currently, a number of sdAbs have been developed against various respiratory RNA viruses and demonstrated potent therapeutic efficacy in mouse models. Here, we review the current status of the development of antiviral sdAb and discuss their potential as therapeutics for respiratory RNA viral diseases.


Introduction
Respiratory diseases caused by various types of virus infections have been the focus of global health concerns and are one of the leading causes of death in developing countries [1]. According to the nucleic acid types, respiratory viruses can be divided into RNA and DNA viruses. However, the primary viruses causing the epidemics of respiratory infections in the last two decades were RNA viruses, such as the severe acute respiratory syndrome coronavirus (SARS-CoV) in 2003, the influenza H1N1 virus in 2009, the Middle East respiratory syndrome coronavirus (MERS-CoV) in 2012, and the SARS-CoV-2 in 2019 [2]. Therefore, the development of effective therapeutics for respiratory RNA viruses is rival to combat infectious diseases.
Respiratory RNA viruses include coronaviruses (SARS-CoV-1, MERS-CoV, and SARS-CoV-2), influenza viruses, respiratory syncytial virus (RSV), and others [3,4]. With the development of innovative recombinant DNA technologies, monoclonal antibodies (mAbs) have been proven effective in controlling respiratory RNA viral diseases. In 1998, the mAb palivizumab targeting RSV fusion (F) protein was approved by the FDA as prophylaxis against serious lower respiratory tract disease caused by RSV in children at high risk [5]. However, the high production costs of mAbs limit their commercial market, and the large size of mAbs leads to their low tissue accessibility and penetration, thus affecting their therapeutic efficacy. These features obstruct the development of mAbs [6]. Single-domain antibodies (sdAbs), consisting of only variable domains, have many advantages compared to mAbs. Their smaller size enables tissue penetration, so they could recognize epitopes that are normally not accessible for mAbs. In addition, the smaller size and higher stability of sdAbs make administration by inhalation possible, which is more suitable for treating As the smallest antibodies, sdAbs have several beneficial characteristics, such as low molecular weight (12)(13)(14)(15), small size (4×2.5 nm) [18], as well as high stability and solubility due to their longer CDR3 loops than conventional antibodies [19,20]. Because of the small size, sdAbs can recognize cavities or hidden epitopes that are not accessible to conventional antibodies [21]. For example, Wu et al. identified several fully human sdAbs that recognize a cryptic epitope located in the SARS-CoV-2 spike trimeric interface and have potent neutralization [16]. Moreover, sdAbs can be easily produced in bacteria, greatly reducing the production cost.
In addition, sdAbs can be easily engineered to be multivalent or multispecific, improving their binding affinity and breath [22] and prolonging half-life in vivo [23]. Lauren et al. reported four sdAbs, designed as SD36, SD38, SD83, and SD84, which could bind to highly conserved epitopes of the influenza A and B virus hemagglutinins (HAs), respectively. By fusing these sdAbs with peptide linkers, two multidomain antibodies (MDAbs) were generated and demonstrated the improved binding breadth and neutralizing potency of individual sdAbs [24]. In another study, a bispecific sdAbs bn03 was constructed by linking two sdAbs (n3113v and n3130v) that recognized two different conserved epitopes on SARS-CoV-2 RBD and exhibited potent neutralizing activity to all SARS-CoV- Currently, sdAbs have been developed to treat and diagnose various diseases [18]. In 2019, FDA approved Sanofi's caplacizumab (an anti-von Willebrand factor nanobody and the first nanobody approved by FDA) for treating acquired thrombotic thrombocytopenic purpura (aTTP), a rare disease characterized by excessive blood clotting in small blood vessels [26]. A nanobody targeting human epidermal growth factor receptor 2 (HER2) is being evaluated in detecting breast-to-brain metastasis by Positron Emission Tomography (PET)/Computed Tomography (CT) imaging in phase II clinical trials (NCT03331601) [27,28]. Furthermore, due to the small size and favorable biophysical characteristics, some studies have reported that directly delivering nanobodies into the lung can block the viral invasion of airway epithelial cells in situ by inhalation [25,29,30]. Therefore, nanobodies can be administered by inhaled delivery and are particularly suitable for treating respiratory diseases caused by respiratory RNA virus infections. Single-domain antibodies inhibit respiratory RNA virus infections by several different mechanisms in parallel with the virus life cycle. Respiratory RNA viruses primarily inoculate through the nose and then enter into host cells. After entry, respiratory RNA viruses multiply in the host cells, such as epithelium cells of the large and small airways, vascular endothelial cells, and alveolar macrophages. The infection of the respiratory tract can induce an immune response, and the immune response may lead to the cytokine storm; finally, the inflammation and necrosis of the epithelium cells may lead to pneumonia. The majority of neutralizing sdAbs inhibit the entry process by blocking the attachment of membrane fusion between virus and host cell. Moreover, some sdAbs neutralize the virus by inhibiting the viral genomic replication in host cells and the release of virion progeny from the host cells.  The process of viral entry usually starts from the attachment to a host cell; hence sdAbs can block viral entry by directly interfering with the interactions between the virus and host receptors, such as the RBD of SARS-CoV-1 and SARS-CoV-2 S proteins [39], and the receptor-binding site of influenza HA [40]. For example, a number of sdAbs targeting the RBD of SARS-CoV-2 S protein can block the interaction of RBD with the receptor angiotensin-converting enzyme 2 (ACE2) [41]. After attachment, the fusion of viral and host membranes is triggered and the viral genome is released into the cell cytoplasm from the endosome. Therefore, sdAbs can also block viral entry by binding to viral proteins that mediate membrane fusion and the release of the viral genome, such as the F protein of RSV [14], the HA and M2 of influenza viruses [42,43], and the S protein of SARS-CoV-2 [44]. Previous studies have reported that one sdAb targeting the non-ACE2 binding site of RBD can neutralize SARS-CoV-2 by inhibiting the conformational change of S protein that is essential for membrane fusion [45]. Moreover, some sdAbs can inhibit viral replication by binding to the viral proteins that are essential for virion release from the host cell, such as the neuraminidase (NA) of influenza viruses [46]. Apart from binding to the viral surface proteins, sdAbs can also inhibit the viral replication by binding to the viral intracellular proteins, such as the nucleoprotein of influenza viruses that mediates the viral nuclear trafficking and packaging [43,47].

Coronaviruses
Coronaviruses (CoVs) are enveloped viruses with a positive-sense RNA genome that usually cause mild to moderate respiratory diseases in humans, which lead to several outbreaks of respiratory diseases. Based on their phylogenetic relationships and genomic structures, CoVs are classified into four genera (Alphacoronavirus, Betacoronavirus, Gammacoronavirus, and Deltacoronavirus). Among them, the highly pathogenic CoVs belong to betacoronaviruses and cause acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) in humans, such as SARS-CoV-1, MERS-CoV, and SARS-CoV-2 [48]. The CoV virion contains at least four structural proteins: spike protein (S), nucleocapsid protein (N), membrane protein (M), and envelope protein (E) [49]. S protein is a class I viral fusion protein that mediates the process of viral entry by binding to the target host cell and initiating fusion with the cell membrane. S is homotrimeric, with each subunit consisting of two domains, S1 and S2. S1 contains an RBD that binds to the receptor of the host cell. Different CoVs enter target cells with distinct receptors. For example, SARS-CoV-1 and SARS-CoV-2 bind to ACE2 for host cell entry, whereas MERS-CoV recognizes dipeptidyl peptidase 4 (DPP4) as its receptor. S2 mediates the subsequent membrane fusion to enable entry to the host cytoplasm [50,51]. Due to these essential roles in viral entry, the S protein is the main target of neutralizing sdAbs.
In 2003, the emergence of SARS-CoV-1 has caused a devastating pandemic. Wrapp et al. reported that two potently neutralizing sdAbs, VHH-72 and VHH-55, were isolated from a llama immunized with prefusion-stabilized SARS-CoV-1 and MERS-CoV spikes. These sdAbs are bound to the SARS-CoV-1 and MERS-CoV RBDs with a high affinity of 1.15 nM and 0.079 nM, respectively ( Table 2). In addition, VHH-72 is also bound to SARS-CoV-2 RBD with an affinity of 38.6 nM. However, VHH-72 must be fused with an Fc domain of human IgG1 with a (GGGGS) 3 linker to neutralize SARS-CoV-2 pseudovirus with an IC 50 of~0.2 µg/mL [39]. Besides, several sdAbs binding to SARS-CoV-1 RBD were also identified, such as 5F8 [52], Fu2 [53], S1-1, S1-RBD-6 [54], and S14 [55] (Table 2).       (Table 2). However, none of them bound to the D539N variant located in RBD, suggesting that VHHs neutralized MERS-CoV most likely by blocking binding to DPP4. NbMS10 was selected from a VHH library by phage display and bound to MERS-CoV RBD-Fc with a high affinity of 0.87 nM. The sdAb neutralized MERS-CoV by blocking the binding between MERS-CoV RBD and DPP4 and the IC 50 of NbMS10 was 3.52 µg/mL. Meanwhile, the dimeric and trimeric NbMS10, generated by linking monomeric NbMS10 with a GGGGS linker, showed higher binding affinity to RBD and cross-neutralizing activity against divergent strains of MERS-CoV. In addition, NbMS10-Fc had potently prophylactic and therapeutic efficacy in hDPP4-Tg mice challenged with lethal MERS-CoV [57,90] (Table 2).

Influenza Viruses
Influenza viruses are negative-sense single-stranded segmented enveloped RNA viruses belonging to the Orthomyxoviridae family and become the most common causes of human respiratory infections with high morbidity and mortality [94]. There are three genera that could infect humans: influenza A virus (IAV), influenza B virus (IBV), and influenza C virus (ICV) ( Table 1). The majority of seasonal influenza viruses are IAVs and IBVs [95], which contain eight RNA segments that encode eight proteins, including two major targets of neutralizing antibodies, HA (a homotrimer that mediates the viral entry process) and NA (critical for viral release) [96]. IAVs are classified into subtypes based on the combination of HA and NA. Furthermore, these subtypes are classified into group 1 (H1, H2, H5, H6, H8, H9, H11, H12, H13, H16, H17) and group 2 (H3, H4, H7, H10, H14,  H15), according to the phylogenetics of HA [97]. Due to the high genome mutation rate and the existence of antigenic drift and shift, new IAV types constantly appear and escape from the currently available neutralizing antibodies of influenza viruses [98]. Therefore, it is necessary to develop universal antibodies recognizing the conserved sites of HA or NA.
The monomer HA contains two subunits, HA1 and HA2. The head domain of the HA1 subunit is antigenic and variable, resulting in specific neutralizing of one influenza virus type for several sdAbs recognizing this region [40,77,78]. However, the receptor-binding site (RBS) is relatively conserved and has been recognized as the main target of broadly neutralizing antibodies. In contrast, the HA2 mediated membrane fusion shows more conserved [99], and, and the majority of broadly neutralizing sdAbs recognize the stem region [24,80]. Laursen et al. reported four sdAbs isolated from the immunized llamas, among which SD36, SD38, and SD83 bound to the conserved stem region of HA, while SD84 bound to the relatively conserved region of the HA head domain. SD36 and SD38 exhibited exquisite neutralizing breadth of IAVs, while SD83 and SD84 potently neutralized IBVs (Table 2). Moreover, the multidomain antibody was engineered by genetically fusing individual sdAbs with peptide linkers and then linking to human IgG1-Fc, named MD3606, exhibited enhanced broadly neutralizing activity against IAVs and IBVs and superior protective activity in mice from influenza B infection [24]. Further, humanized MD3606 expressed using an adeno-associated virus vector showed potent prophylactic efficacy in mice challenged with H1N1, H3N2, or IBV at a dose of 4 × 10 7 to 5 × 10 9 genome copies per mouse by intranasal administration [24]. In another study, R1a-B6 binding to the stem region of HA with high affinity was developed from an immunized alpaca by phage display and demonstrated neutralizing breadth against H1N1, H5N1, and H9N2 viruses. The bivalent R1a-B6 showed more neutralizing strength, including the H2N2 virus [80,82] (Table 2). Moreover, R1a-B6-Fc fusion protein delivered by AAV could be sustained in sera for more than 6 months and protected mice challenged with lethal H1N1 or H5N1 viruses [81]. In contrast, some sdAbs have been reported that bound to the HA stem region; however, they were only effective for a specific strain of influenza virus [78,79].
NA is a tetramer with each monomer consisting of four distinct domains: the catalytic head, the stalk region, the transmembrane region, and the cytoplasmic tail [100]. Due to the low mutation rate of HA, some universal mAbs targeting NA have been isolated from infected or vaccinated people, demonstrating NA as a promising target for broad protection against various influenza viruses [101]. N1-3-VHHb, a bivalent nanobody isolated from immunized alpaca and binding to H5N1 NA, showed potent NA-inhibiting activity and antiviral potential in vitro, as well as prophylactic efficacy in mice, challenged with lethal H5N1 virus administered intranasally at 60 µg [46].
M2 is an ion channel protein activated by the low pH of the endosome and plays essential roles in releasing the viral genome from the endosome to the cytoplasm [43]. The 1-9, His37, and Trp41 residues of M2 in all IAVs are extremely conserved. Therefore, sdAbs targeting these residues of M2 showed universal protection against IAVs [102]. Wei et al. reported a sdAb M2-7A that bound to the recombinant M2 protein and the native M2 on the virion, as well as the M2, expressed on the cell surface, but not to the synthetic ectodomain of M2 (M2e, residues 1-24) ( Table 2). M2-7A broadly inhibited the replication of H3N2 and H1N1 in vitro and protected mice challenged with lethal H1N1 virus in vivo [42]. In addition, some VHHs targeting M2e showed broad binding breadth to most H2N2, H1N1 and H3N2 avian and swine influenza viruses. One of these VHHs, named M2e-VHH-23m, was fused with FcγRIV by a (GGGGS) 3 linker and exhibited potently prophylactic efficacy in mice challenged with lethal H3N2 virus at 50 µg [83]. Furthermore, an mRNA treatment that encoded this bispecific sdAb also showed protection against the same H3N2 virus strain [84] ( Table 2).
Because of the relatively high mutation rate of HA, NA, and M2, the development of truly universal antibodies remains challenging. To overcome this challenge, researchers focus on the less variable proteins, such as the nucleoprotein coating the viral RNA that is critical to transporting viral ribonucleoproteins (vRNPs) into the nucleus. More than 20 sdAbs targeting the nucleoprotein have been identified [47,103,104]. One of them, named αNP-VHH1, exhibited potent antiviral activity by blocking vRNP nuclear import and subsequent viral transcription and replication [47,105] (Table 2). In addition, Schmidt et al. reported a lentiviral screening strategy that developed several nucleoprotein-specific VHHs showing IAV-inhibiting activity by blocking the nuclear import of vRNPs and viral mRNA transcription [104].

Respiratory Syncytial Virus
Respiratory syncytial virus (RSV) is the main cause of respiratory virus infections in young children and elderly people. Apart from a humanized mAb that is licensed for high-risk infants as passive immunoprophylaxis, there are no effective RSV vaccines [38]. RSV is a non-segmented, negative-sense, single-stranded RNA virus belonging to the Pneumoviridae family (Table 1). Based on the antigenic subgroups, RSV is classified into two types: RSV-A and RSV-B. RSV is a filamentous enveloped virus and its envelope contains three proteins on the surface: the glycoprotein that plays an important role in host cell attachment and is the most variable structural protein, the fusion protein (F) that mediates the process of fusion and cell entry, and the small hydrophobic protein that is not required for the entry process [38,106]. The sequences of the F ectodomains share a high similarity (~90%) between RSV-A and RSV-B. Therefore, most sdAbs were developed for RSV F protein. ALX-0171, the most intensively studied nanobody targeting RSV F, is a trimeric nanobody developed by Ablynx with two GS linkers among each subunit ( Table 2). ALX-0171 demonstrated high binding affinity with the antigenic site II epitope of F and broadly neutralizing activity against different clinical RSV isolates in vitro, as well as therapeutic efficacy in cotton rats challenged with RSV Tracy by nebulization administration [14]. In the preclinical study, ALX-0171 also exhibited robust antiviral effects and safety in newborn lambs infected with hRSV-M37 [29]. The phase I/IIa clinical trial showed that ALX-0171 administrated by inhalation at 1.2 mg/kg in infants infected with RSV could reduce the nasal viral titters with no safety concerns (https://www.clinicaltrialsregister.eu/ctr-search/ trial/2014-002841-23/results, accessed on 18 August 2016). However, ALX-0171 showed no significant effect in children who were hospitalized with lower respiratory RSV infection in the phase IIb clinical trial [107]. With extensive research on the structure and function of RSV surface F glycoproteins, a number of sdAbs specifically targeting the prefusion conformation of F protein (PreF) exhibited higher binding affinity and inhibitory activity against RSV in vitro [85][86][87]. For example, F-VHH-4 and F-VHH-L66 could bind to PreF with <18 pM and 154 pM, respectively, and broadly neutralized RSV in vitro ( Table 2). The crystal structures of F-VHH-4 and -L66 showed that they were both bound to a cavity formed by the boundary of two F protomers. In addition, F-VHH-4 demonstrated better prophylactic efficacy than palivizumab in mice challenged with RSV A2 at 10 µg [86]. These PreF-binding sdAbs with picomolar affinity are potential therapeutics for RSV infections.

Other Respiratory RNA Viruses
In addition to CoVs, influenza viruses, and RSV, there are other respiratory RNA viruses, such as HRVs, human metapneumoviruses (hMPVs), and parainfluenza viruses  (Table 1). They are genetically classified into three types (RV-A, RV-B, and RV-C) [33]. The capsid of HRVs consists of four proteins, VP1, VP2, VP3, and VP4. Except for VP4, VP1, VP2, and VP3 are exposed at the viral surface and therefore are the main targets for neutralizing antibodies. The canyon in VP1 is the attachment site for host cell receptors. For most known HRV serotypes (>90%), the intercellular adhesion molecule 1 (ICAM-1) acts as their receptor for viral entry, whereas a small portion of HRVs enter the host cell via the low-density lipoprotein receptor (LDLR) [33,108]. A number of mAbs and Fabs targeting the VP1 [109], VP3 [110], or ICAM-1 [111] have been identified with neutralizing activity, while there are no available sdAbs targeting HRVs. PIVs are the second leading cause of acute respiratory tract infections in children. They are negative-sense single-stranded enveloped RNA viruses belonging to the Paramyxoviridae family [35,112] (Table 1). Based on the genetic diversity, PIVs can be divided into four serotypes, PIV-1, PIV-2, PIV-3, and PIV-4. There are two glycoproteins studded in the viral envelope. One of them is the HA-NA (HN) protein, which serves as the attachment protein by binding to the sialic acid-containing molecules of the host cells. The other one is the F protein responsible for the fusion. These two proteins are exposed on the viral surface and are the main targets for neutralizing antibodies and prophylactic vaccines [112]. There are no effective vaccines or therapeutics for PIV infections currently, and only several antibodies targeting the apex of F protein were developed and exhibited protective activity in cotton rats [113]. hMPVs, a common cause of viral pneumonia among infants and children, are enveloped viruses with negative-sense single-stranded RNA genomes and belong to the Pneumoviridae family [34] ( Table 1). Similar to RSVs, there are three proteins exposed on the hMPV surface, the glycoprotein, the F protein, and the small hydrophobic protein. Among them, the F protein could induce neutralizing antibodies and hence is the target for developing neutralizing antibodies or vaccines [114].

Conclusions and Perspectives
In the past twenty years, respiratory RNA virus infections caused several outbreaks, with high morbidity and mortality [115][116][117]. The main characteristics of respiratory RNA viruses are strong transmission capacity and a high mutation rate. For instance, the current pandemic of SARS-CoV-2 has lasted for more than two years and five variants of concern (VOCs), including Alpha, Beta, Gamma, Delta, and Omicron, have been identified. The latest mutant strain Omicron has been reported to be resistant to most neutralizing antibodies and currently available vaccines [118,119]. Because of these characteristics, the developments of universal vaccines and effective therapeutics specific for respiratory RNA viruses remain a great challenge.
mAb-based therapeutics for respiratory RNA virus infections have been demonstrated to be effective [120]. However, the development of mAbs against viral infections has been hampered due to expensive production costs and limited commercial markets. Since the end of the last century, sdAbs with smaller size, higher stability and solubility, and lower immunogenicity and production costs have attracted more and more attention [18]. Compared to mAbs, sdAbs have advantages that have been confirmed by many studies. Firstly, sdAb can penetrate deep inside the "sterically hidden" interface of the virus and neutralize the virus. As the enveloped virus with flexible viral states, the large size of mAbs might bind the cryptic epitope in one state; however, it exhibited neglectable neutralization. Importantly, these cryptic domains are usually highly conserved and induce a low immune response in humans. Secondly, sdAb can be easily engineered to be multivalent to improve neutralizing potency and breadth [22]. Notably, by inhaled delivery, sdAbs can be directly delivered to the lung as the main infectious tissue in respiratory diseases [25]. Therefore, these sdAbs are promising to develop universal antiviral therapeutics for respiratory RNA virus infections and provide insights into the rational design of effective vaccines.
Despite the superior properties of sdAbs, mAbs are still dominant in the treatment of viral infections. For instance, 6 mAbs have been FDA approved for COVID-19 therapy and more than 50 ongoing clinical trials. The short half-life in vivo may be the limitation of sdAb developments. Currently, many strategies have been used to extend the half-life of sdAbs, such as fusing antiviral sdAbs with anti-human serum albumin nanobodies or IgG1 Fc fragments [121][122][123]. So far, only one sdAb has been approved for treating aTTP, and more than 30 sdAb-based drug candidates are in clinical trials. We expect recent technological advances in the fields of half-life extension, nebulized delivery, and production processes to advance more sdAbs into the clinic for the treatment of respiratory diseases.

Conflicts of Interest:
The authors declare that they have no conflict of interest.