key: cord-0969858-qc0hga3w
authors: Piepenbrink, Michael S.; Park, Jun-Gyu; Oladunni, Fatai S.; Deshpande, Ashlesha; Basu, Madhubanti; Sarkar, Sanghita; Loos, Andreas; Woo, Jennifer; Lovalenti, Phillip; Sloan, Derek; Ye, Chengjin; Chiem, Kevin; Bates, Christopher W.; Burch, Reuben E.; Erdmann, Nathaniel B.; Goepfert, Paul A.; Truong, Vu L.; Walter, Mark R.; Martinez-Sobrido, Luis; Kobie, James J.
title: Therapeutic activity of an inhaled potent SARS-CoV-2 neutralizing human monoclonal antibody in hamsters
date: 2021-02-25
journal: Cell Rep Med
DOI: 10.1016/j.xcrm.2021.100218
sha: 6e24b170e68054953444d748b13c024c9d3b291d
doc_id: 969858
cord_uid: qc0hga3w

SARS-CoV-2 infection results in viral burden in the respiratory tract, enabling transmission and leading to substantial lung pathology. The 1212C2 fully human monoclonal antibody was derived from an IgM memory B cell of a COVID-19 patient, has high affinity for the Spike protein Receptor Binding Domain, neutralizes SARS-CoV-2 and exhibits in vivo prophylactic and therapeutic activity in hamsters when delivered intraperitoneally, reducing upper and lower respiratory viral burden and lung pathology. Inhalation of nebulized 1212C2 at levels as low as 0.6mg/kg, corresponding to 0.03mg/kg of lung deposited dose, reduced viral burden below the detection limit, and mitigated lung pathology. The therapeutic efficacy of an exceedingly low-dose of inhaled 1212C2 supports the rationale for local lung delivery for dose-sparing benefits as compared to the conventional parenteral route of administration. These results suggest clinical development of 1212C2 formulated and delivered via inhalation for the treatment of SARS-CoV-2 infection should be considered.

The SARS-CoV-2 global pandemic has infected over forty-five million people and has resulted in over one million deaths so far. It is expected that new infections will continue for many more months, and the virus may persist endemically for years. Although severe infections and deaths have been reported in all ages and demographics, those over 65 years old and those with pre-existing conditions are at highest risk for death [1] [2] [3] [4] [5] . Moreover, in the US, the number of COVID-19 patients that are hospitalized represent a small fraction (~10%) of the active cases, which implies that the vast majority of COVID-19 symptomatic patients are not hospitalized and need treatment 6-8 . Therefore, facilitating greater treatment coverage will be of importance to control transmissibility and healthcare burden.

Neutralizing antibodies (NAbs) induced either by natural infection or vaccination are likely to be critical for protection from SARS-CoV-2 infection and have been correlated with protection from SARS-CoV-2 in animal studies [9] [10] [11] , and passive transfer of neutralizing mAbs has demonstrated prophylactic and therapeutic activity against SARS-CoV-2 infection [12] [13] [14] .

Emerging results in humans treated with convalescent plasma with high titer NAbs suggest therapeutic activity 15, 16 . The primary target for SARS-CoV-2 neutralizing antibodies is the Receptor Binding Domain (RBD), whereby antibodies are expected to inhibit the binding of the SARS-CoV-2 Spike (S) protein to the host Angiotensin Converting Enzyme 2 (ACE2), preventing viral attachment. Already, several human monoclonal antibodies (hmAbs) have been isolated from patients following SARS-CoV-2 infection that are specific for RBD, neutralize SARS-CoV-2, and have anti-viral activity in animal models 12, 17 . RBD is used predominantly as the target in clinical stage vaccines and antibody candidates, with preliminary positive clinical responses reported 18, 19 .

To obtain more precise resolution of the RBD-specific NAb response, a panel of RBD specific hmAbs were isolated and their molecular features, reactivity profiles and in vitro and in J o u r n a l P r e -p r o o f vivo anti-viral activities were defined. Several high affinity hmAbs with modest somatic hypermutation and potent SARS-CoV-2 neutralizing activity were isolated from IgG, IgA, and IgM memory B cells. In this report, we show that 1212C2 hmAb demonstrated substantial prophylactic and therapeutic activity in hamsters when delivered parenterally. Moreover, when delivered as inhaled liquid aerosols using a commercially available nebulizer, 1212C2 mediated eradication of lung viral load at a substantially higher dosing efficiency than the parenteral route of administration. The combination of a potent SARS-CoV-2 mAb and inhaled, local lung delivery using widely available nebulizers could provide for a promising treatment option.

To identify and isolate RBD-specific B cells, recombinant RBD protein was expressed, biotinylated, and used to form streptavidin-conjugated RBD tetramers to various fluorochromes.

Peripheral blood CD27+ memory B cells binding RBD were single cell sorted from convalescent SARS-CoV-2 patients and the immunoglobulin heavy and light chain variable regions of the B cells cloned to generate IgG1 recombinant hmAbs (Fig 1A) . Twenty hmAbs were isolated that exhibited substantial binding to SARS-CoV-2 RBD. These hmAbs all bound to recombinant SARS-CoV-2 RBD and S1 D614G proteins and exhibited varying reactivity SARS-CoV-2 S1S2 protein ( Fig 1B) . As expected, minimal to no reactivity was observed to SARS-CoV-2 Nucleocapsid (N) protein, SARS-CoV-1, or HepG2 cell lysate for most hmAbs, indicating high specificity for SARS-CoV-2 RBD and minimal off-target binding. REGN10987, REGN10933, and CB6/JS016 hmAbs were synthesized and included as positive controls 17, 20 . To ascertain the avidity of the hmAbs for SARS-CoV-2 RBD, their binding stability was determined in the presence of the chaotropic agent 8M urea. The hmAbs 1206D1, 1212C2, 1212F5, and 1215D1 retained at least 50% of their binding activity to SARS-CoV-2 RBD and S protein (Fig 1C) .

Binding affinity to RBD was further tested for a subset of the hmAbs by surface plasmon resonance (SPR), demonstrating a variety of high affinity hmAbs exhibiting equilibrium dissociation constants (KD) ranging from 71 pM to ~10 nM (Fig 1D and Fig S1) . Several of the hmAbs (1212C2, 1215D1, 1215D5) exhibited ~8-fold or higher affinity to RBD than control hmAbs (REGN10987, REGN10933 20 , and CB6/JS016 17 ), predominantly due to their slower off rates (kd). These results indicate that this panel of hmAbs recognize SARS-CoV-2 RBD specifically and with high affinity.

J o u r n a l P r e -p r o o f

The process of entry into a susceptible host cell is an important determinant of infectivity and pathogenesis of viruses, including coronaviruses 21, 22 . SARS-CoV-2 relies on the ability of its S glycoprotein to bind to the ACE2 receptor through its RBD driving a conformational change that culminates in the fusion of the viral envelope with the host cell membrane, and cell entry 23 . The hmAbs were tested for neutralization of live SARS-CoV-2 using a virus plaque reduction microneutralization (PRMNT) assay we previously described 24 , with the hmAb and virus preincubated prior to culture with susceptible Vero E6 cells (pre-treatment) or allowing virus adsorption to Vero E6 cells to occur for 1 hour prior to addition of the hmAb (post-treatment).

This gives an opportunity for the virus to initiate viral entry by binding to the cell surface receptor, potentially distinguishing the hmAb's ability to preferentially block later steps of virus entry into the cell or by inhibiting the cell-to-cell spread of virus progeny. The panel of hmAbs neutralized SARS-CoV-2 in both pre-and post-treatment conditions at NT 50 of 1 µg/ml or less, with 1212F5, 1212C2, and 1213H7 exhibiting the highest potency, with NT 50 of 100 ng/ml or less (Fig 2A) . The hmAb 1215D1 was more effective in neutralizing in post-treatment (NT 50 = 59 ng/ml) compared to pre-treatment (NT 50 = 226 ng/ml). The 1212C2 hmAb was particularly potent (pre-treatment NT 50 = 10 ng/ml, post-treatment NT 50 = 22 ng/ml), with similar neutralizing activity to CB6/JS06, REGN10987, and REGN10933 (Fig 2B) . Additionally, testing of 1212C2 in a SARS-CoV-2 VSV vectored pseudovirus assay confirmed its potent neutralizing activity (NT 50 = 1.9 ng/ml) that was also similar to REGN10987 and CB6/JS016 ( Table S1 ).

The panel of hmAbs clearly recognized SARS-CoV-2-infected Vero E6 cells as evident by immunofluorescence (Fig 2C and Fig S2) . No background staining of mock-infected cells was evident with any of the hmAbs, consistent with their high affinity specificity for SARS-CoV-2 S.

The hmAbs also exhibited binding to SARS-CoV-2 viral lysate (Fig S3) and SARS-CoV-2 J o u r n a l P r e -p r o o f D614G infected cells (Fig S4) . The neutralizing activity of 1212C2 against SARS-CoV-2 D614G was confirmed (Fig S4) . The ability of the hmAbs to directly inhibit the binding of RBD to ACE2 was determined using HEK293 cells overexpressing ACE2. All of the tested RBD-specific hmAbs inhibited the binding of recombinant SARS-CoV-2 RBD protein to the ACE2 expressing cells (Fig 2D) . Inhibition was nearly complete with the exceptions of 1206D1, 1207B4, and 1215D1. These results demonstrate the potent in vitro neutralizing and binding activity of these SARS-CoV-2 RBD specific hmAbs.

The most potent hmAbs, 1212C2 and 1212F5 were both isolated from IgM+ B cells, belong to the same clonal lineage that utilizes the VH1-2 heavy chain gene, and exhibited modest somatic hypermutation, with 1212C2 further mutated from germline compared to 1212F5 (8.2% vs 6.1% amino acid VH) ( Table 1) . Most of the hmAbs were isolated from IgG1 expressing B cells, while 1215D1 and 1212C8 were isolated from IgA expressing B cells. All hmAbs exhibited somatic hypermutation (2.0%-9.1% VH) suggesting they arose from multiple rounds of germinal center reactions. VH3-66 and Vk1-9 gene usage was dominant among the hmAbs. Targeted VH-deep sequencing of the 1212C2/1212F5 clonal lineage from contemporary peripheral blood B cells identified numerous members (Fig 3) . The lineage was dominated by IgM and IgG expressing B cells, with several expanded nodes of identical clones being expressed by both IgM and IgG expressing B cells.

SPR epitope mapping was performed and used to cluster the hmAbs into five major epitopes Fig 4) . With one exception, all isolated hmAbs are part of the A epitope, that includes 1212C2. Thus, 1212C2 efficiently blocked all hmAbs from binding to RBD, except 1215B11 (E J o u r n a l P r e -p r o o f epitope), whose epitope overlaps with CR0322. Within the footprint of the A epitope there are several sub-epitopes. In particular, we highlight hmAbs 1207B4/1215D1 (B-epitope) and 1213H7 (C-epitope) that exhibit essentially no overlap with one another, based on their ability to simultaneously bind RBD at greater than 90% of their expected binding levels. Consistent with their distinct classification, B-epitope hmAbs exhibit a reduced ability to block RBD attachment to ACE2 expressing cells (Fig 2D) .

The in vivo activity of 1212C2 and 1206D1 was evaluated in the golden Syrian hamster model of SARS-CoV-2 infection 25 . 1212C2 was chosen based on its potent in vitro neutralizing activity and high affinity, while 1206D1 was chosen based on its in vitro neutralizing activity and distinct affinity and reactivity profile from 1212C2.

To test for prophylactic activity, 10 mg/kg of hmAb was administered by intraperitoneal (IP) injection 6 hours prior to intranasal (IN) challenge with SARS-CoV-2. At 2 days post infection (dpi), all PBS control and isotype control hamsters had detectable live virus as measured by plaque assay in their nasal turbinates and lungs. In contrast at 2 dpi, hamsters that received 1212C2 already started to exhibit meaningful viral load reduction in their nasal turbinate and lungs (Fig 5A) . At 4 dpi virus was detected in the nasal turbinates and lungs of all hamsters in the PBS and isotype control groups, although an overall decrease compared to 2 dpi, consistent with the viral dynamics of SARS-CoV-2 infection in hamsters 25 . In comparison to the control groups at 4 dpi, prophylactically treated 1212C2 hamsters exhibited eradication of viral loads in the nasal turbinates and lungs in 3 of 4 animals. Consistent with the viral load reduction, 1212C2 treated animals exhibited significantly less lung pathology compared to the PBS treated group at 2 dpi (p=0.0334) and 4dpi (p=0.0004), with this reduction also reaching significance compared to the isotype control group at 4 dpi (p<0.0001) (Fig 5C) . There was J o u r n a l P r e -p r o o f ~80% reduction in lung pathology at 4 dpi when 1212C2 was given prophylactically. 1206D1 hmAb exhibited modest activity, noted by 50% of hamsters having no detectable virus in the nasal turbinates, and all 1206D1 hamsters having detectable virus in the lungs, although trending to lower titers compared to PBS and isotype control groups. These in vivo results are consistent with the lower in vitro viral neutralization activity of 1206D1 in comparison to 1212C2 (Fig 2A) . Overall, 1212C2 demonstrated substantial prophylactic activity as evident by sterilizing protection in 63% of hamsters.

Therapeutic activity was tested by treatment with 25 mg/kg of 1212C2 6 hours following intranasal infection of hamsters with SARS-CoV-2 and evaluating viral burden at 4 dpi only due to limited availability of hmAb. SARS-CoV-2 was detected in the nasal turbinates in 3 of 4 animals in the PBS and isotype control groups. No virus was detected in the nasal turbinates of the 1212C2 treated hamsters. SARS-CoV-2 was detected in the lungs of all PBS and isotype control treated hamsters, but only in 1 of 4 of the 1212C2 treated hamsters (Fig 5B) . Overall, 1212C2 demonstrated substantial therapeutic activity, reducing virus to undetectable levels in 75% of the treated hamsters. SARS-CoV-2 infection in hamsters results in gross lung lesions, including consolidation, congestion, and atelectasis 25 . Therapeutically, significantly less lung pathology was also observed in the 1212C2 treated hamster group compared to the PBS treated group (p=0.0136), and ~58% reduction in lung pathology in 1212C2 group compared to the control hamsters (Fig 5D) . Importantly, these differences observed in lung lesions are consistent with the viral burden seen in the upper and lower respiratory tract (Fig 5A and 5B) .

Overall, these results demonstrate the ability of 1212C2 to substantially reduce viral burden and lung pathology of SARS-CoV-2 infection when used either prophylactically or therapeutically. (Table S2) . However, 42 hours later, the hmAb was detectable in the BAL, suggesting that IP injected hmAb gradually penetrates the lungs from the serum compartment. Inhalation (IH) administration of the hmAb using liquid aerosols delivered whole body to the animals from a commercially available nebulizer (Aerogen Aeroneb Solo nebulizer) showed that approximately 1.7% of the inhaled dose was deposited in the BAL. The low percentage of the inhaled dose that is deposited in the lungs of hamsters is in line with prior lung uptake studies for the approximately 4 µm diameter droplets produced by the nebulizer used here (volume mean diameter measured by laser diffraction 4.1 µm), and is a result of a higher degree of inertial impaction of liquid aerosols in the upper airways of small animals 27 . Nevertheless, such lung deposited (BAL) dose is still substantially higher at both time points than the BAL concentration achieved with the IP route. A higher lung deposited dose afforded by the inhaled route demonstrates higher delivery efficiency to the lungs than the IP route. At 42 hours post-inhaled dose, most of the mAb is cleared from the BAL (~85% less than 30 minutes post-inhalation), J o u r n a l P r e -p r o o f which appears to be faster clearance rate than an expected lung half-life of ~8 days reported in other studies (45) .

The Fc of 1212C2 was modified with the LALA mutation to reduce FcR binding and subsequent Fc-mediated effector functions 28 , and further modified to increase half-life 29 (referred to as '1212C2-HLE-LALA'). To determine the efficacy of inhaled 1212C2 and evaluate the therapeutic dose-response, hamsters were infected intranasally with 2x10 5 PFU SARS-CoV-2 and treated with a single dose of 1212C2 hmAb or isotype control hmAb 12 hours later using IH or IP routes. As measures of efficacy, body weight was recorded, pulmonary lung lesions were measured, and viral titers quantitated from nasal turbinates and lungs on days 2 and 4 following infection. At 2 dpi virus was detected in the nasal turbinates and lungs of all infected control hamsters (PBS and isotype control hmAb) (Fig 6A) . Virus was not detectable in the lungs of any hamsters treated via inhalation with 1212C2-HLE-LALA at 16.3 mg/kg, and virus was lower in their nasal turbinates compared to controls. In contrast, virus was detectable at 2 dpi in the lungs of 50% of the hamsters treated IP with 1212C2-HLE-LALA at 25 mg/kg, and not significantly decreased in nasal turbinates compared to controls. At 2 dpi, virus was only detectable in the lungs of 25% and 50% of the hamsters treated via inhalation with a lung delivered dose of 3.2 mg/kg and 0.6 mg/kg of 1212C2-HLE-LALA, respectively. At 2 dpi, 16.3 mg/kg and 3.2 mg/kg inhaled 1212C2-HLE-LALA was superior in reducing viral titer in the lungs compared to IP administration. At 4 dpi virus was detected in the nasal turbinates and lungs of all infected control hamsters (saline and isotype hmAb), with a trend toward modest non-specific reduction in viral J o u r n a l P r e -p r o o f titer in nasal turbinates only with isotype control hmAb (Fig 6B) . Viral titers were only sporadically detected in the nasal turbinates of the 1212C2 treated groups, with the exception of detectable virus in all of the 0.6 mg/kg inhaled 1212C2-HLE-LALA group. At 4 dpi there was no detectable viral titer in the lungs of any 1212C2 treated hamsters, including the 0.6 mg/kg IH 1212C2-HLE-LALA group, representing at least a 2-log reduction compared to control groups.

Control PBS treated hamsters exhibited approximately 10% maximum weight loss, 1212C2 treated animals did not have appreciable weight loss (Fig S5) . Lung pathology was evident at 4 dpi, particularly in the control groups, however overall, relative to non-infected hamsters, no lung pathology was evident in 9/24 (37.5%) of the 1212C2 treated hamsters, compared with lung pathology evident in 11/12 (91.7%) of the control treated infected hamsters (Fig 6C) . At 4 dpi lung lesions were significantly decreased in hamsters that were treated with 3.2 mg/kg (p=0.0028) of inhaled 1212C2 hmAb compared to PBS treated hamsters.

With regard to effector function removal brought about by the addition of the LALA mutation, animals treated 1212C2-HLE-LALA appeared to achieve comparable viral reduction as the un-modified 1212C2 mAb. Overall these results confirm the therapeutic activity of 1212C2 hmAb against SARS-CoV-2 infection and suggest increased efficacy of inhaled 1212C2 hmAb, with elimination of viral lung burden with a single inhaled dose of just 0.6 mg/kg. As shown in Table S2 , the lung deposited dose in hamsters is approximately 1.7% of the inhaled dose. This implies that elimination of lung viral burden was achieved at day 4 post-infection with a lung dose of 0.01 mg/kg. Further, if the relevant translation between species to achieve an efficacious dose is a normalization based on a lung-deposited dose per kilogram of lung weight, then, combined with the expectation that the at least 50% of the inhaled dose from a similar commercial nebulizer would be deposited in the lung of a human, the human equivalent efficacious inhaled dose would be 0.03 mg per kg body weight (assumptions: 110g hamster with 1 gram lung weight; 70 kg human with 1000g lung weight).

The SARS-CoV-2 global pandemic continues without an optimal targeted intervention to treat the infection. Our results clearly demonstrate that SARS-CoV-2 infection can result in memory B cells encoding high affinity, high potency NAbs specific for the RBD. The 1212C2 hmAb is able to significantly reduce viral burden in SARS-CoV-2 infected hamsters when used either prophylactically or therapeutically. The in vitro affinity and the neutralization activity of 1212C2 compared favorably to hmAbs that are in late stage clinical trials to treat SARS-CoV-2 (i.e. REGN10987, REGN10933, and CB6/JS016), suggesting that 1212C2 could have clinical activity. Of note is that 1212C2 bound consistently across RBD domain, S1 subunit, and fulllength spike protein (S1S2), which is a property that is additionally different from that of the REGN10987, REGN10933, and CB6/JS016 mAbs (Fig 1B) .

The resulting panel of SARS-CoV-2 neutralizing hmAbs indicate that RBD-specific IgG, IgA, and IgM memory B cells develop after infection, however, exhibit only modest somatic hypermutation, which is consistent with an acute primary immune response. Several hmAbs, including 1212C2, 1212F5, 1213H7, and 1212F2 despite only modest somatic hypermutation (4.1-8.2% VH) have remarkably high affinity for SARS-CoV-2 RBD (104-1180 pM KD) and potent neutralization (<200 ng/ml NT 50 ). The frequent usage of VH3-66 and Vk1-9 among the hmAbs may represent preferential RBD specificity among those germline genes. 1212C2 and 1212F5, which are members of the same clonal lineage, were isolated from IgM memory B cells, and utilize the VH1-2 heavy chain gene. VH1-2 usage is a common feature of other antibodies targeting viral glycoproteins, including HIV envelope [30] [31] [32] . VH1-2 usage is more pronounced among marginal zone B cells than naive B cells or switched memory B cells and increased among splenic marginal zone B cell lymphoma [33] [34] [35] [36] . This may indicate that the 1212C2 lineage arose from a marginal zone B cell response.

J o u r n a l P r e -p r o o f 1212C2 has higher affinity (KD =104 pM) for SARS-CoV-2 RBD compared to 1212F5 (KD =841 pM) which is consistent with its further somatic hypermutation. Additional lineage members were identified, with greater somatic hypermutation, and may be used to guide rational improvement of 1212C2 affinity, neutralization, and ultimate efficacy against SARS-

The resulting hmAbs all neutralized SARS-CoV-2 at NT 50 of 1 µg/ml or less, indicating their utility for investigating the mechanisms and epitopes mediating SARS-CoV-2 RBD targeted anti-viral activity. Although there was overall consistency that the hmAbs neutralized SARS-CoV-2 either when pre-incubated with the virus (pre-treatment) or after the cells were infected with the virus for 1 h (post-treatment), particularly 1212C2, 1212F5, and 1213H7 potently neutralized (NT 50 <100 ng/ml) SARS-CoV-2 in both conditions, discordance was noted. 1206D1, which was tested in vivo, was approximately 2-fold less effective neutralizing in post-treatment than pre-treatment, which may contribute to its limited in vivo activity. In contrast, 1215D1, which had the highest affinity for RBD (KD = 71 pM) is approximately 5-fold more effective in neutralizing in post-treatment than pre-treatment. It is expected that the post-treatment neutralization assay may identify antibodies that are better able to mitigate cell-to-cell virus spreading, including antibodies targeting the fusiogenic activity of the S2 domain.

Subsequently, 1215D1 should be evaluated for its in vivo efficacy, and discern if the posttreatment neutralization assay is a more sensitive indicator of in vivo efficacy. Epitope mapping suggests that 1212C2 has a large footprint on the RBD, blocking the binding of several other neutralizing hmAbs encoded by diverse heavy and light chain variable region genes. Efforts are ongoing to solve the 1212C2 -RBD structure to adequately define the epitope.

The high affinity binding of 1212C2 to SARS-CoV-2 RBD, mediating its ability to block RBD attachment to ACE2, and subsequent potent neutralization of SARS-CoV-2 even when added after virus has been added to culture, demonstrate its direct and substantial anti-viral As the portal of entry for SARS-CoV-2 virus is the respiratory tract, with the lungs serving as the key target organ for pathogenesis, delivering directly the mAbs to the lungs using inhalation is a logical approach. While aerosols delivery of drugs to the lungs is more inefficient in small animals as compared to humans (~1% of the inhaled dose is deposited in the lungs for rodents as compared to ~50% in humans), our data showed that inhaled delivery of mAbs to the lung BAL in hamsters is still substantially more efficient than that achieved using the IP route (Table S2) , When comparing the inhaled route to the IP route in the hamster challenge study, therapeutic efficacy was achieved at a higher efficiency than was observed for the inhaled route.

Inhaled administration of 1212C2 hmAb resulted in sterilizing therapeutic protection at all tested doses, with the lowest inhaled dose of 0.6 mg/kg which corresponds to 0.01 mg/kg of lung deposited dose (or a human equivalent inhaled dose of 0.03 mg/kg). In contrast, mAbs against SARS-CoV-2 that have been reported to date required at least 5 mg/kg or above to achieve therapeutic efficacy when administered parenterally 17, 37-39 . Therefore, inhaled 1212C2 has the potential to profoundly facilitate dose-sparing and treatment coverage compared to conventional parenteral administration. There are several commercial inhaled protein therapeutic products and many more inhaled protein therapeutics that are in various stages of clinical evaluations, most with attractive safety profile. Several mAbs have also been clinically evaluated as inhaled J o u r n a l P r e -p r o o f aerosols with demonstrated preliminary safety and tolerability 6, 40-42 , suggesting the practicality of formulating 1212C2 for self-administered inhalation. Over 90% of all symptomatic COVID-19 patients are not hospitalized, but they all still need treatment to minimize the potential for transmission and limit complications from viral infection. This is a large unmet COVID-19 population and having a convenient self-administered dose that can be done on an outpatient basis or at home using commercially available nebulizer devices could materially impact treatment coverage, reducing transmissibility, and ultimately the viral burden of the population.

Epitope mapping studies were performed by injecting multiple (n=6) hmAbs sequentially over the same immobilized RBD surface. While this mapping format can obscure details of specific hmAb epitopes, our goal was to rapidly identify distinct epitopes that could simultaneously bind to RBD. In aggregate, the mapping studies successfully identify four major Ab classes, with the 1212C2 overlapping all hmAbs, except 1215B11. More precise epitope resolution approaches such as x-ray crystallography would be needed to accurately determine critical binding residues and epitope relation to other RBD-specific neutralizing mAbs reported. The recent emergence of SARS-CoV-2 variants precluded testing of the described hmAbs against these viruses, and such testing would inform the clinical utility of the hmAbs. Only a single animal model, the hamster was used for this study, and testing in another animal model such as the lethal K18 Relationship between lineage members determined by amino acid sequence similarity. 

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, James Kobie (jjkobie@uabmc.edu).

Limited quantities of newly generated materials associated with the paper are available under MTA.

All relevant data are included within the manuscript and are available without restriction from the Lead Contact upon request. 

Peripheral blood mononuclear cells (PBMC) were isolated by density gradient centrifugation and cryopreserved. SARS-CoV-2 recombinant protein was generated, briefly SARS-CoV-2 Spike RBD domain consisted of residues Thr-333 to Thr-531 (uniprot P0DT2). The RBD sequence was cloned in frame with a his8 and avi-tag sequence, respectively (RBDhisavi). RBDhisavi 

Single B cells were sorted using a FACSMelody (BD Biosciences) into 96-well PCR plates and immediately frozen at −80 °C until thawed for reverse transcription and nested PCR performed for IgH, Igλ, and Igκ variable gene transcripts as previously described 30, 43 . Paired heavy and light chain genes cloned into IgG1 expression vectors and were transfected into HEK293T cells and culture supernatant was concentrated using 100,000 MWCO Amicon Ultra centrifugal filters (Millipore-Sigma, Cork, Ireland), and IgG captured and eluted from Magne Protein A beads (Promega, Madison, WI) as previously described 30, 43 . The hmAb 1069D6 or a human myeloma IgG1 (BioXcell, Labanon, NH) were used as an isotype control. 1212C2 was also generated with the incorporation of the LALA mutation to diminish Fc receptor binding 44 and a mutation described to increase half-life 29 . Immunoglobulin sequences were analyzed by IgBlast (www.ncbi.nlm.nih.gov/igblast) and IMGT/V-QUEST (http://www.imgt.org/IMGT_vquest/vquest) to determine which sequences should lead to productive immunoglobulin, to identify the germline V(D)J gene segments with the highest identity, and to scrutinize sequence properties. hpi, infected cells were fixed with 10% neutral formalin for 24 h and were immune-stained using anti-NP monoclonal 1C7C7 antibody 46 . Virus neutralization was evaluated and quantified using ELISPOT, and the percentage of infectivity calculated using sigmoidal dose response curves.

The formula to calculate percent viral infection for each concentration is given as [(Average # of plaques from each treated wells -average # of plaques from "no virus" wells)/(average # of plaques from "virus only" wells -average # of plaques from "no virus" wells)] x 100. A non-linear regression curve fit analysis over the dilution curve can be performed using GraphPad Prism to calculate NT 50. Mock-infected cells and viruses in the absence of hmAb were used as internal controls. hmAbs were also tested using a SARS-CoV-2 Spike protein pseudotyped virus (PsV) containing the gene for firefly luciferase. Virus neutralization can be measured by the reduction of luciferase expression. VeroE6/TMPRSS2 cells were seeded at 2 x 10 4 cells/well in opaque plates (Greiner 655083). The next day, PsV corresponding to 1-10 x 10 6 luciferase units was mixed in Opti-MEM with dilutions of hmAbs and incubated at RT for 1 h. Media was removed from the cells and 100 µl/well of the hmAb/PsV mix was added in triplicates. After 1 h incubation at 37C and 5% CO 2 , another 100 µl of media containing 2% FBS, was added, and cells were incubated for 24 more hrs. After this time, luciferase activity was measured using Passive Lysis Buffer (Promega E1941) and Luciferase substrate (Promega E151A) following the J o u r n a l P r e -p r o o f manufacturer's instructions. Neutralization was calculated as the percent reduction of luciferase readings as compared to no-antibody-controls.

We have previously demonstrated that golden Syrian hamsters (Mesocricetus auratus) are susceptible to SARS-CoV2, showing loss of body weight and viral replication in the lungs and nasal turbinate 25 . In a prophylactic experiment, 6-week old, female hamsters were IP injection of 10 mg/kg of hmAb 1212C2 (n=6), 1206D1 (n=3), human IgG1 Isotype control mAb (BioXcell, Labanon, NH; n=6), or PBS (n=5). Six hours later, hamsters were given 2x10 5 PFU of SARS-CoV-2, intranasally. Three animals were also mock infected with PBS at this time to serve as a negative control. For each day, lungs and nasal turbinates were collected removed and photographed on white paper towels to look for gross pathology using a macroscopic pathology scoring analysis measure the distributions of pathological lesions, including consolidation, congestion, and pneumonic lesions using ImageJ software (NIH). Resulting pathology were represented as the percent of the total lung surface area. Left side of the lungs were used for histopathology and the right side was used for viral titers.

The hmAb 1212C2 was also assessed in an in vivo therapeutic experiment. At the start of the experiment, 12, 6-week old, female hamsters were administered 2x10 5 PFU of SARS-CoV-2, intranasally. Four animals were mock infected with PBS at the same time. Six hours later, animals were given either 25 mg/kg hmAb 1212C2 (n=4), 25 mg/kg isotype control (n=4), or PBS (n=4) IP. Animals were euthanized at 4 dpi and lungs and nasal turbinates collected for histopathology and viral titers as described for the prophylactic experiment.

J o u r n a l P r e -p r o o f

The therapeutic testing of 1212C2 was tested using 5-week old, female hamsters. Hamsters were challenged with 2x10 5 PFU of SARS-CoV2 IN and 12 hours later treated with hmAb IP or IH. The estimated inhaled dose of hmAb is measured by sampling the aerosolized atmosphere the hamsters are exposed at a known fixed flowrate during the entire course of their exposure.

The gas drawn out of the exposure chamber by vacuum passes through a collection filter to capture the aerosolized droplets. The amount of mAb deposited in the filter is determined using A280 following extraction in formulation buffer, which determines the mAb aerosol concentration in the exposure chamber. The approximate inhaled dose over a 30 minutes whole body exposure is then derived from the aerosol concentration and minute volume of the hamsters.

The target high dose level of nonspecific IgG or 1212C2-HLE-LALA was 25 mg/kg. The actual dose measured was 11.30 mg/kg for the non-specific IgG group and 16.3 mg/kg for the 1212C2-HLE-LALA group ( Table 2) .

The mid and low dose groups for 1212C2-HLE-LALA were 5-fold and 25-fold dilutions of the mAb solution used in the high dose, respectively, which correspond to 3.2 mg/kg and 0.6 mg/kg.

10 million PBMCs were thawed and used for RNA isolation (Qiagen, RNeasy Mini Kit).

Removal of any residual genomic DNA was performed using the Turbo DNA-Free kit (Invitrogen) and cDNA was generated using the qScript cDNA synthesis kit (QuantaBio). Two rounds of PCRs were performed to generate the libraries with either global (VH1 -VH6) or individual VH targeted forward primers as previously described 30, 43 with the modifications adapted for Illumina Nextera approach. The final PCR products were gel extracted (QIAQuick, Qiagen, Hilden, Germany) and further purified using the ProNex size-selective purification system (Promega) to select products between 500-700 bp range. Libraries were submitted to J o u r n a l P r e -p r o o f the Heflin Center for Genomic Sciences at the University of Alabama at Birmingham where DNA quantification was made by qPCR and sequenced on an Illumina MiSeq system (Illumina, Inc., CA, USA) using 2 × 300 bp paired-end kits (Illumina MiSeq Reagent Kit v3). Sequence analysis and assembly of lineage trees were performed using an in-house custom analysis pipeline as previously described 43, 47 . All sequences were aligned using IMGT.org/HighVquest 48 . Lineage trees were generated by identifying the lineage (the cluster of sequences with identical VH, JH, and HCDR3 lengths and ≥85% HCDR3 similarity) containing the corresponding hmAb sequence.

Significance was determined using GraphPad Prism, v8.0. Two-tailed t-tests were applied for evaluation of the results between treatments. Bonferroni correction for multi-outcomes was applied to the therapeutic treatment experiment of hamsters utilizing multiple hmAb doses and routes of administration (n= 9 outcomes), p<0.0055 was considered significant. For statistical analysis viral titers were log transformed and undetectable virus was set to the limit of detection (200 PFU/ml).

J o u r n a l P r e -p r o o f

Obesity and COVID-19 Severity in a Designated Hospital

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Preliminary Estimates of the Prevalence of Selected Underlying Health Conditions Among Patients with Coronavirus Disease 2019 -United States

Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Characteristics of Hospitalized COVID-19 Patients Discharged and Experiencing Same-Hospital Readmission -United States

Disparities in COVID-19 Incidence, Hospitalizations, and Testing, by Area-Level Deprivation -Utah

Ad26 vaccine protects against SARS-CoV-2 severe clinical disease in hamsters

Neutralizing antibodies correlate with protection from SARS-CoV-2 in humans during a fishery vessel outbreak with high attack rate

Replication-Competent Vesicular Stomatitis Virus Vaccine Vector Protects against SARS-CoV-2-Mediated Pathogenesis in Mice

Ultrapotent human antibodies protect against SARS-CoV-2 challenge via multiple mechanisms

Structural basis for neutralization of SARS-CoV-2 and SARS-CoV by a potent therapeutic antibody

Potent neutralizing antibodies against multiple epitopes on SARS-CoV-2 spike

Factors Associated with Good Patient Outcomes Following

Convalescent Plasma in COVID-19: A Prospective Phase II Clinical Trial. Infect Dis Ther e-pub ahead of print

Convalescent plasma anti-SARS-CoV-2 spike protein ectodomain and receptor binding domain IgG correlate with virus neutralization

A human neutralizing antibody targets the receptor-binding site of SARS-CoV-2

What Pfizer's landmark COVID vaccine results mean for the pandemic. Nature e-pub ahead of print Nov 9

COVID vaccine excitement builds as Moderna reports third positive result. Nature e-pub ahead of print

Studies in humanized mice and convalescent humans yield a SARS-CoV-2 antibody cocktail

Structure, Function, and Evolution of Coronavirus Spike Proteins

Coronaviruses post-SARS: update on replication and pathogenesis

Cell entry mechanisms of SARS-CoV-2

PRESS)). Rapid in vitro assays for screening neutralizing antibodies and antivirals against SARS-CoV-2

Rescue of SARS-CoV-2 from a Single Bacterial Artificial Chromosome

Randomized, Double-Blind, Placebo-Controlled, Single-Ascending-Dose Study of the Penetration of a Monoclonal Antibody Combination (ASN100) Targeting Staphylococcus aureus Cytotoxins in the Lung Epithelial Lining Fluid of Healthy Volunteers

Regional Deposition of Inhaled Monodisperse Coarse and Fine Aerosol Particles in Small Laboratory Animals

Is COVID-19 receiving ADE from other coronaviruses?

Extending human IgG half-life using structure-guided design

Persistence of HIV-1 Env-Specific Plasmablast Lineages in Plasma Cells after Vaccination in Humans

Characteristics of Plasmablast Repertoire in Chronically HIV-Infected Individuals for Immunoglobulin H and L Chain Profiled by Single-Cell Analysis

Structural basis for HIV-1 gp120 recognition by a germ-line version of a broadly neutralizing antibody

Immunoglobulin gene repertoire diversification and selection in the stomach -from gastritis to gastric lymphomas

analysis of immunoglobulin genes in the tumor B cells in splenic marginal zone lymphoma

Over 30% of patients with splenic marginal zone lymphoma express the same immunoglobulin heavy variable gene: ontogenetic implications

Mouse marginal zone B cells harbor specificities similar to human broadly neutralizing HIV antibodies

Potently neutralizing and protective human antibodies against SARS-CoV-2

REGN-COV2 antibodies prevent and treat SARS-CoV-2 infection in rhesus macaques and hamsters

A Therapeutic Non-self-reactive SARS-CoV-2 Antibody Protects from Lung Pathology in a COVID-19 Hamster Model

Randomized study of the safety and pharmacodynamics of inhaled interleukin-13 monoclonal antibody fragment VR942

Designing inhaled protein therapeutics for topical lung delivery: what are the next steps?

A Highly Potent and Broadly Neutralizing H1 Influenza-Specific Human Monoclonal Antibody

Effector function activities of a panel of mutants of a broadly neutralizing antibody against human immunodeficiency virus type 1

Human monoclonal antibody combination against SARS coronavirus: synergy and coverage of escape mutants

Rapid in vitro assays for screening neutralizing antibodies and antivirals against SARS-CoV-2

Diversity, cellular origin and autoreactivity of antibodysecreting cell population expansions in acute systemic lupus erythematosus

IMGT((R)) tools for the nucleotide analysis of immunoglobulin (IG) and T cell receptor (TR) V-(D)-J repertoires