Spike mutation D614G alters SARS-CoV-2 fitness and neutralization susceptibility

A spike protein mutation D614G became dominant in SARS-CoV-2 during the COVID-19 pandemic. However, the mutational impact on viral spread and vaccine efficacy remains to be defined. Here we engineer the D614G mutation in the SARS-CoV-2 USA-WA1/2020 strain and characterize its effect on viral replication, pathogenesis, and antibody neutralization. The D614G mutation significantly enhances SARS-CoV-2 replication on human lung epithelial cells and primary human airway tissues, through an improved infectivity of virions with the spike receptor-binding domain in an “up” conformation for binding to ACE2 receptor. Hamsters infected with D614 or G614 variants developed similar levels of weight loss. However, the G614 virus produced higher infectious titers in the nasal washes and trachea, but not lungs, than the D614 virus. The hamster results confirm clinical evidence that the D614G mutation enhances viral loads in the upper respiratory tract of COVID-19 patients and may increases transmission. For antibody neutralization, sera from D614 virus-infected hamsters consistently exhibit higher neutralization titers against G614 virus than those against D614 virus, indicating that (i) the mutation may not reduce the ability of vaccines in clinical trials to protect against COVID-19 and (ii) therapeutic antibodies should be tested against the circulating G614 virus before clinical development.


Introduction
Since the emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in China in late 2019 1 , coronavirus disease 2019 (COVID-19) has caused > 25 million con rmed infections and > 850,000 fatalities worldwide. Hospitals and public health systems were overwhelmed rst in Wuhan, followed by Italy, Spain, New York City, and other major cities, before cases peaked in these locations. Although most infections are mild, SARS-CoV-2 can cause severe, life-threatening pneumonia, particularly in older age groups and those with chronic pulmonary and cardiac conditions, diabetes, and other comorbidities. The exact mechanisms of severe disease remain unclear but typically involve a dysregulated, hyperin ammatory response following the initial stages of viral infection 2 . However, in addition to the host response, variation in viral strain phenotypes could also contribute to disease severity and spread e ciency.
Coronaviruses have evolved a genetic proofreading mechanism to maintain their long RNA genomes 3 .
Despite the low sequence diversity of SARS-CoV-2 4 , mutations that mediate amino acid substitutions in the spike protein, which interacts with cellular receptors such as angiotensin-converting enzyme 2 (ACE2) to mediate entry into cells, can strongly in uence host range, tissue tropism, and pathogenesis. During the SARS-CoV outbreak in 2002-2003, one such mutation was shown to mediate adaptation for infection of the intermediate civet host as well as for interhuman transmission 5 . For SARS-CoV-2, analyses of over 28,000 spike protein gene sequences in late May 2020 revealed a D614G amino acid substitution that was rare before March but increased in frequency as the pandemic spread 6 , reaching over 74% of all published sequences by June 2020 7 . The D614G substitution was accompanied by three other mutations: a C-to-T mutation in the 5' untranslated genome region at position 241, a synonymous C-to-T mutation at position 3,037, and a nonsynonymous C-to-T mutation at position 14,408 in the RNAdependent RNA polymerase gene 8 . This set of mutations not only increased globally, but during cocirculation within individual regions during outbreaks, suggesting a tness advantage rather than simply founder effects or genetic drift. The association of spike protein amino acid substitutions with coronavirus transmissibility suggested that the D614G substitution was critical to this putative selective sweep. The correlation of this mutation with higher nasopharyngeal viral RNA loads in COVID-19 patients 6,9 also supported a putative advantage of the mutant in transmission, which is key for viral tness. However, direct measurements of tness were needed to con rm this hypothesis.
Initial phenotypic characterizations of the D614G spike substitution were performed using pseudotyped viruses, whereby vesicular stomatitis virus (VSV) and lentiviral particles incorporating the SARS-CoV-2 spike protein alone were studied by replication kinetics. The production of signi cantly higher pseudotyped viral titers in multiple cell types by the G614 spike variant suggested that this substitution could be associated with enhanced entry into cells and replication in the airways of infected patients 6,7 . However, these results need to be con rmed in studies with authentic SARS-CoV-2 containing the spike 614 variant, and also using in vivo studies with a suitable animal model. Therefore, using an infectious cDNA clone for SARS-CoV-2 10 , we generated the D614G substitution in the January 2020 USA-WA1/2020 strain 11 and performed experimental comparisons using in vitro cell culture, a primary human 3D airway tissue, and a hamster infection model 12 . We also developed a pair of D614 and G614 mNeonGreen SARS-CoV-2 viruses that could be used for rapid neutralization testing of serum specimens and monoclonal antibodies (mAbs). Using the reporter SARS-CoV-2 viruses, we analyzed the effect of D614G mutation on susceptibility to neutralization. Our study has important implications in understanding the evolution and transmission of SARS-CoV-2 as well as the development of COVID-19 vaccines and therapeutic antibodies.

Results
Enhancement of viral replication and infectivity by the spike D614G substitution in human lung epithelial cells. We rst examined the effect of the spike D614G substitution on viral replication in cell culture. A site-directed mutagenesis was performed on an infectious cDNA clone of SARS-CoV-2 to prepare a pair of recombinant isogeneic viruses with spike D614 or G614 (Fig. 1a). Similar infectious amounts of D614 and G614 viruses were recovered from Vero E6 cells (monkey kidney epithelial cells), with viral titers of 1 × 10 8 and 8 × 10 7 plaque-forming units (PFU)/ml, respectively. The two viruses formed similar plaque morphologies (Fig. 1b). In Vero E6 cells, the G614 virus replicated to a higher infectious titer than D614 at 12 h post-infection (hpi), after which the two viruses replicated to comparable levels (Fig. 1c). A similar trend was observed for extracellular viral RNA production from the infected Vero E6 cells (Fig. 1d). To compare the infectivity between the two viruses, we calculated the genomic RNA/PFU ratios; no signi cant differences were found (Fig. 1e), indicating that the D614G mutation does not affect viral replication or virion infectivity on Vero E6 cells.
Next, we compared the replication kinetics of D614 and G614 viruses on the human lung epithelial Calu-3 cells. After infection at a multiplicity of infection (MOI) of 0.01 PFU/ml, the G614 virus produced modest 1.2-, 2.4-, and 1.9-fold more infectious virus than the D614 virus at 24, 36, and 48 hpi, respectively ( Fig. 1f), indicating that D614G enhances viral replication. In contrast, the G614-infected cells produced less (at 24 and 36 hpi) or equivalent (at 48 hpi) extracellular viral RNA compared to D614-infected cells (Fig. 1g). The genomic RNA/PFU ratios of D614 virus were therefore 1.9-to 3.0-fold higher than those of G614 ( Fig. 1h), indicating that the D614G mutation increases the infectivity of SARS-CoV-2 produced from the human lung cell line.
To explore the mechanism of increased infectivity of G614 virus produced from Calu-3 cells, we compared the spike protein processing from D614 and G614 viruses. Virions were puri ed from the culture medium of infected Calu-3 using ultracentrifugation and a sucrose cushion. The pelleted viruses were analyzed for spike protein processing by Western blot, with nucleocapsid included as a loading control. For both viruses, full-length spike was almost completely processed to the S1/S2 cleavage form and S2', with comparable cleavage e ciencies of 93% for D614 and 95% for G614 (Fig. 1i). When virions produced from Vero E6 cells were analyzed, less full-length spike protein was processed to the S1/S2 form, with cleavage e ciencies of 73% for D614 and 67% for G614 (Fig. 1j). These results suggest that (i) more spike protein is cleaved to S1/S2 within virions produced from Calu-3 cells than those produced from Vero E6 cells and (ii) the D614G substitution does not signi cantly affect the spike cleavage ratio.
Increased tness in the hamster upper airway of SARS-CoV-2 with the D614G substitution. The in vivo relevance of the S-D614G mutation was evaluated in the golden Syrian hamster model (Extended Data   Fig. 1a). After intranasally infecting four-to ve-week-old hamsters with 2 × 10 4 PFU of D614 or G614 virus, animals from both groups exhibited similar mean weight losses (Fig. 2a). No visible illness was observed in either infected cohort. On day 2 post-infection (pi), infectious viral titers from nasal washes, trachea, and various lobes of the lung (Extended Data Fig. 1b) were consistently higher in the G614infected subjects compared to the D614-infected animals, although the differences did not reach statistical signi cance (Fig. 2b). The viral titer differences were greater in the upper airway samples than those in the lower airway tissues. On day 4 pi, the differences in infectious viral titers between the two viruses became more signi cant in the upper airway, with 1.3 log 10 PFU/ml higher G614 virus than D614 in nasal wash (Fig. 2c). The trachea had 0.9 log 10 PFU/g higher G614 virus than D614, but the statistical signi cance was lost upon correction for multiple comparisons. The viral loads in various lung lobes were nearly identical, with ≤ 0.1 log 10 PFU/g differential between the two viruses. No infectious virus was detected in any airway tissues on day 7 pi (data not shown). Overall, the results demonstrate that the D614G mutation leads to a higher infectious virus production and shedding in the upper airway of infected hamsters.
We compared the infectivity of the D614 and G614 viruses produced in hamsters by determining their viral RNA levels and viral RNA/PFU ratios. In contrast to the higher infectious titers for G614 than D614 virus, the two viruses produced nearly identical levels of viral RNA across all organs and timepoints ( Fig. 2d). The RNA/PFU ratios of G614 virus were 0.3 log 10 to 0.7 log 10 lower than those of D614 virus across airway tissues (Fig. 2e). On day 4 pi, a 1.1 log 10 lower RNA/PFU ratio was detected for G614 than D614 in nasal wash, while the differences in the trachea and lungs were 0.1 log 10 to 0.3 log 10 (Fig. 2f). On day 7 pi., despite no detectable infectious virus (detection limit 40 PFU/ml), more than 10 8 viral RNA copies/ml were detected in the nasal washes ( Fig. 2d), demonstrating high levels of viral RNA persistence after the clearance of infectious virus; this result recapitulates ndings in COVID-19 patients, who frequently tested positive with RT-PCR for up to several weeks but have low or undetectable infectious virus. One caveat of the above RNA/PFU calculation was that the total RNA could include viral RNAs from both virions and lysed cells during sample processing. Nevertheless, the results suggest that mutation D614G may enhance the infectivity of SARS-CoV-2 in the respiratory tract, particularly in the upper airway of infected animals.
The above results prompted us to directly compare the nesses of D614 and G614 viruses through a competition experiment. This approach has major advantages over performing individual strain infections with numerous host replicates; each competition is internally controlled, eliminating host-tohost variation that can reduce the power of experiments, and the virus strain ratios can be assayed with more precision than individual virus titers. Thus, competition assays have been used for many studies of microbial tness, including viruses [13][14][15][16] . To perform the competition between D614 and G614 variants, we intranasally infected hamsters with equal amounts of the two viruses (10 4 PFU per virus). Since the infecting viruses were prepared from Vero E6 cells with comparable viral RNA/PFU ratios (Fig. 1e), the animals also received equivalent levels of D614 and G614 viral RNA. On days 2, 4, and 7 pi, nasal wash and respiratory organs were harvested and quanti ed for relative amounts of D614 and G614 RNAs by RT-PCR and Sanger sequencing. The ratios of G614/D614 RNA were then calculated from the electropherograms to indicate the relative tness for viral replication. A G614/D614 ratio of > 1.0 or < 1.0 indicates a replication advantage for G614 or D614, respectively. As shown in Fig. 2h-i, all respiratory tissues showed G614/D614 ratios of 1.2 to 2.6 on days 2, 4, and 7 pi, indicating that G614 virus has a consistent advantage over D614 virus when infecting the respiratory tract of hamsters.
Dramatic enhancement of viral replication by spike mutation D614G in a primary human airway tissue model. To further de ne the function of D614G mutation in human respiratory tract, we characterized the replication of D614 and G614 viruses in a primary human airway tissue model (Fig. 3a). This airway model contains human tracheal/bronchial epithelial cells in multilayers which resemble the epithelial tissue of the respiratory tract. The primary tissue is cultured at an air-liquid interface to recapitulate the barrier, microciliary response, and infection of human airway tissues in vivo 17,18 . After infecting the airway tissue at an MOI of 5, both D614 and G614 viruses produced increasing infectious titers from day 1 to 5, up to 7.8 × 10 5 PFU/ml (Fig. 3b), demonstrating that the airway tissue supports SARS-CoV-2 replication. The infectious viral titers of G614 were signi cantly higher (2.1-to 8.6-fold) than those of D614 (Fig. 3b). In contrast, no differences in viral RNA yields were observed between the two viruses ( Fig. 3c). The genomic RNA/PFU ratios of D614 virus were 1.4-to 5.3-fold higher than those of G614 virus (Fig. 3d). Sequencing of the D614 and G614 viruses collected on day 5 pi did not show any other mutations. Collectively, the results demonstrate that substitution D614G enhances viral replication through increased virion infectivity when SARS-CoV-2 replicates on primary human upper airway tissues.
Next, we performed competition experiments to directly compare the replication tness of D614 and G614 viruses in the human airway culture. After infecting with a 1:1 infectious ratio of D614 and G614 viruses (produced from Vero E6 cells), the G614/D614 ratios increased to 1.2, 3.7, 8.2, 8.8, and 13.9 on days 1, 2, 3, 4, and 5 pi, respectively (Fig. 3e). In addition, after infecting the airway culture with 3:1 ratio of D614 and G614 viruses, the G614 variant was able to rapidly overcome its initial de cit to reach a slight advantage with a G614/D614 ratio of 1.2 by day 1 pi, with that advantage increasing to 9.1-fold by day 5 pi (Fig. 3f). Furthermore, when infecting the airway tissue with 9:1 ratio of D614 and G614 viruses, the G614/D614 ratios increased from 1.4 to 5.2 from days 1 to 5 pi (Fig. 3g). A similar competition result was obtained when the experiment was repeated using a different donor-derived human airway culture (Extended Data Fig. 2). These competition results con rm that the G614 virus can rapidly outcompete the D614 virus when infecting human airway tissues, even initially as a minor variant in a mixed population.
Effect of spike mutation D614G on neutralization susceptibility. All of the COVID-19 vaccines currently in clinical trials are based on the original D614 spike sequence 19,20 . An important question is whether substitution D614G could reduce vaccine e cacy, assuming G614 virus continues to circulate. To address this question, we measured the neutralization titers of a panel of sera collected from hamsters that were previously infected with D614 SARS-CoV-2 (Extended Data Fig. 3). Each serum was analyzed by a pair of mNeonGreen reporter SARS-CoV-2 viruses with the D614 or G614 spike (Extended Data Fig. 4) 21 .
The mNeonGreen gene was engineered at the open-reading-frame 7 of the SARS-CoV-2 genome 10 . As shown in Figs. 4a-c, all sera exhibited 1.4-to 2.3-fold higher neutralization titers (mean 1.7-fold) against G614 virus than those against D614 (Extended Data Fig. 5), suggesting that mutation D614G may confer higher susceptibility to serum neutralization.
To further examine the role of D614G mutation in antibody recognition and neutralization, we evaluated a panel of eleven human receptor-binding domain (RBD) mAbs against the D614 and G614 mNeonGreen SARS-CoV-2 viruses. The details of these RBD mAbs are reported elsewhere (An et al., submitted for publication). One mAb (mAb18) showed a 2.1-fold higher potency against G614 than D614 virus, whereas the other ten mAbs exhibited similar neutralization activities against both viruses (Figs. 4d-f and Extended Data Figs. 6 and 7). The results suggest that mutation D614G may modulate spike protein conformation to affect mAb neutralization in an epitope-speci c manner.

Discussion
We demonstrated that the spike substitution D614G enhanced SARS-CoV-2 replication in the upper respiratory tract through increased virion infectivity. Compared with the original D614 virus, the emergent, now dominant G614 virus replicated to a higher level in the human lung Calu-3 cells and primary human upper airway tissues. The replication differences were more dramatically observed in the human airway culture, with up to a 13.9-fold advantage when the two viruses were compared in a head-to-head competition test. The increased replication tness corelated with an enhanced speci c infectivity of the G614 virion. Since previous studies with pseudotyped virus showed that the cleavage e ciency of the spike protein into S1/S2 modulates SARS-CoV-2 infection 22,23 , we compared the S1/S2 ratios between the D614 and G614 virions. Although virions produced from Calu-3 cells had more complete S1/S2 cleavage than those produced form Vero E6 cells, no substantial differences in spike cleavage were detectable between the D614 and G614 virions produced from either cell type, suggesting that the enhanced virion infectivity is not likely due to the D614G-mediated spike cleavage difference. Our results from authentic SARS-CoV-2 are in contrast with previous studies reporting that the D614G mutation changes the cleavage and shedding of spike protein when expressed alone or in the context of pseudotyped virions 24,25 . Mechanistically, two recent studies showed that the D614G mutation abolishes a hydrogen-bond interaction with T859 from a neighboring protomer of the spike trimer 6 , which allosterically promotes the RDB domain to an "up" conformation for receptor ACE2 binding and fusion 7 , leading to an enhanced virion infectivity.
The higher viral loads of G614 in the upper airway of COVID-19 patients 26 and infected hamsters support the role of D614G mutation in viral transmissibility. The robust replication of SARS-CoV-2 in the human upper airway may be partially conferred by a higher ACE2 receptor expression in the nasal cavity compared to that in the lower respiratory tract 27,28 . Patients infected with G614 virus developed higher levels of viral RNA in the nasopharyngeal swabs than those infected with D614 virus, but disease severity is not associated with the D614G mutation 6,7,9 . Our hamster infection model recapitulated these clinical ndings: The G614 virus developed higher infectious titers than the D614 virus in nasal washes and tracheas, but not lungs; no differences in weight loss or signs of disease were observed between the G614-and D614-infected animals. If the lower viral RNA/PFU ratio of the G614 virus observed in our hamster and human airway models could be extrapolated to COVID-19 patients, the modest differences in cycle threshold (Ct) values of RT-qPCR observed in patients' nasopharyngeal swabs would translate to ≥ 10-fold infectious G614 virus, underscoring the potential for enhanced transmission and spread. This potential is further bolstered by the observation that a COVID-19 patient with two distinct populations of SARS-CoV-2 in the throat swabs and sputum samples only transmitted the throat strain to individuals downstream in the transmission chain 26,29 . Similar nasal-driven transmission was recently reported for human in uenza A virus in a ferret model 30 .
Our results showed that G614 virus is consistently more susceptible to neutralization by sera collected from D614 virus-infected hamsters. The increased susceptibility of G614 virus to serum neutralization generated by D614 seems counterintuitive, but could be explained by the D614G-mediated increase in the "up" conformation of the RDB for binding to ACE2 receptor 6,7 . Since current COVID-19 vaccines in clinical trials are based on the original D614 sequence, our neutralization result mitigates the concern that the D614G mutation might compromise the e cacy of vaccines against the circulating G614 virus. Future studies are needed to eliminate this concern by testing human sera collected from the D614 spike vaccinatees. Besides antisera, we also showed that, depending on the epitope locations on RBD, the neutralizing potency of certain mAbs may be affected by the D614G mutation. The results underscore the importance to test therapeutic mAbs against G614 and other newly emerged mutant viruses during preclinical development.
In summary, we have used authentic SARS-CoV-2 to demonstrate that spike substitution D614G enhances viral replication in the upper respiratory tract and increases neutralization susceptibility. These ndings have important implications in understanding the evolution and spread of the ongoing COVID-19 pandemic, vaccine e cacy, and therapeutic antibody development.

Methods
Ethics statement. Hamster studies were performed in accordance with the guidance for the Care and Use of Laboratory Animals of the University of Texas Medical Branch (UTMB). The protocol was approved by the Institutional Animal Care and Use Committee (IACUC) at UTMB. All the hamster operations were performed under anesthesia by iso urane to minimize animal suffering.
Animals and Cells. The Syrian hamsters (HsdHan:AURA strain) were purchased from Envigo (Indianapolis, IN). African green monkey kidney epithelial Vero E6 cells were grown in Dulbecco's modi ed Eagle's medium (DMEM) with 5% fetal bovine serum (FBS; HyClone Laboratories, South Logan, UT) and 1% antibiotic/ streptomycin (Gibco). Human lung adenocarcinoma epithelial Calu-3 2B4 cells were maintained in a high-glucose DMEM supplemented with 10% FBS and 1% penicillin/streptomycin at 37 °C with 5% CO 2 . The EpiAirway system is a primary human airway 3D tissue model purchased from MatTek Life Science (Ashland, MA). This EpiAirway system was maintained with the provided culture medium at 37 °C with 5% CO 2 following the manufacturer's instruction. All other culture medium and supplements were purchased from ThermoFisher Scienti c (Waltham, MA). All cell lines were veri ed and tested negative for mycoplasma.
Generation of SARS-CoV-2 spike D614G mutant viruses. One single-nucleotide substitution was introduced into a subclone puc57-CoV-2-F5-7 containing the spike gene of the SARS-CoV-2 wild type (WT) infectious clone 10 to convert the 614th amino acid from aspartic acid (D) to glycine (G) by overlap fusion PCR. The full-length infectious cDNA clone of SARS-CoV-2 D614G was assembled by in vitro ligation of seven contiguous cDNA fragments following the protocol previously described 10 . For construction of D614G mNeonGreen SARS-CoV-2, seven SARS-CoV-2 genome fragments (F1 to F5, F6 containing D614G mutation, and F7-mNG containing the mNeonGreen reporter gene) were prepared and in vitro ligated as described previously 10 . In vitro transcription was then preformed to synthesize full-length genomic RNA. For recovering the mutant viruses, the RNA transcripts were electroporated into Vero E6 cells. The viruses from electroporated cells were harvested at 40 h post electroporation and served as seed stocks for subsequent experiments. The D614G mutation from the recovered viruses was con rmed by sequence analysis. Viral titers were determined by plaque assay on Vero E6 cells. All virus preparation and experiments were performed in a biosafety level 3 (BSL-3) facilities.
RNA extraction, RT-PCR, and Sanger sequencing. Cell culture supernatants or clari ed tissue homogenates were mixed with a ve-fold excess of TRIzol™ LS Reagent (Thermo Fisher Scienti c, Waltham, MA). Viral RNAs were extracted according to the manufacturer's instructions. The extracted RNAs were dissolved in 20 µl nuclease-free water. Two microliters of RNA samples were used for reverse transcription by using the SuperScript™ IV First-Strand Synthesis System (ThermoFisher Scienti c) with random hexamer primers. Nine DNA fragments anking the entire viral genome were ampli ed by PCR. The resulting DNAs were cleaned up by the QIAquick PCR Puri cation Kit, and the genome sequences were determined by Sanger sequencing at GENEWIZ (South Plain eld, NJ).
The quantify viral RNA samples, quantitative real-time RT-PCR assays were performed using the iTaq SYBR Green One-Step Kit (Bio-Rad) on the LightCycler 480 system (Roche, Indianapolis, IN) following the manufacturers' protocols. Primers are listed in Extended Data Table 1. The absolute quanti cation of viral RNA was determined by a standard curve method using an RNA standard (in vitro transcribed 3,839 bp containing genomic nucleotide positions 26,044 to 29,883 of SARS-CoV-2 genome).
Sequences of the puri ed RT-PCR products were generated using a BigDye Terminator v3.1 cycle sequencing kit (Applied Biosystems, Austin, TX, USA). The sequencing reactions were puri ed using a 96well plate format (EdgeBio, San Jose, CA, USA) and analyzed on a 3500 Genetic Analyzer (Applied Biosystems, Foster City, CA).The peak electropherogram height representing each mutation site and the proportion of each competitor was analyzed using the QSVanalyser program 31 .
Plaque assay. Approximately 1.2 × 10 6 Vero E6 cells were seeded to each well of 6-well plates and cultured at 37 °C, 5% CO 2 for 16 h. Virus was serially diluted in either DMEM with 2% FBS (for viral stocks and in vitro-generated samples) or DPBS (for hamster tissues) and 200 µl was transferred to the monolayers. The viruses were incubated with the cells at 37 °C with 5% CO 2 for 1 h. After the incubation, overlay medium was added to the infected cells per well. The overlay medium contained either DMEM with 2% FBS and 1% sea-plaque agarose (Lonza, Walkersville, MD) in the case of in vitro samples or Opti-MEM with 2% FBS, 1% penicillin/streptomycin, and 0.8% agarose in the case of in vivo samples. After a 2day incubation, plates were stained with neutral red (Sigma-Aldrich, St. Louis, MO) and plaques were counted on a light box. Viral infection on cells. Approximately 3 × 10 5 Vero E6 or Calu-3 cells were seeded onto each well of 12well plates and cultured at 37 °C, 5% CO 2 for 16 h. Either SARS-CoV-2 D614 or G614 virus was inoculated into the cells at an MOI of 0.01. The virus was incubated with the cells at 37 °C for 2 h. After the infection, the cells were washed by DPBS for 3 times to remove the un-attached virus. One milliliter of culture medium was added into each well for the maintenance of the cells. At each time point, 100 µl of culture supernatants were harvested for the real-time qPCR detection and plaque assay. Meanwhile, 100 µl fresh medium was added into each well to replenish the culture volume. The cells were infected in triplicates for each virus. All samples were stored in -80 °C freezer until plaque or RT-PCR analysis.
Virion puri cation and spike protein cleavage analysis. Vero E6 or Calu-3 2B4 cells were infected with D614 or G614 viruses at an MOI of 0.01. At 24 (for Vero) or 48 (Calu-3) hpi, the culture media were collected and clari ed by low speed spin. Virions in the media were pelleted by ultracentrifugation through a 20% sucrose cushion at 26,000 rpm for 3 h at 4 °C by in a Beckman SW28 rotor. The puri ed virions were analyzed by Western blot using polyclonal antibodies against spike protein and nucleocapsid as described previously 32 .
Viral infection in a primary human airway tissue model. The EpiAirway system is a primary human airway 3D mucociliary tissue model consisting of normal, human-derived tracheal/bronchial epithelial (HAE) cells. For viral replication kinetics study, either D614 or G614 virus was inoculated onto the culture at an MOI of 5 in DPBS. After 2 h infection at 37 °C with 5% CO 2 , the inoculum was removed, and the culture was washed three times with DPBS. The infected epithelial cells were maintained without any medium in the apical well, and medium was provided to the culture through the basal well. The infected cells were incubated at 37 °C, 5% CO 2 . From day 1 to day 5, 300 µl DPBS were added onto the apical side of the airway culture and incubated at 37 °C for 30 min to elute the released viruses. All virus samples in DPBS were stored at -80 °C.
Hamster infection. Four-to ve-week-old male golden Syrian hamsters, strain HsdHan:AURA (Envigo, Indianapolis, IN), were inoculated intranasally with 2 × 10 4 PFU SARS-CoV-2 in a 100-µl volume. Eighteen animals received WT D614 virus, 18 received mutant G614 virus, and 18 received a mixture containing 10 4 PFU of D614 virus and 10 4 PFU of G614 virus. The infected animals were weighed and monitored for signs of illness daily. On days 2, 4, and 7 pi, cohorts of 6 infected animals and 4 (days 2 and 4) or 6 (day 7) mock-infected animals were anesthetized with iso urane and nasal washes were collected in 400 µl sterile DPBS. Animals were humanely euthanized immediately following the nasal wash. The trachea and the four lobes of the right lung were harvested in maintenance media (DMEM supplemented with 2% FBS and 1% penicillin/streptomycin) and stored at -80 °C. Samples were subsequently thawed, tissues were homogenized for 1 min at 26 sec-1, and debris was pelleted by centrifugation for 5 min at 16,100 × g. Infectious titers were determined by plaque assay. Genomic RNAs were quanti ed by quantitative RT-PCR (Extended Data Table 1). Ratios of D614/G614 RNA were determined via RT-PCR with quanti cation of Sanger peak heights.
Competition assay. For the competition on primary human airway 3D tissue model, the D614 and G614 mutant viruses were mixed and inoculated onto the cells at a nal MOI of 5. The initial ratio of D614 and G614 viruses is 1:1, 3:1, or 9:1 based on PFU titers determined on Vero E6 cells. The DPBS with viruses was harvested every day from day 1 to 5 following the protocol described above. For the competition in hamsters, 100 µl mixtures of D614 and G614 viruses (total 2 × 10 4 PFU per hamster) were inoculated intranasally into 4-5 weeks old Syrian hamsters. On days 2, 4, and 7 pi, 6 infected hamsters were sampled for competition detection. An aliquot of the inoculum for both hamster and human airway infections was back titered for estimating the initial ratio of viruses. All samples were stored in -80 °C freezer prior to analysis.
Neutralization assay. Neutralization assays were preformed using D614 and G614 mNeonGreen SARS-CoV-2 as previously described 21     Experimental scheme. D614 and G614 viruses were inoculated onto the primary human airway tissues. After incubation at 37°C for 2 h, the culture was washed with DPBS for three times to remove the unattached virus. The culture was maintained at 37°C, 5% CO2 for 5 days. On each day, 300 µl DPBS was added onto the culture. After incubation at 37°C for 30 min, the DPBS containing the eluted viruses was subjected to plaque assay, real-time RT-qPCR, and competition analysis by Sanger sequencing. (b-d) Viral replication and genomic RNA/PFU ratios. Human airway tissues were infected with D614 or G614 virus at an MOI of 5. The amounts of infectious virus (b) and genomic RNA (c) were quanti ed by plaque assay and real-time RT-qPCR, respectively. The genomic RNA/PFU ratio (d) was calculated to indicate virion infectivity. The results were pooled from two independent biological replicates. Data are presented as means ± standard deviations. P values were determined by two-tailed Mann-Whitney test. (e,f) Competition assay. A mixture of D614 and G614 viruses with different initial ratios were inoculated onto the human airway tissues at a total MOI of 5. The initial D614/G614 virus ratio was 1:1 (e), 3:1(f), or 9:1(g). The G614/D614 ratios after competition were measure by Sanger sequencing and analyzed using R statistical software. The distribution of the model-adjusted means is illustrated by catseye plots with shaded +/-standard error (SD) overlaid by scatterplots of subject measures; scatterplots have been randomly jittered horizontally for clarity, and are shown on the log10 scale such that comparisons are a null value of 1. *p < 0.05, ** p< 0.01, *** p< 0.001.

Figure 4
D614G substitution affects the neutralization susceptibility of SARS-CoV-2 to neutralizing sera and mAbs. (a) Neutralizing activities of hamster sera against D614 and G614 mNeonGreen reporter SARS-CoV-2. Eight sera from D614 virus-infected hamsters were tested for neutralizing titers against D614 and