SARS-CoV-2 D614G variant exhibits efficient replication ex vivo and transmission in vivo

Abstract

The spike aspartic acid-614 to glycine (D614G) substitution is prevalent in global severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) strains, but its effects on viral pathogenesis and transmissibility remain unclear. We engineered a SARS-CoV-2 variant containing this substitution. The variant exhibits more efficient infection, replication, and competitive fitness in primary human airway epithelial cells but maintains similar morphology and in vitro neutralization properties, compared with the ancestral wild-type virus. Infection of human angiotensin-converting enzyme 2 (ACE2) transgenic mice and Syrian hamsters with both viruses resulted in similar viral titers in respiratory tissues and pulmonary disease. However, the D614G variant transmits significantly faster and displayed increased competitive fitness than the wild-type virus in hamsters. These data show that the D614G substitution enhances SARS-CoV-2 infectivity, competitive fitness, and transmission in primary human cells and animal models.

Fig. 1. SARS-CoV-2 D614G variant shows enhanced infectivity in immortalized cell lines and replication fitness in upper human respiratory epithelia compared with the ancestral WT virus.

(A) Genomes of recombinant SARS-CoV-2 D614G variants based on the backbone of a D614-form strain WA1. E, envelope; M, membrane; N, nucleocapsid; nLuc, nanoluciferase; S, spike. (B) Entry efficiency of WT-nLuc and D614G-nLuc in multiple susceptible cell lines at a MOI of 0.5. After 1-hour of infection, cells were treated with neutralization Abs to minimize the secondary round of infection. The relative light unit (RLU) representing the nLuc expression level was measured at 8 hours after infection. RLU values were normalized with background (Bkgd) residual luciferase signals in both viral inocula. (C) Growth curves of the two viruses in Vero-E6 (i), Vero-81 (ii), A549-ACE2 (iii), and Huh7 (iv) cell lines at a MOI = 0.5. (D to F) Comparison of 24-, 48-, and 72-hour titers between the two variants in infected primary nasal (D), large airway (E), and small airway (F) cells at a MOI of 0.1. Triplicated titers of the two viruses in the cultures from the same donor were analyzed by paired t test. (G) Schematic of competition assays on LAE cells. Cultures were infected with a 1:1 or 10:1 ratio of WT and D614G mixture at a MOI of 0.5, and the supernatants were serially passaged three times in naïve cultures. PBS, phosphate-buffered saline. (H and I) BtsCI digestion (H) and Sanger sequencing chromatogram (I) of S gene fragments amplified from viral samples in the LAE competition assay. The 1.5-kb fragments containing the residue 614 were amplified from the total RNA of individual samples collected in each passage. P, passage. Data in (B) and (C) are indicated as mean ± SD and were analyzed by unpaired t test between both viruses; data in (D) to (F) were analyzed by paired t test. N.S., not significantly different; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Fig. 2. D614G substitution does not alter SARS-CoV-2 virion morphology, S protein cleavage patterns, and sensitivity to neutralizing Abs.

(A) TEM image of WT and D614G virions on airway epithelial cell surface. Scale bar, 200 nm. (B) SEM images of WT and D614G virions on airway epithelial cell surface. Scale bar, 100 nm. (C) Quantification of S protein on individual SARS-CoV-2 virion projections. The number of S proteins on individual virion projections from different SEM images was quantified manually, n = 20. (D) Western blot analysis of SARS-CoV-2 virions washed from WT- or D614G-infected HNE cultures at 72 hours after infection. Each lane contains a pooled sample from triplicated cultures derived from the same donor. Full-length (FL), S1/S2 cleaved spike protein (S), and nucleocapsid protein (N) were probed. Samples in each pair were loaded on the basis of an equal amount of the N protein. MW, molecular weight. (E and F) S-to-N ratios (E) and FL-to-cleaved S ratios (F) were determined by measuring relative intensity of bands in the Western blot image. (G) Summary of ID₅₀ values of 25 convalescent human sera against WT- and D614G-nLuc viruses. (H) Neutralization curves of three representative human sera. Viral sequence reveals that serum #1 was collected from a COVID-19 patient infected with a D614G SARS-CoV-2 variant. (I and J) Summarized IC₅₀ values (I) and individual neutralization curves (J) of six human neutralizing mAbs against both viruses. Data in (C), (E), and (F) are indicated as mean ± SD and were analyzed by unpaired t test; data in (G) and (I) were analyzed by paired t test. N.S., not significantly different.

Fig. 3. D614G substitution does not alter SARS-CoV-2 pathogenesis in hACE2 mice.

(A) Lung, brain, and nasal turbinate titers of WT and D614G-infected mice were determined on day 2 and day 5. Each mouse was infected with 10⁵ PFU of the virus (n = 5 per group); the plaque assay detection limit (1.7 log₁₀PFU/ml) is indicated as dashed lines. Data were analyzed by unpaired t test. (B) Representative hematoxylin and eosin (H&E) staining and IHC staining of SARS-CoV-2 N protein in the lung tissues collected from infected hACE2 mice harvested at day 2 after infection. Scale bars, 100 μm.

Fig. 4

D614G substitution enhances SARS-CoV-2 transmission in hamsters. (A and B) Viral titers in the lung (A) and nasal turbinates (B) collected from SARS-CoV-2–infected hamsters at days 3 and 6. Each hamster was infected with 10³ PFU of the virus, n = 8 per virus for each time point; the plaque assay detection limit (1 log₁₀PFU/g) is indicated as dashed lines. (C) Body weight of mock-, WT-, and D614G-infected hamsters (n = 4 per group). Hamsters in the body weight study were not subjected to nasal wash sampling. *p < 0.05 (D) IHC staining of SARS-CoV-2 nucleocapsid proteins in representative lung tissues collected from WT- and D614G-infected hamsters at day 3. Scale bars, 100 μm. (E) H&E staining of representative lung tissues collected on days 3, 6, and 9 from hamsters infected with WT or D614G. Scale bars, 1 mm. (F)(i) Quantification of IHC-positive cells in hamster lung tissues, using the following scoring system: 0, no positive cell; 1, <10%; 2, 10 to 50%; and 3, >50% positive cells in each lobe of lung. (ii) The size of pulmonary lesions was determined on the basis of the mean percentage of affected area in each section of lobes from each animal. (iii) Pathological severity scores in infected hamsters, based on the percentage of inflammation area for each section of the five lobes collected from each animal using the following scoring system: 0, no pathological change; 1, affected area (≤10%); 2, affected area (<50% and >10%); and 3, affected area (≥50%). An additional point was added when pulmonary edema and/or alveolar hemorrhage was observed. (G) Viral titers in nasal washes collected from infected and exposed hamster pairs in WT and D614G groups; plaque assay detection limit (1 log₁₀PFU/ml) is indicated as dashed lines. The number of positive hamsters in both exposure groups at day 2 (WT versus D614G = 0 of 8 versus 5 of 8) was analyzed by Fisher’s exact test, P = 0.0256. N.S., not significantly different.