Skip to main content

SARS-CoV-2 Spike protein variant D614G increases infectivity and retains sensitivity to antibodies that target the receptor binding domain

Our take —

This analysis of over 52,000 SARS-CoV-2 genomes, available as a preprint and thus not yet peer reviewed, shows that a genetic variant with a single amino acid change in the spike protein (D614G) has become the dominant infection globally since its discovery in late January 2020. Experiments in human cell cultures indicate that the variant is more infectious than the wildtype (D614), although human antibodies are still capable of neutralizing both spike variants. Additional studies in animal models will be necessary to determine if the D614G variant produces higher viral loads and is more transmissible.

Study design

Ecological, Other

Study population and setting

The authors analyzed SARS-CoV-2 genomes available on GISAID ( as of June 25, 2020 (n = 52,292), representing infections in Africa, South America, North America, Europe, Oceania, and Asia. They analyzed the frequency of a viral variant with an amino acid change (D to G) at position 614 of the spike protein over time, both globally and within geographic regions. They compared the relative infectivity of the D614G variant virus versus the wildtype (D614) using HIV-1 vectors expressing either variant of the SARS-CoV-2 spike protein. Relative infectivity was compared in human lung, colon, and embryonic kidney cells or human T lymphocytes expressing angiotensin-converting enzyme 2 (ACE2; SARS-CoV-2 uses ACE2 as a receptor to enter human cells) and transmembrane protease, serine 2 (TMPRSS2; SARS-CoV-2 spike protein is primed by TMPRSS2). The authors also compared the relative infectivity of D614G variant in human embryonic kidney cells expressing different ACE2 orthologues from human, pangolin, pig, cat, dog, rat, and mouse. To determine the possible effect of D614G on human immunity, the authors tested the ability of four monoclonal antibodies designed to target and neutralize SARS-CoV-2 spike protein. They tested each antibody alone or in combinations in a neutralization assay against wildtype D14 spike variant and the D614G variant.

Summary of Main Findings

The D614G variant was first detected in sequenced genomes in late January 2020 in China and Germany, and has since increased in frequency so that it is the dominant variant globally (>97% of published sequences). Even for Africa and South America where early cases had a higher frequency of D614G, suggesting a founder effect, the frequency has continued to increase. In all human cell types tested, the D614G variant was 4-6 times more infectious than the D614 variant. In embryonic kidney cells or T lymphocytes expressing ACE2 and TMPRSS2, the D614G variant was 9 times more infectious than D614. Similarly, the D614G variant was 2-3 times more infectious than D614 in human embryonic kidney cells expressing human, pangolin, pig, cat, and dog orthologues of ACE2; infection of cells with mouse or rat ACE2 with D614 or D614G was not detectable. All monoclonal antibodies, and combinations thereof, were equally capable of neutralizing D614 and D614G spike protein variants.

Study Strengths

Compared to previous studies, this study uses many more sequences to assess the change in frequency of the D614G variant. It also provides evidence that the D614G is potentially more infectious (at least in cell cultures) than the wildtype D614 variant.


While the selective advantage of the D614G variant suggested by this study is an important finding, there are other neutral factors (founder effects, bottlenecks in transmission) that may have contributed to the spread of the D614G variant. Modeling efforts that account for selection and neutral effects will be needed to ascertain the relative effects of each. Additionally, the authors cannot tell the exact geographical origin of D614G, because of the few viral genomes that were sequenced early in the outbreak. Since the sequences used in this study represent only a miniscule fraction of the total cases (0.5% of almost 10 million cases as of 25 June 2020), more sequencing from banked samples may provide more resolution. Lastly, the experimental model, wherein lentiviral particles pseudotyped with SARS-CoV-2 spike protein are used to infect human cell cultures, cannot fully represent the complexity of SARS-CoV-2 infection in humans. Additional experiments in an animal model with D614 and D614G variants of actual SARS-CoV-2 that demonstrate higher viral loads for the D614G variant (and possibly increased transmissability between animals), or human outbreak data that demonstrates greater transmissability in cases with the D614G variant, will be a necessary next step to verifying any selective advantage claimed from population-level frequency data.

Value added

This study supports earlier reports that the D614G variant of SARS-CoV-2 has increased in frequency during the pandemic and now represents the dominant genetic variant globally, confirms that the D614G variant is more infective in human cell cultures, and demonstrates that human antibodies are still capable of neutralizing the D614G variant as well as the wildtype.

This review was posted on: 3 August 2020