Host cell lectins ASGR1 and DC-SIGN jointly with TMEM106B confer ACE2 independence and imdevimab resistance to SARS-CoV-2 pseudovirus with spike mutation E484D

Abstract

The naturally occurring mutation E484D in the spike (S) protein of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) can render viral entry ACE2 independent and imdevimab resistant. Here, we investigated whether the cellular proteins ASGR1, DC-SIGN, and TMEM106B, which interact with the viral S protein, can contribute to these processes. Employing S protein-pseudotyped particles, we found that expression of ASGR1 or DC-SIGN jointly with TMEM106B allowed for robust entry of mutant E484D into otherwise non-susceptible cells, while this effect was not observed upon separate expression of the single proteins and upon infection with SARS-CoV-2 wild type (WT). Furthermore, expression of ASGR1 or DC-SIGN conferred ACE2 independence and imdevimab resistance to entry of mutant E484D but not WT, and entry under those conditions was dependent on endogenous TMEM106B. These results suggest that engagement of certain cellular lectins can direct SARS-CoV-2 mutant E484D to an ACE2-independent, TMEM106B-dependent entry pathway that is not inhibited by imdevimab.IMPORTANCEThe interaction of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein with the ACE2 receptor determines the viral cell tropism and is the key target of the neutralizing antibody response. Here, we show that SARS-CoV-2 with a single, naturally occurring mutation in the spike protein, E484D, can use the cellular lectins ASGR1 and DC-SIGN in conjunction with TMEM106B for ACE2-independent entry and evasion of therapeutic antibodies. These results suggest that engagement of cellular lectins might modulate target cell choice of SARS-CoV-2 and might allow evasion of certain neutralizing antibodies.

Keywords: SARS-CoV-2; entry; lectin; neutralization; spike.

Fig 1

Huh-7 and NCI-H522 cells allow for ACE2-independent entry of mutant E484D. (A) ACE2 dependence of host cell entry. The indicated cell lines were grown in 96-well plates, incubated with ACE2 antibody (10108-MM36, Sino Biological) for 30 min and inoculated with pseudotypes bearing the indicated viral glycoproteins. Luciferase activity in cell lysates was quantified at 16–20 h postinoculation. The average of three independent experiments ± SEM is shown. Data were normalized against the assay background (i.e., particles bearing no glycoprotein, set as 1). Statistical significance was assessed by two-tailed Student’s t-test with Welch’s correction (P > 0.05, not significant [ns]; P  ≤  0.05, *; P  ≤  0.01, **; P  ≤  0.001, ***; not determined [nd]). (B) Mutation E484D does not allow for S protein binding to ACE2 in complex with an entry-inhibiting antibody. The indicated S proteins were transiently expressed in 293T cells and the cells incubated with soluble ACE2 preincubated with the indicated concentrations of anti-ACE2 antibody. ACE2 binding was detected by incubation with a secondary antibody and the cells analyzed by flow cytometry. Soluble ACE2 binding to cells transfected with empty plasmid served as control. The average ±SEM of three biological replicates conducted with unicate samples is shown. Data were normalized against the assay background (i.e., cells incubated with secondary antibody alone). Statistical significance was assessed by two-tailed Student’s t-test with Welch’s correction (P > 0.05, not significant [ns]; P  ≤  0.05, *; P  ≤  0.01, **).

Fig 2

ACE2-independent entry correlates with resistance to imdevimab and bebtelovimab. Pseudotyped particles were preincubated with the indicated antibodies at the indicated concentrations for 1 h before addition to the indicated target cells. Luciferase activity in cell lysates was quantified at 16–20 h postinoculation. The average of three (two in case of NCI-H522 cells) independent experiments ± SEM is shown. Data were normalized against signals measured in the absence of antibody (= 0% inhibition). Curves were calculated using a non-linear regression model (variable slope).

Fig 3

Directed expression of ASGR1/DC-SIGN jointly with TMEM106B allows for efficient entry into otherwise non-susceptible cells. (A) 293T cells were transfected to express the indicated soluble S proteins (S1 domain protein of B.1 [WT] and B.1 [E484D] S proteins fused to the Fc portion of human immunoglobulin G), and S protein levels in cell lysates and supernatants were analyzed by immunoblot using anti-Fc antibody. The results of a representative experiment are shown and were confirmed in two additional experiments. (B) 293T cells transfected to express ASGR1, DC-SIGN, TMEM106B, or ACE2 were analyzed by immunoblot using anti-c-Myc (DC-SIGN, TMEM106B, and ACE2) and anti-AU1 (ASGR1) antibody. The results of a representative experiment are shown and were confirmed in a separate experiment. (C) 293T cells transfected to express ASGR1, DC-SIGN, TMEM106B, or ACE2 were incubated with the indicated soluble S proteins, and S protein binding was analyzed by flow cytometry. Presented are the average (mean) data from three (DC-SIGN, TMEM106B) or six (ASGR1, ACE2) biological replicates (each conducted with single samples). Statistical significance was assessed by two-tailed Student’s t-test with Welch’s correction (P > 0.05, not significant [ns]; P  ≤  0.05, *; P  ≤  0.01, **). (D) BHK-21 cells were transfected with plasmids encoding ASGR1, DC-SIGN, TMEM106B, and ACE2 either alone or in combination and transduced with SARS-2-S_pp and E484D_pp followed by quantification of luciferase activity in cell lysates. Presented are the average (mean) data from three biological replicates (each conducted with four technical replicates), for which transduction was normalized against signals obtained from control-transfected cells inoculated with the respective pseudotyped particles (background, set as 1). Statistical significance was assessed by two-tailed Student’s t-test with Welch’s correction (P > 0.05, not significant [ns]; P  ≤  0.05, *; P  ≤  0.01, **).

Fig 4

Directed expression of DC-SIGN or ASGR1 allows for imdevimab resistance and ACE2 independence of mutant E484D. For analysis of imdevimab resistance (A), 293T cells were transfected with ASGR1 or DC-SIGN encoding plasmids or empty plasmid as control and inoculated with pseudotypes that harbored the indicated S proteins and were pre-incubated with 1 µg/mL of imdevimab. Luciferase activities in cell lysates were quantified at 16–20 h postinoculation. For analysis of ACE2-independent entry (B), the experiment was conducted as described above, but cells were preincubated with 7 µg/mL anti-ACE2 antibody. Presented are the average (mean) data from three biological replicates (each conducted with four technical replicates), for which transduction was normalized against samples that did not contain antibody (= 100% pseudotype entry). Error bars indicate the SEM. Statistical significance was assessed by two-tailed Student’s t-test with Welch’s correction (P > 0.05, not significant [ns]; P  ≤  0.05, *; P  ≤  0.01, **; P  ≤  0.001, ***).

Fig 5

KO of TMEM106B in 293T cells. (A) TMEM106B expression in 293T WT and TMEM106B KO cells. 293T cells were stably transfected with EV (control) or vector encoding TMEM106B specific guide RNAs followed by antibiotic selection and single cell cloning. Parental (293T), two clones of control 293T cells (EV1, EV2) and two clones of TMEM106B KO cell lines (Clone 1, Clone 2) were lysed and analyzed for TMEM106B expression by immunoblot, using TMEM106B specific antibody. Similar results were obtained in two separate experiments. (B) Control or TMEM106B KO 293T cells transfected with EV or DC-SIGN plasmid or cotransfected with DC-SIGN and TMEM106B plasmids were transduced with particles pseudotyped with EBOV-GP or VSV-G. At 16–20 h postinoculation, transduction efficiency was analyzed by quantifying luciferase activity in cell lysates. Presented are the results of a representative experiment conducted with technical quadruplicates; error bars indicate SD. Transduction was normalized against signals obtained for control particles bearing no viral glycoprotein (background, set as 1). Similar results were obtained in two separate experiments.

Fig 6

Endogenous expression of TMEM106B is required for DC-SIGN-dependent imdevimab resistance and ACE2 independence of mutant E484D. For analysis of ACE2-independent entry (A), 293T control (EV) or TMEM106B KO cells were transiently transfected with EV (control), or DC-SIGN plasmid or cotransfected with DC-SIGN and TMEM106B plasmids, incubated with 7 µg/mL of anti-ACE2 antibody and inoculated with pseudotypes bearing the indicated S proteins followed by quantification of luciferase activities in cell lysates at 16–20 h postinoculation. For analysis of imdevimab resistance (B), the experiment was conducted as described above, but pseudotyped particles were preincubated with 1 µg/mL imdevimab. Presented are the results from a representative experiment carried out with four technical replicates; transduction was normalized against samples that did not contain antibody. Error bars indicate SD, similar results were obtained in a separate experiment. Statistical significance was assessed by two-tailed Student’s t-test with Welch’s correction (P > 0.05, not significant [ns]; P  ≤  0.05, *; P  ≤  0.01, **; P  ≤  0.001, ***).

Fig 7

EGTA blocks ACE2-independent entry into Huh-7 cells. (A) Expression and N-glycosylation of ASGR1 in Huh-7 cells. Untransfected Huh-7 cells and 293T and BHK21 cells transfected with EV or ASGR1 encoding vector were control treated or treated with PNGaseF and analyzed for ASGR1 expression by immunoblot. Expression of β-actin served as loading control. Similar results were obtained in two (all cell lines) to four (293T, BHK21 cells) independent experiments. (B) Huh-7 cells were preincubated with medium alone (control) or medium containing anti-ACE2 antibody, EGTA, or anti-ACE2 antibody combined with EGTA for 1 h followed by inoculation with pseudotypes bearing the indicated S proteins. Luciferase activity in cell lysates was quantified at 16–20 h postinoculation. The results ± SD of a single experiment performed with technical quadruplicates are shown. Similar results were obtained in a separate experiment.