key: cord-0259823-uto9q6s2
authors: Kimura, Izumi; Kosugi, Yusuke; Wu, Jiaqi; Yamasoba, Daichi; Butlertanaka, Erika P; Tanaka, Yuri L; Liu, Yafei; Shirakawa, Kotaro; Kazuma, Yasuhiro; Nomura, Ryosuke; Horisawa, Yoshihito; Tokunaga, Kenzo; Takaori-Kondo, Akifumi; Arase, Hisashi; Saito, Akatsuki; Nakagawa, So; Sato, Kei
title: SARS-CoV-2 Lambda variant exhibits higher infectivity and immune resistance
date: 2021-07-28
journal: bioRxiv
DOI: 10.1101/2021.07.28.454085
sha: 1619d5bd67bce60cb993feaa39d3558e09bab87e
doc_id: 259823
cord_uid: uto9q6s2

SARS-CoV-2 Lambda, a new variant of interest, is now spreading in some South American countries; however, its virological features and evolutionary trait remain unknown. Here we reveal that the spike protein of the Lambda variant is more infectious and it is attributed to the T76I and L452Q mutations. The RSYLTPGD246-253N mutation, a unique 7-amino-acid deletion mutation in the N-terminal domain of the Lambda spike protein, is responsible for evasion from neutralizing antibodies. Since the Lambda variant has dominantly spread according to the increasing frequency of the isolates harboring the RSYLTPGD246-253N mutation, our data suggest that the insertion of the RSYLTPGD246-253N mutation is closely associated with the massive infection spread of the Lambda variant in South America. Highlights Lambda S is highly infectious and T76I and L452Q are responsible for this property Lambda S is more susceptible to an infection-enhancing antibody RSYLTPGD246-253N, L452Q and F490S confer resistance to antiviral immunity Graphical Abstract

vaccination. In this study, we reveal the evolutionary trait of the Lambda variant by 110 molecular phylogenetic analysis. We further demonstrate that the RSYLTPGD246-111 253N mutation, a unique mutation in the NTD of the Lambda S protein, is responsible 112

for the virological phenotype of the Lambda variant that can associate with the 113 massive infection spread mainly in South American countries. 114

Epidemic dynamics of the Lambda variant in South American countries 116 As of June 29, 2021, 1,908 genome sequences of the Lambda variant belonging to 117 the PANGO C.37 lineage have been isolated from 26 countries and deposited in 118 GISAID. Although it is considered that the Lambda variant was first detected in Peru 119

in December 2020 (WHO, 2021a), our in-depth analysis revealed that the Lambda 120 variant was first detected in Argentina on November 8, 2020 (GISAID ID: 121 EPI_ISL_2158693) (Figure 1A ; see STAR★METHODS for the detail). The fact that 122 the percentage of the Lambda sequence is increasing in South American countries 123

including Peru, Chile, and Argentina ( Figure 1A and Table S1) suggests that the 124

Lambda variant is spreading predominantly in these countries (Table S2) . We then 125 generated a maximum likelihood-based phylogenetic tree of the Lambda variant 126 (C.37). Although there are some isolates that have been misclassified as the 127

Lambda, which could be a sister group of the Lambda variant, our phylogenetic tree 128 indicated the monophyly of genuine Lambda variant isolates ( Figure S1 ).

Association of the lambda variant spread with the increasing frequency of the 131 isolates harboring the RSYLTPGD246-253N mutation 132

The S protein of the consensus sequence of the Lambda variant bears six 133 substitution mutations (G75V, T76I, L452Q, F490S, D614G and T859N) and a 7-134 amino-acid deletion in the NTD (RSYLTPGD246-253N) ( Table S3 ). The analysis 135 using the 1,908 sequences of the Lambda variant (C.37 lineage) showed that the six 136 substitution mutations are relatively highly (> 90%) conserved ( Figure 1B and Table  137 S3). Although a large deletion in the NTD, the RSYLTPGD246-253N mutation, is 138 also highly conserved, 287 out of the 1,908 sequences (15.0%) of the Lambda (C.37 139 lineage) genomes do not harbor this mutation ( Figure 1C and Table S3 ). To ask 140

whether the epidemic dynamics of the Lambda variant is associated with the 141 emergence of the RSYLTPGD246-253N mutation, we examined all amino acid 142 replacements in the S protein of the SARS-CoV-2 genomes deposited in the GISAID 143 database (2,084,604 sequences; as of June 29, 2021). The RSYLTPGD246-253N 144 mutation was first found in Argentina on November 8, 2020 (GISAID ID: 145 EPI_ISL_2158693), which is the first isolate of the Lambda variant ( Figure 1A) , 146 suggesting that this deletion event uniquely occurred in the ancestral lineage of the 147 Lambda variant. We then analyzed the molecular evolutionary dynamics of the 148

Lambda variants by performing the Bayesian tip-dating analysis. We showed that 149

the Lambda variant bearing the RSYLTPGD246-253N mutation emerged around 150

July 12, 2020 (95% CI, January 5, 2020 -October 22, 2020) ( Figure 1C and Figure  151 S1B). To infer the population dynamics of the lineage, we performed the Bayesian 152 skyline plot analysis. This analysis showed that the effective population size of the 153 Lambda variant has increased at the beginning of 2021 ( Figure 1D) . Intriguingly, 154

when we plot the proportion of the Lambda variant that bears RSYLTPGD246-253N 155 mutation, it was increased after the emergence of the Lambda variant and closely 156 associated with the increase of effective population size ( Figure 1D) (Table S4) , which is used to be a 166 VOC/VOI by July 6, 2021 (WHO, 2021a), as controls of this experiment. As shown 167

in Figure 2A , the infectivities of the Alpha and Beta variants were significantly lower 168 than that of the parental D614G S, and the infectivities of Gamma variant and the 169 parental D614G S were comparable. On the other hand, the infectivities of the Delta, 170

Epsilon and Lambda variants were significantly higher than that of the parental 171 D614G ( Figure 1A ). This pattern was independent of the input dose of 172 pseudoviruses used ( Figure S2A ).

To assess the effect of a characteristic mutation of the Lambda variant, the 174 RSYLTPGD246-253N mutation (Figure 1) , on viral infectivity, we prepared the 175 pseudovirus with the Lambda S derivative recovering this deletion mutation 176 ("Lambda+N246-253RSYLTPGD"). The infectivity of this mutant was comparable to 177 that of the Lambda S (Figures 2A and S2A sensitive to the vaccine-induced neutralization than the Lambda (P = 0.0016 by 188

Wilcoxon matched-pairs signed rank test) ( Figure 2B) . These results suggest that 189 the Lambda S is highly infectious and resistant to the vaccine-induced humoral 190 immunity, and the robust resistance of the Lambda S to the vaccine-induced 191 neutralization is determined by a large deletion in the NTD. 192 193 Effect of the consensus mutations in the Lambda S on viral infectivity and NAb 194 sensitivity 195 We next plotted six substitution mutations (G75V, T76I neither residues are located on the ACE2 interface ( Figure S2D ). The T859N 203 mutation is in the heptad repeat 1 of the S2 subunit ( Figure 2C ).

To investigate the effects of these seven consensus mutations in the 205 Lambda S on viral infectivity and NAb sensitivity, we prepared the viruses 206 pseudotyped with the D614G S-based derivatives that possess respective mutations 207 of the Lambda variant. Figure 2E showed that the G75V mutation significantly 208 reduces viral infectivity, while the T76I and GT75-76VI mutations significantly 209 increases it. Additionally, 91.5% (1,746/1,908) of the Lambda variant sequence 210 possesses these two mutations, and the phylogenetic tree of the Lambda variant 211

indicated that the variant harboring either G75V or T76I sporadically emerged during 212 the epidemic of the Lambda variant ( Figure S1A ). These findings suggest that T76I 213 is a compensatory mutation to recover the decreased infectivity by the G75V 214 mutation. 215

Similar to the experiment using the pseudoviruses with the Lambda S and 216

Lambda+N246-253RSYLTPGD (Figure 2A) , the insertion of the RSYLTPGD246-217 253N mutation did not affect viral infectivity ( Figure 2E ). The infectivity of the T859N 218 mutation was also similar to that of the parental D614G pseudovirus ( Figure 2E ).

When we focus on the effect of the mutations in the RBD, the L452Q and 220 L452Q/F490S mutations significantly increased viral infectivity, while the F490S sole 221 mutation did not ( Figure 2E ). Taken together, these results suggest that the T76I 222 and L452Q mutations are responsible for the higher infectivity of the Lambda S 223 (Figure 2A) . The effect of each mutation was similar in different amounts of 224 pseudoviruses used and in the target cells without TMPRSS2 expression ( Figure  225 S2E). 226 We next assessed the sensitivity of these pseudoviruses with the mutated 227 S proteins to BNT162b2-induced antisera. As shown in Figure 2F , the G75V, T76I, 228 GT75-76VI and T852N mutations did not affect the vaccine-induced neuralization. 229

On the other hand, the RSYLTPGD246-253N mutant exhibited a significant 230 resistance to the vaccine-induced neuralization (P = 0.027 by Wilcoxon matched-231 pairs signed rank test; Figure 2F ), which is relevant to the experiment with the S 232 proteins of the Lambda variant and the Lambda+N246-253RSYLTPGD derivative 233 ( Figure 2B ). Additionally, we found that the L452Q and F490S mutations confer 234 resistance to the vaccine-induced antisera ( Figure 2F ). The results that the F490S 235 mutation does not affect viral infectivity ( Figure 2E ) but confers the resistance to the 236 vaccine-induced antisera ( Figure 2F ) suggest that this mutation has acquired to be 237 resistant to antiviral humoral immunity. On the other hand, the L452Q mutation not 238

only increases viral infectivity ( Figure 2E ) but also augments the resistance to the 239 vaccine-induced antisera ( Figure 2F ), suggesting that that this mutation can be 240 critical for the viral dissemination in the human population. 241

To further assess the association of the mutations in the Lambda S, 242

particularly those in the NTD, we used two monoclonal antibodies that targets the 243 NTD: an NTD-targeting NAb, clone 4A8 (Chi et al., 2020), and an EAb, clone COV2-244 2490, that recognizes the NTD and enhances viral infectivity (Liu et al., 2021c). As 245 shown in Figure 2G , the 4A8 antibody inhibited the pseudovirus infections of the 246 parental D614G, Lambda+N246-253RSYLTPGD derivative, G75V, T76I and GT75-247 75VI in dose-dependent manners. Intriguingly, the pseudovirues with the S proteins 248 of the Lambda and the RSYLTPGD246-253N mutant were resistant to the antiviral 249 effect mediated by the 4A8 antibody ( Figure 2G ). These results suggest that the 250 RSYLTPGD246-253N mutation critically affects the sensitivity to certain NAbs 251 targeting the NTD. On the other hand, the COV2-2490 antibody enhanced the 252 infectivities of the parental D614G, the Lambda, and the Lambda+N246-253 253RSYLTPGD derivative ( Figure 2H ). Particularly, the infectivities of the Lambda 254 and the Lambda+N246-253RSYLTPGD derivative were more significantly enhanced 255 than the parental D614G ( Figure 2H ). These data suggest that the Lambda S is 256 more susceptible to the EAb-mediated virus infection enhancement. 257

In See also Figure S1 and Tables S1-S3. as the D614G S were prepared. The pseudoviruses were inoculated into HOS-595 ACE2/TMPRSS2 cells at 1,000 ng HIV-1 p24 antigen, and the percentages of 596 infectivity compared to the virus pseudotyped with parental S D614G are shown. 597

(F) Neutralization assay. Neutralization assay was performed using the 598 pseudoviruses used in Figure 2B and 18 BNT162b2-vaccinated sera as described 599

in STAR★METHODS. The raw data are shown in Figure S2B . In A, E and H, assays were performed in triplicate, and the average is shown with 606 SD. Statistically significant differences (*, P < 0.05) versus the D614G S were 607 determined by Student's t test. 608

In B and F, assays were performed in triplicate, and the average is shown with SD. 609 Statistically significant differences were determined by Wilcoxon matched-pairs 610 signed rank test. The P values are indicated in the figure. 611

In G, assays were performed in quadruplicate, and the average is shown with SD. 612 Statistically significant differences (*, P < 0.05) versus the value without antibody 613

were determined by Student's t test. NS, no statistical significance. 614

See also Figure S2 and Table S4 . 615 616 617 Figure S1 . A maximum likelihood-based phylogenetic tree and an evolutionary 618 timetree of the Lambda variant (C.37 lineage) (Related to Figure 1 ).

(A) The 696 SARS-CoV-2 genome sequences were used for the analysis. Wuhan-620

Hu-1 (GISAID ID: EPI_ISL_1532199), located in the bottom of the tree, was 621 indicated by a black star. EPI_ISL_1532199 and EPI_ISL_1093172 belonging to the 622 B.1.1.1 lineage were indicated by grey stars. Red or blue circle on the branch was 623

shown in each internal node if the bootstrap value was ≥ 80 or ≥ 50 (n = 1,000). A 624 color box in pink or pale blue indicates the mutation in the S protein exist or not, 625 respectively. 626 

CoV-2 infections and incidence of presumptive B.1.427/B.1.429 variant among 444 healthcare personnel at a northern California academic medical center

Effectiveness of an Inactivated 448 SARS-CoV-2 Vaccine in Chile

MAFFT multiple sequence alignment software 450 version 7: improvements in performance and usability

In vitro and in vivo functions of 453 SARS-CoV-2 infection-enhancing and neutralizing antibodies

BNT162b2-elicited neutralization of B.1.617 and 456 other SARS-CoV-2 variants

Neutralizing activity of BNT162b2-459 elicited serum

An infectivity-enhancing site on the 462 SARS-CoV-2 spike protein targeted by antibodies

An NTD supersite of attack

N-terminal domain antigenic 466 mapping reveals a site of vulnerability for SARS-CoV-2

CoV-2 spike deletion H69/V70 and its role in the Alpha variant B.1.1.7

IQ-TREE 2: New Models and Efficient Methods 473 for Phylogenetic Inference in the Genomic Era

SARS-CoV-2 B.1.617.2 Delta 476 variant emergence, replication and immune evasion

An emerging SARS-CoV-2 mutant 479 evading cellular immunity and increasing viral infectivity

An evolutionary timetree of the Lambda variant (C.37 lineage). The estimated 627 date of the emergence of the Lambda variant is indicated in the figure. The GISAID 628 ID and sampling date is noted at each terminal node. The three sister sequences of 629 the genuine C

EPI_ISL_1093172 (B.1.1.1 lineage) and EPI_ISL_1534656 (C.37 lineage)] are used 631 as an outgroup and indicated in black. Wuhan-Hu-1 (GISAID ID: EPI_ISL_1532199), 632 the oldest SARS-CoV-2

Virological features of the Lambda S (Related to Figure 2)

Pseudovirus assay. The HIV-1-based reporter viruses pseudotyped with the 638 SARS-CoV-2

Epsilon (B.1.427) and Lambda (C.37) variants as 640 well as the Lambda+N246-253RSYLTPGD derivative were prepared as described 641 in STAR★METHODS

ACE2/TMPRSS2 cells at 4 different doses

and percentages of infectivity compared to the virus pseudotyped with 644 parental S D614G (1,000 ng HIV-1 p24 antigen) are shown. Assays were performed 645 in triplicate. Note that the data of 1,000 ng HIV-1 p24 antigen are identical to those 646 shown in Figure 1B

Eighteen vaccinated sera were used for the neutralization 651 assay. The 50% neutralization titers of respective serum against respective virus are 652 shown. The values are summarized in Figures 2B and 2F. 653 (C) Amino acid and nucleotide sequences of the residues 245-254 of the SARS-654

The amino acid sequences (residues 245-254, bold) and nucleotide 655 sequences of the parental S (top) and the Lambda S (bottom) are shown. The 656 nucleotides shaded

Positions of the residues L452 and F490. The residues L452Q and F490S are 659 labeled in the cocrystal structure of SARS-CoV-2 and human ACE2 (PDB: 6M17) 660

Pseudovirus assay. The HIV-1-based reporter viruses pseudotyped with the 662 SARS-CoV-2 S proteins bearing respective mutations of the Lambda variant as well 663 as the D614G S were prepared. The pseudoviruses were inoculated into HOS-664

250 or 125 ng HIV-1 p24 antigen), and percentages of infectivity compared to 666 the virus pseudotyped with parental S D614G (1,000 ng HIV-1 p24 antigen) in HOS-667 ACE2 cells are shown. Assays were performed in triplicate

All human subjects provided written 705 informed consent

Collection of BNT162b2-Vaccinated Sera 708 Peripheral blood were collected four weeks after the second vaccination of 709

and the sera of 18 vaccinees (average age: 40, range: 710 28-59, 22% male) were isolated from peripheral blood. Sera were inactivated at 56°C 711 for 30 min and stored at -80°C until use

HEK293T cells (a human embryonic kidney cell line; ATCC CRL-3216), and HOS 715 cells (a human osteosarcoma cell line; ATCC CRL-1543) were maintained in 716

Dulbecco's modified Eagle's medium (high glucose) (Wako, Cat# 044-29765) 717 containing 10% fetal calf serum and 1% PS

HOS-ACE2/TMPRSS2 cells, the HOS cells stably expressing human ACE2 and 719 TMPRSS2

HOS-ACE2 cells, the HOS cells stably expressing human ACE2

We 730 confirmed all of them are isolated from humans. To estimate when a Lambda variant 731 harboring the RSYLTPGD246-253N deletion mutation in the S protein occurred, we 732 screened 1,908 Lambda variants by removing genomes 1) containing more than 5 733 undetermined nucleotides at coding regions and 2) having an unknown sampling 734 date. We then collected 644 and 49 viral genomes with and without RSYLTPGD246-735 253N deletion mutation in the Lambda S protein. We used Wuhan-Hu-1 strain 736 isolated in China on

GTR+G substitution model is utilized based on BIC 742 criterion. We found that several sequences without the RSYLTPGD246-253N 743 mutation also clustered with the genomes carrying the RSYLTPGD246-253N 744 mutation (Figure S1A), which could be due to reversible mutation(s) and/or 745 recombination

To estimate the emerging time of the Lambda variant (C.37 lineage), we 748 collected all Lambda sequences carrying the RSYLTPGD246-253N mutation that 749 were sampled in 2020 (2 sequences) and randomly sampled 100 sequences in 2021

We also added the following 4 SARS-CoV-2 genomes as the outgroup: strain 751

EPI_ISL_1532199 and EPI_ISL_1093172) were categorized in the B.1.1.1 755 lineage, although they were previously categorized as the C.37 lineage. We carefully 756 examined these two sequences and found that they could be used as a sister group 757 of the C.37 lineage. As for EPI_ISL_1534645

A timetree was 766 summarized using TreeAnnotator software in the BEAST package and visualized by 767 using FigTree v1.4.4 (Figure 1C and Figure S1B). Reconstruction of the population 768 history, namely the changing on effective population size across time (Figure 1D), 769 was conducted by Bayesian skyline plot using the same software and parameter 770 settings using the sampled Lambda sequences as noted in the tip-dating analysis. 771 772 Protein Structure Homology Model 773 All protein structural analyses were performed using Discovery Studio

2020) was used as the template, and 40 776 homology models of the SARS-CoV-2 S of the Lambda variant were generated using 777

Evaluation of 778 the homology models were performed using PDF total scores and DOPE scores and 779 the best model for the Lambda S was selected

Plasmids expressing the S proteins of the Alpha 787 (B.1.1.7), Beta (B.1.351), Gamma (P.1), Delta (B.1.617.2), Lambda (C.37) variants 788 and the point mutants were generated by site-directed overlap extension PCR using 789 pC-SARS2-S D614G (Ozono et al., 2021) as the template and the following primers 790 listed in Table S4. The resulting PCR fragment was digested with Acc65I or KpnI 791 and NotI and inserted into the corresponding site of the pCAGGS vector

Pseudovirus Assay

Pseudovirus assay was performed as previously described

Briefly, the pseudoviruses, lentivirus (HIV-1)-based, luciferase-799 expressing reporter viruses pseudotyped with the SARS-CoV-2 S protein and its 800 derivatives, HEK293T cells (1 × 10 6 cells) were cotransfected with 1 μg

1 μg of pWPI-Luc2 (Ozono et al., 2020), and 500 ng 802 of plasmids expressing parental S or its derivatives using

At two days posttransfection, the culture 805 supernatants were harvested, centrifuged. The amount of pseudoviruses prepared 806 was quantified using the HiBiT assay as previously described

The pseudoviruses prepared were stored at -80°C until use. 808 For the experiment, HOS-ACE2 cells and HOS-ACE2/TMPRSS2 cells (10,000 809 cells/50 μl) were seeded in 96-well plates and infected with 100 μl of the 810 pseudoviruses prepared at 4 different doses. At two days postinfection, the infected 811 cells were lysed with a One-Glo luciferase assay system (Promega, Cat# E6130), 812 and the luminescent signal was measured using a CentroXS3 plate reader 813 (Berthhold Technologies) or GloMax explorer multimode microplate reader

Antibody Treatment

Briefly, this assay was performed on 819 HOS-ACE2/TMPRSS2 cells using the SARS-CoV-2 S pseudoviruses expressing 820 luciferase (see "Pseudovirus Assay" above). The SARS-CoV-2 S pseudoviruses 821 (counting ~20,000 relative light unit) were incubated with serially diluted heat-822 inactivated human sera, a NTD-targeting NAb clone 4A8

QUANTIFICATION AND STATISTICAL ANALYSIS 831 Data analyses were performed using Prism 9