viruses Article Previous SARS-CoV-2 Infection Increases B.1.1.7 Cross-Neutralization by Vaccinated Individuals Benjamin Trinité 1, *,† , Edwards Pradenas 1,† , Silvia Marfil 1 , Carla Rovirosa 1 , Víctor Urrea 1 , Ferran Tarrés-Freixas 1 , Raquel Ortiz 1 , Jordi Rodon 2 , Júlia Vergara-Alert 2 , Joaquim Segalés 3,4 , Victor Guallar 5,6 , Rosalba Lepore 5 , Nuria Izquierdo-Useros 1 , Glòria Trujillo 7 , Jaume Trapé 7 , Carolina González-Fernández 7 , Antonia Flor 7 , Rafel Pérez-Vidal 7 , Ruth Toledo 8 , Anna Chamorro 8 , Roger Paredes 1,8 , Ignacio Blanco 9 , Eulàlia Grau 1 , Marta Massanella 1 , Jorge Carrillo 1 , Bonaventura Clotet 1,8,10 and Julià Blanco 1,10, * 1 IrsiCaixa AIDS Research Institute, Germans Trias i Pujol Research Institute (IGTP), Can Ruti Campus, Autonomous University of Barcelona (UAB), 08916 Badalona, Spain; epradenas@irsicaixa.es (E.P.); smarfil@irsicaixa.es (S.M.); crovirosa@irsicaixa.es (C.R.); vurrea@irsicaixa.es (V.U.); ftarres@irsicaixa.es (F.T.-F.); rortiz@irsicaixa.es (R.O.); nizquierdo@irsicaixa.es (N.I.-U.); rparedes@irsicaixa.es (R.P.); egrau@irsicaixa.es (E.G.); mmassanella@irsicaixa.es (M.M.); jcarrillo@irsicaixa.es (J.C.); bclotet@irsicaixa.es (B.C.) 2 Institute de Recerca i Tecnologia Agrària (IRTA), Centre de Recerca en Sanitat Animal (CReSA, IRTA-UAB), Campus de la UAB, 08193 Bellaterra, Spain; jordi.rodon@irta.cat (J.R.); julia.vergara@irta.cat (J.V.-A.) 3 Centre de Recerca en Sanitat Animal (CReSA, IRTA-UAB), Campus de la UAB, Autonomous University of   Barcelona (UAB), 08193 Bellaterra, Spain; joaquim.segales@irta.cat 4 Departament de Sanitat i Anatomia Animals, Facultat de Veterinària, UAB, 08193 Bellaterra, Spain Citation: Trinité, B.; Pradenas, E.; 5 Barcelona Supercomputing Center, 08034 Barcelona, Spain; victor.guallar@bsc.es (V.G.); Marfil, S.; Rovirosa, C.; Urrea, V.; alba.lepore@bsc.es (R.L.) Tarrés-Freixas, F.; Ortiz, R.; Rodon, J.; 6 Catalan Institution for Research and Advanced Studies (ICREA), 08010 Barcelona, Spain 7 Fundació Althaia, Hospital de Sant Joan de Déu, 08243 Manresa, Spain; gtrujillo@althaia.cat (G.T.); Vergara-Alert, J.; Segalés, J.; et al. Previous SARS-CoV-2 Infection jtrape@althaia.cat (J.T.); cgonzalezf@althaia.cat (C.G.-F.); aflor@althaia.cat (A.F.); rperez@althaia.cat (R.P.-V.) 8 Infectious Diseases Department, Fight against AIDS Foundation (FLS), Germans Trias i Pujol Hospital, Increases B.1.1.7 Cross-Neutralization 08916 Badalona, Spain; rtoledo@flsida.org (R.T.); achamorro@flsida.org (A.C.) by Vaccinated Individuals. Viruses 9 Germans Trias i Pujol Hospital, 08916 Badalona, Spain; iblanco.germanstrias@gencat.cat 2021, 13, 1135. https://doi.org/ 10 Chair of Infectious Diseases and Immunity, University of Vic–Central University of Catalonia (UVic-UCC), 10.3390/v13061135 08500 Vic, Spain * Correspondence: btrinite@irsicaixa.es (B.T.); jblanco@irsicaixa.es (J.B.); Tel.: +34-934-656-374 (B.T. & J.B.); Academic Editors: Fax: +34-934-653-968 (B.T. & J.B.) Luis Martinez-Sobrido and † Equal contribution. Fernando Almazan Toral Abstract: With the spread of new variants of severe acute respiratory syndrome coronavirus 2 (SARS- Received: 2 May 2021 CoV-2), there is a need to assess the protection conferred by both previous infections and current Accepted: 10 June 2021 vaccination. Here we tested the neutralizing activity of infected and/or vaccinated individuals Published: 12 June 2021 against pseudoviruses expressing the spike of the original SARS-CoV-2 isolate Wuhan-Hu-1 (WH1), the D614G mutant and the B.1.1.7 variant. Our data show that parameters of natural infection (time Publisher’s Note: MDPI stays neutral from infection and nature of the infecting variant) determined cross-neutralization. Uninfected with regard to jurisdictional claims in vaccinees showed a small reduction in neutralization against the B.1.1.7 variant compared to both the published maps and institutional affil- WH1 strain and the D614G mutant. Interestingly, upon vaccination, previously infected individuals iations. developed more robust neutralizing responses against B.1.1.7, suggesting that vaccines can boost the neutralization breadth conferred by natural infection. Keywords: SARS-CoV-2; humoral response; pseudovirus; neutralization; B.1.1.7 variant Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and 1. Introduction conditions of the Creative Commons Attribution (CC BY) license (https:// Early in the COVID-19 pandemic, severe acute respiratory syndrome coronavirus creativecommons.org/licenses/by/ 2 (SARS-CoV-2) variants started to develop regionally and globally. Currently, the rapid 4.0/). spread of the B.1.1.7, or 501Y.V1, variant [1], first reported in the UK, casts doubts on the Viruses 2021, 13, 1135. https://doi.org/10.3390/v13061135 https://www.mdpi.com/journal/viruses spread of the B.1.1.7, or 501Y.V1, variant [1], first reported in the UK, casts doubts on the protection conferred by the neutralizing antibody response acquired during a previous immunization. Besides the D614G mutation, the B.1.1.7 variant contains six non-synony- mous mutations and three deleted amino acids in the spike (S) protein (Figure 1A). The major changes are the mutation N501Y in the receptor-binding domain (RBD); the dele- Viruses 2021, 13, 1135 2 of 12 tion 69–70 which may increase transmissibility [2] and produces a false negative in certain RT-PCR-based diagnostic assays; and the mutation P681H, next to the furin cleavage site, that could impact antigenicity and enhance viral infectivity. Recent studies indicate that protection conferred by the neutralizing antibody response acquired during a previous im- B.1.1.7 is associated with a higher hospitalization risk [3] and higher mortality [4] and munization. Besides the D614G mutation, the B.1.1.7 variant contains six non-synonymous several reports indicate that it has a higher secondary attack rate, making this viral variant mutations and three deleted amino acids in the spike (S) protein (Figure 1A). The ma- 30–50% more transmissible [5]. Importantly, this variant remains susceptible to some jor changes are the mutation N501Y in the receptor-binding domain (RBD); the deletion monoclonal and plasma antibodies from convalescent or vaccinated individuals [6–8]. 69–70 which may increase transmissibility [2] and produces a false negative in certain RT- Here we analyzed cross-neutralizing plasma antibody titers in individuals infected PCR-based diagnostic assays; and the mutation P681H, next to the furin cleavage site, that during both the first and second waves of COVID-19 epidemics in Catalonia (Spain), as could impact antigenicity and enhance viral infectivity. Recent studies indicate that B.1.1.7 well as in vaccinated individuals. The B.1.1.7 variant showed minimal resistance to the is associated with a higher hospitalization risk [3] and higher mortality [4] and several neutralizing capacity from both infected and vaccinated individuals, but its impact was reports indicate that it has a higher secondary attack rate, making this viral variant 30–50% significantly more morepronounced transmissibleon[5]. theImportantly, latter group. Interestingly, this previous variant remains infection susceptible sig- monoclonal to some nificantly improved neutralization and plasma antibodies titers fromagainst this variant convalescent upon vaccination. or vaccinated individuals [6–8]. Figure 1. In vitro infectivity of SARS-CoV-2 variants. (A). The different mutations identified in the B.1.1.7 variant are listed Figure 1. In vitro infectivity of SARS-CoV-2 variants. (A). The different mutations identified in the and their location in the spike protein (side and top views) is shown. This variant also includes the D614G mutation. B.1.1.7 variant are listed and their location in the spike protein (side and top views) is shown. This (B). Spike expression in pseudovirus-producing cells stained with an anti-RBD polyclonal rabbit antibody (See Methods variant also includes the D614G mutation. (B). Spike expression in pseudovirus-producing cells for details). stained with an anti-RBD polyclonal rabbit antibody (See Methods for details). Here we analyzed cross-neutralizing plasma antibody titers in individuals infected 2. Materials and Methods during both the first and second waves of COVID-19 epidemics in Catalonia (Spain), 2.1. Study Overview as wellandasSubjects in vaccinated individuals. The B.1.1.7 variant showed minimal resistance to The studythe was neutralizing approved by capacity from Ethics the Hospital both infected andBoard Committee vaccinated individuals, from Hospital Uni-but its impact was significantly versitari Germans Trias i Pujolmore pronounced (PI-20-122 on the latter and PI-20-217) and group. Interestingly, all participants previous infection provided significantly written informed improved consent before neutralization titers against this variant upon vaccination. inclusion. Plasma samples were obtained from the prospective KING cohort of the HUGTiP 2. Materials (Badalona, Spain) and fromand Methods Althaia (Manresa, Spain). The KING cohort included individ- 2.1. Study Overview uals with a documented positiveand Subjectsresult from nasopharyngeal swab and/or a RT-qPCR positive serologicalThe study was diagnostic test.approved by the Hospital Ethics Committee Board from Hospital Universitari Samples in this study Germans Trias ifrom were collected PujolMarch (PI-20-122 2020 and PI-20-217) to February andthus, 2021; all participants cover- provided written ing the different informed COVID-19 consent outbreaks inbefore inclusion. Catalonia (dadescovid.cat). We analyzed 32 non- Plasma vaccinated individuals samples infected were obtained in March from 2020, using the prospective plasma KING cohort samples collected at a me-of the HUGTiP dian of 48 days (n = 16) or 196 (n = 16) days after symptom onset. We also selected 16included indi- (Badalona, Spain) and from Althaia (Manresa, Spain). The KING cohort vidualsinwith individuals infected a documented August positivesamples 2020 using plasma RT-qPCR result from collected nasopharyngeal at a median of 44 daysswab and/or a positive serological diagnostic test. Samples in this study were collected from March 2020 to February 2021; thus, covering the different COVID-19 outbreaks in Catalonia (dadescovid.cat). We analyzed 32 non- vaccinated individuals infected in March 2020, using plasma samples collected at a median of 48 days (n = 16) or 196 (n = 16) days after symptom onset. We also selected 16 individuals infected in August 2020 using plasma samples collected at a median of 44 days after symptom onset, and 5 patients (n = 13 samples) infected in January 2021 by the B.1.1.7 variant. Finally, 32 individuals having received two doses of Pfizer/BioNTech vaccine were sampled 2 weeks after the second dose. This last group included uninfected and long-term Viruses 2021, 13, 1135 3 of 12 previously-infected individuals. A description of the different groups and subgroups is shown in Table 1. Table 1. Description of participants. Uninfected individuals were included as negative controls for neutralizing activity. All of them showed undetectable neutralizing activity. IQR: interquartile range. Uninfected Infected Non-Vaccinated Vaccinated n=5 n = 53 n = 32 Infection Status Uninfected Infected Infected Infected Infected Uninfected Date of Infection March 2020 August 2020 January 2021 March 2020 Strains D614/G614 20E (EU1) B.1.1.7 D614/G614 Sampling Early Late n = 16 n = 16 n = 16 n=5 n = 16 n = 16 Age 46 65 56 44 79 39 45 (years), median (IQR) (42–52) (55–68) (54–62) (37–54) (60–91) (29–44) (30–61) Gender (female), 4 (80) 4 (25) 7 (44) 8 (50) 2 (40) 11 (69) 12 (75) n (%) Days from symptom 48 196 44 16 324 — — onset, median (IQR) (36–57) (186–207) (37–54) (8–20) * (184–339) Days from vaccination, 13 9 — — — — — median (IQR) (10–14) (7–12) Hospitalized, n (%) — 11 (69) 11 (69) 11 (69) 5 (100) 0 (0) — Severe (%) — 6 (38) 6 (38) 5 (31) 0 (0) 0 (0) — * Data from 13 samples obtained from 5 participants in this group. 2.2. Cell Lines HEK293T cells (presumably of female origin) overexpressing WT human ACE-2 (Integral Molecular, Philadelphia, PA, USA) were used as a target for a SARS-CoV-2 spike-expressing pseudovirus infection. The cells were maintained in T75 flasks with Dulbecco0 s Modified Eagle0 s Medium (DMEM) supplemented with 10% FBS and 1 µg/mL of Puromycin. 2.3. Spike Plasmid Generation SARS-CoV-2.Sct∆19 WH1 and B.1.1.7 were generated (Geneart) from the full protein sequence of the original SARS-Cov-2 isolate Wuhan-Hu-1 (WH1) and the UK variant (B.1.1.7) spike sequences respectively, with the deletion of the last 19 amino acids in C- terminal [9], human-codon optimized and inserted into pcDNA3.1(+). The D614G spike mutant was generated by site-directed mutagenesis as previously described [10]. In brief, SARS-COV-2.Sct∆19 WH1 plasmid was amplified by PCR with Phusion DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA, ref# F-549S) and the following primers: 50 -TACCAGGgCGTGAACTGTACCGAAGTGCC-30 and 50 -GTTCACGcCCTGGTACAGCA CTGCCAC-30 . PCR was 20 cycles with an annealing temperature of 60 ◦ C and an elon- gation temperature of 72 ◦ C. The PCR product was then treated for 3 h with the DpnI restriction enzyme (Thermo Fisher Scientific, ref# ER1705), to eliminate template DNA, and transformed into supercompetent E. coli. The final mutated DNA was then fully sequenced for validation. 2.4. Pseudovirus Generation and Neutralization Assay HIV reporter pseudoviruses expressing SARS-CoV-2 S protein and Luciferase were generated using the defective HIV plasmid pNL4-3.Luc.R-.E- obtained from the NIH AIDS Reagent Program [11]. Expi293F cells were transfected using ExpiFectamine293 Reagent (Thermo Fisher Scientific) with pNL4-3.Luc.R-.E- and SARS-CoV-2.Sct∆19 (WH1, G614 or B.1.1.7), at an 8:1 ratio, respectively. Control pseudoviruses were obtained by replacing the S protein expression plasmid with a VSV-G protein expression plasmid as reported [12]. Supernatants were harvested 48 h after transfection, filtered at 0.45 µm, frozen and titrated on HEK293T cells overexpressing WT human ACE-2. Spike expression in pseudovirus- producing cells was confirmed by flow cytometry, showing comparable expression for all constructs (Figure 1B). Viruses 2021, 13, 1135 4 of 12 Neutralization assays were performed in duplicate. Briefly, in Nunc 96-well cell culture plates (Thermo Fisher Scientific), 200 TCID50 of pseudovirus were preincubated with three-fold serial dilutions (1/60–1/14,580) of heat-inactivated plasma samples for 1 h at 37 ◦ C. Then, 2 × 104 HEK293T/hACE2 cells treated with DEAE-Dextran (Sigma-Aldrich) were added. Results were read after 48 h using the EnSight Multimode Plate Reader and BriteLite Plus Luciferase reagent (Perkin Elmer, Waltham, MA, USA). The values were normalized, and the ID50 (the reciprocal dilution inhibiting 50% of the infection) was calculated by plotting and fitting the log of plasma dilution vs. response to a 4-parameter equation in Prism 8.4.3 (GraphPad Software, San Diego, CA, USA). This neutralization assay had been previously validated in a large subset of samples [13,14]. The lower limit of detection was 60 and the upper limit was 14,580 (reciprocal dilution). 2.5. Flow Cytometry Transfected Expi293F cells were first stained extracellularly with a polyclonal rabbit anti-spike RBD antibody (Sino Biological, Beijing, China, ref# 40592-T62) and a secondary APC labeled anti-rabbit (Jackson Immuno Research, West Grove, PA, USA, ref# 711-605- 152). Cells were then fixed (Life Technologies, Carlsbad, CA, USA, ref# GAS001S100) and stained in permeabilization buffer (Life Technologies, ref# GAS002S100), with a FITC- labeled mouse anti-p24Gag antibody KC57 (Beckman Coulter, Brea, CA, USA, ref# 6604665). Cells were acquired on a BD Celesta flow cytometer with DIVA software and analyzed on FlowJo vX.0.7 (Tree Star, Inc., Ashland, OR, USA). 2.6. Statistical Analysis Continuous variables were described using medians and the interquartile range (IQR, defined by the 25th and 75th percentiles), whereas categorical factors were reported as per- centages over available data. Quantitative variables were compared using the Mann–Whitney test; and percentages using the chi-squared test. The Friedman test with Dunn’s multiple comparison test was used to compare neutralization of different pseudoviruses. Multiple M–W comparisons were corrected by false discovery rate. Analyses were performed with Prism 8.4.3 (GraphPad Software) and R version 4.0 (R Foundation for Statistical Computing). 3. Results 3.1. Global Analysis of Cross-Neutralizing Titers in Vaccinated and Infected Participants As a first approach, we tested the neutralizing antibody response of all participants segregated in two main groups labeled “all vaccinated” and “all infected”. “All vaccinated” included all the vaccinated participants whether they had experienced previous natural infection or not. “All infected” included all non-vaccinated participants who were infected during either the first wave, the second wave or specifically infected by the B.1.1.7 variant. We tested all the plasma samples (n = 98) against pseudoviruses expressing three different spike glycoproteins: a spike corresponding to the original SARS-CoV-2 virus, isolated in Wuhan, and named here WH1; a D614G mutant based on the WH1 spike and a spike including the defining mutations of the B.1.1.7 variant and named B.1.1.7. The neutralization titers of all vaccinated individuals against B.1.1.7 and WH1 were not statistically different (according to the Friedman test with Dunn’s multiple comparison test) (Figure 2A), despite WH1 expressing the spike sequence on which the vaccine was based. In contrast, vaccinated individuals showed significantly higher potency to neutral- ize the intermediate D614G mutant (p < 0.0001) (Figure 2A). A similar analysis, including all non-vaccinated infected individuals, showed similar results (Figure 2B). The highest neutralization (p < 0.0001) was noticed for the D614G mutant, while no significant differ- ences were observed between WH1 and B.1.1.7. Next, we compared cross-neutralization capacities by determining fold change ratios between a spike of interest (Sx , indicated on top) and a reference spike (Sr , indicated in the Y axis of Figure 2C,D, fold change = Sr /Sx ). This ratio is a measure of the loss of neutralizing capacity; a ratio inferior to one indicates a better neutralization of the spike of interest in comparison to the reference spike, and including all non-vaccinated infected individuals, showed similar results (Figure 2B). The highest neutralization (p < 0.0001) was noticed for the D614G mutant, while no significant differences were observed between WH1 and B.1.1.7. Next, we compared cross-neutrali- zation capacities by determining fold change ratios between a spike of interest (Sx, indi- Viruses 2021, 13, 1135 cated on top) and a reference spike (Sr, indicated in the Y axis of Figures 2C,D, fold change 5 of 12 = Sr/Sx). This ratio is a measure of the loss of neutralizing capacity; a ratio inferior to one indicates a better neutralization of the spike of interest in comparison to the reference spike, and vice versa. When B.1.1.7 and WH1 variants were compared, median fold- vice versa. change When was 0.97 B.1.1.7 in the and WH1 vaccinated variants group, were a value compared, higher significantly medianthanfold-change the one ob- was 0.97 in tained inthe vaccinated naturally group, infected a value significantly individuals higher (0.7; p = 0.033, M–W than the test, one obtained Figure in naturally 2C). Similarly, the infected fold change individuals (0.7; p and between B.1.1.7 = 0.033, D614GM–Wwastest, Figure 2C).different significantly Similarly, the fold between thechange all vac-be- cinated and the all infected groups (median values 1.9 vs 1.17, respectively; p < 0.0001, M–all tween B.1.1.7 and D614G was significantly different between the all vaccinated and the Winfected groups test, Figure 2D).(median values Altogether, 1.9data these vs 1.17, respectively; indicate p < 0.0001, that, relative to WH1 M–W and test, D614G,Figure 2D). cross- Altogether, these neutralization data indicate of B.1.1.7 was worse that, inrelative to WH1 vaccinated and D614G, individuals cross-neutralization in comparison to infectedof B.1.1.7 was worse in vaccinated individuals in comparison to infected ones. ones. Figure Figure 2. 2. Global Global analysis analysis ofof neutralizationtiters neutralization titersininSARS-CoV-2 SARS-CoV-2vaccinated vaccinatedandandinfected infectedindividu- individuals. Values of ID als. Values of ID (as reciprocal dilution) are shown for all 50 50 (as reciprocal dilution) are shown for all plasmaplasma samples from (A) vaccinated and (A) vaccinated and (B)(B) infected infected non-vaccinated non-vaccinated individuals individuals against against thethe indicated indicated pseudoviruses. pseudoviruses. Bars Bars indicate indicate median median titer intiter each ingroup each group with awith 95%a confidence 95% confidence interval interval and pand p values values show show the comparison the comparison of of median median titers among the three viruses (Friedman test with Dunn’s multiple titers among the three viruses (Friedman test with Dunn’s multiple comparison test, * p < comparison test, * p <0.05, 0.05, **** p < 0.0001). The corresponding fold-change in neutralization titers between **** p < 0.0001). The corresponding fold-change in neutralization titers between (C) WH1 and B.1.1.7(C) WH1 and B.1.1.7 or (D) D614G and B.1.1.7 is shown (lower is better), comparing the vaccinated (VACC) and or (D) D614G and B.1.1.7 is shown (lower is better), comparing the vaccinated (VACC) and the infected (INF) groups. Bars indicate median in each group with 95% confidence interval and top values indicate the median fold-change between the indicated variants (variants compared are indicated in the graph title and in the Y axis). Fold change medians were compared using the Mann–Whitney test (* p < 0.05, **** p < 0.0001). 3.2. Identification of Parameters Influencing Cross-Neutralization To better understand these differences, we analyzed infected and vaccinated sub- groups. To assess the impact of sequence evolution on the immune responses, infected individuals were divided according to infection date. Individuals infected during the first wave (March 2020) in Spain were initially exposed to the original D614 virus that was rapidly displaced by the G614 variant. An evolving 85 to 22% prevalence for the original variant has been estimated [15]. Individuals infected during the second wave Viruses 2021, 13, 1135 6 of 12 (August 2020) were almost exclusively exposed to the G614-containing 20E (EU1) lin- eage, which accounted for nearly 100% of new infections during the summer of 2020 (https://nextstrain.org/ncov/global, accessed on 31 March 2021). Individuals infected by the B.1.1.7 variant, identified in January 2021, were also analyzed. First, we analyzed the neutralization response specifically in participants infected during the first wave. Individuals sampled 48 days after infection showed a small but significant decrease in neutralization capacity against the B.1.1.7 variant when compared with the D614G mutant (median fold change of 1.53, p = 0.031, Friedman test, Figure 3A,B). When compared to WH1, no significant difference was observed (median = 1.20-fold, p = 0.155). Individuals infected on March 2020 and sampled 6 months later also showed a general decay of the neutralization response as previously reported by us and many others [14,16]. Interestingly though, in these limited cohorts, decay of the neutralization response was only significant when measured against WH1 (p = 0.0374, K-W, Figure 3A) and not against D614G or B.1.1.7. In fact, we observed a general trend of improving cross- neutralization capacities against both D614G and the B.1.1.7 variant when compared with WH1, although it was only significant for the latter (fold change evolving from 1.2 to 0.6, p = 0.024, M–W, Figure 3B). These data suggest that the neutralization response induced by natural infection continued to evolve several months after the initial symptoms. We then looked at the neutralization responses induced during the subsequent waves of infection. Individuals infected in August 2020 or infected by the B.1.1.7 variant be- haved differently from those infected in March 2020, in that no significant differences were detected between neutralizing titers against all three tested spikes (Figure 3C). When comparing ratios with first wave participants, B.1.1.7 infected individuals showed a signifi- cant increase in the neutralization of B.1.1.7 vs D614G (fold change going from 1.53 to 1, p = 0.0006) (Figure 3D). These data illustrate the progressive evolution of the virus infecting the population and its impact on the induced cross-neutralization response. Importantly, individuals infected by the B.1.1.7 variant were still able to cross-neutralize the original WH1 spike and the D614G mutant. Finally, we analyzed vaccinated individuals. Vaccinated individuals were sub-classified according to previous COVID-19 evidence into infected or uninfected. We compared three groups: vaccinated non-infected, vaccinated previously infected during the first wave and the group of first wave infected participants (late plasma sampling) not vaccinated. As we did not have the neutralization levels of vaccinated previously infected individuals prior to vaccination, we used the group of first wave infected (late plasma) individuals as a surrogate. When analyzing specifically vaccinated individuals not previously infected, the decrease in B.1.1.7 cross-neutralization already observed for the all vaccinated group (Figure 2A) was even more clear (Figure 4A). This was significant in comparison to both WH1 (fold change of 2.04, p = 0.0021) and D614G (fold change of 2.65, p = <0.0001). Vac- cinated individuals previously infected did not show such a decrease; in fact, B.1.1.7 neutralization was increased compared to WH1 (fold change of 0.60, p = 0.0034). B.1.1.7 neutralization was still reduced when compared to D614G (fold change of 1.55, p = 0.0027) but, overall, the response in the vaccinated previously infected group was improved in comparison to vaccinated only individuals (Figure 4A,B). Viruses2021, Viruses 2021,13, 13, 1135 x FOR PEER REVIEW 8 of 14 7 of 12 Figure 3. Subgroup analysis of neutralization titers in infected individuals. (A,B). Comparison of plasmas collected at 48 days (early sampling) and >6 month (late sampling) after infection from Viruses 2021, 13, 1135 8 of 12 2 independent groups of participants infected in March 2020 (first wave). Neutralization titers (ID50 expressed as reciprocal dilutions) are shown in (A), and the corresponding ratios between variants (lower is better) are shown in (B). (C,D) Neutralizing titer of individuals infected during the second wave (August 2020) or specifically by the B.1.1.7 variant. Neutralization titers (ID50 expressed as reciprocal dilutions) are shown in (C), and the corresponding ratios between variants (lower is better) are shown in (D). In (D), we also included ratios from the first wave infected individuals (early sampling) for comparison. In (A,C), bars indicate median titer in each group and p values show the comparison of median titers among the three viruses (Friedman test with Dunn’s multiple comparison test; * p < 0.05, **** p < 0.0001). Specifically in (A), p values are also indicated for the comparison of neutralization titers against the same spike between the 2 groups (Kruskal–Wallis test; * p < 0.05). In (B,D), bars indicate the median in each group with a 95% confidence interval and top values indicate the median fold-change between the indicated variants (variants compared are indicated in the graph title and in the Y axis). p values show the comparison of the group sampled at 48 days vs the group sampled at 6 months (Mann–Whitney test; * p < 0.05). In (D), p values show the comparison between individuals from the different infection waves (Kruskal–Wallis with Dunn’s multiple comparison test; *** p < 0.001). Neutralization titers in infected only individuals were generally lower than in the vaccinated groups (Figure 4A) as the former group experienced a decay in the neutralizing response over time (Figure 3A) [14,16] while vaccinated participants were sampled between 1 and 2 weeks after the second shot, when the neutralizing response was maximized [17]. However, in comparison to vaccinated only individuals, the cross-neutralizing response against both D614G and B.1.1.7 was superior to WH1 and the response against B.1.1.7 was only marginally reduced compared to D614G (fold change of 1.28, p = 0.034). Interestingly, the cross-neutralization profile of vaccinated previously infected individuals was very similar to infected only participants as there was no significant difference between the two groups when looking at the ratios of the neutralizing response between variants (Figure 4B). This suggests that vaccination was able to boost the full cross-neutralization breadth of the response previously established by natural infection during the first wave. In fact, in these limited cohorts, the averages of the titers against each variant were all increased by a factor of four when considering the average titers of vaccinated and non-vaccinated previously infected groups (Figure 4A). Viruses Viruses2021, 2021,13, 13,x 1135 FOR PEER REVIEW 10 9ofof1412 Figure 4. Subgroup analysis of neutralization titers in vaccinated individuals. Comparison of plasma Figure 4. Subgroup analysis of neutralization titers in vaccinated individuals. Comparison of from vaccinated plasma individuals, from vaccinated previously individuals, infected previously or not during infected the firstthe or not during wave, first as well as wave, as well late plasma as from first wave infected individuals. (A) Neutralization late plasma from first wave infected individuals. (A) Neutralization titers (ID expressed as reciprocal 50 titers (ID50 expressed as dilutions). recip- Barsdilutions). rocal indicate theBars median indicate titer thein each group median titer inwith eachagroup 95% confidence with a 95% interval. confidence p values show interval. p the comparison values show the of comparison median titers ofagainst medianthe three titers variants against in thevariants the three same groupin the(Friedman same group with Dunn’s (Fried- man with Dunn’s multiple multiple comparison test comparison test ;p*