Abstract
With more human papillomavirus (HPV) virus-like particle (VLP) vaccines to hit the market in future, a monoclonal antibody (mAb) with preferably comparable reactivity against vaccines from different expression systems and bioprocesses is urgently needed for the potency characterization. Among all mAbs against HPV16 collected, rabbit mAb H16.001 is potently neutralizing with the highest affinity, recognizes an immune-dominant epitope, and can comparably react with HPV16 vaccines from various sources. Cryo-electron microscopic (cryo-EM) structure demonstrated that 360 H16.001 Fabs could bind to HPV16 capsid in preferable binding manner without steric hindrance between neighboring Fabs, potentially supporting its identification for VLP structural integrity and utility in monitoring VLP structural probity. This structural analysis indicated that mAb H16.001 afforded unbiased potency characterization for various HPV16 vaccines and was potential for use in vaccine regulation practice. This study also showed a model process for selecting suitable mAbs for potency assays of other vaccines.
Similar content being viewed by others
Introduction
Cervical cancer is the fourth most common female malignancy worldwide and represents a major global health challenge1. Among the human papillomavirus (HPV) types associated with this carcinoma, HPV16 is the most prevalent genotype and accounts for 50% of tumor specimens2.
The major capsid protein L1 of HPV can self-assemble into empty virus-like particles (VLPs) that are virion sized, highly immunogenic, and therefore the immunogens in several HPV vaccines, including Cervarix, Gardasil, Gardasil 9, and Cecolin3,4,5,6,7. These HPV vaccines conferred high protection efficacy against CIN2+ symptom in clinical trials and the protection is believed to be mediated by L1-neutralizing antibodies, which can be detected in the sera and cervicovaginal secretions of naturally infected or vaccinated individuals8,9.
Potency characterization is one of the most critical items in the quality control (QC) and regulatory lot release for HPV vaccines. Vaccine potency assay is usually conducted in vivo by an ED50 (half of effective dosage) value associated with neutralizing antibody elicitation in experimental mouse or alternatively measured in vitro by a neutralizing monoclonal antibody (mAb)-based double-antibody sandwich enzyme-linked immunosorbent assay (ELISA), also referred as in vitro relative potency (IVRP) that is well established with significant correlation to ED50 test. However, it is a challenge to develop a common IVRP test across various HPV vaccines, VLP antigens of which are produced in different expression systems, such as yeast, baculovirus-based insect cell and bacterial, or distinct manufacturing processes and therefore require an mAb that should recognize immunodominant neutralizing epitope and equally react with all sorts of VLPs.
For the purpose of finding an optimal antibody, a panel of mAbs against HPV16 L1 VLPs was characterized by a variety of immune assays. The rabbit mAb H16.001 was the optimal candidate for the assessment of the epitope integrity and antigenicity of VLP vaccine products, in the virtue of its potent neutralizing activity, high affinity in surface plasma resonance (SPR), and comparable reactivity against VLPs derived from different expression systems. An atomic model of HPV16 pseudovirus (PsV) in complex with H16.001 was obtained to demonstrate a unique binding preference for the IVRP assay. This study provides insights for selecting mAb used for a QC reagent for regulatory approval that may cover different manufacturing process and different expression systems.
Results
Neutralization profile of a panel of HPV16 mAbs
To establish a common neutralizing mAb (nAb)-based characterization approach to HPV16 vaccines generating from various expression systems, we first measured the in vitro neutralizing activities of a collection of 13 HPV16 mAbs (H16.001, H16.25F12, H16.26H7, H16.4G12, H16.V5, H16.8A9, H16.17B10, H16.6C7, H16.22G5, H16.25H8, H16.20G3, H16.1D12, and H16.5C10) by a PsV-based neutralization assay. The tests were conducted at a starting mAb concentration of 7.5 μg/mL, subsequent twofold serial dilution and inhibition ratio curves were generated for the 50% maximal inhibitory concentration (IC50) calculation. Six mAbs—H16.001, H16.25F12, H16.26H7, H16.8A9, H16.V5, and H16.4G12—showed excellent neutralizing activities with IC50 values of 0.40, 0.81, 1.29, 6.17, 6.77, and 7.52 ng/mL (Table 1 and Fig. 1), which were chosen for further study, while the rest 7 mAbs, including H16.6C7, H16.22G5, H16.17B10, H16.25H8, H16.1D12, H16.20G3, and H16.5C10, exhibited lower neutralizing activities with IC50 values of 7.62, 10.53, 20.12, 53.45, 192.30, 220.22, and 392.19 ng/mL. Notably, H16.001 had the most potent neutralizing activity, with 18-fold higher than that of the well-known mAb H16.V510,11.
Binding affinity and kinetics of a panel of HPV16 mAbs
Previous site-directed mutagenesis study showed that mAbs H16.25F12, H16.26H7, and H16.4G12—among the top six nAbs in the above-mentioned panel—shared some key epitope amino acids, suggesting that their binding regions are overlapping12. We then selected four representative mAbs, including H16.001, H16.8A9, H16.V5, and H16.4G12, for further binding study. Through SPR assay, as shown in Table 2 and Supplementary Fig. 1, H16.001 showed the highest affinity against HPV16 L1 pentamer as KD of 4.22 × 10−11 M, while H16.8A9, H16.V5, and H16.4G12 possessed relatively high affinities of KD with 1.18 × 10−10, 2.89 × 10−9, and 1.37 × 10−9 M, respectively, although they are lower than that of H16.001. Notably, H16.001 demonstrated the binding affinity as high as 67-fold than H16.V5 that was used in characterization of Gardasil13.
Reactivity of mAbs to HPV16 VLPs from various expression systems
To develop a common IVRP test for HPV16 VLPs derived from different expression systems (Table 3), we sought to identify some mAbs affordable to recognize various source VLPs comparably. The VLP samples Y1.16, and Y2.16 were produced in yeast expression system, E1.16, and E2.16 from Escherichia coli expression system, and I1.16 from insect cell expression system. Interestingly, H16.001 and H16.8A9 reacted with all sorts of HPV16 VLPs comparably as manifested by almost overlapping binding response curves for five HPV16 VLP samples, indicating that H16.001 and H16.8A9 reactivities are tolerant and irrespective of the expression systems that generated the HPV16 VLPs, whereas H16.V5 reacted the best with E. coli-derived E1.16 and worst with insect cell-derived I1.16. H16.4G12 bound with E. coli-derived VLPs (E1.16 and E2.16) preferably (Fig. 2). The commonality of H16.001 or H16.8A9 across distinct expression systems for VLP generation suggested its potential for IVRP test development used in vaccine regulation practice.
Immunodominance in anti-HPV16 sera of mAbs
Immunodominance analysis for an mAb by competition ELISA (c/ELISA) or blocking ELISA (b/ELISA) usually reflect on the major neutralization epitope that generates corresponding antibodies in the anti-sera. We then measured the blocking ratio of the 4 mAbs against 14 guinea pig sera (Supplementary Table 1) from animals vaccinated with preparations of VLPs produced from 5 different manufacturing processes using yeast, E. coli, or insect cell expression through c/ELISA. BID50 value (a serum dilution titer capable of blocking mAb from binding to HPV VLPs by 50%) was used to quantify the dominance of specified mAb. mAb H16.001 had BID50 ranging from 295 to 8356, from 935 to 8356 for H16.8A9, from 100 to 1673 for H16.V5, and from 184 to 25,586 for H16.4G12 as blocking by the guinea pig serum samples in c/ELISA. Meanwhile, the neutralization titers (denoted as 50% maximal dilution concentration (ID50) value) for the sera were measured by PBNA and ranged from 442,588 to 4,425,884. Then correlation between BID50 and ID50 for each serum sample was plotted for the 4 mAbs and analyzed by linear regression fitting; H16.001 and H16.8A9 showed significant correlation between BID50 and ID50 as relatedness coefficient values of 0.417 (P < 0.05) and 0.634 (P < 0.001), respectively (Fig. 3a, b), H16.4G12 showed lower linear correlation with relatedness coefficient value of 0.294 (P < 0.05; Fig. 3d), whereas H16.V5 had no significant correlation (Fig. 3c). These results indicated that H16.001 and H16.8A9 were more immunodominant in vaccinated guinea pig sera with regard to neutralization titer.
Next, we explored the blocking ratio of the 4 mAbs against HPV16 reactive human sera (Supplementary Table 1, N = 50) by b/ELISA in which HPV16 VLPs were premixed with mAb and then reacted with human sera. Notably, the human sera were obtained from individuals vaccinated with a yeast-derived VLP16/18 vaccine under clinical development (ClinicalTrials.gov ID:2011L01085). As shown in Fig. 4, H16.001 reduced 33–60% reactivities against HPV16 VLPs for 27 human sera, 12 of which were inhibited by >50%; H16.8A9 showed the most inhibition ratio for all the 50 human vaccinated sera as 52–91%, demonstrating that H16.8A9 epitope was immunogenic in the vaccine-induced humoral response in human; H16.V5 reduced 32–67% reactivities against HPV16 VLPs for 29 human sera, 7 of which were inhibited by >50%; H16.4G12 reduced 31–55% reactivities against HPV16 VLPs for 21 human sera, 3 of which were inhibited by >50%. Taken together, H16.001 and H16.8A9 identified some immunodominant neutralization epitope and their binding to HPV16 VLPs seemed to be irrespective of the expression systems that generated the VLPs.
Cryo-electron microscopic (cryo-EM) structure of HPV16 PsV in complex with Fab H16.001
As HPV16 VLPs were shown to be extremely heterogeneous, it was not suitable for high-resolution structure determination. To elucidate the potentially binding modality of HPV16.001 that preferably suits for vaccine characterization, we prepared the immune complex (termed HPV-001) by mixing HPV16 PsV with Fab H16.001 at a molar ratio of 1:1.5 for cryo-EM analysis. The cryo-EM micrographs and two-dimensional (2D) classifications analysis revealed that the HPV-001 immune complex have a generally homogenous population with a diameter of ~700 Å. Further, the clearly observed protrusions demonstrated the Fab binding to the viral particle (Fig. 5a and Supplementary Fig. 2a). Finally, we selected a total of 9162 particles to reconstruct the immune-complex structure imposing icosahedral symmetry with cisTEM14. The resolution of the final map was determined at 4.41 Å using the “gold” standard Fourier Shell Correlation (FSC) = 0.143 criterion (Supplementary Fig. 2b)15. The central section of HPV-001 density map shows the high quality of the reconstructions (Fig. 5b). High-resolution cryo-EM density map in radial color revealed H16.001 Fabs could bind both 5-coordinated and 6-coordinated pentamer in HPV16 PsV, and thus a total of 360 Fabs engage to a T = 7 HPV16 capsid. (Fig. 5c, d and Supplementary Movie 1). Further, as the immune-complex structures reported in H16.V5, the Fab-binding occupancies were various within one icosahedral asymmetric unit (loci −1 to −6), especially in locations −1 and −2, which could be a result of steric hindrance (Supplementary Fig. 3)16. Unlike H16.V5-binding pattern, H16.001 Fab densities in 5-coordinated pentamer (location −1) did not overlap with that in 6-coordinated pentamer (loci −2 to −6) (Fig. 5d), which enable us to reconstruct the Fab density map more confidently.
However, such 4.41 Å resolution was still insufficient for model building, which might have resulted from the sample heterogeneity and the flexibility of the virion or the Fab. To conquer these issues and improve the resolution, we extracted the twofold axis region comprising the viral asymmetric unit with a diameter of 451.2 Å as subparticle for further refinement using the localized reconstruction procedure17. The resolution was finally determined to 3.43 Å by “gold” standard FSC = 0.143 criterion (Fig. 6a and Supplementary Fig. 2a). The backbone of the polypeptide, as well as amino acid side chains, can be clearly traced in the capsid and Fab variable domain (Fig. 6b, c), enabling us to build the majority of the capsid protein and the variable domain of H16.001 Fab but excluding the constant domain that was disordered density as shown in Fig. 6b. In an asymmetric unit, each fab binds across three L1 monomers (here termed chain-C, chain-D, and chain-E) with a total buried surface area of ~1183 Å2, with about 73% (~867 Å2) buried by the heavy chain (Fig. 6d–f). Of note, chain-D occupy a dominant role in Fab–capsid interaction with a buried surface of ~762 Å2 (64%), while for chain-C and chain-E it is only ~315 Å2 (27%) and ~106 Å2 (9%), respectively (Fig. 6d–f). Nineteen hydrogen bonds and salt bridges, as well as several van der Waals interactions, establish the elaborated interaction network, which were majorly donated from the DE and FG loops from chain-D, HI loop from chain-C, and FG loop from chain-E (Fig. 6g–i and Supplementary Table 2). Among these interaction sites, the residues D127 from chain-D and S282 from chain-E forming three hydrogen bonds and salt bridges indicated them as critical role in H16.001 Fab binding to the capsid. As compared with other reported structures of HPV-Fab immune complex (HPV-V5 and HPV-U4), the full-occupied binding to whole capsid without any steric hindrance between neighboring Fabs is assumed to be its preference for vaccine characterization such as IVRP, in particular has tolerance upon various expression systems being applied in VLPs production.
Discussion
Here we present comparable biochemical and virological studies of a panel of HPV16 nAbs. Based on our study, H16.001 had the strongest neutralizing activity with 0.40 ng/ml IC50 value (Fig. 1). Meanwhile, H16.001 had the highest binding affinity with a 4.22 nM KD value (Table 2 and Supplementary Fig. 1). It seemed that the top neutralizer tended to be the top binder, which was consistent with a previous study where 8H7 had the smallest NC50 (neutralization) and EC50 (affinity) values18. Notably, our previous study showed that the neutralization efficacy of H16.001 was not significantly affected by all the 31 naturally occurring HPV16 variants12, indicating a broader protection of H16.001. These results suggested that H16.001 was an mAb with more potent neutralizing activity, higher affinity, and broader protection and can be tailored for evaluating the integrity and antigenicity of HPV16 vaccines from different expression systems.
Currently, the IVRP assay, a sandwich-type immunoassay, has been developed as an alternative to the mouse potency assay, which was usually used for lot release of the HPV prophylactic vaccines. And the relative antigenicity measured by the IVRP assay is demonstrated to be a good predictor of in vivo potency13. However, the vaccine manufacturers apply different expression systems to produce their vaccine and develop the IVRP assays based on their own mAbs to evaluate the vaccine, which makes the results incommensurable among different manufacturers’ vaccines. So an mAb that can identically recognize various HPV16 L1 VLPs from different expression systems is urgently needed. Fortunately, our study found that the binding response curves of H16.001 to HPV16 L1 VLPs derived from three expression systems were almost identical, which indicated that H16.001 could identically bind the various HPV16 L1 VLPs (Fig. 2). Structural study showed that five H16.001 Fabs bound to one pentamer with low steric hindrance for neighboring H16.001 Fabs (Fig. 5c), which may partially explain why H16.001 can show same sensitivities to various HPV16 L1 VLPs.
H16.001 and H16.8A9 recognize more immunodominant epitopes than others in HPV vaccine recipients. The PBNA was used to determine the functional activity of total neutralizing antibodies to inhibit a PsV infection, while epitope-specific c/ELISA assay was used to define the neutralizing antibody titers directed to a certain epitope on the viral capsid. It is reasonable to consider that more immunodominant epitope resulted in better correlation between the two methodologies. Our epitope-specific c/ELISA assay showed that antibodies which recognized the same epitope as H16.001 or H16.8A9 correlated better with total neutralizing antibodies in vaccinated guinea pig sera than others (Fig. 3). Meanwhile, 12 and 50 vaccinated human sera were strongly inhibited by mAbs H16.001 and H16.8A9, respectively (Fig. 4). These results suggested that H16.001 or H16.8A9 represented the majority of neutralizing antibodies in HPV16 vaccine recipients, indicating that two antigenic sites were immunodominant in the humoral response to HPV16 L1 VLPs vaccination: one recognized by the H16.001, and the other by H16.8A9.
To date, there are several reported cryo-EM structures of immune complex containing HPV PsV, wherein the highest resolution was only at 4.7 Å and could not afford a confident model building at near atomic level16,19,20,21. We circumvented the resolution bottleneck over 4 Å resolution for HPV capsid immune complex through new emerging subparticle approach. The subparticle extraction from the large H16-001 immune complex with a diameter of ~700 Å allowed local reconstruction and accurate determination of defocus value and eventually improved the resolution from 4.41 Å to 3.43 Å. The approach will determine more structures of immune complexes in expedient and high-resolution manner and definitely accelerate the anti-virus drug and vaccine design. Furthermore, the H16-001 cryo-EM structure revealed a distinctive antibody-binding pattern as compared to mAbs H16.V5 and H16.U4. H16.V5 Fab preferentially binds with the hexavalent capsomer, and H16.U4 Fab preferentially binds with the pentavalent capsomer16, whereas H16.001 Fab can engage both hexavalent and pentavalent capsomeres of HPV capsid (Figs. 5 and 6a, d). Such unique binding mode might be associated with its high potent neutralizing activity over other mAbs. Interaction analysis revealed that H16.001 Fab-binding footprint is mainly located at DE, FG, and HI loops of HPV16 L1, overlapping with those epitopes of mAbs H16.V5, 1A, 14J, and 263A216,21. Intriguingly, these five mAbs recognize three common sites, at S282, N285, and T358, which might be associated with the cellular attachment and entry during HPV16 infection.
Taken together, we identified a rabbit monoclonal antibody H16.001 recognizing a potent neutralizing, high affinity, immunodominant, and broad protection epitope. Importantly, the unique binding modality of this antibody to the HPV16 capsid protein may contribute to be used for in vitro assays to evaluate all HPV vaccine from different expression systems. Further studies need to be conducted to explore the use of H16.001 in practical quality assessment work for HPV vaccines, such as developing a common potency assay.
Methods
Plasmids, mAbs, bulk HPV16 L1 VLP products, HPV16 pentamers, and post-vaccinated human serum samples
The HPV16 packaging plasmid (p16LLw) containing the codon-optimized L1 gene (HPV16 isolate 114K) and L2 gene (isolate 114B) was kindly provided by Dr. John T. Schiller (National Cancer Institute, Bethesda, MD)4,22.
As shown in Table 1, 13 neutralizing mAbs against HPV16 were kindly provided by 5 vaccine manufacturers, with H16.5C10 by Ruike Biotechnology (Jiangsu, China); H16.V5 and horseradish peroxidase (HRP)-H16.V5 by Bowei Biotechnology (Shanghai, China); H16.4G12, H16.6C7, H16.26H7, H16.25F12, H16.25H8, H16.22G5, H16.17B10, H16.20G3, and HRP-H16.4G12 by Beijing Health Guard Biotechnology (Beijing, China); H16.8A9, H16.1D12, and HRP-H16.8A9 by Innovax Biotech Co., Ltd. (Xiamen, China); and H16.001 and HRP-H16.001 by Sino Biological Inc. (Beijing, China).
As shown in Table 3, five kinds of bulk HPV16 L1 VLP products were also kindly provided by the aforementioned vaccine manufacturers, with Y1.16 from Ruike Biotechnology, Y2.16 from Bowei Biotechnology, E1.16 from Beijing Health Guard Biotechnology, E2.16 from Innovax Biotech Co., Ltd., and I1.16 from Sino Biological Inc. Notably, Y1.16 and Y2.16 were produced in yeast expression system, E1.16 and E2.16 from E. coli expression system, and I1.16 from insect cell expression system.
HPV16 pentamers were kindly provided by Innovax Biotech Co., Ltd. Fifty post-vaccinated human serum samples were kindly provided by the vaccine manufacturer Shanghai Zerun Biotechnology (Shanghai, China), which were collected from a phase I clinical trial of a bivalent HPV16/18 vaccine, produced in Pichia Pastoris (ClinicalTrials.gov ID:2011L01085). Vaccines were administered according to their recommended three-dose vaccination schedules (months 0, 2, and 6). Serum samples were collected 7 months after the first immunization.
Fab preparation
To obtain H16.001 Fab, the mAb H16.001 was digested in the mixture solution containing 1‰ (w/w) papain, 20mM L-Cysteine, and 50 mM ethylenediamine tetraacetic acid at 37 °C. Following 12-h incubation, 30 mM iodoacetamide was added to stop the digestion. The resulting Fab fragment was then purified by diethylethanolamine column (TOSOH, Japan) chromatography.
Generation and purification of HPV16 PsVs
293FT cells were grown in complete Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum, 2% HEPES, 1% nonessential amino acids, and 1% penicillin/streptomycin. The reporter plasmid pcDNA3.1-EGFP was constructed by inserting an EGFP reporter gene into the pcDNA3.1 vector (Invitrogen, Carlsbad, CA). HPV16 L1L2 PsVs were generated and titrated as described previously4,23,24. The PsV titers were defined as 50% tissue culture infective dose, calculated with the Reed–Muench method25. For structure determination, PsV particles were purified by sucrose density gradient and subsequent CsCl density gradient ultracentrifugation26.
Sera from immunized guinea pigs with HPV vaccines
The animal study was approved by the Institutional Animal Care and Use Committee of National Institutes for Food and Drug Control. All animals were housed in accordance with relevant guidelines. Five samples of HPV16 VLPs, which were kindly provided by Ruike Biotechnology, Bowei Biotechnology, Health Guard Biotechnology, Innovax Biotech Co., Ltd., and Sino Biological Inc., were mixed with aluminum hydroxide adjuvant, respectively. Ten micrograms of the aforementioned adjuvant-adsorbed HPV16 VLPs were delivered intramuscularly for each guinea pig (200–220 g, female) and three guinea pigs were immunized with each of the vaccines. At the same time, the aluminum hydroxide adjuvant was inoculated into three guinea pigs intramuscularly as a control. Immunization was repeated twice at 3-week intervals, and sera were obtained 3 weeks after the second immunization. All serum samples were inactivated before the first use.
PsV-based neutralization assay
Neutralization properties of the mAbs or sera were assessed as described23,24,27,28. Briefly, the PsVs were mixed for 1 h at 4 °C with medium, serial dilutions of mAbs or sera. 293FT cells were plated 4–6 h before the mixtures were added with 1.5 × 104cells/100 μL/well. The aforementioned combinations were transferred into cell culture plates preseeded with 293FT cells and incubated for 68–72 h. After incubation, the numbers of fluorospots were counted with an ImmunoSpot reader. The mAbs and serum neutralization titers were defined as the IC50 and ID50, respectively, and calculated with the Reed–Muench method25.
SPR binding assay
Affinity measurement was conducted at 25 °C on a BiacoreT200 instrument (GE Healthcare, Uppsala, Sweden). mAb (25 μg/mL) was directly immobilized to the surface of a CM5 sensor chip via covalent amine coupling for 60 s at a flow rate of 5 μL/min. HPV16 pentamer was initially diluted as 20 nM and then serial diluted to five concentrations (10, 5, 2.5, 1.25, and 0.625 nM) in 250 μL by using HBS-EP+ buffer, pH7.4 (GE Healthcare, Uppsala, Sweden). Then HPV16 pentamer analyte with each concentration was loaded onto sensor chip to carry out an association step with mAb ligand for 120 s for determining their relative on-rates; this was followed by a dissociation step for 600 s for determining their relative off-rates. Ten millimoles of glycine-HCl (pH 1.5) was used as regeneration buffer at a flow rate of 10 μL/min for 30 s. The SPR curves were recorded and processed with the subtraction of buffer reference and then evaluated using the BIA evaluation software.
IVRP test
Bulk HPV16 L1 VLP product concentrations were analyzed at the Innovax Biotech Co., Ltd. (Xiamen, China) using a TCA-Lowry assay. And protein concentration was determined relative to a bovine serum albumin (BSA) standard curve.
The IVRP assay is a sandwich immunoassay that uses the same mAb for capture and detection. The assay was performed by first coating Immulon96-well microplate (Thermo Scientific) with the mAb diluted in phosphate-buffered saline (PBS) at a concentration of 2 μg/mL. These antibody-coated plates were referred to as assay plates. The bulk HPV16 L1 VLP products were prepared in 0.5% BSA-PBST at 7 concentrations of 8.2, 24.7, 74.1, 222, 667, 2000, and 6000 ng/mL, and 100 μL of each dilution was added to the assay plate. The plate was allowed to incubate for 1 h at 37 °C and was washed before 100 μL of the diluted HRP-mAb at a concentration of 2 μg/mL was added. All loaded plates were subsequently incubated at 37 °C for 1 h, washed three times with PBST, and developed with TMB. The reaction was stopped with 2 N H2SO4 and the optical density (OD) of each well of the plate was determined on a microplate reader at 450/620 nm.
The immunized guinea pig sera tested by competition ELISA using different mAbs
Immulon96-well microplates (Thermo Scientific) were coated with 100 μL of Y1.16 (2 μg/mL) and incubated overnight at 4 °C. Plates were incubated with 100 μL of guinea pig sera for 1 h at 37 °C. Guinea pig sera were initially diluted at 1:100 and serially diluted twofold in 0.5% BSA-PBST across the plate in duplicates. After incubation, plates were washed with PBST three times and incubated with specific HRP-mAb. After 1-h incubation at 37 °C, TMB and 2 N H2SO4 were added as aforementioned above, and the OD was read at 450/620 nm. BID50 blocking activity is defined above in the “Results” section.
The immunized human sera tested by blocking ELISA using different mAbs
The proportion of vaccinated human sera directed against different HPV16 mAbs was determined by a blocking ELISA, in which Y1.16 was incubated with saturating levels of mAbs prior to the addition of vaccinated sera, and residual binding of human immunoglobulin G (IgG) was detected using HRP-anti-human IgG. Saturating mAbs (100 μL, 5 μg/mL in 0.5% BSA-PBST) were added to VLP-coated plates for 1 h at 37 °C, followed by the addition of 100 μL of vaccinated serum diluted 1:6000 in BSA-PBST for a further 1 h. Dilutions of vaccinated sera were chosen in order to give an ELISA result of approximately 1.0 OD units. Plates were then washed with PBST, and residual binding of human IgG was detected using HRP-conjugated sheep anti-human IgG. To quantitate the residual binding of human IgG, any difference in the level of human IgG binding between the control (0.5% BSA-PBST) and specific mAbs is therefore a measure of the proportion of total HPV16-specific IgG directed against epitopes, which are altered or blocked by binding of the specific mAb. Notably, 30% was used as a cutoff, and sera reducing the photometer extinction ≥30% were defined as positive.
Cryo-EM sample preparation and data acquisition
HPV-001 immune complex was prepared by the mixture of HPV16 PsV with Fab H16.001 with a molar ratio of 540 Fabs to every HPV16 viral particle incubating at 37 °C for 30 min. Aliquots (3 μL) of the immune-complex sample were deposited onto the glow-discharged holey carbon Quantifoil Cu grids (R2/2, 200 mesh, Quantifoil Micro Tools). The grids were blotted for 6 s by filter paper at 100% humidity and 4 °C using Vitrobot Mark IV and then plunged-frozen into liquid ethane. Prepared grids were examined using 300 kV Tecnai F30 microscope (Thermo Scientific) equipped with a Falcon II direct electron detector. The cryo-EM data were collected at a nominal magnification of ×93,000 in 17-frame movie mode, yielding a pixel size of 1.128 Å. The total exposure dose is 30 e− Å−2 and exposure time is 1 s.
Cryo-EM data processing
Beam-induced motion correction was executed by MotionCor229. Each micrograph contrast transfer function was estimated using Gctf30. A total of 12,034 particles (768-by-768 pixels) were semi-automatic boxed-out from 880 micrographs using e2boxer.py31 and subjected to reference-free 2D classification using Relion 3.032. Nine thousand one hundred and sixty-two particles were selected for initial model generation and auto-refined to be 4.41 Å resolution according to the “gold” standard FSC = 0.143 criterion imposing icosahedral I2 symmetry using cisTEM14. For further improving the resolution and conquering the sample heterogeneity and defocus gradient, we applied the localized reconstruction method17 to extract the subparticles for refinement and reconstruction. Briefly, based on the orientation and center parameters of each particle image, which was calculated from the icosahedral symmetry refinement, we extracted the subparticles in a box size of 400 × 400 pixels centered at 290 Å from the center of the particle. This resulted in a total of 274,860 subparticles with their defocus values re-calculated from their locations in each viral particle. Such subparticles were then imported into cisTEM for manual local refinement imposing C2 symmetry, which were determined to be 3.43 Å resolution with FSC cutoff of 0.143. Local resolution variations were estimated using ResMap33.
Model building and refinement
The initial model of H16.001 Fab was generated from homology modeling using the Accelrys Discovery Studio software34. Then we generated the initial template model (enclosing an icosahedral asymmetric unit) by docking the homology Fab models and HPV16 capsid structure (PDB code: 5KEP) into the density map using Chimera35. The template was further refined and rebuilt iteratively against the cryo-EM map using phenix. real_space_refine36 and Coot37. The refinement statistics of the final model is provided in Supplementary Table 3, as validated by MolProbity38. The footprint of Fab H16.001 was generated onto the 2D projection of the stereographic sphere using RIVEM39. Fab–capsid interaction analysis and buried surface area calculation were performed using the PISA server (www.ebi.ac.uk/pdbe/pisa). Structural figures were illustrated by PyMOL40 and Chimera35.
Statistical analyses
Correlation between the inhibition ratio by mAbs and neutralization titer of the vaccinated guinea pig sera was determined using the calculated Pearson coefficient of correlation. All analyses were performed using the GraphPad Prism 7 software.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request. The cryo-EM density map and the corresponding atomic model have been deposited in the Electron Microscopy Data Bank (EMDB, http://www.ebi.ac.uk/pdbe/emdb/) and Protein Bank (PDB, https://www.ebi.ac.uk/pdbe/), respectively. The accession codes are: HPV:001 (EMDB: EMD-30415) and HPV:001 twofold sub-particle reconstruction (EMDB: EMDB-30414, PDB: 7CN2).
References
Bray, F. et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 68, 394–424 (2018).
Li, N., Franceschi, S., Howell-Jones, R., Snijders, P. J. & Clifford, G. M. Human papillomavirus type distribution in 30,848 invasive cervical cancers worldwide: Variation by geographical region, histological type and year of publication. Int. J. Cancer 128, 927–935 (2011).
Kirnbauer, R., Booy, F., Cheng, N., Lowy, D. R. & Schiller, J. T. Papillomavirus L1 major capsid protein self-assembles into virus-like particles that are highly immunogenic. Proc. Natl Acad. Sci. USA 89, 12180–12184 (1992).
Kirnbauer, R. et al. Efficient self-assembly of human papillomavirus type 16 L1 and L1-L2 into virus-like particles. J. Virol. 67, 6929–6936 (1993).
Schiller, J. T., Castellsague, X. & Garland, S. M. A review of clinical trials of human papillomavirus prophylactic vaccines. Vaccine 30(Suppl 5), F123–F138 (2012).
Zhang, Z., Zhang, J., Xia, N. & Zhao, Q. Expanded strain coverage for a highly successful public health tool: prophylactic 9-valent human papillomavirus vaccine. Hum. Vaccines Immunother. 13, 2280–2291 (2017).
Qiao, Y. L. et al. Efficacy, safety, and immunogenicity of an Escherichia coli-produced bivalent human papillomavirus vaccine: an interim analysis of a randomized clinical trial. J. Natl Cancer Inst. https://doi.org/10.1093/jnci/djz074 (2019).
Einstein, M. H. et al. Comparative immunogenicity and safety of human papillomavirus (HPV)-16/18 vaccine and HPV-6/11/16/18 vaccine: follow-up from months 12-24 in a Phase III randomized study of healthy women aged 18-45 years. Hum. Vaccines 7, 1343–1358 (2011).
Draper, E. et al. A randomized, observer-blinded immunogenicity trial of Cervarix((R)) and Gardasil((R)) human papillomavirus vaccines in 12-15 year old girls. PLoS ONE 8, e61825 (2013).
Wang, Z., Christensen, N., Schiller, J. T. & Dillner, J. A monoclonal antibody against intact human papillomavirus type 16 capsids blocks the serological reactivity of most human sera. J. Gen. Virol. 78(Pt 9), 2209–2215 (1997).
Lee, H. et al. A cryo-electron microscopy study identifies the complete H16.V5 epitope and reveals global conformational changes initiated by binding of the neutralizing antibody fragment. J. Virol. 89, 1428–1438 (2015).
Ning, T. et al. Naturally occurring single amino acid substitution in the L1 major capsid protein of human papillomavirus type 16: alteration of susceptibility to antibody-mediated neutralization. J. Infect. Dis. 216, 867–876 (2017).
Shank-Retzlaff, M. et al. Correlation between mouse potency and in vitro relative potency for human papillomavirus type 16 virus-like particles and Gardasil vaccine samples. Hum. Vaccines 1, 191–197 (2005).
Grant, T., Rohou, A. & Grigorieff, N. cisTEM, user-friendly software for single-particle image processing. Elife 7, e35383 (2018).
Scheres, S. H. & Chen, S. Prevention of overfitting in cryo-EM structure determination. Nat. Methods 9, 853–854 (2012).
Guan, J. et al. High-resolution structure analysis of antibody V5 and U4 conformational epitopes on human papillomavirus 16. Viruses 9, 374 (2017).
Ilca, S. L. et al. Localized reconstruction of subunits from electron cryomicroscopy images of macromolecular complexes. Nat. Commun. 6, 8843 (2015).
Zhang, C. et al. Epitope clustering analysis for vaccine-induced human antibodies in relationship to a panel of murine monoclonal antibodies against HPV16 viral capsid. Vaccine 36, 6761–6771 (2018).
Liu, X. et al. Neutralization sites of human papillomavirus-6 relate to virus attachment and entry phase in viral infection. Emerg. Microbes Infect. 8, 1721–1733 (2019).
Li, Z. et al. Crystal structures of two immune complexes identify determinants for viral infectivity and type-specific neutralization of human papillomavirus. MBio 8, e00787–00717 (2017).
Guan, J. et al. Structural comparison of four different antibodies interacting with human papillomavirus 16 and mechanisms of neutralization. Virology 483, 253–263 (2015).
Leder, C., Kleinschmidt, J. A., Wiethe, C. & Muller, M. Enhancement of capsid gene expression: preparing the human papillomavirus type 16 major structural gene L1 for DNA vaccination purposes. J. Virol. 75, 9201–9209 (2001).
Wu, X. L., Zhang, C. T., Zhu, X. K. & Wang, Y. C. Detection of HPV types and neutralizing antibodies in women with genital warts in Tianjin City. China Virol. Sin. 25, 8–17 (2010).
Nie, J., Liu, Y., Huang, W. & Wang, Y. Development of a triple-color pseudovirion-based assay to detect neutralizing antibodies against human papillomavirus. Viruses 8, 107 (2016).
Matumoto, M. A note on some points of calculation method of LD50 by Reed and Muench. Jpn. J. Exp. Med. 20, 175–179 (1949).
Wang, D. et al. [Preparation and cryo-EM structure determination of human papillomavirus 16 pseudovirion derived from suspension-adapted HEK293 cells]. Bing Du Xue Bao = Chin. J. Virol. 32, 551–559 (2016).
Nie, J., Huang, W., Wu, X. & Wang, Y. Optimization and validation of a high throughput method for detecting neutralizing antibodies against human papillomavirus (HPV) based on pseudovirons. J. Med. Virol. 86, 1542–1555 (2014).
Pastrana, D. V. et al. Reactivity of human sera in a sensitive, high-throughput pseudovirus-based papillomavirus neutralization assay for HPV16 and HPV18. Virology 321, 205–216 (2004).
Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. Nat. Methods 14, 331 (2017).
Zhang, K. Gctf: real-time CTF determination and correction. J. Struct. Biol. 193, 1–12 (2016).
Ludtke, S. J. in Methods in Enzymology, Vol. 579 (ed. Crowther, R. A.) Ch. 7 (Elsevier, 2016).
Zivanov, J. et al. New tools for automated high-resolution cryo-EM structure determination in RELION-3. Elife 7, e42166 (2018).
Kucukelbir, A., Sigworth, F. J. & Tagare, H. D. Quantifying the local resolution of cryo-EM density maps. Nat. Methods 11, 63 (2014).
Biovia, D. S. Discovery studio modeling environment. https://www.3ds.com/products-services/biovia/products/molecular-modeling-simulation/biovia-discovery-studio/. (Release, 2017).
Pettersen, E. F. et al. UCSF Chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612 (2004).
Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. Sect. D Biol. Crystallogr. 66, 213–221 (2010).
Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. Sect. D Biol. Crystallogr. 60, 2126–2132 (2004).
Chen, V. B. et al. MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr. Sect. D Biol. Crystallogr. 66, 12–21 (2010).
Xiao, C. & Rossmann, M. G. Interpretation of electron density with stereographic roadmap projections. J. Struct. Biol. 158, 182–187 (2007).
Schrodinger LLC. The PyMOL molecular graphics system, Version 1.0. https://pymol.org/2/ (2010).
Acknowledgements
We thank Dr. John T. Schiller for providing the plasmids encoding the structural genes (L1 and L2), Ruike Biotechnology for providing bulk HPV16 L1 VLP products Y1.16 and mAb H16.5C10, Bowei Biotechnology for access to their bulk HPV16 L1 VLP products Y2.16, mAbs (H16.V5 and HRP-HPV16.V5), Beijing Health Guard Biotechnology for providing their bulk HPV16 L1 VLP products E1.16 and mAbs (H16.4G12, H16.6C7, H16.26H7, H16.25F12, H16.25H8, H16.22G5, H16.17B10, H16.20G3, and HRP-H16.4G12), Innovax Biotech Co., Ltd. for their bulk HPV16 L1 VLP products E2.16 and mAbs (H16.1D12, H16.8A9, and HRP-H16.8A9), Sino Biological Inc. for providing bulk HPV16 L1 VLP products I1.16 and mAbs (H16.001 and HRP-H16.001), and Zerun Biotechnology for access to their 50 post-vaccinated human serum samples. This work was supported by the National Science and Technology Major Projects of Drug Discovery [grant number 2018ZX09101–001].
Author information
Authors and Affiliations
Contributions
Y.W. and S.L. designed, planned the work, and revised the manuscript. W.H., M.H., and T.N. performed the experiment, analyzed the data, and drafted the manuscript. J.N., F.Z., and R.Z. partially performed the affinity experiments. J.N. and Y.Q. partially performed the ELISA experiments. Q.Z. and Y.G. partially performed the cryo-EM structure analysis.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Huang, W., He, M., Ning, T. et al. Structural characterization of a neutralizing mAb H16.001, a potent candidate for a common potency assay for various HPV16 VLPs. npj Vaccines 5, 89 (2020). https://doi.org/10.1038/s41541-020-00236-w
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41541-020-00236-w
- Springer Nature Limited