A Visual Discrimination of Existing States of Virus Capsid Protein by a Giant Molybdate Cluster

doi:10.21203/rs.3.rs-963410/v1

Download PDF

Research Article

A Visual Discrimination of Existing States of Virus Capsid Protein by a Giant Molybdate Cluster

https://doi.org/10.21203/rs.3.rs-963410/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 22 Feb, 2022

Read the published version in Nanomaterials →

Version 1

posted

You are reading this latest preprint version

This study reports a unique phenomenon, the opposite color response of a giant polyoxometalate, (NH₄)₄₂[Mo₁₃₂O₃₇₂(CHCOO)₃₀] (H₂O)₇₂ ([Mo₁₃₂]), to the existing states of human papillomavirus (HPV) major capsid protein, L1-pentamer (L1-p) and virus-like particles (VLPs). The color responses originate from the different assembly forms between [Mo₁₃₂] and the capsid protein, which were inspected and separated by using CsCl gradient centrifugation, and validated in detail by sodium dodecyl sulfate polyacrylamide gel-electrophoresis (SDS-PAGE), dynamic light scattering (DLS) and transmission electron microscopy (TEM) images. The intrinsic mechanism difference between them were investigated in depth by using the XPS-based semi-quantitative analysis and the well designed peptides, which reveal the key points of L1 determining the charge-transfer ratio between Mo(V) to Mo(VI), and consequently the levels of [Mo₁₃₂] hypochromicity in different assemblies. Such unique phenomenon is significant as it supplies a colorimetry approach to directly distinguish the existing states of HPV capsid protein and would be significant in the quality assay of HPV vaccine and existing states of other virus in future.

General Biochemistry

colorimetric discrimination

giant polyoxometalate

HPV capsid protein

hypochromicity of molybdate clusters

quality assay of vaccine

opposite color response

Human papillomavirus (HPVs) are one of the commonly transmitted infections that occur in human bodies,^1,2 while some of their types lead to serious diseases including series verrucas and cancers.³ To prevent the occurrence of the infections effectively, targeted vaccines against the virus-induced diseases are developed recently and a lot of efforts have been made to build a security shielding system.⁴ Virus-like particles (VLPs), formed from the self-assembly of pentamer subunits comprising of major capsid protein L1 without the participation of DNA, have been demonstrated to be an important resource of vaccines because of the similar surface structure and antigenic epitopes to those of real virus. As a kind of prophylactic vaccine, VLPs were proved to be effective for the protection of several subtypes of HPV from infections.^5,6 On one hand, these assembled particulate structures are also resilient to most environmental stresses and provide a strong possibility to elicit potent immune responses efficiently.⁷ On the other hand, during the development of prophylactic vaccines against high risk types of HPVs, VLPs are also adopted directly in mimicking more likely to the neutralizing epitopes, morphology, and keeping the size of native ones.⁸ However, the production and usage of spherical capsids as vaccines deal greatly with the completeness of VLPs, and any disassembly of it may result in losing effectiveness in producing antibodies. Therefore, the quick and convenient detection for the existing state of VLPs and/or its disassembly becomes highly desirable. Since the normal subunits of VLPs are the pentamer of capsid protein L1(L1-p), the sensitive determination of the species can clearly demonstrate the incomplete state of VLPs.⁹ Recently, spectroscopic, dynamic light scattering, and electron microscopic measurements are commonly used physical techniques,^10,11 but they are not convenient enough for in situ quality assay during the preparation of vaccines. Therefore, it is highly important to develop a simple and rapid method for differentiating complete capsids and dissembled pentamers via a quick and direct procedure.

Polyoxometalates (POMs) are a kind of negatively charged inorganic clusters comprising of early transition metal oxides.¹² The uniform shape and nanoscale size of POMs produce important functional properties applicable in the fields of chemistry, materials sciences, biology and medicine.^13,14 Typically, the reduced coordination metal ions allow the clusters with deep color to have extended absorptions in visible region, even much longer wavelengths based on their structure. In the presence of oxidants, the reduced clusters exhibit reversible oxidation property, leading to the shifted absorption at another state.¹⁵ Moreover, among the POM family, the giant molybdate clusters built by Müller’s group firstly possess more beautiful topologic architecture and morphology within the larger sizes over 2.5nm.¹⁶ In contrast to those giant clusters which have been well studied with structural properties, such as self-assemblies into nanoscale capsules and forming co-assemblies together with cationic organic components via electrostatic interactions,¹⁷ their functionalization was less focused on biomaterials.¹⁸ We have reported the electrostatic interaction of the giant clusters with cationic amphiphile for synergistic self-assembly, the behaviors with biomolecules in solutions in adhesion, anti-bacteria, as well as bio-imaging.^19,20 But up to date, there are no reports concerning the utilities of these POMs and other inorganic clusters on the detection of existing state of virus capsids, including the assembly and disassembly, as well as the ionic binding position with capsid protein.

Several Mo(V) ions have been proved to be essential components in giant molybdate POMs, which make them show colors in solid and solution due to their absorptions corresponding to the intervalence charge transfer (IVCT).^21,22 Thus, the valence state change of giant clusters always leads to a color change and as a result, Mo(V) ions can be oxidized into Mo(VI), accompanied by a color fading to brick-red in the presence of oxidants. The color changes are revisable because the Mo(VI) ions at the highest oxidized state are reduced while the cluster structure maintains, just like the oxidation to them at reduced state.^23,24 These properties of POMs together with their charged surface, provide an excellent opportunity in the detection of the existing state of the proteins comprising capsids through the electrostatic interaction with free N-terminals bearing redox features. Thus, the sates of some proteins that can be oxidized and reduced become possible to be checked out through the color change of giant POMs after encountering redox process. In this context, we herein selected a giant brown Keplerate cluster, (NH₄)₄₂[Mo₁₃₂O₃₇₂(CHCOO)₃₀](H₂O)₇₂, abbreviated as [Mo₁₃₂], which contains 72 Mo(VI) centers at full oxidation state and 60 Mo(V) centers at reduced state, to sensitize the status of L1 proteins and the stability of the capsid. Because L1 proteins at the states of assembly and disassembly show different binding modes with [Mo₁₃₂] cluster by exposing and shielding amino residues, we successfully realized the vital identification for the capsid proteins through simple color changes of the solutions (Scheme 1). The present method demonstrated an effective approach to examine the quality of the prophylactic HPV vaccines. Of course, the obtained results are also significant in widening bio-applications, such as qualitative screening of HPV infections and tracking the differentiation process of HPVs in human body by detecting L1 protein, since both the repeating of virus DNA and proliferating involve the production of capsid L1 that can be sensed by the POMs used.

Hypochromicity effect of [Mo ₁₃₂ ] in the detection of L1-p.

The [Mo₁₃₂] (2.5µM) in buffer A shows typical absorption band at 450 nm like the cluster in neutral aqueous solution (Figure 1A), which can be clearly assigned to the intervalence charge transfer (IVCT) between Mo(V) and Mo(VI) centers bridging by O atoms.^21,22 Since the absorption bands of the L1 protein at both states of pentamer and VLP appear at a wavelength less than 300 nm, the changes of absorption spectra of [Mo₁₃₂] mixture with L1-p at visible region is definitely ascribed to source from the inorganic cluster. A gradual hypochromicity of [Mo₁₃₂] in buffer A solution with time in air is observed and nearly 20% absorbance is lost after 4h, when a turning point emerged. After that, no more change occurs for the absorption with time in one day. The reason for the hypochromicity of [Mo₁₃₂] can be attributed to the oxidation of partial Mo(V) fragments in the giant cluster by oxygen in air, according to the opposite observation during its formation.²⁴ As further confirmed in characterization, this color change in aerobic environment does not affect the structure completeness and the detection for proteins because of the large number of reduced Mo(V) fragments in one cluster.

In comparison to [Mo₁₃₂] alone in buffer A, a much stronger hypochromic effect of the cluster occurs in the same solution with the addition of L1-p. A linear decrease of absorbance is observed and more than 80% of the original absorbance disappears until culturing to a turning point at 12h. Further color fading proceeds continuously and more than 90% of absorbance vanishes after 24h (Figure 1B). Interestingly, when [Mo₁₃₂] was added into the solution containing VLP, only a little color degradation occurs with time (Figure 1C) and over 95% absorbance of the inorganic cluster is maintained at the culture time scale. The job plots of [Mo₁₃₂] cluster at the states in the presence of protein L1-p or VLPs show the clear changes of the absorbance at visible region versus the time (Figure 1D).

[Mo₁₃₂] has been demonstrated to form blackberry-like hollow spherical self-assembly in several tens of nanometer in aqueous solution,²⁶ while maintaining a mono-dispersed state for a couple of days. As shown in Figure 2A, the DLS curve of the cluster alone shows a hydrodynamic diameter of 2.7 nm, just like the similar size (2.9 nm) to that calculated from crystal structure, illustrating the mono-dispersion and structural completeness.^27,28 The L1-p in buffer A exists in a size of about 12.0 nm (Figure 2B), being very close to the well dispersed state of pentamer subunit published before.^29,30 When the inorganic cluster was added into a solution containing L1-p, larger aggregates in a diameter of about 73 nm (Figure 2C) formed immediately. Such a size differs from each individual component or typical VLPs, implying the quick interaction between [Mo₁₃₂] and the protein. The incubation for 24 h or longer time does not lead to any precipitation but maintains at a stabilized aggregate with a slight increase in size to 82 nm (Figure 2D). As have been demonstrated the negatively charged POMs bind with some peptides containing basic residues through electrostatic interaction and hydrogen bonding,^31,32 the non-specific interaction can also induce the combination of [Mo₁₃₂] cluster bearing 42 negative charges with L1-p, causing the formation of larger assemblies or aggregates.

Visual detection for isolated L1-p and VLPs.

The absorption spectral changes of [Mo₁₃₂] in L1-p and its VLP assembly (Figure S1) can also be examined in parallel by a simple visual discrimination through a direct observation of color change. As shown in Figure 3, [Mo₁₃₂] is sensitive in detecting the existing states of protein L1. In a regular process, L1 monomer is very unstable and normally exists in a pentamer state, which further self-assembles into an empty capsid, VLP, at the condition of high ionic concentration and low pH, spontaneously.^27,28 To identify the possible influence of [Mo₁₃₂] cluster on the assembly process of L1 pentamer, the L1-p in a large amount of assembly buffer is cultured to allow an assembly of VLP in the presence of [Mo₁₃₂], following a published standard process. After encountering a dialysis procedure, the DLS assay reveals the formation of characteristic assemblies accompanying by a size change from 12 nm at the beginning to an average of 51.8 nm after 24 h incubation (Figure 2E), in perfect agreement with the full-sized scale (50−55 nm) of VLPs comprising of HPV16 L1 protein.^29,30 This result indicates that the addition of [Mo₁₃₂] does not affect the self-assembly of L1-p into VLP. However, after this assembly process, the dynamic size attributing to [Mo₁₃₂] is no longer observed apparently, even after 24h incubation (Figure 2F), implying that most of the clusters are either trapped into inside or onto the outside surfaces of the formed VLPs, or just run away through dialysis. Apparently, the existence of [Mo₁₃₂] does not affect the further assembly of the L1-p as subunit. Importantly, accompanying the self-assembly of L1-p, the faded color of [Mo₁₃₂] recovered because the solution changes from colorless back to brown, the original color of the cluster in solution (Figure 3).

Prevention of [Mo ₁₃₂] hypochromicity by VLPs.

More interestingly, the addition of [Mo₁₃₂] into an assembly buffer containing VLPs does not result in an obvious hypochromicity after 24h incubation in air; even in contrast to L1-p, it shows a much less color fading than that of cluster alone in buffer A encountering the same incubation time in air, indicating the strong inhibition of VLP on color bleaching of [Mo₁₃₂]. Meanwhile, the DLS histogram (Figure 2F) confirms a well-kept full-size of VLPs in 24 h though a slight increase of diameter to ~55.3 nm is observed. To identify whether [Mo₁₃₂] binds to VLPs or just stay alone in solution, the ultracentrifugation of various components in gradient CsCl solution, bearing in sample tubes, is performed based on the principle of size and density dependence on the position to the rotating center, where components with the larger size and higher density will locate at the position closer to the bottom.

After the CsCl gradient ultracentrifugation the photograph of VLP alone in tube #0 shows a sole blue band, which can be well ascribed to the position of complete VLPs, at zone F2 (Figure 4A) in the middle of the centrifugation tube. In the case of VLP mixing with [Mo₁₃₂] in tube #1, beside a narrower blue belt at the position of F2#1 a wide band emerges at the lower position F3#1, suggesting the formation of aggregations with larger size differing from both VLP and L1-p. Because the observed color of the belt near F3#1 is far from that of VLP at F2#1 while the sole VLP does not show any significant hint at F3#0 position, we suggest that the [Mo₁₃₂] trap VLPs to form larger-sized aggregates. As a result, the inhibition to the hypochromicity effect of [Mo₁₃₂] can be explained to derive from the protection of the VLPs from the external oxidation. On the other hand, the photograph of L1-p mixing with [Mo₁₃₂] in tube #2 presents a weak belt with pale color at the position F2#2 similar to that of sole VLP at F2#0, yet no obvious belt emerging at zone F3#2. The result implies that part of L1-p self-assembled into VLPs automatically in the presence of inorganic cluster but almost no formation of the proposed aggregates, comprising of the formed VLPs and [Mo₁₃₂] as that observation at zone F3#1, due to very little quantity of VLPs. When L1-p is mixed with [Mo₁₃₂] in the assembly buffer, as seen in tube #3, both narrow blue belt at zone F2#3 and wide yellowish-brown belt emerge at the same time at F3#3. The identical phenomenon of F3#3 to F3#1 demonstrates the formation of VLP in the presence of inorganic cluster as well as the induced aggregation. All these results together disclose important facts. That is, (1) [Mo₁₃₂] bearing several tens charges can play a role of increasing high-ionic strength in the assembly buffer to promote the assembly of L1-p into VLPs; (2) the color degradation and the possible interaction from clusters do not affect the assembly of L1-p; and (3) there is indeed a cluster induced aggregation of VLPs occur, which is consistent well with the observation in DLS measurement (Figure 2C, 2D).

The belts at different zones in tubes after CsCl gradient ultracentrifugation were taken out and the corresponding protein is assayed via the technique of SDS-PAGE (Figure 4C). By taking two known proteins with molecular weights of 44 and 66 kDa as marker, all products extracted from F2 zones of tube#1–#3 point to a molecular weight of ~53 kDa, which is in perfect consistence with that L1-monomer. Besides emerging in zone F2 of all tubes, the L1 protein also appears at zone F3 of tubes#1 and #3, supporting the assignment for VLPs aggregation with [Mo₁₃₂] at two zones. Meanwhile, the fact that no smaller proteins are checked out in F1 zones of tube#2 and #3 which figures out that all the L1-p have already self-assembled into VLPs, especially for the tube#2 even without the support of assembly buffer. Coincidently, the DLS histograms of the belts taken from tube#1 show the distributions of particle size in consistence with [Mo₁₃₂] at zone F1#1, and the particle sizes corresponding to that of VLPs at zones F2#1 and F3#1 were ascertained (Figure 4C, upper). As the cluster are proposed to be trapped in VLP aggregates, it is possible to find no isolated [Mo₁₃₂] at zone F3#1 due to the strong interaction between two components. For the belts extracted from tube#2 (Figure 4C, middle), besides checking out the particle that can be ascribed to [Mo₁₃₂] at F1#2, we also observed the particle size attributing to VLP at F2#2. Although small size close to [Mo₁₃₂] appears at F3#2, no larger particles corresponding to the induced VLP are observed, excluding the possibility of it there for [Mo₁₃₂]. Again, the DLS histogram of tube #3 (Figure 4C, bottom) show full spectra of size distributions similar to [Mo₁₃₂] at zone F1#3, VLP at zone F2#3, and the cluster-triggered VLPs aggregation at zone F3#3, which are identical to the case observed in tube#1.

To further verify the morphology and completeness of the formed VLPs taken from F2#1-3, TEM imaging of [Mo₁₃₂] alone in buffer A, the mixture of [Mo₁₃₂] with L1-p before and after assembly monitoring, and the mixture of [Mo₁₃₂] and the as-prepared VLP are carried out firstly. Because of the very small size and mono-dispersion, the inorganic cluster (Figure 5A) can be well discerned from that of L1-p or VLPs in solutions (Figure 5C, 5D). In the mixture of L1-p and [Mo₁₃₂], whether the assembly monitoring is performed or not we always obtained spherical particles (Figure 5B, 5C) which can be well ascribed to the formation of VLPs as the match size with VLP cavity. ^29,30 The amplified images (insert of Figure 5B, C(insert not shown in figure 5C) show small particles that are attributed to the inorganic clusters located inside the inner wall. Therefore, the result in Figure 5B provides an additional evidence for the cluster-triggered assembly of VLPs from L1-p even without assembly monitoring. Partial inorganic clusters are observed being encapsulated inside the VLPs during the triggered co-assembly with L1-p, indicating strong interaction between [Mo₁₃₂] and L1-p. This analysis can be further supported by the observation of VLPs after mixing with [Mo₁₃₂] (Figure 5D), where the inorganic clusters mainly locate at the outside surface of VLPs being pretty different from that of co-assembly of the two components.

To get an essential profile, the TEM images for the samples from the zones extracted from F2#0, F2#1 and F2#3 are then obtained with negative staining (Figure 6). It shows that the topography of VLPs and the particle size become more uniform and precise. Besides clear empty shell (Figure 6A), the well mono-dispersive particles with a statistical diameter of about 55 nm (Figure 6B, C) are in high agreement with the reported full-size of VLPs.^29,30. Taken all together, it is concluded that [Mo₁₃₂] can quickly bind with L1-p to form irregular aggregates, and finally lead to the formation of VLP with [Mo₁₃₂] inside after assembly monitoring. However, once [Mo₁₃₂] is mixed with the as-prepared VLPs, it is also assembled together to form [Mo₁₃₂]@VLP where the cluster particles assembled on the surface of VLPs (Figure 6C). Of note, as the molecular weight of [Mo₁₃₂] is much less than that of VLP, the sedimentation coefficient of [Mo₁₃₂]@VLPs is close to that of VLPs and the encapsulation of a few [Mo₁₃₂] clusters does not change much the surface properties of VLPs. So, it is rational to explain that [Mo₁₃₂]@VLPs displays a close level to VLPs in tubes after the CsCl gradient ultracentrifugation (Figure 4A). In fact, the two adjoining lancet belts at the position of F2#2 suggest a fine difference between VLPs and those encapsulated with [Mo₁₃₂]. Moreover, no matter with stain or not the larger aggregates of VLPs induced by [Mo₁₃₂] were shown clearly both in Figure 5 and 6, which support well the observed proteins in F3#1 and F3#3 (Figure 4B).

Colorimetry response mechanism of Mo ₁₃₂ to L1-p and VLP.

Redox nature of [Mo ₁₃₂ ]. The coordination atoms such as Mo and W of POMs at highest oxidation state are known to be photochemically reduced in the presence of reductant and the reduced POMs show oxidation property, like peroxidase, for some organic and bio-molecules.^33,34 For [Mo₁₃₂], it has 60 of reduced Mo(V) atoms and 72 of oxidized Mo(VI), which allow the cluster to be both reduced and oxidized in suitable conditions. As the inter-valent charge transfer absorption of Mo(V) to Mo(VI) emerges at visible region, the [Mo₁₃₂] is normally in yellowish brown. When the Mo(V) atoms are oxidized, however, the color degeneration will occur. As an example, the incubation of it with a weak reductant L-ascorbic acid (Vc) for 24h does not make the [Mo₁₃₂] cluster change to a deep color by a reduction (Figure S2A), but instead of the phenomenon in Figure 1A, the hypochromicity is achieved in aerobic condition by using Vc as a sacrificer against the oxidation of air. To confirm that the hypochromicity is derived from the oxidation of [Mo₁₃₂] and to accelerate the process going on, an irradiation of 365 nm light is then performed in parallel with and without Vc in buffer A (Figure S2B). After 12h, 20% of the absorbance is diminished for the latter case while almost no change was observed for the former.

Interaction of capsid protein for colorimetry change of [Mo ₁₃₂ ]. The non-covalent interactions of POMs with several types of biomolecules are investigated considerably in the past years.^35–37 Based on the structural features of POMs it is obvious that the negatively charged [Mo₁₃₂] mainly provide electrostatic interactions with a variety of cationic species of protein. However, the observed hypochromicity effect here should be essentially conducted by the external oxidation sourcing from the protein rather than the buffer solution or the aerobic environment completely since the color degradation does not occur in such a short time and entirely. To confirm the speculation, several peptides and proteins with specific surface charges are used to examine the hypochromicity effect of [Mo₁₃₂] (Figure S3). The results show that under a neutral condition the negative peptide of pTau-aac (pI=4.5) and protein of BSA (pI=4.6) do not induce much absorption change of [Mo₁₃₂] while the positive charged ones of dTau30 (pI=10.4) and lysozyme (pI=11.0) drive the large decrease of absorption definitely, indicating the vital role of positive partners to the [Mo₁₃₂] hypochromicity. Apparently, it is the exposed basic residues in L1 accelerate the oxidation of [Mo₁₃₂].

The cryo-electron microscopy and image analysis on capsid protein³⁷ reveal that the L1-p bind together to form VLPs via a long segment of the C-terminal (Figure S4A). Particularly, it reveals the sequence after Asp401 at the C-terminal extends from one pentamer to the adjacent one to strengthen the VLPs structure. This segment is dominated by the cationic residues such as arginine and/or lysine (Figure S4B), which are strongly prone to bind with the negatively charged [Mo₁₃₂]. After L1-p assembles into VLP, however, this segment will be embedded in the wall of VLP sphere and the charged environment varies largely (Figure S4C). To further confirm the binding site of [Mo₁₃₂] with capsid protein, the sequence of two peptides, pep1-401 and pep401-495 are constructed and expressed separately, through the identical approaches as the full-length L1.

After DNA sequence assay (Figure S5) and protein purification, each peptide is mixed with [Mo₁₃₂] in buffer A. Different phenomena are achieved for them. Quick decrease of [Mo₁₃₂] absorption at 450 nm is observed in mixing with pep401-495 (Figure 7A). The band disappears almost completely and the solution becomes pale within 90 min. The time-dependent plot of absorption (Figure 7C) reveals that the peptide drives the exhaustive hypochromicity of the cluster much fast than that of L1-p (Figure 1A), which confirms a decisive role of pep401-495 to the hypochromicity. In contrast, mixing pep1-401 with [Mo₁₃₂] leads only to a very slight decrease of absorption at 450 nm (Figure 7B), indicating the feeble contribution of it in L1 to [Mo₁₃₂] hypochromicity. The plots of absorption intensity versus time (Figure 7C) illustrate clearly the responsive differences among pep401-495, pep1-401 and L1-p to [Mo₁₃₂], further demonstrating that it is the binding with basic residues of pep401-495 causing the enhanced hypochromicity by L1. Thus, it can be speculated that during the assembly of L1-p to VLP, the stronger binding affinity between L1-p subunits could force [Mo₁₃₂] to be released from the basic sites of L1-p and attach to other position of VLPs electrostatically. Considering the positive areas at interior surface of VLPs,³⁸ the negatively charged [Mo₁₃₂] are easy to adsorb on inner surface of VLP, which shielding [Mo₁₃₂] from oxidation and consequently show protection of it from hypochromicity. Moreover, the observed larger aggregates of VLPs induced by [Mo₁₃₂] (Figure 5, 6) suggests another protection of [Mo₁₃₂] from hypochromicity.

Redox nature of [Mo ₁₃₂ ] in the presence of L1-p and VLP. X-ray photoelectron spectroscopy (XPS) is used to analyze the redox state of [Mo₁₃₂] (Figure 8A) and that of it in the presence of L1-p (Figure 8B) or VLP (Figure 8C) under aerobic condition. The characteristic Mo_3d doublet, composed of the 3d_5/2 and 3d_3/2 levels resulting from spin-orbit coupling, are shown in the spectra. Good fittings of the data points, corresponding to two possible 3d doublets of Mo in different oxidation states of Mo(VI) and Mo(V), are achieved by means of two pairs of Lorentzian–Gaussian function. The contribution from the peaks centered at 232.4 and 235.5 eV are assigned to Mo(V) while those at 233.6 and 236.6 eV are attributed to Mo(VI).^39,40 From semi-quantitative calculation it shows that the characteristic ratio of peak area for Mo(VI) and Mo(V) is 1.45:1 (Table 1), which is slightly over the value of 1.2 calculated from the ratio of 72 Mo(VI) to 60 Mo(V) in structure. The reason of the positive deviation can be deduced from the aerobic oxidation of partial Mo(V) atoms (approximate 10) in [Mo₁₃₂] solution during sample preparation.

Table 1

Integrated area of the simulated peak for Mo(V) and Mo(VI) from the XPS results in Figure 8A, B, and C, respectively, and the ratio of them.
	[Mo₁₃₂]	[Mo₁₃₂] + L1-p	[Mo₁₃₂] + VLPs
Mo(V)	115359.60+43910.59	15696.10+18827.48	17349.15+21444.59
Mo(VI)	147757.50+82513.98	58321.11+28475.33	26737.09+32049.94
Area ratio Mo(VI) to Mo(V)	1.45:1	2.5:1	1.5:1

The fitting curves of [Mo₁₃₂] in mixing with L1-p (Figure 8B) becomes largely different from [Mo₁₃₂] alone. Besides the binding energy pairs that are ascribed to Mo(V) and Mo(VI), the third coupled binding energy bands are observed which may attribute to the intermediate of Mo during transition. The quantitative calculation reveals the ratio of peak area for Mo(VI) and Mo(V) increases to 2.5:1 (Table 1), indicating that more Mo(V) atoms have been oxidized. Obviously, larger amount of Mo(V) (approximate 31) in [Mo₁₃₂] has been oxidized into Mo(VI) after binding with L1-p, which is a solid evidence for the enhanced hypochromicity in mixture.

Accompanying the assembly of L1-p, the XPS result of [Mo₁₃₂] return back almost to the state of [Mo₁₃₂] alone in solution (Figure 8C). Although the intermediate component of Mo atoms still appears in the fitting-model, the calculation reveals a reduction of band ratio for Mo(VI) to Mo(V) to approximate 1.5:1 (Table 1). That is, 17 of oxidized Mo(VI) are reduced back to Mo(V). As a result, the L1-p present here is just like a sensitizer for [Mo₁₃₂] hypochromicity. It is the electrostatic interaction between [Mo₁₃₂] and pep401-495 connecting the basic residue and [Mo₁₃₂] directly, which allows the oxidation of Mo(V) to Mo(VI) more easily, and consequently the intervalence charge transfer between Mo(V) and Mo(VI) centers is largely weakened and the color disappeared with time finally. As illustrated in Scheme 2, such a process is deteriorated accompanying by the elimination of free basic residues in L1-p, after assembling into VLP.

The opposite color response of a giant polyoxometalate, (NH₄)₄₂[Mo₁₃₂O₃₇₂(CHCOO)₃₀] (H₂O)₇₂ ([Mo₁₃₂]), to the existing states of human papillomavirus (HPV) major capsid protein, L1-pentamer (L1-p) and virus-like particles (VLPs) is reported in the present study, which is originated from the assembly between [Mo₁₃₂] and capsid protein. The assembly with L1-p results in an improved hypochromicity of [Mo₁₃₂] while that with the as-assembled VLPs leads to an obvious protection of [Mo₁₃₂] from the hypochromicity, compared to the cluster alone in solution. Both the size and morphology of the assemblies were inspected and validated well by using CsCl gradient centrifugation, SDS-PAGE, DLS and TEM in detail. Particularly, the in-depth mechanism studies determined by XPS and two well designed peptides from L1 discloses the electrostatic interaction between [Mo₁₃₂] and the peptide of pep401-495 induce the valence state change of Mo(V) to Mo (VI), which facilitate the hypochromicity of [Mo₁₃₂] and the color turns from brick-red to colorless; however, the assembly between [Mo₁₃₂] and VLPs induced the back reduction from Mo(VI) to Mo(V) and the protection of [Mo₁₃₂] from hypochromicity. Therefore, the present study not only reports a unique phenomenon of opposite color responses of Mo₁₃₂ to HPV, 16 L1-p and VLPs (which affords an easily performed colorimetry approach to directly evaluate the states of HPV capsid protein), but also extends potentials of polyoxometalate family in finding new application in medical science and would be possibly extended to other kinds of POMs.

Reagents and Materials
The tryptone was a product of OXOID Ltd (Basingstoke, UK). Isopropyl-β-D-thiogalactoside (IPTG) and kanamycin were purchased from Tianjia Tech, China. Ethylene diamine tetraacetic acid (EDTA), acetic acid, ethanol and methanol were obtained from BCIGC Ltd, China. Coomassie blue G-250 and Tween 80 were produced from DingGuo Ltd (Beijing, China). 3-(N- morpholinyl) propanesulfonic acids (MOPs), and NaCl used for buffer preparation were purchased from Aladdin, China. D,L-dithiothreitol (DTT) and Tris(hydroxymethyl)aminomethane (Tris) are the products of Coolaber Company in China. The purity of all chemicals is higher than 99.9% which are used as received.
Expression, Purification of HPV 16 L1 Protein and Peptides
The coding sequences and preparation of major capsid protein (L1) for HPV 16, were followed the procedures described in previous study where pET-30a vector and BL21 Star^TM (DE3) were employed for better expression.⁴⁰ Briefly, the recombinant Escherichia coli strain BL21 Star^TM (DE3) was cultured at 37 ºC with stirring at 220 rpm, where the bacteria are induced by 0.1 mM IPTG. When OD₆₀₀ reaches 0.6−0.8, the cells were further cultivated at 25 ºC for another 17 h, and then they were harvested by centrifugation. The cell pellets were re-suspended in buffer A (50 mM MOPs, 250 mM NaCl, 10 mM DTT, pH 7.0) with a concentration of 1.0 g in 10 mL. After being lysed by sonication and separated by centrifugation, the supernatant rich with L1 protein was obtained and further stabilized for another 30 min with an increased DTT (20 mM) in buffer A. The purification was performed again by using a cationic exchange column filled with POROS® XS (xk16/10), where the supernatant was feed to the column after equilibrated with binding buffer A. The target protein L1 was finally eluted by 1.2 M NaCl and its final concentration was determined by using a BCA method. The purity of L1 was finally identified by using 12% (wt/vol) SDS-PAGE.
The coding sequences of HPV 16 pep1-401 and pep401-495 which are the same as the corresponding segment of HPV 16 L1 were synthesized by Kumei Company (Changchun, China). The pET-30a vector containing the respect gene was transformed into Escherichia coli strain BL21 Star^TM (DE3) and cultured at 37 ºC. The experimental condition and method used for their expression as well as the following purifications are the same as that of HPV 16 L1 protein described above.
Preparation of (NH ₄ ) ₄₂ [Mo ₁₃₂ O ₃₇₂ (CH ₃ COO) ₃₀ ]∙72H ₂ O
The synthesis and structural characterization of [Mo₁₃₂] cluster follow the procedures similar to those reported.¹⁸ Briefly, 0.8 g N₂H₄·H₂SO₄ was added into 250 mL of H₂O containing 5.6 g of (NH₄)[Mo₇O₂₄]·4H₂O and 12.5 g of CH₃COONH₄. After stirring for 10 min, 83 mL of 50% CH₃COOH (v/v) was added to replace the involved -SO₄²⁻, further stabilizing and giving the product as a crystalline precipitate. After the desiccation in air, the product was characterized by X-ray diffraction, IR and elemental analysis. The 100 µM of stock solution of Mo₁₃₂ was then prepared in deionized water, which was further diluted to the desired concentration when it was ready to use.
Assembly/Disassembly examination of Mo ₁₃₂ and HPV 16 L1
Four typical samples were prepared for the assembly and disassembly, respectively. For the first sample, the stock solution of Mo₁₃₂ was mixed with L1-p in buffer A to get a full binding mixture with the concentration of 2.5 µM for inorganic cluster and 10 µM for protein after stirring at 4 ºC for 24 h; For the second one, it was prepared right after mixing the same amount of Mo₁₃₂ and L1-p used in the first sample, but the mixture was subsequently dialyzed in 1.0 L of an assembly buffer (10 mM phosphate, 500 mM NaCl, 0.03% Tween 80) at 4 ºC for 24 h to induce L1-p assemble into VLPs with the presence of Mo₁₃₂; In the third sample, 10 µM of L1-p was dialyzed in 1.0 L of the assembly buffer to get full-sized empty VLPs without the presence of Mo₁₃₂, which was used as a control group or further mixed with Mo₁₃₂ for a post-assembly as the forth sample after encountering anther 24 h stirring at 4 ºC.
In a parallel experiment, sample 2 was further dialyzed against 1.0 L of disassembly buffer (50 mM Tris, 100 mM NaCl, 2 mM EDTA, 20 mM DTT, 0.01% Tween 80) at 4 ºC for 24 h to record the disassembly behavior of VLPs into L1-p in the presence of Mo₁₃₂. The concentration of Mo₁₃₂ used in all above experiments is the same at 2.5 µM unless it was mentioned, while that for HPV 16 L1 is 10 µM depending on L1 monomer. In addition, the procedures used to monitor the conversion between L1-p and VLPs are the same as reported previously.^27,28
Cesium Chloride Gradient Centrifugation
First, the discontinuous densities of CsCl aqueous solution at 1.50, 1.35, and 1.25 g/mL are prepared respectively. Then, the solutions in a volume of 0.5, 3.0, and 3.5 mL were slowly added into ultracentrifuge tubes in order, respectively. After further addition of the mixture solution of L1-p protein and Mo₁₃₂ with equivalent volume to the upper part separately, the ultracentrifuge tubes were placed in the SW40 rotor and centrifuged at 220,000×g for 3.5 h at 4 ºC. Finally, the interest bands and the counterparts were collected by a needle and further dialyzed in PB buffer for 24 h to remove the attached CsCl. The products were identified by 12% SDS-PAGE, dynamic light scattering (DLS), and transmission electron microscopy (TEM).
Instruments
UV-vis absorption spectra were recorded on a Shimadzu (Japan) RF-5301PC spectrometer for the assembly and/or disassembly of HPV 16 L1 protein or peptides. 1.0 mL target solution was added in a quartz cuvette and the spectra at the range of 300−700 nm were collected for each sample. Size distribution of particles was assayed by DLS. Briefly, after a pre-filtration by placing the sample in cuvette in a type of PCS1115, the particle sizes were tested on a Malvern Zetasizer Nano-ZS 90 (USA) at 25ºC. All data were repeated for each sample in three parallel sets. The procedures in the sample preparation for TEM measurement were the same as those described before.^21,22 The samples without staining were firstly spotted on copper grids coated with carbon and formvar. After drying in air for 2 min, the measurements were carried out on a H-7650 transmission electron microscope (Hitachi, Japan) under an accelerating voltage of 80 kV. The images of samples with stain of 2% phosphotungstate were collected under an accelerating voltage of 120 kV.

Conflicts of Interest

The authors declare that they have no competing interests.

Funding

We greatly appreciate the financial supports from NSFC (Nos. 21875085 and 21373101) and the Open Project from the State Key Laboratory of Supramolecular Structure and Materials, Jilin University.

Supporting Information

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/ >>>. Absorption spectra of (1) [Mo₁₃₂] and VLPs in disassembly buffer, (2) [Mo₁₃₂] in the presence of vitamin C, (3) [Mo₁₃₂] in the presence of control peptides and proteins; TEM images with stain for (1) VLPs, (2) the mixture of [Mo₁₃₂] and L1-p, (3) [Mo₁₃₂] and the as-prepared VLPs in assembly buffer; 3D illustrations of HPV16 L1-p and its neighbors; Agarose gel electrophoresis of two peptides derived from HPV 16 L1 to assay their DNA sequences (PDF).

Frega, ; Stentella, P.; Ioris, A. D.; Piazze, J. J.; Fambrini, M.; Marchionni, M.; Cosmi, E. V. Young women, cervical intraepithelial neoplasia and human papillomavirus: risk factors for persistence and recurrence. Cancer Lett.2003, 196(2), 127–134.
Colón-López, V.; Ortiz, A. P.; Palefsky, J. Burden of human papillomavirus infection and related comorbidities in men: implications for research, disease prevention and health promotion among hispanic men. R. Health Sci. J.2010, 29(3), 232–240.
Dunjic, M.; Stanisic, S.; Krstic, D.; Stanisic, M.; Dunjic, M. Integrative approach to diagnosis of genital human papillomaviruses (HPV) infection of female. Electro-Ther. Res.2014, 39(3-4), 229−239.
Pogoda, C. S., Roden, R. B. S., Garcea, R. L., Immunizing against anogenital cancer: HPV vaccines. PLoS Pathog.2016, 12(5), e1005587.
Harper, D. M.; Franco, E. L.; Wheeler, C.; Ferris, D. G.; Jenkins, D.; Schuind, A.; Zahaf, T.; Innis, B.; Naud, P.; De Carvalho, N. S.; Roteli-Martins, C. M.; Teixeira, J.; Blatter, M. M.; Korn, A. P.; Quint, W.; Dubin, G. GlaxoSmithKline HPV Vaccine Study Group. Efficacy of a Bivalent L1 Virus-Like Particle Vaccine in Prevention of Infection with Human Papillomavirus Types 16 and 18 in Young Women: A Randomised Controlled Trial. Lancet2004, 364, 1757−1765.
Einstein, M. H. Acquired Immune Response to Oncogenic Human Papillomavirus Associated with Prophylactic Cervical Cancer Vaccines. Cancer Immunol. Immunother.2008, 57, 443−451.
Lua, L. H. L.; Connors, N. K.; Sainsbury, F.; Chuan, Y. P.; Wibowo, N.; Middelberg, A. P. J. Bioengineering virus-like particles as vaccines. Bioeng.2014, 111(3), 425−440.
McCarthy, M. P.; White, W. I.; Palmer-Hill, F.; Koenig, S.; Suzich, J. A. Quantitative disassembly and reassembly of human papillomavirus type 11 viruslike particles in vitro. Virol.1998, 72, 32-41.
Zdanowicz, M.; Chroboczek, J. Virus-like particles as drug delivery vectors. Acta Biochim. Pol.2016, 63, 469−473.
Huber B., Schellenbacher C., Shafti-Keramat S., Jindra C., Christensen N., Kirnbauer R. Chimeric L2-Based Virus-Like Particle (VLP) Vaccines Targeting Cutaneous Human Papillomaviruses (HPV). PLoS One2017, 12(1): e0169533.
Mach, H.; Volkin, D. B.; Troutman, R. D.; Wang, B.; Luo, Z.; Jansen, K. U.; Shi, L. Disassembly and reassembly of yeast-derived recombinant human papillomavirus virus-like particles (HPV VLPs). Pharm. Sci-Us2006, 95, 2195–2206.
Molina Sánchez, P. Polyoxometalate Self-Assembly: From Molecules to Hybrid Materials. University of Glasgow2013.
Wu, L., Liang, J. Polyoxometalates and Their Complexes Toward Biological Application. Supramolecular Chemistry of Biomimetic Systems, (Ed.: Li, J.), Springer Nature Singapore Pte Ltd., 2017, pp. 311–354.
Bijelic, A.; Aureliano, M.; Rompel, A. Polyoxometalates as potential next-generation metallodrugs in the combat against cancer. Chem. Int. Ed.2019, 58, 2980–2999.
Liu, T.; Diemann, E.; Li, H.; Dress, A. W. M.; Müller, A. Self-assembly in aqueous solution of wheel-shaped Mo₁₅₄ oxide clusters into vesicles. Nature2003, 426(6962), 59–62.
Müller, A.; Rehder, D. Molecular Metal Oxides in Protein Cages/Cavities. Coordination Chemistry in Protein Cages. Coordination Chemistry in Protein Cages: Principles, Design, and Applications (Eds.: Ueno, T.; Watanabe, Y.), John Wiley & Sons, Inc., 2013, pp. 25–42.
Ren, H.; Shehzad, F. K.; Zhou, Y.; Zhang, L.; Iqbal, A.; Long, Y. Incorporation of Keplerate-type Mo–O based macroanions into Zn₂Al-LDH results in the formation of all-inorganic composite films with remarkable third-order optical nonlinearity. Dalton Trans.2018, 47, 6184–6188.
Zhang, S.; Chen, H.; Zhang, G.; Kong, X.; Yin, S.; Li, B.; Wu, L. An ultra-small thermosensitive nanocomposite with a Mo₁₅₄-core as a comprehensive platform for NIR-triggered photothermal-chemotherapy. Mater. Chem. B2018, 6, 241–248.
Li B.; Li W.; Li H.; Wu L. Ionic Complexes of Metal Oxide Clusters for Versatile Self-Assemblies. Chem. Res.2017, 50, 1391–1399.
Li J.; Chen Z.; Zhou M.; Jing J.; Li W.; Wang Y.; Wu L.; Wang L.; Wang Y.; Lee M. Polyoxometalate-Driven Self-Assembly of Short Peptides into Multivalent Nanofibers with Enhanced Antibacterial Activity. Chem. Int. Ed.2016, 55, 2592–2595.
Müller A.; Serain, C. Soluble molybdenum blues-"des Pudels Kern". Chem. Res.2000, 33, 2–10.
Müller, A.; Krickemeyer, E.; Meyer, J.; Bögge, H.; Peters, F.; Plass, W.; Diemann, E.; Dillinger, S.; Nonnenbruch, F.; Randerath, M.; Menke, C. [Mo₁₅₄(NO)₁₄O₄₂₀(OH)₂₈(H₂O)₇₀]^{(25 ± 5)−}: A Water-Soluble Big Wheel with More than 700 Atoms and a Relative Molecular Mass of About 24000. Chem. Int. Ed. Engl. 1995, 34, 2122–2124.
Müller, A.; Krickemeyer, E.; Bögge, H.; Schmidtmann, M.; Peters, F. Organizational Forms of Matter: An Inorganic Super Fullerene and Keplerate Based on Molybdenum Oxide. Chem. Int. Ed.1998, 37, 3359–3363.
Müller, A.; Sarkar, S.; Shah, S. Q. N.; Bögge, H.; Schmidtmann, M.; Kögerler, P.; Hauptfleisch, B.; Trautwein, A. X.; Schünemann, V. Archimedean Synthesis and Magic Numbers: "Sizing" Giant Molybdenum-Oxide-Based Molecular Spheres of the Keplerate Type. Chem. Int. Ed.1999, 38(21), 3238–3241.
Guo, J.; Liu, W.; Dong, X.; Lei, H.; Li, X.M.; Sun B.; Yin, Y. H. Expression and purification of human papillomavirus type 52 virus-like particles in Escherichia coli. Biotechnology2019, 2, 127–132.
Liu, T.; Langston, M. L. K.; Li, D.; Pigga, J. M.; Pichon, C.; Todea, A. M.; Müller, A. Self-recognition among different polyprotic macroions during assembly processes in dilute solution. Science2011, 331(6024), 1590–1592.
Fu, D.-Y.; Zhang, S.; Qu, Z.; Yu, X.; Wu, Y.; Wu, L. Hybrid Assembly toward Enhanced Thermal Stability of Virus-like Particles and Antibacterial Activity of Polyoxometalates. ACS Appl. Mater. Interfaces2018, 10, 6137–6145.
Jin, S.; Zheng, D.-D.; Sun, B.; Yu, X.; Zha, X.; Liu, Y.; Wu, S.; Wu, Y. Controlled Hybrid-Assembly of HPV16/18 L1 Bi VLPs in Vitro. ACS Appl. Mater. Interfaces. ACS Appl. Mater. Interfaces2016, 8, 34244–34251.
Zheng, D.-D.; Fu, D.-Y.; Wu, Y.; Sun, Y.-L.; Tan, L.-L.; Zhou, T.; Ma, S.-Q.; Zha,; Yang, Y.-W. Efficient inhibition of human papillomavirus 16 L1 pentamer formation by a carboxylatopillarene and a p-sulfonatocalixarene. Chem. Commun.2014, 50, 3201–3203.
Chen, X. S.; Garcea, R. L.; Goldberg, I.; Casini, G.; Harrison, S. C. Structure of Small Virus-like Particles Assembled from the L1 Protein of Human Papillomavirus 16. Cell2000, 5, 557–567.
Zhang, T.; Li, H.-W. Wu, Y.; Wang, Y.; Wu, L. Self-Assembly of an Europium-Containing Polyoxometalate and the Arginine/Lysine-Rich Peptides from Human Papillomavirus Capsid Protein L1 in Forming Luminescence-Enhanced Hybrid Nanospheres. Phys. Chem. C 2015, 119, 8321–8328.
Zhang, T.; Li, H.-W.; Wu, Y.; Wang, Y.; Wu, L. The Two-Step Assemblies of Basic-Amino-Acid-Rich Peptide with a Highly Charged Polyoxometalate. -Eur. J.2015, 21, 9028–9033.
Bernardini, G.; Wedd, A. G.; Zhao, C.; Bond, A. M. Photochemical oxidation of water and reduction of polyoxometalate anions at interfaces of water with ionic liquids or diethylether. Natl. Acad. Sci. U.S.A. 2012, 109, 11552–11557.
Zhang, J.; Bond, A. M.; MacFarlane, D. R.; Forsyth, S. A.; Pringle, J. M.; Mariotti, A. W. A.; Glowinski, A. F.; Wedd, A. G. Voltammetric Studies on the Reduction of Polyoxometalate Anions in Ionic Liquids. Chem.2005, 44, 5123–5132.
Bijelic, A.; Rompel, A. The use of polyoxometalates in protein crystallography – An attempt to widen a well-known bottleneck. Chem. Rev.2015, 299, 22–38.
Bijelic, A.; Rompel, A. Ten Good Reasons for the Use of the Tellurium-Centered Anderson–Evans Polyoxotungstate in Protein Crystallography. Chem. Res.2017, 50, 1441–1448.
Cardone, G.; Moyer, A. L.; Cheng, N.; Thompson, C. D.; Dvoretzky, I.; Lowy, D. R.; Schiller, J. T.; Steven, A. C.; Buck, C. B.; Trus, B. L. Maturation of the Human Papillomavirus 16 Capsid. mBio2014, 5(4), e01104–e01114.
Li, M.; Cripe, T. P.; Estes, P. A.; Lyon, M. K.; Rose, R. C.; Garcea, R. L. Expression of the human papillomavirus type 11 capsid protein in escherichia coli: characterization of protein domains involved in DNA binding and capsid assembly. Virol, 1997, 71, 2988–2995.
Newton, G. N.; Cameron, J. M.; Wales, D. J. Shining a light on the photo-sensitisation of organic-inorganic hybrid polyoxometalates. Dalton Trans.2018, 47, 5120–5136.
Shi,; Zhou, Y.; Zhang, L.; Yang, D.; Mu, C.; Ren, H.; Shehzada, F. K.; Li J. Fabrication and optical nonlinearities of composite films derived from the water-soluble Kepleratetype polyoxometalate and chloroform-soluble porphyrin. Dalton Trans.2015, 44, 4102–4107.

No competing interests reported.

scheme1.png
Scheme 1. Schematic illustration of the opposite colorimetry responses of [Mo132] to HPV L1-p and VLPs.
scheme2.png
Scheme 2. Schematic drawing of the chromic mechanism of [Mo132] induced by protein.
SupplimentaryMaterials.pdf

Download PDF

Journal Publication

published 22 Feb, 2022

Read the published version in Nanomaterials →

Version 1

posted

You are reading this latest preprint version

A Visual Discrimination of Existing States of Virus Capsid Protein by a Giant Molybdate Cluster

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Results And Discussion

3. Conclusions

4. Experimental Section

Declarations

Conflicts of Interest

Funding

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1