Polysaccharide-Based Organic Frameworks with Embedded Nanoparticles: Advanced SPR Study on the Antiviral Activity of Gold Composites Derived from Glucuronoxylomannan

: The nanosized composites based on the natural polysaccharides and nanoparticles of noble metals are promising candidates for efﬁcient antiviral drugs. However, the complexity of such objects, their diversity and novelty necessitate the development of new analytical methods for investigation of such supramolecular architectures. In this work, which was recently developed for SPR-based instrumentation, the concept of variative refraction (DViFA, density variations in ﬁxed architectures) was used to elucidate the mechanism of the antiviral action of a polysaccharide with gold nanoparticles grown in it. The SPR data were conﬁrmed by direct biological tests: the effect of the native polysaccharide glucuronoxylomannan (GXM) obtained from the fungus Ganoderma adspersum and gold nanocomposites thereon on the infection of Datura stramonium with tobacco mosaic virus (TMV) was investigated. Both drugs suppress the development of viral infections. However, if for high concentrations the characteristic activity of the composite is somewhat lower than for GXM, then with an increase in dilution, the effectiveness of the composite increases signiﬁcantly, up to a twofold excess. It has been reasonably suggested that the mechanism of antiviral action is associated with the formation of clusters of viruses that are no longer capable of infecting cells. The results of molecular analysis and in vivo studies suggest that polysaccharide matrices with embedded gold nanoparticles have a stronger antiviral effect at low concentrations in comparison with natural polysaccharides. The mechanism of this action is due the fact that the metal composite induces the aggregation of viral particles into clusters plant cells.


Introduction
Nanosized composites combined organic compounds and inorganic nanoparticles extend our capabilities to form supramolecular architectures of advanced functionality. Complex macromolecules of biological origin, which can act as a matrix for the synthesis of such objects, are of particular interest. In this work, the polysaccharide glucuronoxylomannan with known immunomodulatory activity was chosen as such a matrix [1].
Earlier, we considered the potential of polysaccharide glucuronoxylomannan as an antiphytoviral agent. It was demonstrated that GXM isolated from Tremella mesenterica culture can suppress TMV infection in Nicotiana tabacum and Datura stramonium plants [2]. It was shown that GXM affects both the virus before infection and the processes that are proceeding immediately in the cell. In particular, polysaccharide can suppress virus reproduction and induce plant resistance to pathogens. It was suggested that GXM sterically block the virions, thereby suppressing its ability to infect [3]. In this regard, it is reasonable to evaluate the antiviral effectiveness of GXM nanocomposites with gold and well-known inhibitory agents, the effect of which is often associated with the activity of this metal ions in an aqueous medium [4]. It should be emphasized that the "ionic model" of the antiviral action of nanoparticles finds fewer and fewer supporters among researchers: the authors of Chem. Proc. 2021, 5, 38 2 of 6 many publications increasingly prefer various physical processes in the field of contact of nanoparticles with biological objects when describing one or another effect of nanoobjects on biological systems.
Indeed, the nanosized composites turned out to be a fundamentally new objects with unusual properties-in particular, their toxicity, which does not correlate with the properties of the material in the atomic-molecular state or in a solid. This is well illustrated by the example of the effect of nano-objects on viruses and bacteria, for which a large amount of experimental material has been collected. It is reasonable to conclude that the antiviral effects of nanoparticles are due to some physical phenomenon, rather than a chemical interaction of one type or another. Nanoscale analytes induce the change of the toxicity paradigm: physical effects come to the forefront, not the features of the chemical structure. This means that we need take into account not only the chemical composition of the object, but also its geometric characteristics, such as its size and shape [5].
The aim of this work was to study GXM nanocomposites with gold nanoparticles in order to elucidate their potential as antiviral drugs using the example of the TMV virions.
Since macromolecules of polysaccharides are complex branched structures, it is not possible to uniformly introduce pre-synthesized metal nanoparticles into these structures. However, taking into account the fact that polysaccharides contain a sufficiently large set of functional groups with the characteristics necessary for carrying out the corresponding redox reactions, in this work the synthesis of metal nanoparticles was carried out directly inside the polysaccharide matrix.
This polysaccharide can act both as a reducing agent and as a stabilizer due to the presence of the corresponding functional groups in the molecule. GXM consists of a linear backbone of (1 → 3)-linked α-D-mannose with mainly xylose and glucuronic acid in the side chains [6]. Molecules of glucuronic acid contains carboxylic acid group, which gives acidic properties to GXM (acid polysaccharide). In order to clarify the peculiarities of the interaction of composites with viral particles at the molecular level, an analysis of their interactions with the tobacco mosaic virus was carried out using the SPR method. Antiviral activity of the complex was also tested in vivo at the Datura stramonium L.

Methods
Preparation of GXM. The principle of the method of obtaining preparations of β-glucan from the mycelium of the fungus G. adspersum was the same as in obtaining β(1 → 3)β(1 → 6)-bound glucan ("ganoderan") from other species of Ganoderma sp. Namely: polysaccharide preparations were obtained from lyophilized mycelium of the fungus by sequential aqueous, alkaline and acid extraction [7]. First, to remove low molecular weight compounds to the dry mycelium of the fungus G. adspersum, thoroughly ground in a porcelain mortar, add 85% ethanol solution (1:5, v/v) and boil for 3 h. The alcohol extraction procedure was repeated three times. Each time the precipitate was separated by centrifugation (10,000× g, 15-20 min) and was used in further work. To obtain the aqueous fraction of β-glucan, water (1:5) was added to the mycelium purified from low molecular weight impurities and was boiled for 3 h. The procedure was repeated 5 times. The extracts were collected by centrifugation (10,000× g, 15-20 min) and combined. The resulting extract was dialyzed against running and distilled water and then evaporated to a minimum volume (1:5) on a rotary evaporator 1/5 by volume of a mixture of isoamyl alcohol and chloroform (1:10) was then added to the concentrate, the mixture was shaken vigorously for 10 min and then centrifuged (10,000× g, 20 min) to separate the aqueous and organic phases. Deproteinization of the extract was repeated once more, the organic extracts were removed, the aqueous ones were combined and dried in freeze-drying [8].
Synthesis of AuNPs. First, GXM water solution was prepared by dissolution of 3 mg GXM in 2.9 mL H 2 O, and then aqueous solution of HAuCl 4 (0.1 mL, 30 mM) was added to it at violent stirring. The mixture was stirred during 1 min at room temperature and heated to 100 • C with boiling for 10 min. The AuNPs formation was confirmed by UV-Vis Spectroscopy and TEM.
Instrumentation and Measurements. The morphological, optical, and spectroscopic properties of AuNPs were examined using the following measurements. UV-vis spectra were acquired with Umico UV-Vis spectrophotometer (data not shown). TEM was performed at 100 kV using a JEOL-1011 (JEM, Japan) instrument (data not shown). Scanning spectrometer "BioHelper" (ISP NASU, Kiev, Ukraine) was used for SPR measurements with standard chips (50 nm Au/1.5 nm Cr/glass (n = 1.61)) and the protein A immobilization protocol described in detail in [9]. SPR measurements were carried out in a static mode without a sample flow in an open cell configuration (400 µL).

Molecular Level Analysis: Advanced SPR Study
Conventional SPR studies typically use "2D" interfacial architectures whose thickness is significantly smaller than the penetration depth of an exponentially decaying evanescent wave in a dielectric medium [10,11]. The use of SPR techniques for the investigation of "bigger" objects (e.g., intact virus particles so called virions, cells etc.) has some limitations, since their characteristic size exceeds tens of nanometers, which is required to ensure an adequacy of the quasi-linear approximation in SPR sensing.
In the conventional approach, the SPR response depends on the effective thickness of the analyte layer that is bound to the receptors on the surface-the density of both layers (receptor and analyte) is uniform-i.e., the variation of the SPR signal is due to the changes of the thickness. However, an SPR shift depends also on the change of the refractive index within the layer. Therefore, variations of the layer density can also affect the response value due to the variations of the refractive index inside the layer ("variative" refraction). One of the possible mechanisms is changing the packing of the objects of different sizes and shapes within interfacial architectures on the surface (e.g., a virus, an antibody, a small molecule etc.) If the thickness of the surface layer is fixed due to the constant form of the biggest interacting components (e.g., virion), the SPR shift is a single-valued function of the molecular assembly compactness. Exactly such an unconventional approach DViFA (density variations in fixed architectures) has been proposed by us early [12][13][14] for quantitative detection of native virus particles t (V) using SPR techniques.
Typical protocol includes the incubation of a virus-bearing material (V) with a definite concentration of a specific antibody (Ab) followed by the injection of statistically formed V-Ab complexes into SPR instrument with a sensitive surface covered by protein A. A shift of the minimum of SPR signal depended on the compactness of the V-Ab monolayer on the surface. To realize such an approach, we took advantage of (1) specificity of protein A from Staphylococcus aureus to the Fc-fragment of immunoglobulins as well as development of (2) statistically stable distribution of complexes with different numbers Abs per V specific for different Ab/V ratios.
To refine the DViFA approach, the investigation of TMV interaction with serum containing specific antibodies was performed, both with and without the presence of GXM-Au composites in the sample during the incubation phase. We investigated how the nanocomposite affects the interaction of the virus with specific antibodies. During the measurements, the concentration of TMV particles was constant (at constant concentration of 100 µg/mL), and the concentration of antibodies was changed by serial dilutions of serum in the range from 1:25 to 1:1600. The results obtained were compared with the data for a virus at the same concentration preincubated with GXM-Au complex (Figure 1 At dilution levels between 1:25 and 1:200, the antibodies fill the free surface of the V-Ab cylinders. Such complexes experience a mutual orienting effect on the surface modified by protein A and their configuration is typical for the classical "car parking" problem. The general structure of such an ensemble of viral particles on the surface remains practically unchanged up to a dilution of antibodies of about 1:200. The decrease in the SPR response in this range is due to a decrease in the number of antibodies associated with the virus, i.e., density of surface architecture. Some stabilization of the response in the range from 1:100 to 1:200 can be associated with two competing processes: on the one hand, a decrease in the number of surface-bound antibodies decreases the density, and on the other hand, it increases through some convergence of individual viral particles. In the presence of a GXM-Au composite, all the considered effects are less pronounced and are observed at low dilutions, since the GXM composite blocks some of the surface binding sites of Ab on the virion surface. At a dilution level of 1:200 and 1:100 for a virus without and in the presence of a nanocomposite, respectively, the amount of antibodies becomes insufficient to keep the virus "on side" on the surface and some of the viral particles become "at end", fixing pointwise on the surface with a chain of virus-antibody-protein A. In the absence of a GXM-Au composite, this process continues with further dilution (1:400-1:1600), leading to a decrease in the mass (number of attached virus particles) on the surface.
A radically different situation is observed in the case of the presence of a GXM gold nanocomposite: a decrease in the amount of antibodies at dilutions greater than 1:200 leads to an increase and subsequent stabilization of the response for dilutions up to 1:1600. In this case, the stabilization of the response occurs approximately at the same level, which corresponded to the densest structure of viral particles in the position "on the side". Since the diameter of the virus particle coated with antibodies is about 40 nm, and the length of the virus particle is 300 nm, then in the orientation "on side" and "at end" the surface ensemble of viral particles practically completely overlap the region of the highest concentration of the evanescent wave (less then 100 nm due to its exponential decay with At dilution levels between 1:25 and 1:200, the antibodies fill the free surface of the V-Ab cylinders. Such complexes experience a mutual orienting effect on the surface modified by protein A and their configuration is typical for the classical "car parking" problem. The general structure of such an ensemble of viral particles on the surface remains practically unchanged up to a dilution of antibodies of about 1:200. The decrease in the SPR response in this range is due to a decrease in the number of antibodies associated with the virus, i.e., density of surface architecture. Some stabilization of the response in the range from 1:100 to 1:200 can be associated with two competing processes: on the one hand, a decrease in the number of surface-bound antibodies decreases the density, and on the other hand, it increases through some convergence of individual viral particles. In the presence of a GXM-Au composite, all the considered effects are less pronounced and are observed at low dilutions, since the GXM composite blocks some of the surface binding sites of Ab on the virion surface. At a dilution level of 1:200 and 1:100 for a virus without and in the presence of a nanocomposite, respectively, the amount of antibodies becomes insufficient to keep the virus "on side" on the surface and some of the viral particles become "at end", fixing pointwise on the surface with a chain of virus-antibody-protein A. In the absence of a GXM-Au composite, this process continues with further dilution (1:400-1:1600), leading to a decrease in the mass (number of attached virus particles) on the surface.
A radically different situation is observed in the case of the presence of a GXM gold nanocomposite: a decrease in the amount of antibodies at dilutions greater than 1:200 leads to an increase and subsequent stabilization of the response for dilutions up to 1:1600. In this case, the stabilization of the response occurs approximately at the same level, which corresponded to the densest structure of viral particles in the position "on the side". Since the diameter of the virus particle coated with antibodies is about 40 nm, and the length of the virus particle is 300 nm, then in the orientation "on side" and "at end" the surface ensemble of viral particles practically completely overlap the region of the highest concentration of the evanescent wave (less then 100 nm due to its exponential decay with distance from the surface). All this suggests that the presence of a polysaccharide nanocomposite stimulates the aggregation of virus particles into large clusters that are attached to the surface by only a few antibody molecules. Summarizing the obtained results, it can be argued that the nanocomposite stimulates the aggregation of viral particles into clusters, preventing their "independent" functioning.

Biological Experiments In Vivo
To verify the antiviral activity of the nanosized composite, the classical biological test has been performed. Aqueous solutions of Au-GXM (in the concentration range 1-500 µg/mL) were added to a suspension of TMV (10 µg/mL) and the mixture was incubated for 30 min at room temperature. Then, the left halves Datura stramonium L. were inoculated with mixture, whereas the right halves were infected with the virus at the same concentration without a composite.
The degree of viral infection suppression (in percentage terms) was calculated from the number of necrotic local lesions on the test and control leaves using the following expression [8]: where I is the degree of viral inhibition in percent; C-local lesions number on the control half; P-number of local lesions on the test half. The results of the calculation of necrosis were subjected to statistical processing by parametric criteria, calculating their average number (M) and the ratio of these data in the experiment and control, as well as the average error of the ratio (m). On the graph, the data of statistical evaluation of the results were expressed as M ± m ( Figure 2).
distance from the surface). All this suggests that the presence of a polysaccharide nanocomposite stimulates the aggregation of virus particles into large clusters that are attached to the surface by only a few antibody molecules. Summarizing the obtained results, it can be argued that the nanocomposite stimulates the aggregation of viral particles into clusters, preventing their "independent" functioning.

Biological Experiments In Vivo
To verify the antiviral activity of the nanosized composite, the classical biological test has been performed. Aqueous solutions of Au-GXM (in the concentration range 1-500 μg/mL) were added to a suspension of TMV (10 μg/mL) and the mixture was incubated for 30 min at room temperature. Then, the left halves Datura stramonium L. were inoculated with mixture, whereas the right halves were infected with the virus at the same concentration without a composite.
The degree of viral infection suppression (in percentage terms) was calculated from the number of necrotic local lesions on the test and control leaves using the following expression [8]: where І is the degree of viral inhibition in percent; C-local lesions number on the control half; P-number of local lesions on the test half. The results of the calculation of necrosis were subjected to statistical processing by parametric criteria, calculating their average number (M) and the ratio of these data in the experiment and control, as well as the average error of the ratio (m). On the graph, the data of statistical evaluation of the results were expressed as M ± m ( Figure 2). The in-vivo experiment showed that both native GXM and GXM-Au composites inhibit the development of viral infection: this is manifested in a much smaller number of necrosis on the experimental halves of the leaves of the studied plants (see Insert in Figure  2). In order to compare the effectiveness of potent antiviral drugs, a concentration dependence of the percentage of necrosis on the experimental leaf halves to the necrosis on The in-vivo experiment showed that both native GXM and GXM-Au composites inhibit the development of viral infection: this is manifested in a much smaller number of necrosis on the experimental halves of the leaves of the studied plants (see Insert in Figure 2).
In order to compare the effectiveness of potent antiviral drugs, a concentration dependence of the percentage of necrosis on the experimental leaf halves to the necrosis on the control leaf halves on the drug concentration was constructed (i.e., the lower the value, the more effectively the virus is suppressed). For the preparation of native GXM, a clear linear (exponential in linear coordinates) dependence of the degree of viral infection suppression on the concentration was demonstrated. The difference in the concentration range of 0.5 mg/mL-0.001 mg/mL is c.a. 50%. It was shown that at high concentrations (0.5 mg/mL) the activity of the GXM is higher than the activity of the composite. However, the effectiveness of the GXM-Au composite is significantly less dependent on the concen-tration of the drug. This leads to the fact that the antiviral activity of the composite is much higher at low concentrations. In particular, for a concentration of 0.001 mg/mL, the degree of viral infection suppression was more than twice better than that for Au composites in respect to native glucan.
The results of molecular analysis and in vivo studies suggest that polysaccharide matrices with embedded gold nanoparticles have a stronger antiviral effect at low concentrations in comparison with natural polysaccharides. The mechanism of this action is due to the fact that the metal composite induces the aggregation of viral particles into clusters incapable of subsequent infection of plant cells.