Susceptibility of SARS COV-2 nucleocapsid and spike proteins to reactive oxygen species and role in inflammation

Chemiluminescence was used to test the susceptibility of the SARS-CoV-2 N and S proteins to oxidation by reactive oxygen species (ROS) at pH 7.4 and pH 8.5. The Fenton's system generates various ROS (H2O2, •OH, –OH, •OOH). All proteins were found to significantly suppress oxidation (the viral proteins exhibited 25–60% effect compared to albumin). In the second system, H2O2 was used both as a strong oxidant and as a ROS. A similar effect was observed (30–70%); N protein approached the effect of albumin at physiological pH (∼45%). In the O2.--generation system, albumin was most effective in the suppression of generated radicals (75%, pH 7.4). The viral proteins were more susceptible to oxidation (inhibition effect no more than 20%, compared to albumin). The standard antioxidant assay confirmed the strong antioxidant capacity of both viral proteins (1.5–1.7 fold higher than albumin). These results demonstrate the effective and significant inhibition of ROS-induced oxidation by the proteins. Obviously, the viral proteins could not be involved in the oxidative stress reactions during the course of the infection. They even suppress the metabolites involved in its progression. These results can be explained by their structure. Probably, an evolutionary self-defense mechanism of the virus has been developed.


Introduction
The pathology of SARS CoV-2 and associated complications results from complex biological phenomena such as apoptosis induction, macrophage activation, oxidative tissue damage, and higher content of pro-inflammatory cytokines [1][2][3][4]. The pathogenesis also involves injury of non-infected cells by superoxide anion radicals derived from activated phagocytes infiltrated into the virus-infected organs [5,6]. These superoxide anion-mediated pathways are part of the immune mechanisms responsible for the extensive tissue injury observed during severe virus infection and its complications induced by oxidative stress [7,8]. In particular, reactive oxygen species (ROS), such as hydrogen peroxide (H 2 O 2 ), superoxide anion radicals (O 2 .-), singlet oxygen, and other radicals are considered to be agents attacking polyunsaturated fatty acids in cell membranes and giving rise to lipid peroxidation [1,9].
In fact, free radical processes are part of the body's natural reactions to metabolize oxygen and the toxic products of metabolism, renewal, and immune response. Normally, there is a balance between their production and elimination by the antioxidant systems [1,10,11].
Healthy cells do not undergo severe lipid peroxidation in vivo because of their extremely efficient protective mechanisms against damage. Such antioxidant systems include enzymes such as superoxide dismutase, glutathione-peroxidase, and catalase, as well as nonenzymatic protectors such as vitamin E, vitamin C, ß-carotene, uric acid, various proteins such as albumin and bilirubin, and many other substances [7,9]. In general, antioxidants are compounds that can delay, inhibit, or prevent the oxidation of substances, trapping free radicals and reducing oxidative stress [12][13][14][15]. The antioxidant capacity of the body is constantly provoked by the processes of free radical oxidation, especially in the development of viral infections. For these reasons, the combination of antioxidants with anti-viral drugs could definitely improve the conventional therapy of SARS CoV-2.
This research focused on testing two of the structural proteins of SARS CoV-2 -nucleocapsid (N) and spike (S) proteins ( Fig. 1), for their susceptibility to free radical oxidation and ROS; namely H 2 O 2 , O 2 .-, hydroxyl radicals ( . OH) and hydroperoxyl radicals ( . OOH), using albumin as a reference protein.
The antioxidant properties of albumin and other human proteins are well known. Albumin represents the major and predominant antioxidant in blood plasma, a body compartment known to be exposed to continuous oxidative stress. A large proportion of the plasma antioxidant properties can be attributed to albumin [16].
SARS-CoV-2 belongs to the group of beta-coronaviruses. A coronavirus contains four structural proteins, including spike (S), envelope (E), membrane (M), and nucleocapsid (N) proteins ( Fig. 1) [17]. SARS-CoV-2 infection begins with the attachment of S proteins to host cell receptors, leading to structural changes in the S protein, followed by the fusion of the virus and host membranes. SARS-CoV-2 replicates in the cytoplasm of the infected cells, where the viral genome is first cleaved from the associated viral N proteins by cellular proteases [18]. The N protein has various roles in the life cycle of SARS-CoV-2: to facilitate viral RNA production, suppress the host cell innate immune responses, and package viral genomic RNA into developing virions [19].
In order to test the susceptibility of SARS-CoV-2 N and S proteins to ROS and elucidate their role in the inflammation process and the immune response, we used an ex vivo chemiluminescence assay, applying buffer model systems at pH 7.4 (physiological) and pH 8.5 (enabling the achievement of comparable differences). The chosen test reactions generated various free radicals and ROS. Having in mind that H 2 O 2 , O 2 .-, . OH and . OOH are part of the inflammation, the immune response, and the first line of cellular defense, the tests clarified which of them are most damaging and destructive to those viral proteins. Such assays, testing the above-mentioned specific viral proteins, have not been utilized so far. As far as ROS are part of the cell immune response of the white blood cells, these model reactions can be informative in comparing the intensity of the response in various in vitro or in vivo testing systems.

Chemiluminescence:
The kinetics of lucigenin-enhanced chemiluminescence was examined ex vivo for three chemical systems with ROS. The chosen reactions were tested at 25 • C/pH 7.4 -physiological, and 25 • C/pH 8.5. It is expected that the results obtained at higher temperatures (37-38 • C) would be of the same order. The only thing changed would be the speed of the reactions.
The kinetics was registered for 3 min using LUMIstar Omega luminometer, BMG Labtech, Germany. Each sample was prepared as a mixture containing the buffer medium, the chemiluminescent probe lucigenin, and chemical reagents for ROS. The test mixtures contained one of the structural viral proteins (1.04 pmol) preset before the starter of the reaction. One type of control was the reaction with albumin (identical concentrations). Another type of control (blank) was a reaction without any protein.
Testing model systems. Antioxidant assay: Cayman's Antioxidant Assay measures the total antioxidant capacity. The reaction is based on the ability of the antioxidants in the sample to inhibit the oxidation of ABTS (2,2 ′ -Azino-di-[3-ethylbenzthiazoline sulphonate]) to ABTS by metmyoglobin. The amount of ABTS produced can be monitored by reading the absorbance at 405 nm. The antioxidants in the sample suppress the absorbance at 405 nm to a degree that is proportional to their concentration. The capacity of the tested antioxidants to prevent ABTS oxidation is compared to Trolox. It is a water-soluble tocopherol analogue. The activity is Statistics: All results were obtained as triple-reproducible measurements. Statistical calculations (Student's t-test) were processed by the Origin 8.5 (OriginLab) and Microsoft Office Excel 2010 (Microsoft Corp.) software. Data were accepted as statistically significant at p ≤ 0.05.

Results and discussion
ROS in the body are the result of the cellular immune response to protect against infection on the one side and, on the other, a cause of cell damage and pathology. These model systems can be informative with regards to the extent to which the tested N and S viral proteins are involved in these oxidation reactions, the accumulation of toxic products, and the development of oxidative stress during COVID infection.
The chemiluminescence method allows tracing the concentration and kinetics of free radical formation with small amounts of the sample. It is easy to determine the quantum yield of these reactions (480-580 nm). This is a suitable, reliable, highly sensitive, express, and relatively inexpensive method for in vitro/ex vivo testing. It can monitor the dynamics of free radical reactions and determine their prooxidant/antioxidant activity. The spontaneous chemiluminescence can be enhanced by physical and chemical activators (probes), such as luminol, lucigenin, and many other specific or nonspecific traps of hydroxyl and/or superoxide anion radicals and other ROS. The reactions are accompanied by light emission [13,20,21]. As with any other method, the chemiluminescent assay has its limitations. But our careful optimization and validation of the assay conditions as well as performance ensured the presentation of accurate and reliable results.
In the chemiluminescence model systems used in the study the concentrations given below were final.
• System 1 (Fig. 2) The sample contained 0.2 mol of sodium hydrogen phosphate buffer with the appropriate pH, Fenton's reagent: FeSO 4 (5.10 -4 mol) -H 2 O 2 (1.5%), lucigenin (10 -4 mol), and the protein tested (1.04 pmol). It is well known that the interaction between Fe 2+ ions and H 2 O 2 results in the generation of highly reactive, short-living radicals. The chemiluminescence emission from this reaction was much higher than the one obtained from the other mixtures.
At pH 8.5, the inhibition of the reaction by all three proteins tested was immediate and maintained with time. Their effect in comparison to the control (blank) reaction was as follows: albumin -36%, N protein -42% and S proteinalmost 60%. The kinetic curves were smooth and looked like a plateau.
At physiological pH 7.4, the inhibition effect was maintained but exhausted with time. Most effective was the S protein, whose suppression was highest and stable with time. In comparison to the control reaction, the effects were as follows: albumin -8%, N protein -15% and S protein -45%.
In Fenton's system various ROS are generated -H 2 O 2, • OH, -OH, • OOH. All three proteins demonstrated effective inhibition or neutralization against them, as the viral proteins were better suppressors of the oxidation.
• System 2 ( Fig. 3) The sample contained 0.2 mol of sodium hydrogen phosphate buffer with an appropriate pH, H 2 O 2 (1.5%), the chemiluminescent probe lucigenin (10 -4 mol), and the protein tested (1.04 pmol). In this system, hydrogen peroxide plays both roles -of a strong oxidizing agent and of a ROS. Its major role in the organism is that of a signaling molecule. In this model system, an analogous effect was observed with time: strong and effective inhibition of the luminescent signal and H 2 O 2 oxidation in both physiological and alkaline media.
At pH 8.5, the effect in comparison to the control reaction was as follows: albumin -46%, N protein -57% and S protein -70%. At pH 7.4, the effect in comparison to the control reaction was as follows: albumin -37%, N protein -32% and S proteinmore than 50%. The interesting fact here is that the N protein had an effect very similar to that of albumin but it was a worse inhibitor of oxidation under physiological conditions.
Multiple isoforms of NADP.H oxidase and xanthine oxidase represent the main sources of ROS in immune cells. These enzymes catalyze the synthesis of superoxide anion radicals via single electron reduction of oxygen [25].
At pH 8.5, all three proteins suppressed the generation of superoxide anion radicals over time; albumin was the most effective. The inhibition in comparison to the control reaction was as follows: albuminless than 30%, N protein -57% and S protein -70%.
At physiological pH 7.4, albumin demonstrated the best inhibition of the oxidation process; the viral proteins showed very mild and unstable suppression of the luminescence signal that made them susceptible to destruction by those radicals. The inhibition in comparison to the control (blank) reaction was as follows: albumin -76%, N protein -22% and S protein -13%.
The observed closeness to the blank control luminescence intensities, in some cases of the ex vivo assays, can be explained by the exhaustion of the antioxidant capacity of the tested protein -albumin at pН 7.4 in system (1) and the viral proteins at рН 7.4 in system (3).
In order to confirm the antioxidant properties of SARS CoV-2 N and S proteins, we tested their antioxidant activity by a standard method compared to Trolox (a water-soluble analogue of vitamin E) and albumin ( Table 1). The results confirmed their antioxidant activity in the following order: S protein > N protein ≫ human albumin. The standard antioxidant assay confirmed the strong antioxidant capacity of both viral proteins (1.5-1.7 times higher than that of human albumin).
In conclusion, at both investigated acidities, physiological and alkaline (favoring oxidation reactions, generation of ROS, and luminescence), effective and significant inhibition of the oxidation by ROS was observed by all three tested proteins. This means that both albumin (a major plasma protein, used here as a control) and SARS CoV-2 viral proteins do not participate in oxidative stress or free radical defense reactions during the inflammation process in the course of this viral infection. On the contrary, the viral proteins have a suppressive effect over the ROS metabolites, which stimulate these chain reactions; they do not relate to the severity and development of COVID pathology per se. The deleterious effects of the disease and the development of the infection must be sought and are probably caused by: -the activity of the virus as an intact particle -the weakening of the immune system struggling with the infection -the induced oxidative stress, both as a result of the pathology and the activated immune response.
The obtained results can be explained by the structure of the tested proteins. They all contain the amino acids cysteine and methionine (-SH functional groups). These have proven antioxidant and inhibitory effects on reactions related to oxidative stress in the organism. Their ratio in the viral proteins to that in albumin is as follows: N protein -cysteine 0:35 methionine 7:7, and S protein -cysteine 40:35 methionine 14:7.
It should be noted that the observed considerable differences in the effects and antioxidant activity could not be due to the chelation activity of the amino acid histidine as part of the structure of the tested proteins. The number of histidine residues is as follows: albumin -16, S protein -17, N protein -17. That number could be taken into account only for the inhibition effect observed in system 1, Fenton's system, but cannot explain the differences in the suppressive effect on the generation of ROS.
The applied protein concentrations were equal and extremely low -1.04 pmol. That concentration was chosen as indicative of biochemical sensitivity in the pmol range. In fact, the concentrations of the viral proteins in the blood plasma of SARS-CoV-2 patients vary in the pmol range, starting from ~10 pmol. On the other hand, the average reference concentration of albumin in blood plasma ranges between 33 and 52 g/l. This is another support of the high-presented activity of the tested proteins under physiological conditions.
Probably, the effective defense mechanism of the viral N and S proteins against ROS is a strategy developed and maintained in the evolutionary process against the very aggressive and destructive products of the immune system in the first line of defense, the inflammation process.

Funding
The authors acknowledge the financial support of the National Science Fund of Bulgaria, project: Study of the interaction of specific structural proteins of SARS-CoV-2 with biologically active molecules and their application for the creation of rapid antigen tests for early diagnosis of COVID-19, contract number: ДК1/10 29.03.2021.

Contributions
Elitsa Pavlova has designed the study, carried out all the experiments and conclusions, and written the manuscript; Petya Genova-Kalou and Georgi Dyankov have supported the study financially, discussed the obtained results, and made corrections to the manuscript.

Table 1
The antioxidant activity of the tested proteins, compared to Trolox at 405 nm (p ≤ 0.001).

Declaration of competing interest
Elitsa Pavlova 1 , Petya Genova-Kalou 2 , Georgi Dyankov 3 , 1 Sofia University "St. Kliment Ohridski", 2 National Centre of Infectious and Parasitic Diseases, 3 Bulgarian Academy of Sciences, declare that they have no conflict of interests and that Sofia University, the National Centre of Infectious and Parasitic Diseases and the Bulgarian Academy of Sciences have also approved its publication.

Data availability
Data will be made available on request.