key: cord-0864329-3njr8f7y
authors: Qin, Meng; Cao, Zheng; Wen, Jing; Yu, Qingsong; Liu, Chaoyong; Wang, Fang; Yang, Fengmei; Li, Yanyan; Fishbein, Gregory; Yan, Sen; Xu, Bin; Hou, Yi; Ning, Zhenbo; Nie, Kaili; Jiang, Ni; Liu, Zhen; Wu, Jun; Yu, Yanting; Li, Heng; Zheng, Huiwen; Li, Jing; Jin, Weihua; Pan, Sheng; Wang, Shuai; Chen, Jianfeng; Gan, Zhihua; He, Zhanlong; Lu, Yunfeng
title: An Antioxidant Enzyme Therapeutic for COVID-19
date: 2020-07-15
journal: bioRxiv
DOI: 10.1101/2020.07.15.205211
sha: 98248f61b94b7b1c49a77eb920cf6953c573ca40
doc_id: 864329
cord_uid: 3njr8f7y

The COVID-19 pandemic has taken a significant toll on people worldwide, and there are currently no specific antivirus drugs or vaccines. We report herein a therapeutic based on catalase, an antioxidant enzyme that can effectively breakdown hydrogen peroxide and minimize the downstream reactive oxygen species, which are excessively produced resulting from the infection and inflammatory process. Catalase assists to regulate production of cytokines, protect oxidative injury, and repress replication of SARS-CoV-2, as demonstrated in human leukocytes and alveolar epithelial cells, and rhesus macaques, without noticeable toxicity. Such a therapeutic can be readily manufactured at low cost as a potential treatment for COVID-19.

The COVID-19 pandemic has taken a significant toll on people worldwide, and there 23 are currently no specific antivirus drugs or vaccines. We report herein a therapeutic based 24 on catalase, an antioxidant enzyme that can effectively breakdown hydrogen peroxide and 25 minimize the downstream reactive oxygen species, which are excessively produced resulting 26 from the infection and inflammatory process. Catalase assists to regulate production of 27 cytokines, protect oxidative injury, and repress replication of SARS-CoV-2, as demonstrated 28

in human leukocytes and alveolar epithelial cells, and rhesus macaques, without noticeable 29 toxicity. Such a therapeutic can be readily manufactured at low cost as a potential treatment 30

for COVID-19. 31

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in over 32 ten million COVID-19 cases globally. Broad-spectrum antiviral drugs (e.g., nucleoside analogues 33

and HIV-protease inhibitors) are being utilized to attenuate the infection. However, current 34 management is supportive, and without specific antivirus drugs or vaccine against COVID-19(1). 35

While the pathogenesis of COVID-19 remains elusive, accumulating evidence suggests that a 36 subgroup of patients with severe COVID-19 might have cytokine storm syndrome(2, 3). Cytokine 37 storm is a serious immune dysregulation resultant from overproduction of cytokines, which often 38 occurs during virus infection(4), organ transplant(5), immunotherapy(6), and autoimmune 39 diseases (7) , and may result in death if untreated(8). Treatment of hyperinflammation and 40

immunosuppression are highly recommended to address the immediate need to reduce mortality(2). 41

Current immunosuppression options include steroids(9), intravenous immunoglobulin(10), 42 selective cytokine blockade (e.g., anakinra(11) or tocilizumab(12)), and Janus kinase 43 inhibition(13). 44

In light of the findings that elevated levels of reactive oxygen species (ROS) is strongly 45 correlated with inflammation,(14) oxidative injury,(15) as well as viral infection and 46

replication(16-18), we speculate that regulating the ROS level in COVID-19 patients could be 47 effective for the treatment of hyperinflammation, protection of tissues from oxidative injury, and 48

repression of viral replication. As illustrated in Scheme 1A, after infection of SARS-CoV-2, 49

leukocytes are attracted to affected sites releasing cytokines and ROS. An increasing ROS level 50 promotes viral replication, causes oxidative injury, and induces cell apoptosis through DNA 51 damage, lipid peroxidation and protein oxidation, which further exacerbates the immune response. 52

As a result, an increasing number of leukocytes are recruited, further releasing ROS and cytokines, 53

resulting in hyperinflammation and cytokine storm syndrome. 54

ROS are a class of partially reduced metabolites of oxygen that possess strong oxidizing 55 capability, which are generated as byproducts of cellular metabolism through the electron transport 56 chains in mitochondria and cytochrome P450(19). The other major source are oxidases(15) (e.g., 57

NAPDH oxidase), which are ubiquitously present in a variety of cells, particularly phagocytes and 58 endothelial cells. As shown in Scheme 1B, partial reduction of O 2 in these processes generates 59

superoxide anions (·O 2 -), which are rapidly converted to hydrogen peroxide (H 2 O 2 ) mediated by 60 superoxide dismutase (SOD crosslinker. These molecules are enriched around the catalase molecules through noncovalent 80

interactions; subsequent polymerization grows a thin polymeric shell around individual catalase 81 molecules, forming nanocapsules denoted as n(CAT). The thin shell protects the enzyme, while 82

allowing H 2 O 2 to rapidly transport through, endowing n(CAT) with high enzyme activity, 83

augmented stability, and improved plasma half-life. 84

As shown in Fig. 1A, B , n(CAT) shows a size distribution centered at 25 nm and a zeta 85 potential of 1.5 mV, in comparison with those of native catalase (10 nm and -4.0 mV); TEM image 86 confirms that n(CAT) has an average size of 20~30 nm (Fig. 1C) . Compared with native catalase, 87 n(CAT) exhibits a similar enzyme activity ( fig. S1A ), yet with significantly improved enzyme 88 stability. As shown in Fig. 1E , F, n(CAT) and native catalase retain 90% and 52% of the activity 89 after incubation in PBS at 37 °C for 24 h, respectively, indicating improved thermal stability. After 90 incubation in PBS with 50 µg/mL trypsin at 37 °C for 2 h, n(CAT) and native catalase retain 87% 91

and 30% of the activity, respectively, suggesting improved protease stability. In addition, n(CAT) 92

in solution retains 100% of the activity after storage at 4 °C and 25 °C for 3 mo. (fig. S1B) ; after 93 freeze drying, n(CAT) retains more than 90% of the activity ( fig. S1C ). Such characteristics are 94 critical for the transport and distribution of n(CAT). 95

The ability of n(CAT) to protect lung tissues from oxidative injury was examined in human HPAEpiC were cultured with 20 µg/mL of n(CAT) for 12 h, after which 1,000 µM H 2 O 2 was 102 added to the media and cultured for 24 h (Fig. 1F) . The cells without n(CAT) show a cell viability 103 of 63%, while the cells with n(CAT) retain 100% of the cell viability, demonstrating an ability to 104 protect the cells from oxidative injury. In addition, HPAEpiC were incubated with 1000 µM H 2 O 2 105

for 24 h to induce cell injury, after which the injured cells were incubated with 20 µg/mL of n(CAT) 106

for 12 h (Fig. 1G) . Culturing the injured cells with n(CAT) increases the cell viability from 50% 107 to 73%, indicating an ability of n(CAT) to resuscitate injured cells. Similar protective and 108 resuscitative effects were also observed with lower n(CAT) concentrations ( fig. S2B, C) . 19 patients, in whom the severity is strongly correlated to the level of cytokines, such as tumor 117

necrosis factor a (TNF-a) and interleukin 10 (IL-10)(30, 31). Regulating the production of 118 cytokines, in this context, is critical to reinstate immune homeostasis, and anti-cytokine therapy 119 (e.g., TNF-a antagonist) has been suggested for alleviation of hyperinflammation in severe 120 cases(32). 121

In light of these findings, the ability of n(CAT) to regulate cytokine production was studied 122 in human leukocytes (white blood cells, WBC). Leukocytes were cultured with 123 lipopolysaccharides (LPS, a bacterial endotoxin that activates leukocytes) with and without 124 n(CAT). Fig. 1H , I show the concentration of TNFa and IL-10 in the culture media. Culturing 125 the leukocytes with LPS without n(CAT) significantly increases the production of TNF-a and IL-126 10 (P value 0.0001). Moreover, the cultures with n(CAT) show dramatically lower concentrations 127 of TNF-a and IL-10 (P value 0.01 to 0.001), that are comparable with those of the control cells 128

(resting leukocytes). This ex vivo study suggests that n(CAT) can downregulate the production of 129

TNF-a and IL-10 by activated leukocytes, indicating a potential use of n(CAT) as an 130

immunoregulator for hyperinflammation. 131

To further elucidate the immunoregulatory effect, leukocytes were cultured with injured 132

HPAEpiC, of which cell injury was induced by H 2 O 2 (Control #1, cell viability 85%). As shown 133

in Fig. 1J , culturing the cells with leukocytes reduces the viability to 71%. Furthermore, adding 134 8, 16, and 40 µg/mL n(CAT) increases the viability to 82, 89, and 91%, respectively, which are 135 comparable to those of Control #2 (leukocytes with untreated-HPAEpiC, 91% cell viability). This 136 finding indicates that n(CAT) can not only protect, but also resuscitate, the injured alveolar cells, 137

which is consistent with the observation presented in Fig. 1G . Furthermore, HPAEpiC was 138 cultured with leukocytes activated by LPS. As shown in Fig. 1K This study suggests that n(CAT) can also protect healthy alveolar cells from injury by activated 144

leukocytes, indicating an anti-inflammatory effect. 145 (AUC) indicates that the mice that received n(CAT) had a significantly increased body exposure 154

to catalase compared to the mice with native CAT (~ 2.5-fold increase) (Fig. 2D) . The following 155

were all within the normal ranges: the plasma levels of alanine aminotransferase, aspartate receiving n(CAT) exhibit significantly higher fluorescent intensity after 6 h and 48 h (Fig. 2E) , 164

which is confirmed by their fluorescent intensity plot after 48 h (Fig. 2F) . Except the lung, other 165 organs (heart, liver, spleen, and kidney) after 48 h show negligible fluorescent signal, indicating 166 that the as-administered n(CAT) was mainly retained within the lung. H&E stained sections of 167 the main organs do not show any noticeable tissue damage ( fig. S5) . 168

The ability of n(CAT) to repress the replication of SARS-CoV-2 was examined in rhesus 169 macaques. As illustrated in Fig. 3A , at day 0, all of the animals were inoculated with SARS-CoV-170 2 through the intranasal route. For the control group (C1, C2), two animals received 10 mL PBS 171 though inhalation at day 2, 4, and 6, respectively. For the nebulization group, three animals (N1, 172 N2, N3) received 5 mg of n(CAT) (10 mL) through inhalation at day 2, 4, and 6. For the 173 intravenous group, two animals (I1, I2) received 10 mL PBS though inhalation and 5 mg/kg of 174 n(CAT) intravenously at day 2, 4, and 6. Except N3 (sacrificed at day 21), the other animals were 175 sacrificed at day 7. 176 Fig. 3B shows the viral loads in nasal swabs for the control and nebulized group. N1 177 exhibits a viral load that is similar to C1 and C2 at day 1 and 2, after which the viral load rapidly 178 decreases and becomes significantly lower than the control group. N3 shows a similar viral load 179

to the control group at day 1, after which the viral load remains significantly lower than the control 180 group. It is worth noting that the viral load of N3 at day 2 is lower than the control group. 181

Nevertheless, the oral swabs confirmed that N3 was successfully infected, indicating an individual 182 difference ( fig. S6) . N2 shows similar viral loads to the control group from day 1 to 7. Fig. 3C  183 shows viral loads in the nasal swabs for the control and intravenous group. I1 exhibits a similar 184 viral load to the control group at day 1 and 2, after which the viral load rapidly decreases and 185 remains significantly lower than the control group. I2 also shows a similar viral load to the control 186 group at day 1, after which the viral load remains significantly lower than the control group. 187

Similarly, I2 shows a lower viral load than the control group, yet the oral swabs confirmed its 188 active infection. (fig. S10 ). There is no evidence of eosinophilia or vasculitis, 212

and no viral cytopathic effect is identified. The H&E staining of other major organs also shows 213 no tissue injury for both the control and inhaled group ( fig. S11) , confirming the biosafety of 214 n(CAT) administered through intravenous injection or inhalation. In addition, Fig. 4H also  215 presents a representative H&E section (a) and immunohistochemistry for SARS-CoV-2 216 nucleocapsid protein (b) of the lung LN in one animal from the control group (C1). Reactive 217 follicular hyperplasia could be observed in the H&E section, and scattered positive mononuclear 218 cells (black arrows) indicate the SARS-CoV-2 infection in the lymph node. 219

The action mechanism of n(CAT) is unclear. In addition to being a weapon against 220

pathogens, ROS also serve as signaling molecules in numerous physiological processes.(33) For 221 example, it has been documented that H 2 O 2 generation after wounding is required for the 222 recruitment of leukocytes to the wound(34), and ROS is necessary for the release of pro-223 inflammatory cytokines to modulate an appropriate immune response(22). Eliminating the H 2 O 2 224 excessively produced during inflammation also minimizes the downstream ROS, which assists to 225 downregulate production of cytokines, mitigate recruitment of excessive leukocytes, and repress 226

replication of the viruses. It is also worth noting that immunosuppressive steroids, such as 227 prednisone and dexamethasone, are proven to be effective for treatment of hyperinflammation in 228

severe COVID-19 patients(9). Glucocorticoids constitute powerful, broad-spectrum anti-229

inflammatory agents that regulate cytokine production, but their utilization is complicated by an 230 equally broad range of adverse effects(35, 36). For instance, in a retrospective study of 539 231 patients with SARS who received corticosteroid treatment, one-fourth of the patients developed 232

osteonecrosis of the femoral head(37). We speculate that n(CAT) could also regulate cytokine 233 production, but through a different pathway -reinstating immune homeostasis through eliminating 234 excessively produced ROS. 235

In conclusion, we have shown the anti-inflammatory effect and ability of catalase to 236 regulate cytokine production in leukocytes, protect alveolar cells from oxidative injury, and repress 237 the replication of SARS-CoV-2 in rhesus macaques without noticeable toxicity. Moreover, it is 238 worth noting that catalase is safe and commonly used as a food additive and dietary supplement, 239

and that pilot-scale manufacturing of n(CAT) has been successfully demonstrated. In contrast to 240 the current focus on vaccines and antiviral drugs, this may provide an effective therapeutic solution 241

for the pandemic, as well as treatment of hyperinflammation in general. 

Cytokines and autoimmunity

Into the eye of the cytokine storm. Microbiol

A cheap steroid is the first drug shown to reduce death in COVID-19 263 patients

Immunoglobulins in the treatment of COVID-19 infection: Proceed 265 with caution!

Anakinra for severe forms of COVID-19: a cohort study

Tocilizumab in patients with severe COVID-19: a retrospective cohort 269 study

The Janus kinase 1/2 inhibitor ruxolitinib in COVID-19 with severe 271 systemic hyperinflammation

Reactive oxygen species in 273 inflammation and tissue injury

ROS function in redox signaling and oxidative stress

Influenza virus replication in lung epithelial cells depends on redox-277 sensitive pathways activated by NOX4-derived ROS

Inhibition of Nox2 oxidase activity ameliorates influenza A virus-induced 279 lung inflammation

Oxidative stress favours 281 herpes virus infection in vertebrates: a meta-analysis

Reactive oxygen species and 283 antioxidant defense in human gastrointestinal diseases

Free radicals: properties, sources, targets, and 286 their implication in various diseases

Elevated exhalation of hydrogen peroxide and thiobarbituric acid 288 reactive substances in patients with community acquired pneumonia

Beyond oxidative stress: an immunologist's guide to 291 reactive oxygen species

Reactive oxygen species, heat stress and oxidative-induced 293 mitochondrial damage. A review

Catalase delivery for inhibiting ROS-mediated 295 tissue injury and tumor metastasis

PEGylated human catalase elicits potent therapeutic effects on H1N1 297 influenza-induced pneumonia in mice

Phosphorylcholine polymer nanocapsules prolong the circulation time and 300 reduce the immunogenicity of therapeutic proteins

A novel intracellular protein delivery platform based on single-protein 302 nanocapsules SUPPLEMENTARY INFORMATION

Pathological inflammation in patients with COVID-19: a key role 304 for monocytes and macrophages

The trinity of COVID-19: 306 immunity, inflammation and intervention

Correlation Analysis Between Disease Severity and Inflammation-related 308 Parameters in Patients with COVID-19 Pneumonia. medRxiv (2020)

Targeting Cytokine 311 Storm to Manage Patients with COVID-19: A Mini-Review

TNF-alpha antagonist survival rate in a cohort of rheumatoid arthritis 314 patients observed under conditions of standard clinical practice

Reactive oxygen species (ROS) as pleiotropic physiological signalling 317 agents

A tissue-scale gradient of 319 hydrogen peroxide mediates rapid wound detection in zebrafish

Therapeutic glucocorticoids: mechanisms of actions 322 in rheumatic diseases

Caution against corticosteroid-based COVID-19 324 treatment

The influence of age, gender and treatment with steroids on the incidence 326 of osteonecrosis of the femoral head during the management of severe acute respiratory 327 syndrome: a retrospective study

** ****