Viral burden, inflammatory milieu and CD8+ T‐cell responses to influenza virus in a second‐generation thiazolide (RM‐5061) and oseltamivir combination therapy study

Abstract Background Influenza viruses cause significant morbidity and mortality, especially in young children, elderly, pregnant women and individuals with co‐morbidities. Patients with severe influenza disease are typically treated with one neuraminidase inhibitor, oseltamivir or zanamivir. These antivirals need to be taken early to be most effective and often lead to the emergence of drug resistance and/or decreased drug susceptibility. Combining oseltamivir with another antiviral with an alternative mode of action has the potential to improve clinical effectiveness and reduce drug resistance. Methods In this study, we utilized a host‐targeting molecule RM‐5061, a second‐generation thiazolide, in combination with oseltamivir to determine whether these compounds could reduce viral burden and understand their effects on the immune response to influenza virus infection in mice, compared with either monotherapy or placebo. Results The combination of RM‐5061 and OST administered for 5 days after influenza infection reduced viral burden at day 5 post‐infection, when compared to placebo and RM‐5061 monotherapy, but was not significantly different from oseltamivir monotherapy. The inflammatory cytokine milieu was also reduced in animals which received a combination therapy when compared to RM‐5061 and placebo‐treated animals. Antiviral treatment in all groups led to a reduction in CD8+ T‐cell responses in the BAL when compared to placebo. Conclusions To our knowledge, this is the first time a combination of a host‐targeting compound, RM‐5061, and neuraminidase inhibitor, OST, has been tested in vivo. This antiviral combination was safe in mice and led to reduced inflammatory responses following viral infection when compared to untreated animals.


| BACKG ROU N D
Influenza viruses remain a global health burden, with influenza A (IAV) and B (IBV) strains co-circulating and causing an excess 650 000 deaths annually. 1,2 Infants, the elderly, pregnant women, obese and indigenous populations are particularly at risk of severe disease outcomes. Antivirals are used to mitigate the disease and alleviate symptoms caused by influenza viruses. These compounds typically target various aspects of the viral replication cycle. Amantadine and rimantadine target the M2 ion channel of the influenza virus disrupting viral uncoating following entry into the host cell, 3 while the neuraminidase inhibitors (NAIs), with oseltamivir (OST) being the most frequently prescribed, inhibit the release of newly formed virions. 4 More recently a new influenza antiviral, baloxavir marboxil, which inhibits the polymerase acidic (PA) endonuclease, stopping cleavage of mRNA during the cap snatching process, has been licensed for use in otherwise healthy and high-risk individuals in Japan, the United States and other countries. 5 These antiviral drugs directly target the influenza virus, and therefore drug-resistant viruses can emerge.
This has been observed most dramatically with the widespread circulation of IAVs resistant to the adamantanes, and in 2008-2009 the global circulation of oseltamivir-resistant H1N1 viruses. 6 Currently, the frequency of A/(H1N1)pdm09 viruses with reduced susceptibility to OST is between 1% and 3%. 7,8 An alternate approach, and one which is successful in HIV and malaria, is to use a combination of two or more drugs. Combination therapy can enable rapid viral clearance facilitating better clinical outcomes and reducing the likelihood of the emergence of viral resistance. These compounds can target host factors or viral proteins, as exemplified by the combination of NAIs and M2e ion channel blockers reducing viral burden by ~100% in vitro. 9 In mice, a 1000fold reduction was observed in viral burden in animals treated with OST and rimantadine when compared to placebo-treated animals. 10 Previously, it has been shown that tizoxanide, the active metabolite of the anti-parasitic drug nitazoxanide [11][12][13] (NTZ), was active against circulating influenza viruses, including those resistant to NAIs. 14 Nitazoxanide, a first-generation thiazolide, was licensed for use in the treatment of gut parasitic infections in the 1970s, but more recently has been repurposed to combat viral infections. Nitazoxanide can alter mitochondrial respiration 15 and inhibit terminal hemagglutinin glycosylation in the endoplasmic reticulum thereby preventing viral exit. 16 In the ferret model, we have shown that the combination of OST and NTZ reduced the incidence of infection and that if infected, virus remained in the upper respiratory tract and did not enter the lungs which is typically associated with severe disease. 17 In this study, we investigated the effects of RM-5061, a second-generation thiazolide, developed to improve systemic absorption, 18 in combination with OST on reducing viral burden and subsequent effects on the immune response to influenza virus infection.
We hypothesized that the combination of RM-5061 + OST would have a greater impact on reducing viral burden and potentially altering immune responses to influenza virus infections via targeting different mechanisms of action in the host and viral replication cycle.
Our results show that, the overall cytokine response was lower in animals undergoing RM-5061 combination therapy when compared to glucose and RM-5061 monotherapy but was comparable with OST-treated animals. Antiviral treatment in all the groups led to a significant reduction in the influenza-specific CD8 + T-cell responses in the bronchioalveolar lavage fluid (BAL). Overall, in this study, we show that the combination of RM-5061 + OST is as effective as OST monotherapy at reducing viral burden and inflammatory response in severe influenza disease.

| Mice
C57Bl/6 mice were bred and housed in the Biological Research

| Viral infection and harvesting of tissues
Mice were anaesthetized with 2% of isoflurane in oxygen and infected intranasally with 1 x 10 4 pfu of the laboratory-adapted strain of IAV A/Hong Kong-x31 (A/HKx31, H3N2) in 30 µL PBS. Mice were culled by CO 2 asphyxiation on day 5 (d5) or day 7 (d7) post-infection (pi) for immune analyses. BAL was performed by washing lungs of IAV-infected mice with 1 mL HBSS. Lungs were perfused by cutting the pulmonary artery and injecting 5 mL of PBS into the right atrium of the heart before harvest. Single cell solutions from mLN and spleen were prepared by meshing organs through a 70 µm cell strainer. Lungs were processed with a tissue homogenizer, cell debris was spun down and supernatants were stored at −80°C until analysis for cytokine amounts and viral titres.

| Drug treatment
Mice were treated twice daily via oral gavage with 3 mg of RM-5061, 1 mg of Ost, or 3 mg of RM + 5061 + 1 mg of Ost in 180 µL of 50% Glucose or 50% Glucose only per dose.

| Plaque assays
Plaque assays were performed in MDCK cell lines by seeding 1.2 x 10 6 cells into 6-well plates in 3 mL of MDCK medium and rested overnight at 37°C and 5% CO 2 . Lung homogenates from infected mice were diluted in 10x serial dilution and incubated with MDCK cell monolayers for 1 hour at 37°C for virus attachment. Agarose overlay medium was then added and plaques for plaque-forming units (pfu) per lung were counted after 3 days.

| Cytokine bead array
Cytokine concentrations in supernatants from lung homogenates were quantified using the mouse BD Bioscience cytometric bead array following the manufacturer's instructions. Data were acquired using FACSCantoII and analysed using the flowjo software vs10 (Tree Star) and Prism 7 (graphpad software).

| OST monotherapy and OST + RM-5061 combination reduced body weight loss in mice following A/HKx31 influenza virus infection
Mice were treated twice daily with either RM-5061 (3.4 mg), OST (1 mg), RM-5061 (3.4 mg) + OST (1 mg) or 50% Glucose only (placebo). The first dose was administered 2h prior to intranasal infection with 1 x 10 4 pfu A/HKx31. Mice were monitored daily following viral infections, and animals were euthanized at d5 to d7 p.i.to determine viral burden ( Figure 1A). Animals treated with either OST or RM-5061 + OST maintained their overall body weight following infection when compared to the RM-5061 and placebo groups ( Figure 1B).
Animals receiving RM-5061 + OST and OST (mean viral titre 4.1 and 4.1 log 10 pfu/lung, respectively) had significantly reduced viral titres in comparison to placebo and RM-5061 monotherapy (mean viral titres 5.5 and 5.6 log 10 pfu/lung, respectively; Figure 1C). By d7, no differences in viral titres were evident between all the treatment groups, suggesting an early effect of RM-5061 + OST and OST either on viral clearance, or by directly suppressing viral replication when antiviral treatment was administered for up to d5 p.i.. Despite observing a reduction in viral burden at d5 p.i., we found no differences in the levels of eosinophils, DCs and macrophages in the BAL (Figure 2A,C). However, we identified significantly lower frequencies of neutrophils in RM-5061 + OST (mean 6.8%) and OST (mean 2.5%)-treated mice compared with placebo (mean 13.9%) and/or RM-5061 monotherapy (mean 9.5%) F I G U R E 2 Reduced NK cells and neutrophils in RM-5061 + OST-treated and OST-treated animals. Innate immune responses across different drug treatments and A/HKx31 infection were determined by staining for eosinophils, neutrophils, DCs, NK cells and macrophages in BAL and spleen. Frequencies at (A) d5 and (B) d7 as well as (C) total numbers of the cell subsets were graphed. Data are from two independent experiments with 5 mice per treatment group. One-way ANOVA was performed between groups (*P ≤ .05; **P ≤ .01, ***P ≤ .001) groups in BAL on d5 p.i. (Figure 2A) but not on d7 p.i. (Figure 2B).

| Reduced NK cells and neutrophils in OST and RM-5061 + OST-treated animals
Although not significant, RM-5061 monotherapy reduced the levels of neutrophils by 4.4% in comparison to placebo (Figure 2A).
The percentages of NK cells were significantly reduced on d5 in RM-5061 + OST-treated mice (mean 1.1%) and OST monotherapy (mean 1.1%) groups compared with RM-5061 monotherapy (mean 5.5%)-treated mice (Figure 2A). There was an overall trend for animals treated with RM-5061 + OST to have lower total numbers of innate immune cells in BAL and spleen on d5 p.i. when compared to other treatment groups ( Figure 2C); however, this result was not statistically significant.
On d5 pi, we found significantly elevated levels of NK cells in spleens of the OST group (mean 3.9 x 10 5 ) and on d7 pi for the RM-5061 + OST group (mean 4.9 x 10 5 ) when compared to placebo (d5, 1.2 x 10 5 ; d7, 1.6 x 10 5 ) and RM-5061 (d5, 1.4 x 10 5 ; d7, 1.5 x 10 5 )-treated animals ( Figure 2C). Trends in the different innate cell numbers between animals receiving either placebo or the combination RM-5061 + OST were seen in the BAL on d5 p.i. (Figure 2C), suggesting a potential early effect of RM-5061 + OST and OST therapies on innate immune responses during influenza infection.

| Reduced RANTES in lungs of RM-5061 + OST and OST-treated mice during the course of influenza virus infection
Given the important role inflammatory cytokines and chemokines play in both influenza-induced morbidity and mortality 19 and recovery from influenza virus infection, 20 Figure 3C). In comparison with placebo animals (mean 2451 pg/mL), the levels of RANTES on d7 pi. were significantly reduced in RM-5061 + OST-treated mice (mean 1331 pg/mL). Overall, combination RM-5061 + OST or OST monotherapy leads to a significant reduction in the amounts of cytokines and chemokines in the lung.  Figure 4A).

| Reduced CD8 + T-cell numbers in BAL of RM-5061 and/or OST-treated mice on d7 p.i
There was a trend for RM-5061 + OST-treated animals to have lower percentages ( Figure 4B) and total number of cells ( Figure 4C) in the BAL on d5 p.i. compared with placebo. A significant difference was observed in the combination group on d7 p.i. (Figure 4B,C), with additionally RM-5601 and OST alone showing significant lower CD8 + T-cell numbers in the BAL on d7 p.i. (Figure 4C), whereas this effect was not present in the spleen on d7 pi. (Figure 4C). This late effect of OST or RM-5061 + OST combination therapy was also seen for NKT cells, with reduced NK T-cell percentages in the BAL of OST and RM-5061 + OST groups on d7 p.i. (Figure 4B). By d7, no differences in the draining mLN were detected in total CD8 + T, NKT and γδ T cells amongst all treatment groups (data not shown).

| D ISCUSS I ON
In this study, we investigated the effects of a combination treatment of RM-5061 + OST on influenza viral replication and the subsequent effects on the influenza-specific immune responses. We found that the combination therapy was as effective as OST monotherapy in reducing viral load, inflammation and innate immunity on d5 at the site of infection, and consequently, antigen-specific CD8 + T-cell responses. RM-5061 + OST and OST-treated animals had an overall reduction in the amount of cytokines present in the lungs with RM-5061 + OST-treated animals having significantly reduced levels of RANTES. We also found a reduction in the numbers of influenzaspecific CD8 + T-cell responses in all treatment groups. To our knowledge, this is the first time these two antiviral compounds have been tested in combination in vivo.
RM-5061 is a second-generation thiazolide that targets host-specific factors inhibiting hemagglutinin maturation via the endoplasmic reticulum protein ERp57. 28 First-generation thiazolides, such as nitazoxanide, were licensed to treat parasitic infections and more recently repurposed for the treatment of viral respiratory infections. 14,28 Second-generation thiazolides were developed for better systemic absorption to increase antiviral activity, 28 therefore, initial studies with the second-generation thiazolide RM-5061 showed up to sevenfold higher blood concentration and improved bioavailability in rats, compared to its progenitor compounds. 18 Toxicity studies have been completed in dogs and rats where up to 75 mg/kg/d or 25 mg/kg/d once daily for 28 consecutive days with no adverse effects identified. 28 Similarly, in this study, we found that the administration of 3 mg/kg/d in mice was safely tolerated in these animals.
Combination therapy is often considered as an approach to enhance virologic activity and reduce the incidence of resistance.
In this study, we used a neuraminidase inhibitor, OST and a host Interestingly, we did not find an overall reduction in the amount of cytokines present in the lungs of treated animals but a significant reduction was observed in the levels of RANTES in RM-5061 + OST at d5 and d7 p.i.. RANTES is a key innate cytokine produced by infected epithelial cells, T cells and other immune cells 29 and results in the recruitment and activation of T cells, granulocytes and macrophages the lung.
In our study, the numbers of influenza-specific CD8 + T cells in mice treated with antivirals were significantly lower than those in placebo-treated animals. It has been previously reported that OST treatment decreases the magnitude of the influenza-specific immune response generated during the acute phase of viral infection but that these cells were capable of mounting a robust recall response upon secondary infection with a heterologous strain of the influenza virus. 30 Overall, our study provides evidence that a combination therapy with host targeting compound, RM-5061 and neuraminidase inhib-

CO N FLI C T O F I NTE R E S T S
ACH is an employee of Hoffmann La-Roche. Romark Laboratories provided funds for research.