Deciphering the antiviral potential of tomatidine towards mosquito-borne viral infections
Troost-Kind, Berit
DOI:
10.33612/diss.161786279
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2021
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Troost-Kind, B. (2021). Deciphering the antiviral potential of tomatidine towards mosquito-borne viral infections. University of Groningen. https://doi.org/10.33612/diss.161786279
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Chapter
7
Summarizing discussion and
In recent decades, several arthropod-borne viruses (arboviruses) have re-emerged and caused millions of infections worldwide1,2. Among these viruses are mosquito-borne viruses such as
dengue virus (DENV), West Nile virus (WNV), chikungunya virus (CHIKV) and Zika virus (ZIKV). DENV was first isolated in 1943 in Japan. To date, DENV is the most common arboviral infection worldwide and is estimated to infect 390 million individuals every year3.
Infected individuals can either experience an asymptomatic infection or develop dengue fever, a self-limiting febrile illness with symptoms such as high fever, rash, severe headache as well as muscle and bone pain4–6. Each year an estimated 96 million individuals experience dengue
fever. In approximately 0.5 to 1 million cases, dengue fever progresses to potentially fatal complications (severe dengue), which are characterized by capillary leakage, pleural effusion, severe bleeding and organ impairment7.
WNV was first isolated in 1937 in the West Nile district of Uganda and has globally expanded within the last three decades causing thousands of infections8. WNV infection causes a
self-limiting dengue fever-like illness in approximately 20% of the infected individuals. In approximately 1% of symptomatic infections, the virus causes inflammation of the central nervous system, leading to various neurological symptoms which can be fatal8. WNV infection
is the most important cause of viral encephalitis in the USA9. CHIKV was first identified
in Tanzania between 1952 and 1953 during an outbreak of febrile polyarthralgia10,11. Within
the last two decades the virus has re-emerged and caused large outbreaks in Africa, Asia and the Americas involving millions of infections. The infection manifests as symptomatic disease in 85% of cases with symptoms such as fever, polyarthralgia, myalgia, rash, headaches and nausea6. Importantly, in 30 to 40% of symptomatic infections, the disease progresses to
chronic polyarthralgia, which can persist for months to years after the infection12. The most
recently re-emerged arboviral infection that became clinically important is ZIKV. ZIKV was originally isolated from a Nigerian woman 195413. Since the first major outbreak on the
Yap Island in 2007, various outbreaks have been globally reported again involving millions of infections. The acute symptoms of ZIKV resemble those of the other mosquito-borne infections. However, the main complication of this infection is the risk for microcephaly and other brain malformations in the fetus of pregnant women and the development of Guillain-Barre syndrome in adults13.
Arboviral infections cause a substantial burden on human health and have large economic consequences for the affected countries mainly due to the lack of effective preventative or therapeutic options. Indeed, a study from 2011 estimated the economic burden of dengue disease in the Americas to be US$2.1 billion per year14. Likewise, the costs of the CHIKV
outbreak in 2014/2015 in the U.S. Virgin islands were estimated between US$14 and US$33 million15. These calculations include factors such as missed days at work and direct healthcare
costs. Whereas these cost estimations provide a general idea of the economic impact of the diseases, it is believed that due to the difficulties in diagnosing these infections and the underreporting, their true financial burden is significantly higher16.
The high burden of disease, large economic consequences and expanding geographical spread of arboviral infections highlight the need to develop efficient strategies to combat infection and disease. In theory, there are two main antiviral treatment strategies to alleviate disease symptoms and progression. First, an antiviral treatment can be prescribed prophylactically in outbreak situations. This strategy has already been applied before during Influenza virus outbreaks17. Prophylactic treatment likely maximises the antiviral effect however toxic
7
be applied therapeutically upon diagnosis. In case of therapeutic drugs towards DENV and CHIKV, treatment should ideally start during the early phase of infection, i.e. within 48 h after the onset of symptoms to quickly control and reduce peak viremia and the associated risk to develop severe or chronic disease18–20. Early therapeutic treatment has been proven
successful for other viruses such as the use of oseltamivir to treat influenza infection21. For
CHIKV, it might be possible to still see beneficial antiviral effects when treatment is started later in infection as it may still help to completely clear the virus to alleviate or even prevent chronic disease symptoms20,22. Thus far, only few antiviral drug candidates, the majority
towards DENV, have been tested in clinical trials and none of them showed a beneficial effect23–30.
In this thesis, we investigated whether the plant-derived, steroidal alkaloid tomatidine has antiviral potential towards the flaviviruses DENV, WNV, ZIKV and the alphavirus CHIKV. We provided in vitro evidence that tomatidine exhibits antiviral activity towards all four DENV serotypes, ZIKV, CHIKV but not WNV. Furthermore, to gain insights into the mode of action of tomatidine, we determined which host factors and viral factors are influenced by tomatidine and which structural regions of the compound are important for its antiviral activity towards DENV and CHIKV. Lastly, the in vivo potency of tomatidine towards DENV was studied in a non-lethal mouse model. Below, I will summarize and discuss the individual findings of this thesis including future perspectives under the following sections:
I. Tomatidine as an antiviral drug: Its potential and next steps II. Antiviral treatment for DENV and CHIKV: How close are we?
Tomatidine as an antiviral drug: Its potential and next steps
Lessons learned on the antiviral mechanism of tomatidine
In chapter 3, we explored the antiviral activity of tomatidine towards re-emerging clinically relevant flavivirus infections such as DENV, WNV and ZIKV. We demonstrated that tomatidine has a potent antiviral effect towards all four DENV serotypes. The tomatidine concentration that reduced virus particle production with 50% (EC50) was 2.08 µM for DENV-1, 0.82 µM for DENV-2, 4.87 µM for DENV-3 and 2.50 µM for DENV-4 in human hepatic Huh7 cells following infection at multiplicity of infection (MOI) 1. Moreover, a significant antiviral effect of tomatidine was observed towards ZIKV. Tomatidine did not exhibit antiviral activity towards WNV. Time-of-addition studies identified that tomatidine may act at multiple stages of the DENV replication cycle. Since potent antiviral activity is still observed when tomatidine is added up to 12 hours post-infection (hpi), i.e. long after virus attachment and entry into the host cell, the main antiviral target of tomatidine is likely after virus cell entry. This hypothesis is supported by the finding that no difference is observed in the number of DENV-infected cells when tomatidine is added at the point of infection or 2 at hpi, i.e. when cell entry has already occurred. Tomatidine was earlier described to inhibit the expression of activating transcription factor 4 (ATF4), which is an important regulator of cellular homeostasis upon stress and was reported to translocate to the nucleus in DENV-infected cells. Therefore, we explored whether the antiviral activity of tomatidine is related to ATF4. ATF4 knockdown in Huh7 cells led to a 2-fold reduction in the antiviral
potency of tomatidine, indicating that ATF4 might indeed be partly involved in the antiviral activity of tomatidine. Nevertheless, ATF4 is not the sole factor involved as more than 10-fold reduction in virus particle production is seen in presence of tomatidine. In conclusion, we identified tomatidine as a potent inhibitor of DENV replication in vitro.
In chapter 4, we evaluated the antiviral potential of tomatidine towards CHIKV. We found potent antiviral activity towards different genotypes of CHIKV with EC50 values of 1.3 µM for an East/Central/South African genotype, 2 µM for an Asian genotype and 3.8 µM for a West African genotype following infection of Huh7 cells at MOI 1. Similar to DENV, time-of-addition studies in CHIKV-infected Huh7 cells revealed that tomatidine mainly acts at post-entry stages of the CHIKV replication cycle. We also examined how long the antiviral activity could be sustained in Huh7 cells. We found that tomatidine controls CHIKV infection for up to three replication rounds without replenishment of the compound. We also evaluated the antiviral activity of commercially available structural derivatives of tomatidine to identify the structural region of the compound responsible for its antiviral effect. The structural derivative solasodine contains an additional double-bond in the steroidal ring structure and sarsasapogenin lacks the basic nitrogen of the spiroaminoketal group in tomatidine. Both derivatives showed efficient, albeit less potent antiviral activity towards CHIKV, suggesting that the basic nitrogen and the steroidal ring structure of tomatidine may be relevant for its antiviral activity. Altogether, this in vitro study identifies tomatidine as a promising antiviral drug candidate towards CHIKV.
In chapter 5, we further evaluated the underlying antiviral mechanism of tomatidine towards CHIKV. DENV was not included due to time constraints. We identified four cellular proteins, p62, C98, metallothionein and thioredoxin-related transmembrane protein 2, which were differentially expressed in the presence of tomatidine. Two proteins, p62 and CD98 were successfully validated via western blot analysis. However, subsequent knockdown studies indicated that these proteins are not involved in the antiviral mechanism of tomatidine. Furthermore, transfection of in vitro transcribed viral RNA into Huh7 cells treated with tomatidine confirmed that tomatidine mainly acts after virus cell entry. Tomatidine was found to reduce CHIKV replication in a trans-replicase system and reduced intracellular CHIKV RNA expression levels at 6 and 8 hpi. No effect on viral RNA replication was seen when tomatidine was added at 6 hpi, i.e. when RNA replication has been already initiated. In line with the inhibitory effect on intracellular viral RNA expression, a significant reduction in the expression of the CHIKV proteins capsid, Envelope-1 and non-structural protein (NS) 2 was seen. Taken together, this study shows that tomatidine predominantly acts after viral cell entry but prior to active RNA replication. More specifically, we postulate that tomatidine interferes with efficient viral RNA translation and/or hampers the early stages of RNA replication i.e. the formation of the RNA replication complex. Moreover, CHIKV did not develop resistance towards tomatidine after a maximum of 15 passages, indicating that tomatidine has a high genetic barrier to resistance.
Further studies are required to decipher the exact molecular mechanisms by which tomatidine exerts its antiviral effect towards CHIKV and DENV. The initial stages of RNA translation and replication and the host factors involved in these processes are currently poorly understood. Based on our findings, it is possible that the antiviral mechanism of tomatidine is comparable for both viruses. We have observed a similar pattern in the time-of-addition study, which showed that tomatidine predominantly acts after virus cell entry. Furthermore, a reduction in the number of infected cells was detected for both viruses. However, to verify whether DENV
7
and CHIKV share the antiviral mechanism of tomatidine, experiments similar to those conducted for CHIKV in chapter 5 should be performed for DENV. To further delineate the mechanism involved, future studies should focus on the factors involved in the process of RNA translation and replication. To confirm that tomatidine directly interferes with viral RNA translation, tomatidine should be added to infected cells at 6 hpi (i.e. at a condition in which tomatidine no longer interferes with viral RNA levels) and the viral protein expression levels should be assessed at 9 hpi. Although it is technically challenging to study viral RNA translation early in infection due to sensitivity issues, novel methods such as single-molecule sensitivity fluorescence in situ hybridization analysis may be used to track single viral RNA molecules within the cell. This may help to determine whether the initial viral RNA trafficking to the ribosomes is affected by tomatidine31,32. Another rather novel approach to determine
the global translational state of a cell by detecting ribosome locations on cellular RNA is ribosome profiling33. This approach has been previously used to investigate the translational
landscape of DENV and ZIKV and may be applied to test whether tomatidine affects translation by preventing the binding of the ribosomes to the viral RNA34,35. To determine the
potential effect of tomatidine on replication, a double-stranded RNA (dsRNA) staining could be performed to see whether dsRNA intermediates are formed in the presence of tomatidine. Electron microscopy could be applied to test whether the virus-specific formation of dsRNA-containing spherules still occurs in the presence of tomatidine36,37. These experiments may
help to decipher the antiviral mechanism of tomatidine in the future.
In chapter 5, we only found four differentially expressed proteins in the presence of tomatidine and subsequent experiments revealed that the two validated proteins are not involved in the antiviral activity of tomatidine towards CHIKV. The inability to identify host factor associated with the antiviral activity of tomatidine, may be due to our choice of technical approach. The unlabelled mass spectrometry method is particularly useful for the detection of major proteomic changes. Thus, it is possible that we missed minor but relevant alterations in the proteome upon tomatidine treatment. A more sensitive mass spectrometry approach such as SILAC labelling, could be applied to identify smaller tomatidine-induced proteomic changes38. Alternatively, tomatidine may influence protein functionality rather than its
expression. To test this, metabolomic studies may be applied to identify potential tomatidine-induced changes in protein function. Importantly, we cannot rule out the possibility that the antiviral effect of tomatidine could stem from its ability to modulate the cellular lipids rather than proteins. Tomatidine has indeed been shown to alleviate hyperlipidemia in the context of atherosclerosis in vivo39. Lipids play an important role in DENV and CHIKV infectivity,
translation, replication and morphogenesis40,41. Thus, to investigate whether tomatidine
modulates the cellular lipid metabolism, lipidomic studies may be applied.
Tomatidine has been proposed to exhibit various health benefitting activities including anti-carcinogenic, cardioprotective and anti-atherosclerotic activity as well as the ability to counteract age-related muscle atrophy and weakness39,42–45. Tomatidine has been studied
most extensively for its anti-inflammatory properties in vitro and in vivo46–48. As an example,
tomatidine was described to supress the NF-kB and JNK pathway in murine macrophages46.
Moreover, tomatidine reduced airway hyperresponsiveness and lung infiltration of eosinophiles in an asthma mouse model48. The ability of tomatidine to modulate the cellular inflammatory
response may also be involved in the antiviral mechanism of tomatidine. This however has not been investigated in this thesis. Therefore, it may also be interesting for future studies to investigate the role of the NF-kB and JNK pathway in the antiviral activity of tomatidine.
Altogether, these studies will not only shed light on the mode of action of tomatidine but may also reveal new insights in the replicative cycle of these viruses.
Lessons learned on the in vivo potency of tomatidine
The strong antiviral potential of tomatidine towards DENV, ZIKV and CHIKV in vitro, prompted us to evaluate whether tomatidine is able to exert antiviral activity in vivo. We first chose to evaluate the in vivo potency towards DENV due to the availability of the non-lethal AG129 DENV mouse model via collaboration. In chapter 6, the first pilot experiment is presented. Tomatidine treatment (50 mg/kg) was initiated one day prior to DENV infection and was given in one dose of 50 mg/kg every 24 h or in two doses of 25 mg/kg every 12 h for the duration of the experiment. The vehicle control group received the solvent (corn oil) only. The mice were inoculated with DENV via intraperitoneal injection with 102 PFU/mL.
Samples of plasma, spleen, kidney and liver were taken 5 days post-infection and the viral RNA copy numbers were determined. No difference in the viral RNA load was observed in plasma, spleen, kidney and liver between the vehicle control and the tomatidine-treated groups. Subsequent quantification of the tomatidine concentration in the plasma samples of tomatidine-treated mice indicated an average concentration of 59 ng/mL or 0.13 µM. This plasma concentration is approximately 6 times lower than the in vitro determined EC50 value of tomatidine in Huh7 cells, which was 0.82 µM for DENV-2. Whereas it is not possible to directly correlate the in vitro determined EC50 value with the effective in vivo dose of a drug, it is possible that the tomatidine plasma concentration in the mice is too low to exert an antiviral effect. Future research is needed to prove or disprove the antiviral potential of tomatidine towards DENV and CHIKV in vivo.
The most commonly used DENV models to evaluate antiviral drug candidates in vivo are the lethal and non-lethal AG129 mouse models49. AG129 mice are deficient in their interferon
I and II receptors and thus facilitate high viremia as well as the development of symptoms such as thrombocytopenia, vascular leakage and, depending on the model, death. Whether or not the AG129 mouse model is lethal mainly depends on the used viral dose and the virus strain49. In humans, antiviral treatment can be given prophylactically during outbreak
situations or therapeutically early after diagnosis i.e. after disease onset. When studying these two treatment approaches in vivo, the choice of model is important. To study the prophylactic effect of an antiviral drug, both, the lethal and non-lethal model are suitable. In this case, treatment should be initiated prior to infection. The advantage of the lethal model is the presence of high viremia and severe disease manifestations in the mice49. The advantage of
the non-lethal model is that it allows to determine the antiviral effects on the time to resolve viremia, which is an important aspect to study regarding human infection50. In our study,
we used the non-lethal AG129 model to investigate the prophylactic antiviral potential of tomatidine in vivo. To evaluate the antiviral activity of a lead compound under therapeutic conditions, it is important to consider that in human infection the onset of symptoms starts when viremia reaches it plateau51. Therefore, it is of particular importance to investigate
the time to resolve viremia and antiviral treatment should be initiated after peak viremia has been established. To study the antiviral activity under therapeutic conditions, the non-lethal model is most suitable as it allows to study the time to resolve viremia. This cannot be investigated using the lethal model as the mice succumb to infection quickly after peak viremia is established49. For future studies on tomatidine, it would also be interesting to test
7
DENV-induced symptoms in mice46,48. To do so, the lethal AG129 mouse model would be
most suitable as the mice develop DENV-related symptoms.
We have not yet been able to test the in vivo potency of tomatidine towards CHIKV. The CHIKV literature suggests a correlation between viral load and chronic disease onset20. Thus,
also here it is of importance to test the direct early effect of tomatidine on the peak viremia. Due to the unique pathology of CHIKV on the joints it is also interesting to investigate the antiviral effect on joint swelling, histological damage and inflammation. Currently, the most commonly used model to investigate antiviral treatments for CHIKV in vivo is the C57BL/6 arthritis mouse model52. The advantage of this model is that the used mice are
immunocompetent, which allows to investigate antiviral treatments in the presence of an intact interferon response and is therefore more clinically relevant than immunocompromised mouse models. Moreover, CHIKV infection via the footpad induces food-pat inflammation and swelling, which peaks around 3 and 7 days post-infection. Joint pain and inflammation is a major hallmark of human CHIKV infection. Hence, this model allows for investigation of the virus titer and the CHIKV-associated joint inflammation52. I therefore propose to use this
model to study the antiviral activity of tomatidine towards CHIKV in vivo.
Given the reported anti-inflammatory and anti-atherosclerotic effects of tomatidine and its protective effect on muscle atrophy, it would also be interesting to examine its potential to interfere with chronic CHIKV infection39,46,48. The mechanism of chronic chikungunya fever
is still not very well understood. It has been demonstrated that CHIKV RNA and antigen can be detected in synovial tissue samples of chronic chikungunya fever patients and CHIKV antigen was detected in perivascular macrophages20. Moreover, CHIKV antigen was detected
in the muscle satellite cells of a chronic chikungunya fever patient22. So far, however, attempts
to isolate infectious virus from these tissues were unsuccessful. Therefore, it is not known whether CHIKV persists in the tissues by active replication. The best way to investigate the beneficial effect of tomatidine on chronic chikungunya fever would be a chronic/persistent model52. An example is the chronic C57BL/6 mouse model where mice are infected with a
clinical CHIKV isolate via the footpad53. This model is based on the above described arthritis
model and would therefore allow to test the antiviral and anti-inflammatory activity of an antiviral treatment at the same time. After an initial systemic viral infection, the virus is cleared from most tissues of the mouse by week 4 of infection. Intriguingly, however, follow-up testing revealed that CHIKV RNA can still be detected 16 weeks post-infection in joint-associated tissues alongside a chronic inflammation of the joints53. Thus, this model may shed
light on the potential protective effect of tomatidine on chronic CHIKV infection in vivo. Given the relatively low plasma concentration of tomatidine in mice, another aspect to consider in follow-up experiments is the route of administration. Common routes of administration for tomatidine are food supplementation, oral gavage and intraperitoneal injection45,48. In chapter
6, we used oral gavage and after 6 days of treatment (50 mg/kg daily) we measured 59 ng/mL
tomatidine in the plasma. A report by Ebert et al. demonstrated that food supplementation of tomatidine to the standard chow (0.05% tomatidine) for 2 months resulted in a tomatidine plasma concentration of 287 ng/mL45. This shows that higher plasma concentrations of
tomatidine can be reached via this experimental set-up. However, this concentration still remains below the in vitro determined EC50 value of tomatidine (0.82 µM) for DENV-254.
Thus, oral administration of tomatidine via supplementation of the standard chow or via oral gavage dissolved in corn oil may not lead to a sufficient plasma concentration of tomatidine to exert its antiviral activity. The low bioavailability may be explained by the hydrophobic nature
of tomatidine. Therefore, alternative delivery strategies such as lipid nanoparticles could also be explored. This was already done by Dorsaz et al, who investigated the anti-fungal activity of tomatidine in vivo55. After initial unsuccessful attempts to demonstrate the anti-fungal
potential of tomatidine in vivo using different co-solvents, the authors hypothesised that delivery of tomatidine via tomatidine-containing nanoparticles may increase bioavailability. Indeed, with this new delivery approach, a significant reduction in fungal burden in the kidney of tomatidine-treated BALB/c mice was observed. Hence, an alternative delivery strategy of tomatidine may help to enhance its bioavailability.
Next to the delivery strategy, the treatment regime might also further enhance the potency of tomatidine. In this context, pharmacokinetic studies may help to gain a better understanding of the pharmacokinetic profile, stability and toxicity of tomatidine.
Pharmacokinetic studies are important in antiviral research as they help to understand how and in which time-frame the compound is processed in the body. Thus, it gives information on the biodistribution of a compound and how often it needs to be administered in order to accumulate in the body. The plasma level of tomatidine has only been determined in one study (287 ng/mL) after 2 month of treatment supplemented to the standard chow45. This
demonstrates that tomatidine may accumulate in mouse plasma via oral treatment but does not allow for any direct conclusions on its biodistribution. In our durability experiments in
chapter 3, we observed an anti-CHIKV effect for up to 24 hpi without replenishment of
tomatidine. This may indicate that the compound is stable up to 24 h. This however needs to be further validated in vivo. Based on the scarce literature on the pharmacokinetics of tomatidine, in vivo studies are needed to enhance our knowledge regarding the stability, absorption, distribution, metabolism and the elimination of tomatidine from the mouse body. This can be tested by measuring the tomatidine concentration in blood, plasma and urine at different time points after administration56,57.
On top of this, it is important to consider the safety profile of tomatidine. To the best of my knowledge, two studies, both from the Western Regional Research Center in California, investigated the toxicity profile of tomatidine in pregnant and non-pregnant adult female Swiss-Weber mice using various parameters including increased liver weight (hepatomegaly) as readout58,59. Tomatidine was administered by supplementation to the food
at a concentration of 2.4 mmol/kg for 7 or 14 days and showed a promising safety profile in non-pregnant and pregnant mice. These preliminary toxicity studies suggest a promising and safe toxicity profile for tomatidine. Nevertheless, studies in different mouse strains such as the immunocompromised AG129 mice or C57BL/6 mice and with a greater concentration range would be beneficial to further assess the safety of tomatidine. Given the many reported biological activities of tomatidine, it may also be beneficial to include other readouts such as histological and functional analysis of different organs and alterations in body weight and composition (muscle to fat ratio).
Based on its properties as a steroid, tomatidine is already freely available online as a food supplement to enhance muscle growth. Here a maximum intake of 300 mg tomatidine per day is recommended for the food supplement ‘titan’. It is important to note that the safety and maximal safe dosage of tomatidine in humans have not yet been studied. Therefore, safety studies in humans are required to ensure a safe application of the compound in humans. Altogether, adjusting the delivery strategy as well as optimizing the treatment regime by pharmacokinetic studies should allow us to properly evaluate the in vivo antiviral potential of
7
tomatidine using the above described mouse models.
Antiviral treatment for DENV and CHIKV: How close are we?
The drastic re-emergence of DENV and CHIKV in the last decades has large consequences on the overall human health and economic status of the affected countries14,15. However,
despite increasing efforts in scientific research and the numerous antiviral compounds that have been identified in vitro, there is still no antiviral compound available to treat these infections.
Thus far, most antiviral drug candidates that have been tested in clinical trials are repurposed drugs, which require lower costs and less time for clinical studies19,60. Among them are
chloroquine, prednisolone and celgosivir, which are approved for other medical conditions or infectious diseases such as malaria, inflammatory diseases or diabetes60. Unfortunately,
none of the tested antivirals showed sufficient antiviral activity in clinical trials. Therefore, the search for effective antiviral treatments towards these viruses continues and various promising drug candidates are in pre-clinical and clinical development.
A repurposed lead compound for the treatment of DENV is celgosivir, a drug originally developed to treat diabetes60. Celgosivir did not show any clear antiviral activity in clinical
trials in DENV patients treated during peak viremia, i.e. less than 48 h after fever onset26.
However, a follow-up in vivo study found celgosivir to be effective in a non-lethal AG129 mouse model when treatment started during peak viremia using a different treatment regime61.
Thus, a new clinical study was planned but recently withdrawn due to a lack of funding (NCT02569827). Another promising repurposed lead compound is ivermectin, which is an FDA-approved antiparasitic drug mainly used for the treatment of onchocerciasis, scabies and lice62–64. Ivermectin was identified as a potential DENV inhibitor by in silico docking and
inhibits the DENV NS3 helicase65. Ivermectin also interferes with the interaction of NS5 with
importin, which plays a role in the trafficking of proteins between nucleus and cytoplasm66.
Based on these findings, a phase III randomized clinical trial with DENV patients has been performed by Mahidol University and the Ministry of Health of Thailand (NCT02045069). Here, one daily dose of 400 µg/kg ivermectin was given. Unfortunately, no decrease in fever clearance time or time to resolve viremia was observed67. Nevertheless, ivermectin treatment
significantly decreased the NS1 clearance time. Furthermore, an increased proportion of NS1-negative patients was seen at the time of discharge when compared to the placebo group67. Pharmacokinetic studies on ivermectin may help to optimize the treatment regime
and clinical efficacy to further evaluate its anti-DENV potential67. The host-directed inhibitor
UV-4B is another lead compound for anti-DENV treatment. UV-4B inhibits the cellular a-glucosidase, thereby interfering with proper folding and maturation of the viral proteins28.
The compound showed potent antiviral activity in vitro and in vivo28. Moreover, a high barrier
of resistance was demonstrated by treating DENV-infected cells for 38 cycles with UV-4B without resistance development68. This was further validated in vivo, where drug efficacy
was maintained through five passages in mice69. A phase I clinical trial in healthy subjects
indicated good tolerance as well as the absence of serious side effects after one single dose (3 to 1000 mg) of UV-4B (NCT02061358)68. Further clinical trials are needed to assess its
antiviral potential and safety in DENV-patients.
For CHIKV, the identification of antiviral treatments also greatly focused on drug repurposing. Recently, Karlas et al. performed a human genome-wide loss-of-function
study using reported small molecule inhibitors to identify pro-and antiviral host factors and associated antiviral drug candidates70. In this process, the FLT4 inhibitor tivozanib, the
calmodulin inhibitor pimozide and the fatty acid synthesis inhibitor 5-tetradecyloxy-2-furoic acid (TOFA) were identified as potent inhibitors of CHIKV RNA synthesis and viral release
in vitro without showing any cytotoxicity. All three compounds also reduced the viral load in vivo in C57BL/6 mice when treated prophylactically for at least 2 days before infection
with CHIKV70. Interestingly, combinational drug treatment of TOFA and pimozide further
increased antiviral potency in vitro and in vivo70. In this mouse model, significant reduction
in viral load and joint swelling was observed even when the treatment was initiated post-infection. While the antiviral potential of TOFA and pimozide still needs to be evaluated in clinical trials, the low toxicity together with the antiviral effect when administered post-infection makes this drug combination a promising antiviral treatment option to combat CHIKV infection. The uridine monophosphate prodrug sofosbuvir is another promising drug candidate towards CHIKV, which is already approved for the treatment of hepatitis C virus71. Sofosbuvir was found to inhibit the RNA-dependent RNA polymerase in vitro and
was shown to prevent paw swelling and oedema in an arthralgia model of CHIKV infection and protected neonate mice from CHIKV-induced mortality. Another anti-CHIKV lead compound is ribavirin, a drug that is approved for the treatment of respiratory syncytial virus in infants and chronic hepatitis C virus infection72,73. Ribavirin shows potent antiviral activity
towards CHIKV in vitro alone and in combination with IFN-a2b administration. Moreover, a synergistic effect of ribavirin with doxycycline was demonstrated in vitro and in vivo74. The
antiviral target of ribavirin is the host enzyme inosine monophosphate dehydrogenase which leads to GTP depletion and inhibition of viral RNA capping and RNA synthesis. An early phase I clinical trial on ribavirin and CHIKV-infected individuals reported improvement in joint pains and tissue swelling in a fraction of patients30. The limited patient number (20
patients) and the lack of a placebo group are drawbacks of this trial. Nevertheless, it represents a first positive and promising outcome in CHIKV-infected patients and emphasizes the importance to further evaluate ribavirin as an anti-CHIKV compound in clinical trials. Furthermore, the observed synergistic effect of ribavirin with IFN-a2b or doxycycline is also an attractive treatment option to be pursued.
Major challenges still to overcome
In the last decades, there has been enormous progress in the field of antiviral research towards DENV and CHIKV. This progress has, among others, been driven by technical advances which allow for high throughput screenings and the progress in understanding the virus-host interactions that occur during DENV and CHIKV replication. Nevertheless, there are still some challenges that need to be overcome to successfully develop effective antiviral compounds towards DENV and CHIKV.
One concern for the development of antivirals towards RNA-viruses is the development of antiviral resistance. Especially for virus-directed antivirals this is a problem as for many drug candidates only few mutations are needed to acquire resistance75. Combinational therapy has
recently become a popular way to reduce resistance development. Moreover, combinational therapy has been shown to enhance antiviral efficacy while being able to lower the effective dose of the individual compounds, which also reduces the risk for toxic side effects76,77. This
is particularly advantageous for host-directed antivirals which often have a lower therapeutic window compared to direct-acting antivirals. Future experiments should focus on testing
7
combinations of antiviral drug candidates to lower the risk of antiviral resistance and to potentially decrease the required drug concentration. For tomatidine, a high genetic barrier to resistance was observed and future experiments using drug combinations will reveal if its antiviral efficacy can be enhanced by combinational therapy.
Whereas for DENV novel computer-based in silico approaches have contributed greatly to the identification of antiviral drug candidates, for CHIKV these technical advances are still scarce. Many computer-based in silico approaches rely on the crystal structure of the viral proteins and allow for the investigation of the structure-activity relationship of a drug and aid the generation of optimised derivatives19. Whereas for DENV the crystal structure of all
viral proteins has been resolved, structural information of CHIKV proteins is only available for the envelope and capsid proteins as well as NS2 and NS378–81. The scarce availability of
structural information on the CHIKV proteins has hampered the progress in identifying and optimizing antiviral drug-candidates. Therefore, resolving the crystal structure of other viral proteins will facilitate subsequent in silico docking studies. Altogether, gaining structural information on the CHIKV proteins will allow us to determine if tomatidine directly interacts with the different CHIKV proteins.
Moreover, the often observed conflicting results in vitro and in pre-clinical and clinical trials may be caused by a limited understanding of the antiviral mechanism involved82. Thus,
understanding of the underlying mode of action of an antiviral drug may guide the design of an appropriate treatment regime for subsequent pre-clinical studies. I therefore advocate a more thorough understanding of the antiviral mechanism before proceeding to pre-clinical and clinical studies. In line with this, a major focus of this thesis was to identify the antiviral mechanism of tomatidine from the virus perspective but also with regard to potential host factors and structural elements of the compound that may be relevant for its antiviral activity. Further optimisation of mouse models to better resemble the human natural infection may also help to translate pre-clinical studies to clinical trial outcomes. For DENV, humanized mouse models are available which among others are generated by transplantation of hematopoietic stem cells49,83. Further optimization of these models in terms of their immune responses and
the development of severe disease manifestations, may provide a better and more relevant platform to study anti-DENV treatment options. For CHIKV, further improvement of the chronic mouse model regarding different hallmarks of chronic CHIKV infection will also aid to develop a more adequate model to test antivirals during chronic CHIKV infection52.
Alternatively, the development of a humanized mouse model for CHIKV infection may also be beneficial.
In conclusion, the low transferability of antiviral efficacy from in vitro studies to pre-clinical and clinical trials is a major challenge in antiviral research. Future studies should focus on the investigation of the mode of action and pharmacokinetic profile of compounds with the aim to increase pre-clinical efficacy. Also, optimization of DENV and CHIKV mouse models will aid in improving the translation from pre-clinical to clinical drug efficacy. Lastly, emphasis should be on combinational therapy to avoid antiviral resistance. These studies will help to develop effective antiviral treatments towards DENV and CHIKV to fight the debilitating consequences of these infectious diseases around the globe.
References
1. Leyssen, P., De Clercq, E. & Neyts, J. Perspectives for the treatment of infections with Flaviviridae. Clin. Microbiol. Rev. 13, 67–82, table of contents (2000).
2. Fros, J. J. & Pijlman, G. P. Alphavirus Infection: Host Cell Shut-Off and Inhibition of Antiviral Responses. Viruses 8, 166 (2016).
3. Bhatt, S. et al. The global distribution and burden of dengue. Nature 496, 504–507 (2013).
4. WHO | Dengue guidelines for diagnosis, treatment, prevention and control: new edition. WHO (2011).
5. Halstead, S. B. Neutralization and antibody-dependent enhancement of dengue viruses. Adv. Virus Res. 60, 421–67 (2003).
6. Galán-Huerta, K. A. et al. Molecular and Clinical Characterization of Chikungunya Virus Infections in Southeast Mexico. Viruses 10, 248 (2018).
7. Katzelnick, L. C. et al. Antibody-dependent enhancement of severe dengue disease in humans. Science. 358, 929–932 (2017).
8. Mavrouli, M. et al. Reemergence of west nile virus infections in Southern Greece, 2017. Am. J. Trop. Med. Hyg. 100, 420–426 (2019).
9. McDonald, E. et al. West Nile virus and other domestic nationally notifiable arboviral diseases — United States, 2018. Am. J. Transplant. 19, 2949–2954 (2019).
10. Thiberville, S.-D. et al. Chikungunya fever: Epidemiology, clinical syndrome, pathogenesis and therapy. Antiviral Res. 99, 345–370 (2013).
11. Weaver, S. C. & Lecuit, M. Chikungunya Virus and the Global Spread of a Mosquito-Borne Disease. N. Engl. J. Med. 372, 1231–1239 (2015).
12. Schwartz, O. & Albert, M. L. Biology and pathogenesis of chikungunya virus. Nat.
Rev. Microbiol. 8, 491–500 (2010).
13. Agumadu, V. C. & Ramphul, K. Zika Virus: A Review of Literature. Cureus 10, e3025 (2018).
14. Shepard, D. S., Coudeville, L., Halasa, Y. A., Zambrano, B. & Dayan, G. H. Economic impact of dengue illness in the Americas. Am. J. Trop. Med. Hyg. 84, 200–207 (2011). 15. Feldstein, L. R. et al. Estimating the cost of illness and burden of disease associated with the 2014–2015 chikungunya outbreak in the U.S. Virgin Islands. PLoS Negl. Trop. Dis.
13, e0007563 (2019).
16. Murray, N. E. A., Quam, M. B. & Wilder-Smith, A. Epidemiology of dengue: past, present and future prospects. Clin. Epidemiol. 5, 299–309 (2013).
17. Kandel, R. & Hartshorn, K. L. Prophylaxis and treatment of influenza virus infection.
BioDrugs 15, 303–323 (2001).
18. Beesetti, H., Khanna, N. & Swaminathan, S. Investigational drugs in early development for treating dengue infection. Expert Opin. Investig. Drugs 25, 1059–1069
7
(2016).
19. Kovacikova, K. & van Hemert, M. J. Small-Molecule Inhibitors of Chikungunya Virus: Mechanisms of Action and Antiviral Drug Resistance. Antimicrob. Agents Chemother.
64, e01788-20 (2020).
20. Hoarau, J.-J. et al. Persistent Chronic Inflammation and Infection by Chikungunya Arthritogenic Alphavirus in Spite of a Robust Host Immune Response. J. Immunol. 184, 5914–5927 (2010).
21. McLean, H. Q., Belongia, E. A., Kieke, B. A., Meece, J. K. & Fry, A. M. Impact of Late Oseltamivir Treatment on Influenza Symptoms in the Outpatient Setting: Results of a Randomized Trial. Open Forum Infect. Dis. 2, ofv100 (2015).
22. Ozden, S. et al. Human Muscle Satellite Cells as Targets of Chikungunya Virus Infection. PLoS One 2, e527 (2007).
23. Nguyen, N. M. et al. A randomized, double-blind placebo controlled trial of balapiravir, a polymerase inhibitor, in Adult dengue patients. J. Infect. Dis. 207, 1442–1450 (2013).
24. Tricou, V. et al. A randomized controlled trial of chloroquine for the treatment of dengue in vietnamese adults. PLoS Negl. Trop. Dis. 4, e785 (2010).
25. Whitehorn, J. et al. Lovastatin for the Treatment of Adult Patients with Dengue: A Randomized, Double-Blind, Placebo-Controlled Trial. Clin. Infect. Dis. 62, 468–476 (2015). 26. Low, J. G. et al. Efficacy and safety of celgosivir in patients with dengue fever (CELADEN): A phase 1b, randomised, double-blind, placebo-controlled, proof-of-concept trial. Lancet Infect. Dis. 14, 706–715 (2014).
27. Tam, D. T. H. et al. Effects of short-course oral corticosteroid therapy in early dengue infection in vietnamese patients: A randomized, placebo-controlled trial. Clin. Infect.
Dis. 55, 1216–1224 (2012).
28. Warfield, K. L. et al. Inhibition of endoplasmic reticulum glucosidases is required for in vitro and in vivo dengue antiviral activity by the iminosugar UV-4. Antiviral Res. 129, 93–98 (2016).
29. Lamballerie, X. De et al. On Chikungunya Acute Infection and Chloroquine Treatment. Vector-Borne Zoonotic Dis. 8, 837–840 (2008).
30. Ravichandran, R. & Manian, M. Ribavirin therapy for Chikungunya arthritis. J.
Infect. Dev. Ctries. 2, 140 (2008).
31. Raquin, V. et al. Detection of dengue group viruses by fluorescence in situ hybridization. Parasites and Vectors 5, 1 (2012).
32. Chou, Y. ying et al. Colocalization of Different Influenza Viral RNA Segments in the Cytoplasm before Viral Budding as Shown by Single-molecule Sensitivity FISH Analysis. PLoS Pathog. 9, e1003358 (2013).
33. Stern-Ginossar, N. & Ingolia, N. T. Ribosome Profiling as a Tool to Decipher Viral Complexity. Annu. Rev. Virol. 2, 335–349 (2015).
Reticulum-Associated Translation Machinery as a Strategy for Co-opting Host Cell Protein Synthesis.
J. Virol. 92, 1–19 (2018).
35. Irigoyen, N. et al. The translational landscape of Zika virus during infection of mammalian and insect cells. bioRxiv (2017).
36. Junjhon, J. et al. Ultrastructural Characterization and Three-Dimensional Architecture of Replication Sites in Dengue Virus-Infected Mosquito Cells. J. Virol. 88, 4687–4697 (2014).
37. Jin, J. et al. Neutralizing Antibodies Inhibit Chikungunya Virus Budding at the Plasma Membrane. Cell Host Microbe 24, 417-428.e5 (2018).
38. Megger, D. A. et al. Comparison of label-free and label-based strategies for proteome analysis of hepatoma cell lines. Biochim. Biophys. Acta - Proteins Proteomics 1844, 967–976 (2014).
39. Fujiwara, Y. et al. Tomatidine, a tomato sapogenol, ameliorates hyperlipidemia and atherosclerosis in ApoE-deficient mice by inhibiting acyl-CoA:cholesterol acyl-transferase (ACAT). J. Agric. Food Chem. 60, 2472–2479 (2012).
40. Ng, C. G. et al. Effect of host cell lipid metabolism on alphavirus replication, virion morphogenesis, and infectivity. Proc. Natl. Acad. Sci. U. S. A. 105, 16326–16331 (2008). 41. Wichit, S. et al. Imipramine Inhibits Chikungunya Virus Replication in Human Skin Fibroblasts through Interference with Intracellular Cholesterol Trafficking. Sci. Rep. 7, 1–12 (2017).
42. Hsieh, M.-H. et al. Tomatidine Represses Invasion and Migration of Human Osteosarcoma U2OS and HOS Cells by Suppression of Presenilin 1 and c-Raf–MEK–ERK Pathway. Molecules 25, 326 (2020).
43. Yan, K.-H. et al. Tomatidine inhibits invasion of human lung adenocarcinoma cell A549 by reducing matrix metalloproteinases expression. Chem. Biol. Interact. 203, 580–587 (2013).
44. Friedman, M. Anticarcinogenic, cardioprotective, and other health benefits of tomato compounds lycopene, a-tomatine, and tomatidine in pure form and in fresh and processed tomatoes. J. Agric. Food Chem. 61, 9534–9550 (2013).
45. Ebert, S. M. et al. Identification and small molecule inhibition of an activating transcription factor 4 (ATF4)-dependent pathway to age-related skeletal muscle weakness and atrophy. J. Biol. Chem. 290, 25497–25511 (2015).
46. Chiu, F. L. & Lin, J. K. Tomatidine inhibits iNOS and COX-2 through suppression of NF-kB and JNK pathways in LPS-stimulated mouse macrophages. FEBS Lett. 582, 2407– 2412 (2008).
47. Chu, X. et al. Tomatidine suppresses inflammation in primary articular chondrocytes and attenuates cartilage degradation in osteoarthritic rats. Aging (Albany. NY). 12, 12799– 12811 (2020).
48. Kuo, C. Y. et al. Tomatidine Attenuates Airway Hyperresponsiveness and Inflammation by Suppressing Th2 Cytokines in a Mouse Model of Asthma. Mediators
7
49. Chan, K. W. K., Watanabe, S., Kavishna, R., Alonso, S. & Vasudevan, S. G. Animal models for studying dengue pathogenesis and therapy. Antiviral Research 123, 5–14 (2015). 50. Low, J. G. H., Ooi, E. E. & Vasudevan, S. G. Current Status of Dengue Therapeutics Research and Development. J. Infect. Dis. 215, S96–S102 (2017).
51. Simmons, C. P. et al. Recent advances in dengue pathogenesis and clinical management. Vaccine 33, 7061–7068 (2015).
52. Haese, N. N. et al. Animal models of chikungunya virus infection and disease. J.
Infect. Dis. 214, S482–S487 (2016).
53. Hawman, D. W. et al. Chronic Joint Disease Caused by Persistent Chikungunya Virus Infection Is Controlled by the Adaptive Immune Response. J. Virol. 87, 13878–13888 (2013).
54. Diosa-Toro, M. et al. Tomatidine, a novel antiviral compound towards dengue virus.
Antiviral Res. 161, 90–99 (2019).
55. Dorsaz, S. et al. Identification and mode of action of a plant natural product targeting human fungal pathogens. Antimicrob. Agents Chemother. 61, e00829-17 (2017).
56. Jiménez de Oya, N. et al. Targeting host metabolism by inhibition of acetyl-Coenzyme A carboxylase reduces flavivirus infection in mouse models. Emerg. Microbes
Infect. 8, 624–636 (2019).
57. Leblanc, A. F. et al. Murine Pharmacokinetic Studies. Bio-Protocol 8, e3056 (2018). 58. Friedman, M., Henika, P. R. & Mackey, B. E. Feeding of potato, tomato and eggplant alkaloids affects food consumption and body and liver weights in mice. J. Nutr. 126, 989–999 (1996).
59. Friedman, M., Henika, P. R. & Mackey, B. E. Effect of feeding solanidine, solasodine and tomatidine to non-pregnant and pregnant mice. Food Chem. Toxicol. 41, 61–71 (2003). 60. Botta, L., Rivara, M., Zuliani, V. & Radi, M. Drug repurposing approaches to fight Dengue virus infection and related diseases. Front. Biosci. - Landmark 23, 997–1019 (2018). 61. Watanabe, S. et al. Optimizing celgosivir therapy in mouse models of dengue virus infection of serotypes 1 and 2: The search for a window for potential therapeutic efficacy.
Antiviral Res. 127, 10–9 (2016).
62. Strycharz, J. P., Yoon, K. S. & Clark, J. M. A New Ivermectin Formulation Topically Kills Permethrin-Resistant Human Head Lice (Anoplura: Pediculidae). J. Med. Entomol 45, 75–81 (2008).
63. Victoria, J. & Trujillo, R. Topical Ivermectin: A New Successful Treatment for Scabies. Pediatr. Dermatol. 18, 63–65 (2001).
64. Babalola, O. E. Ocular onchocerciasis: Current management and future prospects.
Clin. Ophthalmol. 5, 1479–1491 (2011).
65. Mastrangelo, E. et al. Ivermectin is a potent inhibitor of flavivirus replication specifically targeting NS3 helicase activity: New prospects for an old drug. J. Antimicrob.
66. Wagstaff, K. M., Sivakumaran, H., Heaton, S. M., Harrich, D. & Jans, D. A. Ivermectin is a specific inhibitor of importin a/b-mediated nuclear import able to inhibit replication of HIV-1 and dengue virus. Biochem. J. 443, 851–856 (2012).
67. Yamasmith, E. et al. Efficacy and Safety of Ivermectin against Dengue Infection: A Phase III, Randomized, Double-blind, Placebo-controlled Trial. 34th Annu. Meet. R. Coll.
Physicians Thail. (2018).
68. Low, J. G., Gatsinga, R., Vasudevan, S. G. & Sampath, A. Dengue antiviral development: A continuing journey. Adv. Exp. Med. Biol. 1062, 319–332 (2018).
69. Plummer, E. et al. Dengue Virus Evolution under a Host-Targeted Antiviral. J.
Virol. 89, 5592–5601 (2015).
70. Karlas, A. et al. A human genome-wide loss-of-function screen identifies effective chikungunya antiviral drugs. Nat. Commun. 7, 11320 (2016).
71. Ferreira, A. C. et al. Beyond members of the Flaviviridae family, sofosbuvir also inhibits chikungunya virus replication. Antimicrob. Agents Chemother. 63, e01389-18 (2019). 72. Li, G. & De Clercq, E. Current therapy for chronic hepatitis C: The role of direct-acting antivirals. Antiviral Research 142, 83–122 (2017).
73. Hall, C. B. et al. Aerosolized ribavirin treatment of infants with respiratory syncytial viral infection. A randomized double-blind study. N. Engl. J. Med. 308, 1443–1447 (1983). 74. Rothan, H. A. et al. A combination of doxycycline and ribavirin alleviated chikungunya infection. PLoS One 10, e0126360 (2015).
75. Irwin, K. K., Renzette, N., Kowalik, T. F. & Jensen, J. D. Antiviral drug resistance as an adaptive process. Virus Evol. 2, vew014 (2016).
76. Boldescu, V., Behnam, M. A. M., Vasilakis, N. & Klein, C. D. Broad-spectrum agents for flaviviral infections: Dengue, Zika and beyond. Nat. Rev. Drug Discov. 16, 565–586 (2017).
77. Kaufmann, S. H. E., Dorhoi, A., Hotchkiss, R. S. & Bartenschlager, R. Host-directed therapies for bacterial and viral infections. Nat. Rev. Drug Discov. 17, 35–56 (2018).
78. Law, Y.-S. et al. Structural insights into RNA recognition by the Chikungunya virus nsP2 helicase. Proc. Natl. Acad. Sci. 116, 9558–9567 (2019).
79. Narwal, M. et al. Crystal structure of chikungunya virus nsP2 cysteine protease reveals a putative flexible loop blocking its active site. Int. J. Biol. Macromol. 116, 451–462 (2018).
80. Malet, H. et al. The Crystal Structures of Chikungunya and Venezuelan Equine Encephalitis Virus nsP3 Macro Domains Define a Conserved Adenosine Binding Pocket. J.
Virol. 83, 6534–6545 (2009).
81. Yap, M. L. et al. Structural studies of Chikungunya virus maturation. Proc. Natl.
Acad. Sci. U. S. A. 114, 13703–13707 (2017).
82. Abdelnabi, R., Neyts, J. & Delang, L. Towards antivirals against chikungunya virus.
7
83. Bente, D. A., Melkus, M. W., Garcia, J. V. & Rico-Hesse, R. Dengue Fever in Humanized NOD/SCID Mice. J. Virol. 79, 13797–13799 (2005).
