Evaluation of Hazara Orthonairovirus infection in tick cell lines as a model to study persistent infection of Crimean-Congo Hemorrhagic Fever Virus in ticks

PhD Course: Molecular Medicine Curriculum: Biomedicine

XXXII cycle

Evaluation of Hazara Orthonairovirus infection in tick cell lines as a model to study persistent infection of Crimean-Congo Hemorrhagic Fever Virus in ticks

Coordinator: Ch.mo Prof. Stefano Piccolo Supervisor: Prof. Arianna Calistri

Co-Supervisor: Dott. Cristiano Salata

Abstract

Crimean-Congo hemorrhagic fever (CCHF) is a severe disease for humans caused by CCHF Orthonairovirus (CCHFV), a class 4 pathogen. Ticks of the genus Hyalomma are the viral

tick blood feeding.

It has been shown that CCHFV can persistently infects Hyalomma-derived tick cell lines without any cytopathic effect. However, the mechanism allowing the establishment of a persistent viral infection in ticks is still unknown. It has been recently reported that Hazara Orthonairovirus (HAZV) can be used as a BSL-2 viral model instead of CCHFV to study viral/vector interaction. The aim of our study is to elucidate the mechanism allowing the establishment of the CCHFV persistent infection in ticks by using HAZV as a model.

We used classical and molecular methods applied to virology to characterize the establishment of the HAZV persistent infection in two Hyalomma anatolicum-derived cell lines, Hae/CTVM8 and Hae/CTVM9.

As for CCHFV, we showed that HAZV persistently infects tick cells without any sign of cytopathic effect, and that cells can be cultured for more than two years. The persistent infection is established in roughly 10 days post-infection and viral titer is maintained at lower level in comparison to the earlier time points. Interestingly, short viral derived DNA forms (vDNAs) start to be detected in parallel with the beginning of viral replication and are maintained in persistently infected cells. Experiments with the antiretroviral drug AZT suggest that vDNAs are produced by a reverse transcriptase activity; furthermore, we collected evidence that vDNAs are not integrated and require an active viral genome replication. Morover, the presence of vDNA correlates with the downregulation of viral replication to promote viral tolerance and cell survival.

In conclusion, vDNAs synthesis might represent a strategy to control viral replication in ticks,

as recently demonstrated in insect, allowing the persistent infection in viral vectors.

Summary

1. Introduction ... 1

1.1 Crimean-Congo hemorrhagic fever (CCHF) ... 1

1.2 The etiologic agent: Crimean-Congo hemorrhagic fever virus (CCHFV) ... 4

1.3 Transmission, distribution and risk factors ... 9

1.4 Ticks as CCHFV vector ... 13

1.5 Hazara Orthonairovirus (HAZV): a model for CCHFV infection ... 15

1.6 Tick cell lines as a model for studying CCHFV ... 16

2. Aim of the study ... 19

3. Material and methods ... 21

3.1 Cell culture ... 21

3.2 Cell lines infections ... 22

3.3 HAZV production ... 22

3.4 DNA extraction ... 23

3.5 DNase treatment ... 23

3.6 RNase treatment ... 23

3.7 Polymerase Chain Reaction (PCR) ... 23

3.8 RNA extraction ... 25

3.9 RT-PCR ... 25

3.10 MTT test (Methyl-Thiazole Tetrazolium Test) ... 26

3.11 Azidothymidine (AZT) ... 26

3.12 Ribavirin ... 26

3.13 UV-inactivated HAZV ... 27

3.14 Viral titration ... 27

3.15 Cloning ... 27

3.16 Real Time PCR ... 28

3.17 Proteins extraction ... 29

3.18 In vitro Reverse Transcriptase assay ... 29

4. Results ... 31

4.1 Tick cells infected with HAZV develop a persistent infection ... 32

4.2 Presence of HAZV-derived vDNAs in infected tick cells ... 35

4.3 A reverse transcriptase activity in tick cells allows vDNAs synthesis ... 37

4.4 vDNAs are involved in the persistent viral infection and cell survival ... 41

4.5 The synthesis of vDNAs require the replication of viral genome ... 45

5. Discussion ... 51

6. References ... 58

1. Introduction

1.1 Crimean-Congo hemorrhagic fever (CCHF)

Crimean-Congo hemorrhagic fever (CCHF) is a vector-borne zoonosis. CCHF is present in a wide geographic area extending from the western of China to Africa, including the Middle East, and the south Asian regions as well as the southern and eastern European countries. CCHF is endemic in 47 countries including Asia, Africa, Middle East, and Europe. The more focalised areas of risk encompass various regions of South East Asia, Sub-Saharan Africa, Black Sea, and Central Asia (Messina et al., 2015).

During the last decade CCHF virus (CCHFV) has been detected in Western Europe (Estrada- Peña et al. 2012) demonstrating that this virus is not restricted to Eastern Europe. A viral strain circulating in south-western Mediterranean is a reason of concern about the spread of the virus into northern latitudes (Gale et al. 2011).

Crimean-Congo hemorrhagic fever is easily transmissible and has a fatality up to 30%. The

incidence of CCHF has been increasing since 2008 (Wahid et al., 2019).

Figure 1. Geographic distribution of CCHFV (Spengler et al., 2017)

The etiologic agent of CCHF is a virus named Crimea-Congo hemorrhagic fever virus (CCHFV), recently re-named Crimea-Congo hemorrhagic fever Ortonairovirus, and its diffusion is correlated with distribution of some species of tick belonging to Ixodid order, in particular those belonging to the genus Hyalomma (Bente et al., 2013).

CCHFV infection is rare in humans, because humans are not part of normal zoonotic cycle of this virus. Normally, people who come to rural areas or people who are in contact with wildlife are more exposed to CCHFV infection, and also people belonging to working categories sharing the same environment of the tick which serves as CCHFV natural host. In most cases, infection is caused by the bite of infected tick, but contact with liquids derived from infected animals or people, can also be a source of disease transmission (Bente et al., 2013).

Most of the CCHFV infections spontaneously resolve after non-specific fever. However, some individuals develop a severe hemorrhagic fever disease.

CCHFV is also included in the list of infectious agents for potential use as a biological weapon,

and also in the list of diseases for which the Revised International Health Regulations call for

assessment of the CCHF event for possible notification to the World Health Organization (Maltezou et al., 2009).

CCHFV is easily transmitted from one area to another because of its vector and some evidences suggest the horizontal and vertical transmission of this virus. In fact, female ticks can transmit the virus to their eggs and during fertilization the female can infect the male or vice versa.

Virus enter into human body through tick's bite or via direct contact with infected animal blood (veterinarians, slaughter house workers, and farmers have a higher probability to contract the virus). Person-to-person transmission can occur through different body fluids, specifically blood, and saliva and, recently it has been demonstrated that sexual transmission can also occur (Bodur et al., 2010; Pshenichnaya et al., 2016).

Figure 2. CCHFV transmission and reservoir (Spengler et al., 2017).

CCHFV was first recognised in Tajikistan in the 12th Century. In 1944, the evidence of CCHF

was reported from the Crimean region of the former Soviet Union, followed by several other

cases reported in several southern Soviet Republics, South Africa and Bulgaria during next few

decades. The virus derived its name “Crimean-Congo hemorrhagic fever” because same disease

was reported in 1969 in Congo region.

Since 2000, small outbreaks of CCHFV have documented in Greece, India, some Balkan countries, Pakistan, Turkey, Spain, Sudan, Uganda, Georgia, and Iran. Since August 2016, an increased incidence of CCHFV has been reported in Pakistan (Wahid et al., 2019).

The CCHFV pathogenesis is largely unknown. It has been reported that CCHFV can disable the host immune system. The upregulation of pro-inflammatory cytokines and soluble molecules activate endothelial cells, which in severe cases may produce: vasodilation, hypotension, increased vascular permeability, shock, multiple organ failure, and death. CCHFV impairs innate immune response and delays adaptive immune system and cause systematic spread as well as uncontrolled replication of virus throughout host body accompanied with the delayed induction of interferon, increased level of IL-6, IL-1, TNF-alpha, apoptosis of lymphocytes, partial activation of macrophages and dendritic cells, lympho-histiocytosis, hemophagocytosis, and weak antibody response (Wahid et al., 2019).

During the pre-hemorrhagic period, patients experience fever, chills, polyneuritis, abdominal pain, back pain, loss of appetite, poor vision, vomiting, nausea, laboured breathing, diarrhea, loss of hearing, and loss of memory. Several cardiovascular as well as neuropsychiatric changes such as mood swings, confusion, violent behaviour, and aggression have also been reported.

Hemorrhage, dehydration, anemia, myocardial infarction, pleural effusion, kidney failure, lung edema, capillary fragility, thrombocytopenia, capillary toxicosis occurs in severe cases of CCHF (Akinci et al., 2013; Bente et al., 2013; Kilinc et al., 2016).

Early and differential diagnosis is always very important. The gold standard is a novel one-step rRT-PCR assay as sensitive, specific, simple, reliable, rapid, and repeatable tool for the detection of the CCHFV RNA that is detectable until day 16 of illness onset (Zahraei et al., 2016).

There are not approved specific drugs to treat the CCHFV infection but supportive therapy including the administration of erythrocytes, thrombocytes, and fresh frozen plasma acts as an important strategy mitigate CCHF symptoms at an early stage. In vitro, it has been demonstrated the synergistic effect of combination of chloroquine or chlorpromazine and ribavirin against CCHFV (Ferraris et al., 2015).

1.2 The etiologic agent: Crimean-Congo hemorrhagic fever virus (CCHFV)

CCHFV belongs to the genus Orthonairovirus of the Nairoviridae family (formerly

Electron microscopy studies of nairoviruses structure, report spherical particles of relatively uniform size with a diameter of ~90–100 nm with small (<10 nm) spike-like projections from the surface of the particle (Zivcec, et al., 2016).

Figure 3. Crimean-Congo hemorrhagic Fever Virus (ECDC)

CCHFV has a circular negative single-stranded RNA genome composed from three segments:

small (S, 1672 b), medium (M, 5366 b) and large (L, 12108 b).

Figure 4. Crimean-Congo hemorrhagic fever virus (Bente et al.,2013).

The S segment codifies for the nucleoprotein (N) (Papa et al., 2017), the M segment codify for a polypeptide precursor that subsequently is processed in two different membrane glycoprotein, called Gn and Gc, responsible for viral adhesion to the host cells (Zivcec et al., 2016). The L segment codify for the L protein that is a RNA-polymerase-RNA-dependent that is essential for the viral replication (Papa et al., 2017).

In addition, a second protein, the non-structural S (NSS), is encoded by the S segment in the opposite orientation relative to the nucleoprotein (NP) gene (Zivcec, et al., 2016).

Each genomic segment contains 5' and 3' complementary non-coding regions (NCRs), which

are composed by 9 nucleotides contributing to the circularization of the genomic RNA

segments. These NCRs work also as a promoter, in fact they are required for the binding of the

viral RNA-polymerase (Zivcec, et al., 2016).

Figure 5. CCHFV genomic organization (Zivcec, et al., 2016).

Phylogenetic analysis of L and S RNA segments of CCHFV revealed seven different genetic clades based on geographical regions: two Asian, two European, and three African. Recent studies had traced the ancestral origin of all clades 1000 years back, possibly in Africa (Chinikar et al., 2010). CCHFV migrated to Middle East and then travelled in two directions leading to two Asian clades; first on scattering in Central Asia and China, while the second one in Pakistan and Iran. Two highly divergent strains that were assumed Turkish invaded Europe finally (Wahid et al., 2019).

To enter cells, viruses bind to a variety of host cell receptors, which determine the entry mechanism, host range and pathogenesis of each virus (Marsh and Helenius, 2006).

The cellular receptor required for CCHFV entry have not been identified. A functional interaction has been suggested between CCHFV Gc and cell surface nucleolin (Xiao et al., 2011).

It is believed that the initial binding of CCHFV to cell surface is mediated by the glycoproteins Gn and/or Gc. However, the details of specific glycoprotein involvement in viral attachment, internalization, and fusion remain unknown. It is suspected that Gc is responsible for binding to the cellular receptors. The Gc of CCHFV is also thought to mediate fusion (Zivcec et al., 2016).

Several viruses use the benefits of endocytosis to enter into target cells, since it provides a

protected environment and permits intracellular movement. The most commonly used cellular

endocytosis pathway is clathrin-dependent endocytosis (Marsh and Helenius, 2006).

It has been demonstrated that CCHFV entry is mediated by clathrin-dependent pathways.

Clathrin-dependent entry was confirmed using chemical inhibitors of clathrin and endosome acidification. It has been also showed that cholesterol is involved in the early events of the CCHFV replication cycle (Simon et al., 2009).

Figure 6. CCHFV virion and replication cycle (Zivcec, et al., 2016).

After cell entry and fusion, the genomic RNPs are released into the cytosol and the encapsidated vRNA is used as a template for the L protein to synthesize viral mRNA.

To initiate viral mRNA synthesis, the L protein uses m7G capped primers that are snatched from cellular mRNA by an endonuclease domain located in the L protein. CCHFV L protein contains a residue (D693) that is predicted to coordinate a Mn2+ critical for the cap snatching activity, as demonstrated for endonucleases of other L proteins.

The replication of genomic RNPs is a process requiring the replication and encapsidation of

uncapped, negative sense vRNA and positive sense complementary RNA (cRNA). The

encapsidated forms of the cRNA and vRNA are respectively defined as antigenomic and

genomic RNPs. Since CCHFV is a negative strand RNA virus, genomic RNPs are used as

template to synthesize capped mRNA and produce antigenomic RNPs. During the replication

of the genomic RNPs, a cRNA is synthesized by the L protein and NP subunits are added to the elongating strands to obtain antigenomic RNPs, and in turn the cRNA of the antigenomic RNPs is used a template to obtain genomic RNPs (Zivcec, et al., 2016).

The replication involving viral components, is followed by viral protein processing and maturation.

The Gn and Gc maturation is very different and very complex compared with glycoprotein processing in other Bunyadavirales (Sanchez et al., 2002). The Gn and Gc maturation allows to obtain the working form of these two glycoprotein. To yield the complete set of glycoproteins, is heavily glycosylated and subsequently cleaved by host proteases. The productive maturation in the ER is followed by PreGn and PreGc transport to the Golgi complex (Zivcec, et al., 2016).

Following assembly, CCHFV particles are released by exocytosis often in the absence of detectable cytopathology and egress occurs from the basolateral membrane in polarized epithelial cells (Connolly-Andersen et al., 2007).

1.3 Transmission, distribution and risk factors

As mentioned before, the diffusion of CCHFV is strictly correlated with the distribution of some species of ticks, in particular ticks belonging to Hyalomma genus.

The genus Hyalomma consists of 30 species of which Hyalomma marginatum is the almost

exclusive CCHFV vector in Europe. H. marginatum is abundant in almost all Balkan countries,

and also in Turkey, Russia, Ukraine, Italy, Spain, and Portugal. This species is also found in

southern France, Cyprus, Israel, and along the northern seaside of all Mediterranean African

countries (Iori and Di Paolo, 1999; Vorou et al., 2007; Pavlidou et al., 2008). In 2006

(Kampen et al., 2007; Nijhof et al., 2007). Although CCHFV has been detected in other tick

species (e.g. Dermacentor, Rhipicephalus, Ixodes, Boophilus), CCHFV cycle in nature is

sustained almost exclusively by Hyalomma ticks.

Figure 7. Virus/Host/CCHFV interaction (Papa et al., 2017).

There are a lot of important factors involved in the distribution of the genus Hyalomma. The most important are temperature and climatic factors, densities of wild and domestic animals or also war and human behaviour.

In fact, these ticks, become more active in spring when daily temperatures exceed 10.5 °C, with optimal temperature between 22–27 °C and 75–100% relative humidity. In Turkey, increased tick activity in summer, coincided with CCHF incidence in humans, with most cases concentrated from May through July, when average temperatures are between 15–20 °C. In Iran, CCHF incidence is higher when temperatures are between 30–40 °C and with maximum relative humidity levels of 20–50%. CCHF incidence in humans is positively related to temperature in Turkey, Iran and Bulgaria and maximum relative humidity in Iran. Higher temperatures and humidity facilitate higher tick development rates, virus transmission, increased tick questing activity, and higher host parasite loads, all of which promote virus circulation (Esser et al., 2019).

The temperature limits explain the consistent association of CCHF cases with warmer

temperatures and the season. Compared to other ticks, Hyalomma ticks adapt better to dry

climates. These characteristics are in accordance with activation of Hyalomma ticks in

temperate areas in late spring and their continuous activation throughout summer to early

autumn (Vorou et al., 2007; Randolph and Rogers 2010).

Hyalomma ticks are found mainly in areas of small and relatively dry vegetation that are

relatively easily accessible by humans, and not in forest-type vegetation (Maltezou and Papa, 2010).

Another important factor for tick distribution is the increased densities of wild and domestic ungulates and of livestock: this because ticks feed on the blood of these animals.

Instead, human population density has not been shown to correlate with CCHF incidence, but activities that increase human interaction with potentially infectious animals did, such as hunting practices and animal husbandry. In fact, the major human risk groups for CCHFV are people who come into close contact with livestock, such as farmers and abattoir employees (Esser et al., 2019). Environmental and social changes that increase the risk of CCHFV in humans may have varying causes, including war conflicts (e.g. past outbreaks of CCHFV occurred in Crimea during World War II and in Kosovo in 1999), changes in animal husbandry practices, agricultural reforms, and abandonment of arable land (Estrada-Peña et al., 2007; Volk et al., 2010; Hoch et al., 2016).

Currently Turkey is experiencing the largest ever recorded CCHF outbreak since the identification of this disease in 1944. Following several decades of serological evidence of zoonotic circulation of the virus, the first human cases occurred in 2002 in northeast Anatolia.

Since then, more than 2500 cases have been recorded with a mean incidence rate of 3.37/100000 habitants, although there were hyper-endemic regions with incidence rates up to 190.81/100000 habitants (Maltezou and Papa, 2010).

The emergence of CCHF in Turkey was potentially driven by anthropogenic factors, at least in the very first events. During 1995 and 2000, agricultural activities and hunting were abandoned in the core CCHF endemic areas in northeast Anatolia, allowing the increased reproduction of hares. When humans returned to agricultural lands in 2001, they were exposed to a much higher population of infected ticks (Ergonul 2006; Estrada-Pena et al., 2007). In practice, emergence of CCHF in this area constituted the tip of an iceberg that was already there for decades, whereas suitable ecological and climate parameters allowed further spatial and temporal expansion of the outbreak (Estrada-Pena et al., 2007).

As for all vector-borne diseases, the establishment and maintenance of endemic CCHF within

a region require environmental and climate factors and human behaviour that favour an efficient

contact between an abundant burden of competent vectors, hosts with relatively high prevalence

of infection, and humans. Given the multiple animal species which may act as hosts for CCHF,

in practice the possibility of CCHF becoming endemic within a region resolves to a question of

abundance of Hyalomma ticks in a favourable environment in terms of temperature and

humidity, a criterion which is already fulfilled by almost all southern European countries along the Mediterranean Sea.

The spread of pathogens is typically characterized by the basic reproduction number, called R0.

R0 is defined as the expected number of secondary cases produced by a single infected animal in a wholly susceptible population when a homogeneous and well-mixed host population exists.

This quantity is very useful to the theory and management of infectious diseases. In the CCHFV case, R0 is represented by the number of infected Hyalomma ticks arising from introduction of a single infected tick into a population of susceptible ticks. When R0 is ≥1, this means that virus can persist and circulate in the tick population, in the other case (when R0 is £1) it does not establish (Estrada-Peña et al., 2012)

CCHFV could be introduced from an endemic to a non-endemic area either by infected vectors, following expansion of their geographical range, or by movement of animal hosts. Climate changes could make an environment more or less favourable for a vector. A model that studied the impact of various climate scenarios on the habitat areas of various tick species predicted that an increase in temperature and decrease in rainfall in the Mediterranean region will produce a sharp increase in the extend of suitable habitat for Hyalomma marginatum, resulting in its northward expansion, with the highest impact at the margins of the current distribution range (Estrada-Pena et al., 2007).

Since quantitative studies of interactions among non-biological factors have not been conducted so far, the impact of climate change on the risk of CCHF infection remains speculative.

In contrast to mosquitoes, ticks do not move over long distances, thus it is highly unlikely to account for CCHFV transfer far away by their own. On the other hand, infected ticks could be carried by animals or birds over long distances. Hares have large home ranges and theoretically could be implicated in CCHFV importation to a naıve area as livestock through (legal or illegal) trade could do. However, the emergence of a novel CCHF focus would require the importation of a significant viral load through animals infested with CCHFV-infected ticks. The short-term viremia of infected animals is an intrinsic limit, if importation of an infected animal is an isolated event. Thus, as viremia is short, emergence of CCHF in a novel focus as a result of tick-infested animals would also require the presence of suitable Hyalomma vectors for the ticks to establish themselves and for the virus to become endemic and not just cause an outbreak.

Overall, the contribution of each parameter in the onset and spread of an outbreak as well as

the transmission rate of the disease remain largely unknown (Maltezou and Papa, 2010).

Given the presence of suitable vectors, hosts, and climate parameters in southern Europe Mediterranean countries, it appears possible for CCHF to emerge in these areas within the near future.

1.4 Ticks as CCHFV vector

As mentioned before, ticks are the most important vector and the only natural reservoir of CCHFV: for that, is very important to understand the mechanisms allowing viral dissemination through ticks and all the biological issues involved in this process.

One of the very important point is that ticks, like other invertebrates, do not present adaptive immunity, and they rely on innate immune response (e.g. phagocytosis, encapsulation, nodulation, and secretion of humoral factors in the hemolymph) (McNally et al., 2012).

Another important mechanism of innate antiviral defence of ticks (and of arthropods in general) to fight arboviruses, is represented by the RNA interference (RNAi).

However, the exact role of RNAi pathway in tick-CCHFV interactions remains to be clarify.

Following the tick's bite and the blood meal carried out on an ungulate or a little mammalian, CCHFV comes into contact with the tick and evades the tick humoral and cellular immune responses and replicates in the lining of the tick’s midgut; then it is disseminated to the hemolymph and infects various tissues, with highest viral titers being observed in the proliferating tissues (e.g., salivary glands and reproductive tissues) (Dickson and Turell, 1992).

It has been recently demonstrated that many mutations in CCHFV were recovered from ticks after only a single transstadial transmission, whereas no mutations were detected in CCHFV recovered from the mammalian host, with greater viral intra-host diversity in the tick rather than the vertebrate host (Xia et al., 2016).

Ticks are a very common vectors for several pathogen and for that, CCHFV is generally not the sole microbe in ticks; endosymbionts and several pathogens may be present and transmitted at the same time (Papa et al., 2017). Viral infections in ticks are not entirely silent as CCHFV infection and may affect the tick survival, behaviour and gene expression (McNally et al., 2012). It has been also showed that the microbiome has an effect on tick fitness and pathogen infection and transmission.

The general issues concerning tick-host interactions likely can be applied to CCHFV. The first

contact between the tick and the host occurs during the tick bite and during the tick feeding on

the host.

The introduction of salivary gland secretions into the feeding lesion is the major, if not exclusive, route via which pathogens and toxins access the vertebrate host and mediate the host reactions (Kaufman, 1989). Despite the host’s hemostatic, inflammatory and immune responses, the tick remains attached for blood-feeding via the pharmacy located in its salivary glands and secreted in saliva. Anticoagulants, cytolytic substances, vasoactive mediators (such as prostaglandins) are among the secreted agents.

Time of attachment may also affect the level of tick-host interaction. Abiotic factors are involved indirectly in the tick-host interactions by playing a role in the abundance and aggressiveness of ticks, thus affecting the chance of a host to be bitten by ticks (Papa et al., 2017).

Figure 8. CCHFV/Host/Vector interaction (Papa et al., 2017)

In contrast to almost all other hemorrhagic fever viruses that have relatively narrow geographical ranges, CCHFV demonstrates the widest geographical distribution among all tick- borne diseases, being detected in more than 30 countries.

Finally, several Hyalomma species, have a two-host life-cycle in which ticks remain on their

host from larva to nymph. These ticks may be attached to their first host for several weeks. The

two-host life-cycle of this kind of ticks in combination with trans-ovarial transmission of

CCHFV favours long-distance dispersal of both vector and virus via migratory birds or livestock trade. Although no evidence exists that wild birds become viremic, immature ticks are frequently introduced into northern latitudes by passerine birds during spring migration (Esser et al., 2019).

Major migratory bird fly paths may explain CCHFV lineage links between West and South Africa (Deyde et al., 2006).

1.5 Hazara Orthonairovirus (HAZV): a model for CCHFV infection

In recent years, the occurrence of CCHFV-mediated disease has been newly reported in many Mediterranean countries (Bente et al., 2013), likely as a consequence of the increasingly broad habitat and population size of its tick vector, with the increases possibly occurring in response to climate change (Estrada-Peña et al., 2013). CCHFV is now recognized as a potential threat to human health in the densely populated regions of Northern Europe (Maltezou and Papa, 2010).

CCHFV is a risk group 4 pathogen and the restrictions imposed for CCHFV manipulation in appropriate bio-security laboratories, are motivated by the potential lethality of the infection, by the fact that preventative or therapeutic measures do not exist and that the virus is not endemic in many western countries and by the possibility of an inter-individual transmission.

A virus called Hazara Orthonairovirus (HAZV), belongs to the same serogroup of CCHFV and is closely related with CCHFV.

HAZV has not been associated with serious human disease and is classified as a risk group 2 pathogen. HAZV infection of type 1 interferon receptor-deficient mice shares clinical features of CCHFV-mediated disease in humans and represents an accessible CCHFV infection model.

Taken together, these findings suggest that HAZV is a useful BSL-2 surrogate that can be used to study the molecular, cellular, and disease biology of the highly pathogenic CCHFV, as well as of other Nairoviruses responsible for serious human and animal diseases (Surtees et al., 2016).

Over the last few years, several systems have been developed that partially overcome the limitations imposed on CCHFV manipulation.

In contrast with CCHFV pathogenicity, HAZV is non-pathogenic (biosafety level 2) to humans

and thus represents an excellent model to study aspects of CCHFV biology under conditions of

more-accessible biological containment (Surtees et al., 2016).

1.6 Tick cell lines as a model for studying CCHFV

Ticks are the most important and devastating ectoparasites in the field of livestock and human diseases. In fact, ticks are second only to mosquitoes as vector of human pathogens. but despite this, tick cell lines are lagged behind the other insect cell lines.

Tick cell lines used nowadays are provided by the “Tick Cell Biobank”.

The Tick Cell Biobank was founded in 2009 as a collaboration between the Universities of Edinburgh, Minnesota and Texas, with the goal to create a repository for existing tick cell lines, a resource to support their successful dissemination worldwide and a facility for establishment of new tick cell lines.

CCHFV has been detected in or isolated from over 30 species of ticks (Shepherd et al., 1989;

Whitehouse 2004). Ixodid tick belonging to Hyalomma genus, are the main vectors of CCHFV (Hoogstraal 1979). Natural hosts of CCHFV maybe are small-sized or medium-sized mammals but this virus also replicates in large mammals such as domestic cattle (Whitehouse 2004).

CCHFV is non-pathogenic in its natural hosts, but highly pathogenic in humans. Unfortunately, there is insufficient knowledge of the role ticks play in the pathogenesis, transmission, and perpetuation of CCHFV in nature. One of the most important topic on arthropod infection by arboviruses of the last few years, is about interactions between arboviruses and immune response of arthropod vectors, which might be crucial in understanding viral host range, persistence, and transmission.

An interesting in vitro model to study CCHFV/vector interaction is represented by tick cell lines.

Since the first tick cell lines were established nearly 50 years ago, they have been applied to isolate, propagate and study tick-borne viruses and bacteria, and in numerous studies on tick- pathogen interactions (Bell-Sakyi et al, 2011).

Nowadays, all tick cell lines available are phenotypically and genotypically heterogeneous, because of their multiple partial or complete tick derivation. This fact has advantages but also disadvantages and there is currently not an alternative to this heterogeneous cell lines.

Tick cell lines are all relatively slow-growing, tolerate high densities and grow in three dimensions; although cryopreservation can be challenging, Ixodid tick cells frozen and thawed rapidly can have good viability after many years in liquid nitrogen (Bell-Sakyi et al, 2011).

There are some studies about infection of mosquito cells with arboviruses in which is

demonstrated that infection does not produce any cytopathic effect. Similar results have been

observed in tick cells infected with arboviruses that often become persistently infected (Bell-

Sakyi et al, 2011). Usually in infected tick cells viral titers achieved were lower than those normally seen in mammalian cell cultures demonstrating that tick cells tolerate viral infection because of their capability to monitor infection. This tolerance is also supported by the observation that in infected mammalian cells we can see cell death following infection while tick cells tolerate the viral infection without displaying any obvious cytopathic effect (Karlberg et al., 2011).

Figure 9. Tick cell lines: Hae/CTVM9 on the left (A), Hae/CTVM8 on the right side (B)

The increasing availability of complete or partial tick genome sequences, especially from important arbovirus vector species, contribute to analysis of virus–cell interactions.

But it also important to translate the findings from in vitro studies using tick cell lines to the in vivo studies using whole ticks, and this is particularly difficult due to highly pathogenic virus CCHFV.

However, cell lines and molecular tools are available now to study the interactions between tick-borne arboviruses such as CCHFV and the cells of their arthropod vectors (Bell-Sakyi et

Tick cell and tissue culture has progressed a long way since the first short-term cultures.

Continuous cell lines have been established from many of the economically important tick species prevalent in Europe, Africa and the Americas. The numbers of scientists and research institutes worldwide utilising tick cell lines, and the resultant outputs, are increasing exponentially. The Tick Cell Biobank has made an important contribution to this expansion.

In the long term, this will build and expand local and global capacity for vector-borne disease

research, leading to both locally-generated and international solutions (Bell-Sakyi et al, 2011).

2. Aim of the study

Crimean-Congo hemorrhagic fever (CCHF) is a severe tick-borne disease caused by CCHF virus (CCHFV). Ticks belonging to Ixodid order, in particular of the genus Hyalomma are the main vectors and reservoir of CCHFV.

Humans are not part of the normal zoonotic cycle of this virus and for that, humans are not adapted to this virus. In fact, CCHFV infection is asymptomatic on ticks while in humans can causes a hemorrhagic disease.

It has been observed in insects that some RNA viruses induces persistent infection in vector cells promoting host survival and viral persistence in the infected host (Goic et al., 2016; Nag and Kramer, 2017). To date, it has been observed that CCHFV establishes a persistent infection that does not cause any cytopathic effect in tick cells (Salata et al., 2018), but the mechanism allowing the persistent viral infection in tick is fully unknown.

CCHFV is classified as a risk group 4 human pathogen, representing a problem to study this virus because of restriction imposed for its manipulation. However, the closely related HAZV is not pathogenic and it has been proposed to be a good surrogate for the research on CCHFV.

The aim of this study is to evaluate the mechanism that allow the establishment of the persistent viral infection in ticks using HAZV as a model (that can be used in BSL 2 laboratory).

In this study we examine the presence of a small fragments of viral genome transformed in DNA (vDNAs) in tick cells after HAZV infection as a mechanism for cell defence against the viral infection. In fact, the presence of vDNAs induces persistent infection that promotes host survival and viral persistence in infected host in some insects. (Goic et al., 2016).

This persistent infection can allow vector survival and thus may be important for arbovirus dissemination and transmission.

3. Material and methods

3.1 Cell culture

Tick cells. Hae/CTVM8 and Hae/CTVM9 cell lines, were a kind gift from Lesley Bell-Sakyi.

Cell lines were maintained between 28°C and 32°C in an atmosphere of ordinary air in sealed flasks or flat-sided tubes.

Culture media used for Hae/CTVM8 cell line was prepared with 1:1 L-15 (additioned as reported in table 1)/H-Lac (additioned as reported in table 2) while culture media used for Hae/CTVM9 cell line were prepared with 1:1 L-15 (additioned as reported in table 1)/Minimal Essential Medium (MEM).

L15 and MEM culture media as well as media components are purchased by Gibco, UK.

Table 1.

Table 2.

VERO cells. VERO cells are cells derived from Cercopithecus aethiops’s kidney (ATCC®

CCL-81™). This cells are maintained at 37°C with 5% CO

and 5% O

in Dulbecco’s modified

L-15 (Leibovitz) medium 70 ml

Tryptose phosphate broth 10 ml

Foetal bovine serum (heat-inactivated) 20 ml

L-glutamine 200 mM 1 ml

Penicillin (10000units)/streptomycin (10000µg) 1 ml

Hanks balanced salt solution 75 ml

Lactalbumin hydrolysate 10% solution 5 ml Foetal bovine serum (heat inactivated) 20 ml

L-glutamine 200 mM 1 ml

Penicillin (10000units)/streptomycin (10000µg) 1 ml

eagle medium (DMEM) (Gibco, UK) supplemented with 10% foetal bovine serum heat- inactivated (FBSi) and 1% Penicillin (10000units)/streptomycin (10000µg) (P/S).

SW13 cells. SW13 cells are cells derived from Homo sapiens (ATCC® CCL-105™). This cells are maintained at 37°C without CO

and O

in Leibovitz's L-15 Medium (Gibco, UK) supplemented with 10% FBSi and 1% P/S.

3.2 Cell lines infections

VERO cells infection. A confluent flask of VERO cells was infected at the multiplicity of infection (MOI) of 1 or 0.1 focus forming unit (FFU)/cell in DMEM supplemented with 2%

v/v of FBSi and 1% v/v P/S. After 1 hour incubation, the viral inoculum was removed and replaced with DMEM supplemented with 2% v/v FBSi.

Tick cells infection. Hae/CTVM8 and Hae/CTVM9 cell lines were infected adding HAZV in cell line-specific culture medium. Cells lines were infected at the multiplicity of infection (MOI) of 1 or 0.1 FFU/cell and, after 1 hour incubation, the viral inoculum was removed and replaced with complete specific culture medium.

3.3 HAZV production

3,5x10

SW13 cells were seeded in T75 flask. After 24h incubation, cells were infected with

HAZV at the MOI of 0.1 in Leibovitz's L-15 Medium (Gibco, UK) supplemented with 2% FBSi

and 1% P/S. After 1 hour, viral inoculum was removed and replaced with culture medium with

2% FBSi. After this, cells were placed at 37°C for 48 hours and then the supernatant was

collected and viral titration was performed performing immunofluorescence assay.

3.4 DNA extraction

DNA was extracted from infected tick cells pellet using the DNeasy Blood & Tissue Kit (QIAGEN) following handbook’s instruction.

3.5 DNase treatment

For DNase treatment the DNase I recombinant (Roche) was employed.

Briefly, to degrade DNA, a mix containing 3µl of nucleic acids, 2µl of 10x incubation buffer, 1µl of recombinant DNase I, and 13µl of RNase and DNase free water, was incubated for 10 minutes at 30°C. Next, reaction was stopped by adding 2µl of 0,2M EDTA (pH 8.0) to a final concentration of 8mM and heating up to 75°C for 10 minutes.

3.6 RNase treatment

For RNase treatment RNase A (Roche) was employed.

The reaction was performed following handbook’s instructions.

3.7 Polymerase Chain Reaction (PCR)

PCRs were performed on DNA extracted from infected tick cells with 9 pairs of primers mapping within the S segment of HAZV genome. Primers used in this study were designed on the genomic S segment of HAZV, using the program “Primer 3” available online. Primers’

name and their sequences are shown in Table 3.

S1-fw CCT CAC AAC ATC AGC GAG AA S1-rev CCT CAA TGA GTG CTG AGC TG S1b-fw GAG AGG GAC GCC ATC TAC AG S1b-rev CAG ACC TTC ACG GTG TCA GA S2-fw TCT GAC ACC GTG AAG GTC TG S2-rev CAC TGT CCC TGG CAC CTT AT

S3-fw TAT GAT TCG ACG CAG GAA CA S3-rev GCC TGG TCC ATC ACG TCT TT S4-fw GCA TTG CAC AAC TTG AAG GA S4-rev CTG CCT TCC AAA GCC AGT AG S5-fw CCA GTC ACC TTC CCA TCT GT S5-rev AGC TCG GAC ATT CTT CCA GA S6-fw AAA TGG GGA AAA AGG CTC AT S6-rev TTG AGG ACT GCC TTT GTG TG S7-fw CAC ACA AAG GCA GTC CTC AA S7-rev GCC GCA ATC CTA GAT GAT GT S8-fw CGC CTC TGA ACA TCT TCT CC S8-rev GCA CAG CCC CAA ATT TTA CA

Table 3. Primers name and sequences

Each PCR was performed as follows:

MIX

5uL 10x PCR buffer with MgCl

1uL dNTPs mix 2uL Primer forward 2uL Primer reverse 0.5uL Taq Gold 36.5uL H

O DEPC 50ng DNA

Cycling conditions are shown in the following table:

10’ 95°C

15’’ 95°C

30’’ 60°C

45’’ 72°C

5’ 72°C

Twenty microliters of each PCR product were loaded in a 2% (g/v) agarose gel.

3.8 RNA extraction

RNA was extracted from cell pellet by using the RNeasy Mini Kit (QIAGEN) following handbook’s instruction.

3.9 RT-PCR

Before RT-PCR, RNA samples were treated for 30’ at 37°C with DNase I (Roche). Next, DNase was inactivated by incubation for 15’ at 95°C.

RT-PCR mix was as follows:

RNA 2-4 µg

MgCl

11 µl

Buffer 10x without Mg

5 µl

dNTPs 2,5 mM 10 µl

RNAseIN 1,5 µl

MuLV 1,5 µl

Random Examer 2,5 µl

H

O DEPC -

Final Volume 50 µl

Cycling conditions are shown in the following table:

25°C 10’

42°C 60’

95°C 3’

4°C Forever

3.10 MTT test (Methyl-Thiazole Tetrazolium Test)

The MTT assay is a colorimetric in vitro assay that measures cell viability based on the mitochondrial metabolism.

In a 96-wells plate, 100000 Hae/CTVM8 cells/well were seeded in the specific culture medium.

Control was represented by wells containing only the cell-free medium. Subsequently plates were placed at 31°C in dry air condition. After the incubation period dependent on the different condition analysed, 10 µl of solution 1, or MTT Reagent, were added to each well to induce the formation of formazan salts, and after maximum 4 hours at 37°C (5% O

, 5% CO

), when purple precipitates were formed, 100 µl of solution 2 (10% SDS in 0.01M HCl), were added to solubilized the formazan salts (insoluble). Then the plate was incubated overnight at 37°C (5%

O

, 5% CO

). Next, the plate was read by the spectrophotometer at 620 nm to quantify the formazan salts which are directly proportional to the mitochondrial metabolism and thus to the amount of metabolic active cells.

3.11 Azidothymidine (AZT)

Azidothymidine (AZT) was purchased by Sigma-Aldrich.

1 gram of AZT was dissolved in dimethylsulfoxide (DMSO) to a final concentration of 5M.

For the experiments the stock solution was diluted in medium culture needed by cells used until the concentration required for the experimental procedure.

3.12 Ribavirin

Ribavirin was purchased by Sigma-Aldrich.

1 gram of Ribavirin was dissolved in water to a final concentration of 100 mM. For the

experiments the stock solution was diluted in medium culture needed by cells used until the

concentration required for the experimental procedure.

3.13 UV-inactivated HAZV

To inactivate HAZV virus, a slight layer of viral suspension has been placed on a dish. A UV lamp was placed at a very close distance to the dish containing viruses. The virus has been exposed to UV rays for 1 minute.

3.14 Viral titration

Immunofluorescence was used to determine viral load in supernatants or in viral stock.

In a 96-wells plate, 20000 VERO cells/well were seeded in 150 µl/well of DMEM 10% FBSi and incubated at 37°C (%5 O

, 5% CO

). After 24 hours of incubation at 37°C (%5 O

, 5%

CO

), VERO cells were infected with serial dilutions of the supernatant (150 µl/well) using DMEM 2% FBSi. After 1 hour incubation at 37°C (%5 O

, 5% CO

), supernatant was removed and 150 µl/well of DMEM 2% FBS were added. After another overnight incubation at 37°C (%5 O

, 5% CO

), medium was removed and cells were fixed using 60 µl/well of acetone:methanol (with a ration of 1:1) for 45 minutes al -20°C.

After this, nonspecific sites were blocked using 60 µl/well of PBS 1x/2,5% BSA incubating 1 hour at room temperature. Subsequently, the plate was incubated 1 hour and 30 minutes at 37°C with 60 µl/well of anti-nucleoprotein antibody (a-NP) diluted 1:200 in PBS 1x/2,5%BSA/0,1%

TWEEN. After this incubation, the plate was incubated 1 hour at 37°C in the dark with 60 µl/well of Alexa Fluor

488 goat anti-rabbit IgG (Invitrogen) diluted 1:800 in PBS 1x/2,5%

BSA/0,1% TWEEN/DAPI 1:1000.

After this last incubation it is possible to count foci formed by the viral infection using fluorescence microscope and calculate viral load as follow:

FFU/mL= n°foci/diluition x 0,06mL

3.15 Cloning

In order to obtain a positive PCR control and to prepare the real-time PCR standard curve, the

PCR product for the detection of HAZV was cloned by the using the CloneJET PCR cloning

kit which offers a high efficiency for cloning of PCR products.

The vector used is pJET1.2 /blunt (Figure 1). The cloning was performed following handbook’s instructions.

Figure 1. pJET1.2 vector

The bacterial colonies that were obtained, were screened to evaluate if the cloning was successful. From one of the colonies that gave a positive result, the plasmid was prepared according to the manufacturer's instructions.

To perform a qRT-PCR a serial dilution of this plasmid were used.

3.16 Real Time PCR

Real Time PCR was performed using two pairs of primers called HAZV-F/HAZV-R (mapping within an HAZV gene) and EF1A-Fw/EF1A-Rev (mapping within a tick cells gene). Primers and probe sequences are shown in the following tables.

Probe 6-FAM-TGTGAGTGGCAGAGAATGTGGTGGGT-TAMRA

EF1A-Fw GGGAGGTGAATAGTTAGGGT

EF1A- Rev GCGCCAGTCATCTAAGGAA

Probe 6-FAM-TGAAGGATGGGTCAAAGA-MGB

HAZV-F CAAGGCAAGCATTGCACAAC

HAZV-R GCTTTCTCTCACCCCTTTTAGGA

Mix used for these reactions is:

12,5µl MasterMix TaqMan 2x 1µl Primer Forward 18µM 1µl PrimerReverse 18µM 9,3µl H

O

0,2µl Probe 25µM

3.17 Proteins extraction

To extract proteins cells were lysed in CHAPS lysis buffer (10mM Tris-HCl pH 7.5, 400mM NaCl, 0,7 mMnCl

1mM MgCl2, 1mM EGTA, 0,5% CHAPS, 10% glycerol, freshly supplemented with complete EDTA-free protease inhibitors cocktail and 1mM DTT). After incubation at 4°C for 10 minutes, cell debris were removed by centrifugation at 15000 rpm for 10 minutes at 4°C. Supernatants were transferred to clean tube. Total protein concentration was determined by using Micro BCA protein Assay Kit (Thermo Scientific), following the manufactures’ instructions.

3.18 In vitro Reverse Transcriptase assay

Reverse transcriptase assays were carried out for 15 minutes at 25°C in a total volume for

reaction of 50 µl containing 4µg of protein sample, 320ng of PAGE-purified oligo(dT)18,

500ng of poly(rA), and 1µl of 84 Ci/mmol 3H-dTTP in 50 mM Tris-HCl (pH 7.5), 50mM KCl,

5mM MgCl

, 5mM DTT and 0,1% Triton X-100. After this, all of 50 µl of the reaction were

spotted on an ion paper that retains incorporated nucleotides but not free dNTPs. Paper were

washed 3 times (10 minutes each wash) with saline-sodium citrate buffer and were soaked into

4mL of liquid scintillation cocktail. Radioactivity was measured using Scintillator (TRI-carb

2819 TR, Perkin Elmer).

4. Results

In the field of emerging viral diseases, the study of the interplay between virus and vector is becoming central to understand the mechanisms allowing viral persistence and spreading to humans. The persistent infection in vectors, increases the chance of viral dissemination to humans and guarantees the persistence of the virus in the environment. The molecular mechanisms allowing viral persistence in the vector are at the moment largely unknown.

The manipulation of CCHFV is very difficult, due to the fact that its use is limited to BSL-4 laboratory.

To get light in the virus/vector interaction, in this study we selected tick cell lines in combination with HAZV (BSL-2 pathogen) as model.

This is possible because HAZV belongs to the same CCHFV serogroup and shows similar disease progression to that seen in Crimean-Congo Hemorrhagic Fever infections in suckling mice and in the interferon receptor knockout mouse model (Bereczky et al., 2010).

With this research activity, we wanted to characterize the interaction between tick cells and

HAZV, as a model for CCHFV and ticks, to try to understand in depth the molecular

mechanisms that can help this dangerous pathogen to persist in its reservoir and in the

environment.

4.1 Tick cells infected with HAZV develop a persistent infection

First of all, we optimized the protocol for tick cells infection with HAZV.

Then, to evaluate HAZV replication in Hae/CTVM8 and Hae/CTVM9 tick cell lines, we performed tick cells infection with MOI 0.1 and MOI 1.

Viral replication was monitored evaluating the viral progeny release and the yield of intracellular viral RNA. We collected the culture supernatant and the tick cells pellets every 24 hours from T0 (namely after 1 hour incubation of tick cells with HAZV) until 7 days post infection.

The viral titer was evaluated in the supernatants determining the focus-forming unit per millilitre (FFU/mL). As indicated in Figure 1 we observed that HAZV replicates faster in Hae/CTVM9 cells during the earlier time point then decreased. In contrast, in Hae/CTVM8 cells viral titer increased slowly but continuously. Finally, at 7 days post infection, the viral titer was very similar in the supernatants collected from both the two tick cell lines.

Figure 1. Viral load in infected Hae/CTVM8 and Hae/CTVM9 (MOI 1 and MOI 0.1) at 1, 2, 3, 4 and 7 days post infection.

In the other hand, total RNA was extracted from cell pellets and analysed by a reverse transcription combined with a real-time PCR approach (real-time RT-PCR) to evaluate the kinetic of HAZV RNA replication in the two different tick cell lines. The amount of viral RNA was normalized to expression of the putative translation elongation factor EF-1 alfa (EF1A)

0 2 4 6 8

10³ 10⁴ 10⁵ 10⁶ 10⁷ 10⁸ 10⁹

60 70 Days p.i.

Viral titer (FFU/mL) HAE/CTVM8 MOI 0.1

HAE/CTVM8 MOI 1 HAE/CTVM9 MOI 0.1 HAE/CTVM9 MOI 1

endogenous gene of H. anatolicum ticks and expressed as fold changes with respect to the initial virus inoculum, set to 1 at day 0 (Figure 2).

Figure 2. Kinetic of infection of HAZV in Hae/CTVM8 and Hae/CTVM9 tick cell lines. The two tick cell lines were both infected with MOI 1and MOI 0.1. The viral RNA yield was evaluated by real-time RT-PCR.

Accordingly, with the previous data of viral titration, as shown in Figure 2, HAZV replicates faster in Hae/CTVM9 cells than in Hae/CTVM8 cells but, 7 days post infection, the yield of intracellular viral RNA is quite similar suggesting that all the cells are infected.

To investigate whether the different behaviour of the viral replication in the two cell lines is associate to cytopathic effects, we monitored the infected cells by using a phase-contrast microscope and the MTT assay to evaluate cellular metabolic activity as a measure of viability.

Microscopic observations did not show cytopathic effects (data not shown) as well as the MTT assay (Figure 3) while human SW13 cells, used as positive control, showed a strong cytopathic effects.

0 2 4 6 8

0.1 1 10 100 1000 10000

Days p.i.

Fold change

HAE/CTVM8 MOI 0.1 HAE/CTVM8 MOI 1 HAE/CTVM9 MOI 0.1 HAE/CTVM9 MOI 1

Figure 3. Percentage of vitality of infected tick cells (Hae/CTVM8 and Hae/CTVM9) and infected human cells (SW13) obtained by MTT assay.

Although our data suggested a different susceptibility of the two tick cell lines, HAZV established a persistent infection and it was possible to culture the infected cells for long time, more than 800 days in the case of Hae/CTVM8 cells.

The persistent infection of Hae/CTVM8 and Hae/CTVM9 cells was demonstrated by PCR approaches or determining the viral titer: persistent infected cells produced viral particles, even if with lower titer than the one measured at the earlier days post infection (table 1).

Days P.I. Viral titer Intracellular viral RNA

15 2.5x10

+

30 2.5x10

+

60 6x10

+

312 4x10

+

715 5x10

+

Table 1. Viral titers and presence of intracellular viral RNA at different days post infection measured in chronically HAZV-infected Hae/CTVM8.

0 2 4 6 8

0 50 100 150

Days p.i.

Cell vitality (%)

HAE/CTVM8 MOI 0.1 HAE/CTVM8 MOI 1 HAE/CTVM9 MOI 0.1 HAE/CTVM9 MOI 1 SW13 MOI 0.1 SW13 MOI 1

4.2 Presence of HAZV-derived vDNAs in infected tick cells

Recently, few studies showed that flies and mosquitos infected with RNA viruses produce viral derived DNA (vDNA) probably through the activity of an endogenous reverse transcriptase.

These vDNAs correspond to small fragments of the viral genome. It has been demonstrated, for instance, that these vDNAs molecules enhance RNAi-mediated antiviral immune response and are fundamental for persistent viral infection in insects (Goic et al., 2013; Goic et al., 2016;

Nag and Kramer, 2017).

Firstly, to determine if vDNAs are formed also in tick cells infected with HAZV, 9 pairs of primers mapping within the S segment of the HAZV ssRNA genome, were designed. These 9 pairs of primers were designed to produce amplicons in a range between 152 and 237 base pairs (Figure 4).

Figure 4. Scheme of the S segment of the HAZV genome and the 9 pairs of primers designed for its amplification.

To optimize the PCR conditions, was used the cDNA derived from the reverse transcription of total RNA extracted from VERO cells infected with HAZV (MOI 1).

Then, both tick cell lines (Hae/CTVM8 and Hae/CTVM9) were infected with HAZV at the MOI of 1 and 0.1. At different time-points (24 hours, 7 days and 15 days) post infection, DNA was extracted from tick cells and analysed performing nine PCR reactions to investigate the presence of vDNAs. As DNA extraction and amplification control, we used primers for the putative translation elongation factor EF-1 alfa (EF1A) endogenous gene of H. anatolicum tick.

As a negative control we used DNA extracted from not infected tick cells.

S segment

As showed in the Figure 5 and in the Figure 6, in infected Hae/CTVM8 and Hae/CTVM9 vDNAs were detectable. These vDNAs appear early after the infection (at 24 hours post infection) and they increase up to 7 days to decrease following the next week. However, in persistently infected cells we were able to detect always 4-9 fragments.

To verify the specificity of vDNA detection in tick cells and not in mammalian cell lines, VERO cells were infected with HAZV (MOI 1). The DNA was extracted 72 hours post infection and the PCR did not show the presence of any vDNAs.

As additional control of specificity, to be sure that vDNAs were obtained by the amplification of DNA fragments derived by the viral RNA genome, DNA extracted from infected tick cells was treated with DNase or RNase before performing the PCR: when RNase treatment was performed, vDNAs were amplified and on the contrary, when DNase treatment was performed vDNAs were not amplified.

Our results clearly indicate that HAZV infection induces the formation of vDNAs in infected Hae/CTVM8 and Hae/CTVM9 cells indicating that the generation of vDNAs is not cell line specific. In addition, this phenomenon seems to be independent from the employed MOI (Figure 6).

Results are summarized in Figure 6 and an example of PCR for vDNAs is shown in Figure 5.

Figure 5. Detection of HAZV-derived vDNAs in infected Hae/CTVM8 and Hae/CTVM9 (MOI 0.1) at 7 days post infection.

MVIII 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

Hae/CTVM8 Hae/CTVM9

152-237 bp

Figure 6. Detection of vDNAs in Hae/CTVM8 and Hae/CTVM9 cells infected with HAZV at MOI 1 and MOI 0.1. These data refer to three independent experiments. Reported values are the number of vDNAs detected at each time point normalized on the 9 fragment detectable and express as percentage.

Of note, vDNAs seem to be produced quickly in Hae/CTVM9 than in Hae/CTVM8 cells (Figure 5), probably it is related to the faster viral replication observed in Hae/CTVM9 cells (Figure 2).

4.3 A reverse transcriptase activity in tick cells allows vDNAs synthesis

The synthesis of vDNAs molecules from a viral RNA genome maybe dependent on reverse transcriptase activity, possibly originating from host-specific endogenous retrotransposons (Goic et al., 2016).

The further experiments reported in this thesis were performed using only the Hae/CTVM8 cells, which are easier to handle and grow better in comparison to Hae/CTVM9 cells.

To evaluate if a reverse transcriptase activity is detectable in tick cells and if it is modulated during viral infection, Hae/CTVM8 cells were infected with HAZV at MOI 0.1 and the RT activity was evaluated performing an RT-assay at different time points post infection.

The RT-assay results shown in the Figure 7 demonstrated a detectable reverse transcriptase activity in tick cells, and this activity does not appear significantly modulated during the HAZV infection.

1 7 15

0 20 40 60 80

Days p.i.

vDNAs (%) HAE/CTVM8 MOI 1

HAE/CTVM9 MOI 1 HAE/CTVM8 MOI 0.1 HAE/CTVM9 MOI 0.1

Figure 7. Kinetic of the reverse transcriptase activity in HAZV-infected tick cells from 24 hours to 15 days post infection.

(N.I.= not infected tick cells; h= hours; P.I.= post-infection;)

Data represent the average value and the standard deviation obtained by two independent experiments.

Thus, we decided to test the impact of the known reverse transcriptase inhibitor Azidothymidine (AZT) on vDNAs production.

AZT is a nucleoside analogue of thymidine that acts as an efficient inhibitor of the reverse- transcriptase of retroviruses. AZT is incorporated into the nascent DNA chain and stops its synthesis because does not have 3'-hydroxy group for attachment of the subsequent nucleoside triphosphate to be incorporated. AZT structure is shown in Figure 8. AZT was the first drug employed against HIV-1 replication, but under the correct salt, pH and concentration conditions it has been shown to inhibit reverse transcriptase of different origins (Goic et al., 2013)

0 2 4 6 8

0 500 1000 1500 2000

14 16

Days p.i.

Counts per minutes

Figure 8. Azidothymidine (AZT) structure

First of all, an MTT test was performed to evaluate tick cells viability in the presence of different concentration of AZT. Specifically, 1 mM, 5 mM and 10 mM were tested. From the results shown in the Figure 9, tick cells showed a moderate effect on the vitality in the presence of 1mM and 5mM AZT while at 10mM we observed a reduction of vitality of 53%.

Figure 9. MMT assay on tick cells 72 hours post-treatment with different concentration of AZT: 1mM, 5mM and 10 mM. These data refer to three independent experiments.

To evaluate the effect of the AZT on the vDNAs synthesis, we performed a time of addition experiment by adding 5mM AZT on Hae/CTVM8 cell line during the infection with HAZV.

1 5 10

0 20 40 60 80 100

[AZT]

Cell vitality (%)

To this end, three different conditions were evaluated: in the first one, after a pre-treatment with 5 mM AZT for 6 hours, a mix containing HAZV (MOI 1) and AZT was added to the cells (Pre+post). In the second one, tick cells were pre-treated with 5 mM AZT for 6 hours. Next, AZT was removed and cells were infected with HAZV MOI 1 (Pre). In the third one, pre- treatment was not performed and the mix containing 5 mM AZT and HAZV (MOI 1) was directly added to tick cells (Post) (Figure 10).

Figure 10. Time line of the time addition experiments to investigate the effect of AZT on the formation of vDNAs.

As a positive control for vDNAs detection, Hae/CTVM8 cells were infected with HAZV (MOI 1) without any AZT treatment.

Seven days later, the time point corresponding to the pick of vDNAs formation (figure 6), tick cells were collected and vDNAs formations was evaluated.

PCRs performed on DNA extracted from infected tick cells treated with AZT, show a reduction at least of the 55% in vDNAs formation following the AZT treatment. The strongest condition (reduction of ≈ 70%) was observed with addition of AZT before and during infection (Figure 11).

6h 7 days

AZT AZT+HAZV PCR

Viral titration

6h 7 days

AZT HAZV PCR

Viral titration

6h 7 days

/ AZT+HAZV PCR

Viral titration

Pre

Post

Figure 11. Effect of different AZT treatments on the vDNAs formation. The presence of vDNAs is expressed in percentage. These data refer to two independent experiments.

These findings support the idea that vDNAs are formed through a reverse transcriptase activity present in tick cells that acts directly on the viral genome during its replication.

As mentioned above, the 9 pairs of primers used to detect the presence of vDNAs, cover the entire sequence of the S segment of the HAZV genome. In order to evaluate the efficiency of the RT activity in the conversion of viral RNA in vDNAs, we tried different combination of primers to detect the presence of big amplicon or the detection of a vDNAs corresponding to the entire S segment. Results we obtained showed that the primers combinations failed to generate PCR products (data not shown). This fact suggests that vDNAs were derived from multiple single reverse transcription reactions involving small parts of the viral RNA.

4. 4 vDNAs are involved in the persistent viral infection and cell survival

As above reported, vDNAs are detectable only in tick cells infected with HAZV and not in uninfected tick cells or in infected mammalian cells. Ticks are the only natural reservoir and the vector of CCHFV and when they are infected by the virus, don’t show any deleterious effects. Therefore, to evaluate the importance of vDNAs in the establishment of the persistent

pre pre+post

post 0

20 40 60 80

AZT treatment

vDNAs (%)