Emerging flaviviruses (genus Flavivirus
, family Flaviviridae
) are transmitted by blood-sucking arthropods, such as ticks or mosquitoes, and are causative agents of serious human diseases such as dengue fever, yellow fever, West Nile fever, Japanese encephalitis, and tick-borne encephalitis (1
). More than 400 million clinical cases of flavivirus-induced infections are reported annually worldwide, which are in many cases fatal (2
). No approved therapy is currently available against infections caused by medically important flaviviruses. Development of new and effective antiviral drugs and therapeutic strategies for infections caused by emerging flaviviruses and other viruses responsible for life-threatening diseases is extremely important. It is particularly crucial in the current era of increased global travel, as well as emerging issues with increasing numbers of zoonotic infections due to the loss of animal habitats, and the growing spread of viral vectors as a result of climate change.
Nucleoside/nucleotide analogues can alter essential biochemical processes by sufficiently mimicking the structure of natural nucleosides/nucleotides for cellular or viral enzyme recognition. This capability makes these compounds attractive candidates for treating various diseases, including those resulting from viral (3
), bacterial (5
), fungal (6
), or parasitic infections (7
), or various types of cancers (9
). As antivirals, nucleoside analogues operate via numerous modes of action, among which suppression of viral nucleic acid synthesis is considered particularly important due to their highly specific interactions with viral polymerases that subsequently result in premature DNA/RNA chain termination (11
). Other modes of action for antiviral nucleosides include: (i) blocking the viral methyltransferases responsible for viral RNA methylation and capping (12
); (ii) suppression of de novo
nucleotide biosynthesis and depletion of the cellular nucleotide pool (14
); (iii) accumulation of mutations in viral genomes, leading to error catastrophe (15
); and (iv) immunomodulation that promotes the Th1 lymphocyte–based antiviral response (17
Nucleoside analogues modified with a fluoro substituent at different positions of a sugar ring and/or heterocyclic (purine/pyrimidine) nucleobase were initially synthesized in the 1970s. Soon after, researchers noted unique properties that the fluorine imparts to the nucleoside scaffold (18
). In drug design, fluorine is often used as an isosteric replacement because of its similar size to hydrogen, as well as its similar electronegativity to the hydroxyl moiety in ribo/deoxyribonucleosides (20
). Because of the exquisite electronegativity of fluorine, this substituent significantly influences the conformational properties of the nucleoside sugar ring by “locking” it into a specific conformation, e.g., in C2′
-endo or other variations of the envelope/half-chair pentose conformations. This effect can substantially influence the recognition of a nucleoside analogue by DNA/RNA polymerases, reverse transcriptases, and nucleoside/nucleotide kinases. These differences occur because each enzyme prefers a different nucleoside/nucleotide conformation, with the ultimate result being efficient enzyme inhibition and cessation of viral replication (21–23
). Furthermore, fluorine increases the stability of neighboring bonds (e.g., N
-glycosidic or phosphoester bonds), which renders fluoro-modified nucleosides resistant to unwanted catabolic degradation by nucleoside phosphorylases, esterases, and other intracellular hydrolases (21
). Finally, fluorine, when incorporated into a pyrimidine or purine base, considerably alters the steric and electronic properties of the base, as well as the hydrogen bonding interactions between the enzyme active site and nucleoside analogue (25–27
). Based on these unique properties, use of fluorine substituents is widely agreed to be an advantageous drug modification, and numerous fluorine-substituted nucleoside-based drugs have been developed. Many of these candidates have shown potent anticancer activity (e.g., gemcitabine [2′-dideoxy-2′,2′-difluorocytidine] [28
] or floxuridine [2′-deoxy-5-fluorouridine] [29
]), while others have been approved to treat serious and life-threatening viral infections, e.g., sofosbuvir, a McGuigan ProTide of 2′-fluoro-2′-methyluridine for treatment of chronic hepatitis C (HCV) infections (30
Here, we have evaluated the antiflaviviral activity of a series of 28 nucleoside analogues modified with a fluoro substituent at different positions of the ribose ring (predominantly at C2′ and C3′) and/or at the C2 or C5 position of the heterocyclic nucleobase. We also tested several fluoro-modified arabino nucleosides. Our antiviral screening revealed that the vast majority of the investigated compounds showed no antiflaviviral effect or were substantially cytotoxic at the tested concentrations. Among the compounds we tested, however, 3′-deoxy-3′-fluoroadenosine exerted a high antiflaviviral potency in vitro, showing low-micromolar antiviral effects against tick-borne encephalitis virus (TBEV), Zika virus (ZIKV), and West Nile virus (WNV). 3′-Deoxy-3′-fluoroadenosine displayed observable cytostatic effects at high concentrations, but was well tolerated in the tested cell lines at compound levels of <12.5 μM. In addition, a quantitative phase imaging (QPI) approach based on high-resolution holographic microscopy was developed and optimized for advanced characterization/description of antiviral efficacy and cytotoxicity of 3′-deoxy-3′-fluoroadenosine in cell culture. Finally, we also demonstrated an antiviral effect of 3′-deoxy-3′-fluoroadenosine in mouse models of TBEV and WNV infection. To the best of our knowledge, this study is one of the few to describe an antiviral effect of a fluoro-substituted nucleoside against emerging flaviviruses. It also is the first to demonstrate antiviral activity of fluorinated nucleosides against TBEV, a virus responsible for serious neuroinfections in Europe and Northeast Asia. Moreover, our work demonstrates that fluoro-modified nucleoside scaffolds represent an interesting group of bioactive molecules characterized by unique structural properties with potential for use in antiviral research, drug development, and structure optimization as prospective broad-spectrum antivirals.
Fluoro-substituted nucleosides show unique chemical, biochemical, and biological properties and are widely used as treatment for numerous viral, bacterial, fungal, and protozoal infections, in addition to many cancers (18
). So far, however, only a few of these compounds have been tested against flaviviruses to assess their antiviral in vitro
and in vivo
). In this study, we evaluated the antiviral activity of a series of 28 fluoro-modified nucleosides against TBEV in the PS cell line. To our surprise, we found little to no inhibition of virus-induced CPE formation by almost all of the tested compounds, and some of them proved to be highly cytotoxic. No antiviral activity or high toxicity was observed in particular with nucleosides modified with a fluoro-substituent at C2′
, or C5
. Inactivity of 2′-fluoro-modified nucleosides against TBEV could be explained by elimination of the 2′-hydroxy hydrogen bond donor, which results in a dramatic decrease in hydrogen-bonding capacity with the polymerase or the incoming nucleoside triphosphate and the inability of such nucleosides to trigger inhibition of viral RNA synthesis (35
Changes in stereochemistry at the C2
′ position from the ribo-configuration to the arabino-orientation also can lead to compound inactivity, as illustrated with a series of 2′-deoxy-2′-fluoroarabinonucleosides tested in this study. Similarly, the addition of a fluoro-substituent to the C2/C5
positions of a purine/pyrimidine yielded compounds with no anti-TBEV efficacy, likely due to electronic or steric hindrance with the viral NS5 RdRp active site (35
). Alternatively, inactivity of these nucleoside analogues may be the result of inefficient compound uptake mediated by specific nucleoside transporters, low conversion efficacy into active (phosphorylated) forms by cellular kinases, or a high degradation rate caused by nucleoside/nucleotide catabolic enzymes (4
). As a result, those mechanisms could be interesting targets for further investigation.
As might be expected, nucleosides lacking the 3′-hydroxyl group could be potent inhibitors of flaviviral NS5 RdRp-mediated RNA synthesis, acting as obligate RNA chain terminators (11
). Indeed, 3′-deoxy-3′-fluoroguanosine 5′-triphosphate was previously described as interacting directly with NS5B RdRp of HCV, resulting in suppression of viral RNA synthesis by disruption of further extension of the replicating viral RNA (35
). The potency to inhibit HCV NS5B polymerase activity was also observed in 3′-deoxy-3′-fluoroadenosine 5′-triphosphate; however, the inhibitory activity was considerably lower than its guanosine counterpart (35
). Our in vitro
antiviral assays revealed that two 3′-deoxy-3′-fluoro–substituted nucleosides, 3′-deoxy-3′-fluoroguanosine and 3′-deoxy-3′-fluorouridine, did not suppress multiplication of TBEV in PS cells. In contrast, 3′-deoxy-3′-fluoroadenosine, a nucleoside first synthesized in the late 1980s at the Rega Institute for Medical Research in Belgium (36
), showed potent, low-micromolar antiviral inhibition of in vitro
TBEV replication. Moreover, this compound suppressed replication of two other medically important flaviviruses, WNV and ZIKV, in PS cells, as well as in primary HBCA cells. The antiviral effect was stable over time, with inhibition of viral replication observed at 48 and 72 h p.i., and the compound was highly active even if added to PS cells at 2 h after infection. The ability of 3′-deoxy-3′-fluoroadenosine to inhibit multiple arthropod-borne flaviviruses demonstrates potent, broad-spectrum antiviral activity across the genus Flavivirus
. Broad-spectrum inhibitory effects for 3′-deoxy-3′-fluoroadenosine have also been described in earlier papers for numerous RNA viruses of the Picornaviridae
, and Arenaviridae
Notably, we did not observe any measurable in vitro
cytotoxicity of 3′-deoxy-3′-fluoroadenosine in PS and HBCA cells within compound concentrations used for antiviral activity evaluation (0 to 25 μM). This result is in agreement with earlier studies showing a minimum cytotoxic concentration of 40 μM for primary rabbit kidney cells, Vero cells, and HeLa cells (36
). However, 3′-deoxy-3′-fluoroadenosine did have observable cytostatic effects at 25 μM in PS cells and at 12.5 μM and 25 μM in HBCA cells. Cytostatic properties of 3′-deoxy-3′-fluoroadenosine were previously described for murine leukemia cells (L1210), human B-lymphoblast (Raji) cells, human T-lymphocyte (H9) cells, and human T4-lymphocytes (MT-4), acting at concentrations ranging from 1.6 to 23 μM (37
It could be speculated that the cytostatic activity is related to low selectivity of 3′-deoxy-3′-fluoroadenosine 5′-triphosphate for viral RdRps over other types of polymerases. The extremely broad range of viruses sensitive to this nucleoside, including vaccinia virus among DNA viruses, supports this hypothesis (36
). In accordance with this claim, 3′-deoxy-3′-fluoroadenosine has also been described as terminating RNA synthesis catalyzed by DNA-dependent RNA polymerase from E. coli
). In addition, 3′-deoxy-3′-fluoroadenosine inhibited incorporation of 3
H-labeled uridine into cellular RNA, resulted in abrogation of cellular RNA synthesis in actively growing Vero cells (38
). Alternatively, the broad range of antiviral activity of 3′-deoxy-3′-fluoroadenosine and the reported suppression of cellular RNA synthesis could also be explained as a result of compound-mediated inhibition of purine/pyrimidine biosynthesis and nucleotide depletion, as was recently described for a broad-spectrum antiviral and anticancer drug gemcitabine (2′-dideoxy-2′,2′-difluorocytidine) (41
). In order to explain a broad spectrum antiviral efficacy of 3′-deoxy-3′-fluoroadenosine, this compound was also tested for its potential inhibition of S-adenosylhomocysteine (SAH) hydrolase. However, no inhibitory activity to SAH hydrolase was demonstrated, unless it was used at a rather high concentration (36
Based on our observation that 3′-deoxy-3′-fluoroadenosine showed a strong antiviral potency in cell culture (in vitro
), we evaluated its activity in vivo
using mouse models of WNV and TBEV infection. Treatment of WNV-infected mice with 25 mg/kg of 3′-deoxy-3′-fluoroadenosine twice a day resulted in a significant decrease of mortality and in a substantial elimination of clinical signs of neuroinfection. In TBEV-infected mice, the antiviral effect of 3′-deoxy-3′-fluoroadenosine (25 mg/kg/2×day) was not as strong; however, the compound administration resulted in significantly longer survival time and slower progress of development of clinical signs compared with control animals. A lower efficacy of 3′-deoxy-3′-fluoroadenosine against TBEV (strain Hypr) in vivo
could be explained by a rapid/aggressive course of TBEV infection in mice, where the average survival time of TBEV-infected mice was shorter than that of WNV-infected mice (9.0 ± 1.0 days versus 12 ± 1.6 days, respectively). 3′-Deoxy-3′-fluoroadenosine was previously demonstrated to show high efficacy in suppressing the formation of tail lesions in vaccinia virus–infected rodents. The compound, administered intravenously for 5 days at concentrations of 50 and 100 mg/kg/day, reduced the number of pox tail lesions by 25% and 80%, respectively. A dose of 200 mg/kg/day was reported to be lethal for mice (36
Previous studies have shown that nucleosides with antiviral activity against arthropod-borne flaviviruses contain one of the following sugar ring modifications: (i) a methyl substitution at C2
), (ii) an ethynyl substitution at C2
), or (iii) an azido substitution at C4
). An observable (but weaker) antiflaviviral effect also has been reported for nucleosides with (iv) a cyano substitution at C1
). These chemical modifications are used in combination with appropriate heterocyclic bases or their modified counterparts (e.g., with 7-deazaadenine) and/or with modifications of N
-glycosidic bond (e.g., switching to a C
-glycosidic bond) (4
). Herein we have described the first results showing that a fluoro-substitution at the C3
′ position of adenosine represents another useful nucleoside modification that exhibited low-micromolar inhibitory activity against emerging flavivirus replication. While the increased cytostatic activity of 3′-deoxy-3′-fluoroadenosine in vitro
limits its potential use for further clinical applications; however, additional chemical modifications could improve this compound’s cytotoxic/cytostatic profile. This is exemplified by 2′-deoxy-2′-fluorinated nucleosides, in which the addition of C2
′ methyl and C5
′ phosphoramidate groups led to increased nucleoside selectivity toward viral RdRp, manifested as high inhibitory potency and low toxicity. These characteristics have been demonstrated for the anti-HCV drug sofosbuvir (47
), illustrating the considerable potential of fluoro-substituted nucleoside analogues as prospective starting points in developing effective antivirals for treating flavivirus-associated diseases.
MATERIALS AND METHODS
This study was carried out in strict accordance with Czech law and guidelines for the use of experimental animals and protection of animals against cruelty (Animal Welfare Act 246/1992 Coll.). All procedures were reviewed by the local ethics committee and approved by the Ministry of Agriculture of the Czech Republic (permit no. 13522/2019-MZE-18134).
Cell cultures, virus strains, and antiviral compounds.
PS cells (31
) were cultured at 37°C in Leibovitz (L-15) medium. HBCA (ScienCell, Carlsbad, CA) cells were cultivated in Astrocyte medium (Thermo Fisher Scientific). The media were supplemented with 3% (L-15) and 6% (Astrocyte medium) newborn calf serum and 100 U/ml penicillin, 100 μg/ml streptomycin, and 1% glutamine (Sigma-Aldrich, Prague, Czech Republic).
The following emerging flaviviruses were tested: TBEV (strains Hypr and Neudoerfl, both members of the West European TBEV subtype, provided by the Collection of Arboviruses, Institute of Parasitology, Biology Center of the Czech Academy of Sciences, Ceske Budejovice, Czech Republic http://www.arboviruscollection.cz/index.php?lang=en
), ZIKV (African strain MR-766 and Brazilian strain Paraiba_01, kindly provided by Carla Torres Braconi and Paolo M. de A. Zanotto, University of Sao Paolo), and WNV (Eg-101, a member of genomic lineage 1, originally isolated from human serum in Egypt, and 13–104, a representative of genomic lineage 2, isolated from the Culex modestus
mosquito in the Czech Republic).
Nucleoside analogues were purchased from Carbosynth (Compton, UK) (2′-deoxy-2′-fluoroadenosine, 2′-deoxy-2′-fluoroguanosine, 2′-deoxy-2′-fluorocytidine, 2′-deoxy-2′-fluorouridine, 2′-deoxy-2′-fluoro-2-fluoroadenosine, 2′-deoxy-2′-fluoro-N2-isobutyrylguanosine, 2′-deoxy-2′-fluoroisoguanosine, 2′-deoxy-2′-fluoro-5-methylcytidine, 2′-deoxy-2′-fluoro-5-iodouridine, 2′-deoxy-2′-fluoro-5-methyluridine, 2′-deoxy-2′-fluoroarabinoadenosine, 2′-deoxy-2′-fluoroarabinoguanosine, 2′-deoxy-2′-fluoroarabinocytidine, 2′-deoxy-2′-fluoroarabinouridine, 2′-deoxy-2′-fluoro-5-iodoarabinouridine, 2′-deoxy-2′-fluoro-5-methylarabinouridine, 2′-deoxy-2′-fluoro-5-ethylarabinouridine, 3′-deoxy-3′-fluoroadenosine, 3′-deoxy-3′-fluoroguanosine, 3′-deoxy-3′-fluorouridine, 2′,3′-dideoxy-3′-fluoroguanosine, and 2′,3′-dideoxy-3′-fluorouridine) and from Sigma-Aldrich (Prague, Czech Republic) (2-fluoroadenosine, 5-fluorocytidine, 2′-deoxy-5-fluorocytidine, capecitabine, 5-fluorouridine, and 2′-deoxy-5-fluorouridine [floxuridine]). Test compounds were solubilized in 100% dimethyl sulfoxide (DMSO) to yield 10 mM stock solutions.
Initial antiviral screening with CPE reduction assay.
The 28 fluoro-substituted nucleosides were first screened at a single concentration of 25 μM for their ability to inhibit cytopathic effects (CPE) mediated by TBEV infection (strain Hypr) in PS cells. The cells were seeded in 96-well plates (approximately 2 × 104
cells per well) and incubated for 24 h to form a confluent monolayer. Following incubation, the medium was aspirated from the wells and replaced with 200 μl of fresh medium containing 25 μM of the test compound (three wells per compound) and incubated for an additional 24 h (i.e., 24-h pretreatment). DMSO was added to virus-infected cells as a negative control at a final concentration of 0.5% (vol/vol). After 24 h, the medium was aspirated again and replaced with 200 μl of compound-containing medium (25 μM) inoculated with TBEV at a multiplicity of infection (MOI) of 0.1. The CPE was monitored visually using the Olympus BX-5 microscope equipped with an Olympus DP-70 CCD camera. At 72 h p.i., cell monolayers were stained by naphthalene black. The rate of CPE was expressed in terms of cell viability as the absorbance at 540 nm by compound-treated cells relative to the absorbance by DMSO-treated cells, as described previously (49
). To analyze the anti-TBEV activity of 3′-deoxy-3′-fluoroadenosine, 3′-deoxy-3′-fluoroguanosine, and 3′-deoxy-3′-fluorouridine in more detail, we treated PS cell monolayers with these compounds at concentrations of 0, 6.25, 12.5, and 25 μM and incubated for 24 h. Then, the growth medium was aspirated and replaced with fresh medium containing the same concentrations of the tested compounds and TBEV inoculum (strain Hypr, MOI of 0.1). Viral titers were estimated from the collected media using the plaque assay after a 72-h cultivation.
Optimization of the compound treatment regimen.
PS cells were seeded in 96-well plates (approximately 2 × 104 cells per well) and incubated for 24 h to form a confluent monolayer. For testing the optimal compound treatment regimen, we performed three independent experiments, differing in the time of drug addition to infected PS cells. (i) For a pretreatment assay, 200 μl of medium with 3′-deoxy-3′-fluoroadenosine at a concentration range of 0 to 25 μM (2-fold dilutions, three wells per concentration) was added to PS cell monolayers at 24 h prior to infection. After a 24-h incubation, medium was aspirated and replaced with 200 μl of fresh compound-containing medium in the same concentration range and inoculated with TBEV (Hypr or Neudoerfl) at an MOI of 0.1. Cells were then incubated for an additional 72 h. (ii) For a simultaneous treatment, medium containing 3′-deoxy-3′-fluoroadenosine at a concentration range of 0 to 25 μM inoculated with TBEV (MOI of 0.1) was added to cells and incubated for 72 h. (iii) For the posttreatment assay, PS cells were first infected with TBEV (MOI of 0.1), and after 2 h (the time needed for virus adsorption and internalization), medium containing 3′-deoxy-3′-fluoroadenosine at a concentration range of 0 to 25 μM was added to the infected cells and incubated for 72 h. Following incubation, the viral titers were determined from the collected supernatant media by a plaque assay and used to construct dose-response curves.
Dose-response studies using the viral titer reduction assay.
To study the dose-response effect of 3′-deoxy-3′-fluoroadenosine, we used a viral titer reduction assay. PS or HBCA cell monolayers were pretreated with medium containing 3′-deoxy-3′-fluoroadenosine at a concentration range of 0 to 25 μM (2-fold dilutions, three wells per compound) for 24 h. The medium was aspirated, replaced with 200 μl of fresh medium containing the compound at the same concentration range, inoculated with TBEV (strains Hypr and Neudoerfl), ZIKV (strains MR-766 and Paraiba_1), or WNV (strains Eg-101 and 13–104) at an MOI of 0.1, and incubated for an additional 48 or 72 h. Viral titers were determined from the collected supernatant media by a plaque assay and used to construct dose-response and inhibition curves. The data used for TBEV strains Hypr and Neudoerfl (24 h pretreatment) in Fig. 2E
were reused in Fig. 3B
, as those were two identical experiments. The viral titers obtained at 72 h p.i. were used for calculation of the 50% effective concentrations (EC50
; the concentration of compound required to inhibit the viral titer by 50% compared to the control value). The individual points of the inhibition curves were calculated by transformation of virus titer values to percent inhibition according to equation 1
where A is the virus titer for the individual compound concentrations tested (0.8 to 25 μM) (PFU/ml) and B is the virus titer for a compound concentration of 0 μM (PFU/ml).
Plaque assays were performed in PS cells (to determine TBEV titers) or Vero cells (for ZIKV and WNV titers) as described previously (42
). Briefly, 10-fold dilutions of TBEV, WNV, or ZIKV were prepared in 24-well tissue culture plates, and PS (for TBEV) or Vero (for WNV and ZIKV) cells were added to each well (0.6 to 1.5 × 105
cells per well). After a 4-h incubation, the suspension was overlaid with 1.5 % (wt/vol) carboxymethylcellulose in L-15 (for PS) or Dulbecco’s modified Eagle’s medium (DMEM) (for Vero). Following a 5-day incubation at 37°C, the infected plates were washed with phosphate-buffered saline, and the cell monolayers were stained with naphthalene black. The virus titer was expressed as PFU/ml.
To measure the compound-induced inhibition of viral surface antigen expression, a cell-based flavivirus immunostaining assay was performed as previously described (42
). Briefly, PS cells were seeded onto 96-well microtitration plates and treated with the test compound at a concentration range of 0 to 25 μM (2-fold dilutions, three wells per concentration) for 24 h. After a 24-h pretreatment, the cell monolayers were infected with the appropriate flaviviruses at an MOI of 0.1 and cultured for 3 days at 37°C. After cold acetone-methanol (1:1) fixation and blocking with 10% fetal bovine serum, we incubated the cells with a mouse monoclonal antibody targeting the flavivirus group surface antigen (protein E) (1:250; antibody clone D1-4G2-4-15; Sigma-Aldrich, Prague, Czech Republic) and subsequently labeled it with an anti-mouse goat secondary antibody conjugated with fluorescein isothiocyanate (FITC; 1:500) by incubation for 1 h at 37°C. The cells were counterstained with 4′,6-diamidino-2-phenylindole (DAPI; 1 μg/ml) for visualization of the cell nuclei and the fluorescence signal was recorded with an Olympus IX71 epifluorescence microscope.
PS or HBCA cell monolayers in 96-well plates were treated with 3′-deoxy-3′-fluoroadenosine at a concentration range of 0 to 25 μM (2-fold dilutions, three wells per concentration) and cultured for 96 h (a 24-h pretreatment followed by a 72-h incubation, i.e., the same time duration as for antiviral assays). The cytotoxic/cytostatic activity of 3′-deoxy-3′-fluoroadenosine was determined in terms of (i) cell death or (ii) relative percentage of cell abundance using the CytoTox 96 non-radioactive cytotoxicity assay (Promega, Fitchburg, WI, USA) following the manufacturer’s instructions. This assay is based on quantitative measurement of lactate dehydrogenase, a stable cytosolic enzyme that is released upon cell lysis. Cell death was estimated as the percentage of colorimetric absorbance at 490 nm by the compound-treated cells relative to the absorbance by totally lysed (chemically killed) cells. We assessed relative percentage of cell abundance as the percentage of colorimetric absorbance at 490 nm by lysed compound-treated cells relative to the absorbance by lysed mock-treated cells.
Quantitative phase imaging.
Quantitative phase imaging (QPI) was performed using a multimodal holographic microscope, Q-PHASE (TELIGHT a.s., Brno, Czech Republic). PS cells were seeded in flow chambers μ-Slide I Luer Family (Ibidi, Martinsried, Germany) (approximately 2 × 104 cells per chamber) and cultivated for 24 h to form a confluent monolayer. We performed the following independent experiments based on: (i) mock-infected and mock-treated cells as a negative control (growth medium was aspirated from the chamber and replaced with fresh medium); (ii) TBEV-infected and mock-treated cells (growth medium was aspirated from the chamber and replaced with fresh medium containing TBEV [Hypr] of MOI = 0.1); (iii) TBEV-infected cells treated with 3′-deoxy-3′-fluoroadenosine at a concentration of 12.5 μM (growth medium was aspirated from the chamber and replaced with fresh medium containing TBEV [Hypr, MOI = 0.1] and 12.5 μM of the compound; virus and compound were added to the cells in the same time); (iv) mock-infected and compound-treated cells (12.5 μM) (growth medium was aspirated and replaced with fresh medium containing 12.5 μM of the compound); (v) mock-infected and compound-treated cells (25 μM) (growth medium was aspirated and replaced with fresh medium with 25 μM of the compound); and (vi) heat-killed cells used as a model of lytic cell death (a positive control). Cells were cultivated in the chambers for 40 h postinfection and the QPI parameters, such as covered area, cell dry mass, cell speed, circularity, density, and cell dynamic score were continuously monitored. To maintain ideal cultivation conditions (37°C, 60% humidified air) during time-lapse experiments, cells were placed in a gas chamber (H201-Mad City Labs [MCL]-Z100/500 piezo Z-stages; Okolab, Ottaviano NA, Italy). To image enough cells in one field of view, we chose the Nikon Plan 10×/0.30. For each treatment, nine fields of view were observed with a frame rate of 3 min/frame for 40 h.
Holograms were captured using a CCD camera (XIMEA MR4021 MC-VELETA). Complete QPI reconstruction and image processing were performed with Q-PHASE control software. Cell dry mass values were derived according to references 51
from the phase (equation 2
), where m
is cell dry mass density (in pg/μm2
is detected phase (in rad), λ
is wavelength in μm (0.65 μm in Q-PHASE), and α
is specific refraction increment (∼0.18 μm3
/pg). All values in the formula except the φ
are constant. The value of φ
(phase) is measured directly by the microscope.
Integrated phase shift through a cell is proportional to its dry mass, which enables studying changes in cell mass distribution (52
Image analysis was performed with customized MATLAB software developed by our laboratory. The analysis process consists of segmentation, the interconnection of matching cells in adjacent time frames, and extraction of the analyzed dynamical and morphological cell features. The details of the cell segmentation algorithm that we used are described in reference 53
. Cell death detection and the distinction between apoptotic and lytic cell death using QPI have been described in detail (54
We evaluated the anti-TBEV effect of 3′-deoxy-3′-fluoroadenosine using five groups of 6-week-old female BALB/c mice (purchased from AnLab, Prague, Czech Republic) as follows: group 1 (n = 10) intraperitoneally injected with 3′-deoxy-3′-fluoroadenosine at 25 mg/kg/2×day to evaluate the compound toxicity; group 2 (n = 10) was subcutaneously injected with TBEV strain Hypr (103 PFU/mouse) and treated intraperitoneally with 200 μl of 3′-deoxy-3′-fluoroadenosine at 25 mg/kg/2×day (treatment started simultaneously with infection); group 3 (n = 10) was subcutaneously injected with TBEV strain Hypr (103 PFU/mouse) and treated with vehicle (control animals); group 4 (n = 10) was subcutaneously injected with WNV strain Eg-101 (103 PFU/mouse) and treated intraperitoneally with 200 μl of 3′-deoxy-3′-fluoroadenosine at 25 mg/kg/2×day (treatment started simultaneously with infection); and group 5 (n = 10) was subcutaneously injected with WNV strain Eg-101 (103 PFU/mouse) and treated with vehicle (control animals). 3′-Deoxy-3′-fluoroadenosine was freshly solubilized in sterile saline buffer before each injection and administered to the animals twice daily for 6 days. The clinical scores and survival rates of virus-infected mice were monitored daily throughout the experiment for 28 days. Illness signs were evaluated as follows: 0 for no symptoms; 1 for ruffled fur; 2 for slowing of activity or hunched posture; 3 for asthenia or mild paralysis; 4 for lethargy, tremor, or complete paralysis of the limbs; and 5 for death. All mice exhibiting disease consistent with a clinical score of 4 were terminated humanely (cervical dislocation) immediately upon detection.
Data are expressed as mean ± standard deviation (SD) and the significance of differences between groups was evaluated using the one sample Wilcoxon test (to compare clinical scores of treated mice with control animals) or ANOVA followed by Tukey-Kramer posttest (to compare multiple QPI parameters in compound-treated/TBEV-infected cells with controls). Survival rates were analyzed using the logrank Mantel-Cox test. All tests were performed with GraphPad Prism 7.04 (GraphPad Software, Inc., San Diego, CA, USA). P < 0.05 was considered significant.