Pharmacological Characterization of TP0591816, a Novel Macrocyclic Respiratory Syncytial Virus Fusion Inhibitor with Antiviral Activity against F Protein Mutants

Human respiratory syncytial virus (RSV) is a major cause of lower respiratory tract infections in early childhood. However, no vaccines have yet been approved for prevention of RSV infection, and the treatment options are limited. Therefore, development of effective and safe anti-RSV drugs is needed. In this study, we evaluated the antiviral activity and mechanism of action of a novel macrocyclic anti-RSV compound, TP0591816. TP0591816 showed significant antiviral activities against both subgroup A and subgroup B RSV, while exerting no cytotoxicity.

ABSTRACT

Human respiratory syncytial virus (RSV) is a major cause of lower respiratory tract infections in early childhood. However, no vaccines have yet been approved for prevention of RSV infection, and the treatment options are limited. Therefore, development of effective and safe anti-RSV drugs is needed. In this study, we evaluated the antiviral activity and mechanism of action of a novel macrocyclic anti-RSV compound, TP0591816. TP0591816 showed significant antiviral activities against both subgroup A and subgroup B RSV, while exerting no cytotoxicity. Notably, the antiviral activity of TP0591816 was maintained against a known fusion inhibitor-resistant RSV strain with a mutation in the cysteine-rich region or in heptad repeat B. Results of a time-of-addition assay and a temperature shift assay indicated that TP0591816 inhibited fusion of RSV with the cell membrane during viral entry. In addition, TP0591816 added after cell infection also inhibited cell-cell fusion. A TP0591816-resistant virus strain selected by serial passage had an L141F mutation, but no mutation in the cysteine-rich region or in heptad repeat B in the fusion (F) protein. Treatment with TP0591816 reduced lung virus titers in a dose-dependent manner in a mouse model of RSV infection. Furthermore, the estimated effective dose of TP0591816 for use against F protein mutants was thought to be clinically realistic and potentially tolerable. Taken together, these findings suggest that TP0591816 is a promising novel candidate for the treatment of resistant RSV infection.

INTRODUCTION

Human respiratory syncytial virus (RSV) is a common cause of bronchiolitis and pneumonia in early childhood (1). An estimated 33.8 million children under 5 years old are reported to suffer from acute lower respiratory infection (ALRI) caused by RSV (accounting for 22% of all cases of ALRI); over 3.4 million of them need hospitalization, and 66,000 to 199,000 die annually (2). Moreover, development of severe RSV bronchiolitis under 1 year of age is associated with an increased risk of development of allergic asthma in early adulthood (3). In adults, advanced age, immunocompromised status, and the presence of cardiopulmonary disease are reported as high-risk factors for the development of severe RSV infection (4). However, no vaccines for prevention of RSV infection have been approved yet and the treatment options are limited. Palivizumab, which is a monoclonal antibody directed against the RSV fusion (F) protein, has been demonstrated as being effective for preventing hospitalization in children with RSV infection, but it is only used for high-risk patients because of the high cost (5, 6). Although a guanosine analogue, ribavirin, has also been demonstrated to show antiviral activity against RSV, it has relatively high toxicity, and its clinical efficacy remains controversial (7, 8). Therefore, new effective and safe drugs are needed to treat clinically significant RSV infections.

RSV attaches to the host cells via glycoprotein (G) and enters the cells by fusion of the viral envelope with the cell membrane mediated by the F protein (9). After invasion, the RNA-dependent RNA polymerase, which consists of a large (L) protein and phosphoprotein (P), carries out both transcription and replication using the genomic RNA-nucleoprotein (N) complex as the template (10), the viruses spread by cell-cell fusion mediated by the F protein, and a syncytium is formed (11). It is thought that for controlling RSV infection, these viral factors that are needed for productive viral infection need to be inhibited, and so far, various anti-RSV compounds targeting the F protein (presatovir [formerly GS-5806], JNJ-53718678, RV521, ziresovir [AK0529], BMS-433771, and VP-14637) (12–17), L protein (lumicitabine [ALS-8176] and PC786) (18, 19), and N protein (RSV604 and EDP-938) (20, 21) have been reported.

Among these classes of RSV inhibitors, fusion inhibitors have the advantage of being able to act without penetrating the cells—they inhibit membrane fusion outside the cells as the virus begins to invade. For this reason, fusion inhibitors are expected to have therapeutic potential, and RSV inhibitors are the test drugs in most clinical trials (12–15, 22, 23). However, it is known that this class of drugs often loses antiviral activity even with a single point mutation of the virus. Fusion inhibitor-resistant isolates most often show resistance-associated mutations in the following three regions: (i) the fusion peptide, 140 to 144 amino acids (aa) long; (ii) the cysteine-rich region, 392 to 401 aa long; and (iii) heptad repeat B, 486 to 489 aa long; the resistant mutants exhibit wide cross-resistance to other fusion inhibitors (24). Therefore, development of resistance is of critical potential concern when developing a new fusion inhibitor.

We have synthesized a number of small-molecule RSV inhibitors (25, 26), and identified some macrocyclic pyrazolo[1,5-a]pyrimidine derivatives that appear to have antiviral activity against RSV with the D486N mutation in the F protein that is resistant to known fusion inhibitors (15, 26–31). We further optimized these compounds with improved activities against the D486N mutant and obtained TP0591816 (Fig. 1). Here, we report the results of our evaluation of the antiviral activity against not only the D486N mutant, but also against known fusion inhibitor-resistant RSV with mutations other than D486N in the F protein and the mechanism of action of TP0591816. In addition, the in vivo efficacy of TP0591816 was investigated in a mouse model of RSV infection. Furthermore, analyses were conducted to predict the clinically effective dose of TP0591816 for cases of infection with the resistant virus.

FIG 1.

Chemical structure of TP0591816.

RESULTS

Antiviral activity and cytotoxicity of TP0591816.

The antiviral activities of TP591816 against wild-type and some RSV F protein mutants were evaluated by 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide salt (XTT) assay using the HEp-2 cell line (Table 1). TP0591816 showed potent antiviral activity against RSV A2 (subgroup A) and RSV 18537 (subgroup B), with 50% effective concentrations (EC₅₀) of 0.255 and 0.0824 nM, respectively. These results demonstrate that TP0591816 has more potent antiviral activity than the other known RSV inhibitors tested in this study. In addition, TP0591816 exhibited excellent antiviral activity against the D486N mutant RSV A2, with an EC₅₀ of 0.549 nM. Moreover, it also showed antiviral activity against F488L and T400I mutants, which are resistant to the known fusion inhibitors, with EC₅₀ values of 5.09 and 2.97 nM, respectively. TP0591816 showed no cytotoxicity in uninfected HEp-2 cells, with a 50% cytotoxic concentration (CC₅₀) of >30,000 nM, similar to that of other RSV inhibitors.

TABLE 1.

Antiviral activities against wild-type and F protein mutant RSV and cytotoxicity of TP0591816 and known RSV inhibitors

Compound	EC50 (nM)a for:					CC50 (nM)a
	A2 (subgroup A)				18537 (subgroup B)
	WTb	D486N	F488L	T400I	WTb
TP0591816	0.255	0.549	5.09	2.97	0.0824	>30,000
BMS-433771	50.0	>1,000	>1,000	>1,000	68.4	>30,000
VP-14637	0.467	>1,000	>1,000	>1,000	0.458	>30,000
Ribavirin	11,700	12,700	13,700	9,710	8,450	159,000

^aAll 50% effective concentration (EC₅₀) and 50% cytotoxic concentration (CC₅₀) values are represented as the geometric means of the results of at least four independent experiments.

^bWT, wild type.

Inhibition of fusion of the viral envelope with the cell membrane by TP0591816.

In order to explore the step of the viral life cycle that is blocked by TP0591816, a time-of-addition experiment was performed. TP0591816 and other known fusion inhibitors were added to HEp-2 cells at 0, 6, or 24 h postinfection (hpi). Addition of TP0591816 to HEp-2 cells at 0 hpi resulted in a reduction of the virus titer in the supernatant by approximately 4 log compared to that of the untreated control. However, the antiviral activity of TP0591816 was much lower when it was added at 6 and 24 hpi (Fig. 2). The antiviral activities of the known fusion inhibitors BMS-433771 and VP-14637 were also attenuated when they were added at 6 and 24 hpi. These results suggest that TP0591816 inhibits viral entry into the cells, an early step in the viral life cycle.

FIG 2.

Effect of time of addition of TP0591816 and known fusion inhibitors on the antiviral efficacy. HEp-2 cells were infected with RSV A2, and 4.5 nM TP0591816, 900 nM BMS-433771, or 9 nM VP-14637 was added at 0, 6, or 24 h postinfection (hpi). The virus titers in the supernatants were determined at 3 days postinfection (dpi). Data are represented as the means ± standard errors of the means (SEM) from three independent experiments performed in triplicate.

A temperature shift assay was performed to investigate the detailed mechanism of inhibition of viral entry into the cells by TP0591816. RSV can attach to cells at 4°C, but fusion requires a temperature greater than 18°C (32). Therefore, temperature shift assays can clarify whether a test compound inhibits viral attachment to a cell or fusion of the viral envelope to the cell membrane. Addition of TP0591816 at 4°C did not reduce the virus titer compared to that observed in the untreated control, but addition after shift of the temperature to 37°C led to a reduction of the viral titer, as in the case of the presence of the drug throughout the experiment (Fig. 3). BMS-433771 and VP-14637 also inhibited viral infection only when added at 37°C. In contrast, heparin, which is known as an RSV attachment inhibitor (33), did not inhibit viral infection at 37°C but inhibited it at 4°C. Collectively, these data reveal that TP0591816 inhibits fusion of the viral envelope with the cell membrane at the time of cell entry.

FIG 3.

Temperature shift assay of TP0591816 and known fusion inhibitors. HEp-2 cells were infected with RSV A2 at 4°C for 1 h, and the temperature was shifted to 37°C; 4.5 nM TP0591816, 900 nM BMS-433771, or 9 nM VP-14637 was added only at 4°C, only at 37°C, or for the entire duration of the experiment (4°C and 37°C). The virus titers in the supernatants were determined at 3 dpi. Data are represented as the means ± SEM from three independent experiments performed in triplicate.

Inhibitory effect of TP0591816 on syncytium formation.

The mechanism of inhibition by TP0591816 of RSV fusion with the cell membrane was confirmed in a syncytium formation assay. TP0591816 was added at 24 hpi to ensure viral entry into the cells, and the cell-cell fusion was evaluated. The cell-cell fusion could be completely blocked by addition of 1.5 nM TP0591816 and partially blocked by addition of the drug at a dose of 0.75 nM (Fig. 4). BMS-433771 and VP-14637 also inhibited syncytium formation. On the other hand, ribavirin, which inhibits viral RNA synthesis at an early step of the replication phase, could not suppress syncytium formation. These results suggest that TP0591816 targets the RSV F protein and can inhibit F protein-mediated cell-cell fusion.

FIG 4.

Inhibitory effect of TP0591816 and known RSV inhibitors on syncytium formation. HEp-2 cells were infected with RSV A2, and 0.75 or 1.5 nM TP0591816, 300 nM BMS-433771, 3 nM VP-14637, or 100,000 nM ribavirin was added at 24 hpi. The inhibition of syncytium formation was evaluated at 3 dpi. Syncytium scores are shown as follows: −, no syncytium; +, <1/3; ++, ∼1/3 to 2/3; +++, ≥2/3 of syncytium compared to uninfected control. Each image is representative of two independent experiments.

In vitro selection of TP0591816-resistant viruses.

To support the proposed mechanism of action of TP0591816, resistant RSV was selected in HEp-2 cells in the presence of TP0591816. RSV was serially passaged with increasing inhibitor concentrations, and resistant viruses that could grow in the presence of an almost 300-fold higher EC₅₀ of TP0591816 were obtained. Genotypic analysis of this TP0591816-resistant virus strain identified the mutation L141F in the F protein. L141F mutation is known to confer resistance to inhibitors of F protein (presatovir and RV521) (14, 22, 23).

The susceptibilities of the TP0591816-resistant L141F mutant virus to TP0591816 and other RSV inhibitors were evaluated. The L141F mutant exhibited a >4,720-fold reduction in the susceptibility to TP0591816 (Table 2). The L141F mutant also exhibited reduced susceptibility to BMS-433771 and VP-14637 while maintaining sensitivity to other class of inhibitors, such as ribavirin and palivizumab. These results confirmed that the target of TP0591816 was the F protein.

TABLE 2.

Susceptibility testing of TP0591816-resistant virus

Compound	EC50 (nM)a for:			EC50 fold difference (×)b for:
	Parent virus	Untreated control virus	TP0591816-resistant virus (L141F)	Untreated control virus	TP0591816-resistant virus (L141F)
TP0591816	0.212	0.459	>1,000	2.17	>4,720
BMS-433771	54.8	242	>1,000	4.41	>18.3
VP-14637	0.502	0.846	5.57	1.69	11.1
Ribavirin	11,900	17,000	14,000	1.42	1.17
Palivizumab	0.317	0.657	0.398	2.07	1.26

^aAll EC₅₀ values are represented as the geometric means of the results of at least three independent experiments. Values are expressed as nM for all compounds except palivizumab (μg/ml).

^bRelative to the value for the parent virus.

Efficacy of TP0591816 in a mouse model.

After it was confirmed that TP0591816 showed significant activity in vitro, its in vivo efficacy was examined in a mouse model of RSV infection. TP0591816 was administered subcutaneously to mice at 1 h before RSV inoculation, and the lung virus titers were determined by plaque assay at 4 days postinfection (dpi). Treatment with TP0591816 at 1, 10, and 100 mg/kg reduced the lung virus titers in a dose-dependent manner by 0.72, 0.92, and 2.04 log, respectively, compared to the titers in the vehicle control (Fig. 5A). The reductions in the lung virus titers were statistically significant at all doses. In addition, lung weights of the infected mice were also reduced by TP0591816 (Fig. 5B).

FIG 5.

Efficacy of TP0591816 in a mouse model. TP0591816 was administered subcutaneously to mice at doses of 1, 10, and 100 mg/kg at 1 h before virus inoculation. The mice were then inoculated intranasally with 1 × 10⁷ PFU of RSV A2, and the lung virus titers (A) and lung weights (B) were determined at 4 dpi. Data are represented as the means ± SEM (n = 8 mice per group). **, P < 0.01 versus vehicle (Steel’s test).

Prediction of the clinical efficacy of TP0591816.

Since, as shown above, TP0591816 has the potential to suppress infection by F protein mutants of RSV, we attempted to estimate the clinically effective dose of TP0591816. First, the pharmacokinetic parameters of TP0591816 were evaluated in vitro (Table 3). The apparent permeability (P_app) of TP0591816 at a pH of 6.2 was calculated by parallel artificial membrane permeability assay (PAMPA) to be 63.7 × 10⁻⁶cm/sec. In the metabolic stability study using human liver microsomes, disappearance of TP0591816 followed the first-order reaction, and the hepatic intrinsic clearance (CL_int) was calculated to be 182 μl/min/mg protein. The fraction unbound in incubated human liver microsomes (f_u,mic) at 0.25 mg microsomal protein/ml was estimated to be 0.966. The percentage of plasma protein binding of TP0591816 in humans was determined to be 90.7%, and the unbound fraction (f_u,p) was calculated to be 0.093. Then, the pharmacokinetics of TP0591816 were evaluated in vivo in the monkey (Table 4). The plasma concentration-time profiles of TP0591816 were analyzed, and the pharmacokinetic parameters were calculated. The steady-state volume of distribution (Vd_ss) and renal clearance (CL_R) of TP0591816 were calculated to be 1,800 ml/kg and 241 ml/h/kg, respectively. Third, physiologically based pharmacokinetic (PBPK) modeling was performed using the physicochemical properties, in vitro data, and predicted pharmacokinetic parameters (Table 3), and the plasma concentration-time profile of TP0591816 in humans was simulated in this model. Finally, the clinically effective dose twice a day (BID) was calculated as the dose at which the trough plasma concentration yielded plasma-adjusted 95% effective concentrations (paEC₉₅). The estimated oral BID doses of TP0591816 for treatment of infection by RSV A2 and the D486N mutant of RSV were 26.9 or 55.5 mg, respectively. It was also predicted that TP0591816 would also be effective for the treatment of infections caused by T400I or F488L mutants (Table 5).

TABLE 3.

Input parameters used for TP0591816 in a human PBPK model

Parametera	Value or condition
Physicochemical properties and blood binding
Mol wt	539.03
Compound type	Neutral
logD (pH 7.4)	1.54
fu,p	0.093
B/Pb	0.7
Absorption
Absorption model	ADAM
Formulation	Solution
fu,gutb	1
Peff,man (10−4 cm/sec)c	5.32
Fac	0.995
ka (h−1)c	2.32
PAMPA Papp (10−6 cm/sec)	63.7
Distribution
Distribution model	Full PBPK model
Vdss (liter/kg)d	1.45
Elimination
Clearance type	Enzyme kinetics
CLint (μl/min/mg protein)	182
fm,CYP3A4b	1
fu,micd	0.966
CLR (liter/h)d	7.88

^aD, distribution coefficient; f_u,p, fraction unbound in plasma; B/P, blood-to-plasma ratio; ADAM, advanced dissolution, absorption, and metabolism; f_u,gut, fraction unbound in the gut; P_eff,man, effective permeability in humans; Fa, fraction absorbed; ka, absorption rate constant; PAMPA, parallel artificial membrane permeability assay; P_app, apparent permeability; PBPK, physiologically based pharmacokinetic; Vd_ss, steady-state volume of distribution; CL_int, hepatic intrinsic clearance; f_m,CYP3A4, fraction metabolized by CYP3A4; f_u,mic, fraction unbound in fraction unbound in human liver microsomes; CL_R, renal clearance.

^bAssumed.

^cCalculated from the PAMPA in Simcyp.

^dPredicted.

TABLE 4.

Pharmacokinetic parameters of TP0591816 in the male cynomolgus monkey^b

Parametera	Value
CLtot (ml/h/kg)	1,770 ± 170
Vdss (ml/kg)	1,800 ± 220
t1/2 (h)	1.3 ± 0.3
AUC0–∞ (h · ng/ml)	140 ± 13
Urinary excretion (% of dose)	13.6 ± 1.8
CLR (ml/h/kg)	241

^aCL_tot, total clearance from plasma; Vd_ss, steady-state volume of distribution; t_1/2, terminal-phase half-life; AUC_0–∞, area under the concentration-time curve from time zero to infinity; CL_R, renal clearance.

^bTP0591816 was administered intravenously to male cynomolgus monkeys at a dose of 0.25 mg/kg. Data are represented as the means ± SD (n = 3 monkeys per group).

TABLE 5.

paEC₉₅ and predicted clinical effective dose of TP0591816

Parameter	F protein type
	WTa	D486N	F488L	T400I
paEC95 (ng/ml)	3.23	6.67	115	84.6
Predicted dose (mg BID)b	26.9	55.5	955	705

^aWT, wild-type.

^bDose required to achieve a trough plasma concentration greater than the paEC₉₅.

DISCUSSION

While fusion inhibitors are expected to become effective agents for treating RSV infection, one of the important problems that need to be overcome is that they exhibit broad cross-resistance with each other (24). Recently, the binding modes of several fusion inhibitors were investigated based on the X-ray crystal structure, and it was revealed that all of the inhibitors bind to the same site of the 3-fold-symmetric pocket of the prefusion RSV F protein despite their dissimilar structures (34). For this reason, fusion inhibitors of different structural types induced the same mutations in the common regions of the F protein, and such mutations cause broad cross-resistance (24, 35). Among these mutations, D486N and F488L in heptad repeat B and T400I in the cysteine-rich region are some of the most commonly reported. The D486N mutation was induced by JNJ-2408068, ziresovir, and TMC353121 in in vitro experiments (15, 27, 28). Also, the F488L mutation was induced by presatovir and VP-14637 (29, 36), and the T400I mutation was induced by VP-14637 and P13 in in vitro experiments (36, 37). In addition, all of these mutations were selected by presatovir in clinical trials that included RSV-challenged volunteers or RSV-infected patients (22, 23). Our results suggested that the antiviral activity of TP0591816 was retained against RSV harboring the D486N, F488L, or T400I mutations. On the other hand, BMS-433771 and VP-14637 showed no activity against the resistant mutants. To the best of our knowledge, TP0591816 is the first fusion inhibitor identified that exhibits antiviral activity against these F protein mutants.

The results of the time-of-addition assay and the temperature shift assay indicate that TP0591816 inhibits the fusion of RSV with the cell membrane during viral cell entry. In addition, TP0591816 also inhibited cell-cell fusion when added after infection in the syncytium formation assay. These data suggest that TP0591816 inhibits the functions of the F protein. Genetic analysis of TP0591816-resistant mutants revealed that TP0591816 induced L141F mutation in the fusion peptide but no mutation in the cysteine-rich region or heptad repeat B, where the known fusion inhibitors are known to induce mutations. These results suggest that TP0591816 binds to a similar site of RSV F protein as the known fusion inhibitors, but that its manner of binding to this site is slightly different. While the binding conformation of TP0591816 was not investigated in this study, docking models of a macrocyclic pyrazolo[1,5-a]pyrimidine derivative, a prototype of TP0591816, with antiviral activity against the D486N mutant indicated that the compound was located in almost the same position in the wild-type and D486N F protein (26). From this information, it is supposed that TP0591816 can bind to the protein in the D486N, F488L, and T400I mutants in a similar manner to that in wild-type RSV. However, further studies are needed to elucidate the details of the binding mode of TP0591816 to understand the reason for its potent antiviral activity against these mutants.

TP0591816 not only showed antiviral activity in vitro, but also in a mouse infection model. TP0591816 reduced not only virus titers but also lung weights in RSV-infected mice. RSV infection induced inflammation in the lung (38), and it has been reported that lung weights of mice infected with RSV or influenza virus increase due to infiltrated inflammatory cells (39, 40). The results of our in vivo studies suggest that TP0591816 reduces RSV and, as a consequence, can suppress inflammation that causes damage to the respiratory system. The clinically effective doses of TP0591816 were predicted from the in vitro antiviral activities and the predicted plasma concentration-time profile, based on the paEC₉₅. In the presatovir challenge study, the antiviral activity was found to be dependent on the trough plasma concentration (12). Although the trough plasma concentration of cohort 7 (10 mg as a loading dose, followed by 4 daily doses of 5 mg) in the challenge study did not achieve paEC₉₅, the regimen significantly reduced both viral load and symptom score. Then, we set paEC₉₅ as the effective trough plasma concentration. The predicted effective doses BID of TP0591816 were 26.9 mg for wild-type RSV and 55.5 to 955 mg for F protein mutants. Fusion inhibitors are well tolerated in general; for example, the leading compounds JNJ-53718678 and ziresovir showed no serious adverse events at doses of 1,000 and 1,200 mg in healthy adult subjects in a phase 1 study (41, 42). Also, the toxicity of TP0591816 is thought to be low because its cytotoxicity is as low as those of the other fusion inhibitors (CC₅₀: >30,000 nM). Since TP0591816 is safe, similarly to the known fusion inhibitors, the predicted effective doses of TP0591816 are considered to be clinically realistic and tolerated.

In conclusion, we have identified a novel macrocyclic RSV fusion inhibitor, TP0591816, that appears to exert significant antiviral activity against not only wild-type RSV but also against a known fusion inhibitor-resistant RSV strain. Furthermore, TP0591816 also showed efficacy in a mouse model and the promise to treat patients with resistant RSV infection. Taken together, these findings suggest that TP0591816 is a promising new clinical candidate to overcome infection with fusion inhibitor-resistant RSV strains.

MATERIALS AND METHODS

Compounds.

TP0591816 was synthesized by the Department of Medicinal Chemistry, Taisho Pharmaceutical Co., Ltd. (Saitama, Japan). BMS-433771, VP-14637, ribavirin, and Synagis (palivizumab) were purchased from Sigma-Aldrich (St. Louis, MO), Cayman Chemical (Ann Arbor, MI), Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan), and AbbVie GK (Tokyo, Japan), respectively.

Cells and virus.

HEp-2 cells were purchased from DS Pharma Biomedical Co., Ltd. (Osaka, Japan), and cultured in minimum essential medium (MEM) supplemented with 10% fetal bovine serum (FBS), 50 μg/ml gentamicin, and 600 μg/ml l-glutamine. RSV A2 (ATCC VR-1540, subgroup A) and RSV 18537 (ATCC VR-1580, subgroup B) were purchased from American Type Culture Collection (Manassas, VA). RSV A2 with the T400I, D486N, or F488L mutation in the F protein were selected in the presence of a pyrazolo[1,5-a]pyrimidine derivative 9c or 14e as described in our previous report (25).

Antiviral assay and cytotoxicity assay.

HEp-2 cells were seeded onto 96-well plates at 1 × 10⁴ cells/well and cultured overnight, followed by washing with phosphate-buffered saline (PBS). Each test compound was diluted with MEM supplemented with 2% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 300 μg/ml l-glutamine (maintenance medium). The diluted compounds were added, and the cells were then infected with RSV A2 at a multiplicity of infection (MOI) of 0.5 or with the other strains at an MOI of 0.75. After incubation at 37°C in the presence of 5% CO₂ for 4 days, the RSV-induced cytopathic effect (CPE) was determined by XTT assay using a cell proliferation kit II (Sigma-Aldrich). The cytotoxicity assay was performed in a similar manner but without the viral infection. The EC₅₀ and EC₉₅ values of the test compound were calculated by the least-squares method.

Time-of-addition assay.

HEp-2 cells were seeded onto 12-well plates and cultured until confluence and then were infected with RSV A2 at an MOI of 0.5. After 1 h of adsorption, the infected cells were washed three times with PBS, and each test compound was added at 0, 6, or 24 hpi. After incubation at 37°C in the presence of 5% CO₂ for 3 days, the culture supernatants were collected, and the virus titers were determined by plaque assay. Serial 10-fold dilutions of the supernatants were added to the confluent HEp-2 cells in the 12-well plates. After 1 h of adsorption, culture medium containing 0.75% methylcellulose was added. After incubation at 37°C in the presence of 5% CO₂ for 5 days, the cells were fixed with formalin and stained with hematoxylin and eosin. The virus titers were calculated by scoring the PFU.

Temperature shift assay.

HEp-2 cells were seeded on to 12-well plates and cultured until confluence and then infected with RSV A2 at an MOI of 0.5 in culture medium containing (for adding the test compound at 4°C) or not containing (for adding the test compound at 37°C) each test compound. After 1 h of adsorption at 4°C, the infected cells were washed three times with ice-cold PBS, and fresh medium containing (for adding the test compound at 37°C) or not containing (for adding the test compound at 4°C) each test compound was added. After incubation at 37°C in the presence of 5% CO₂ for 3 days, the culture supernatants were collected and the virus titers were determined by plaque assay as described above.

Syncytium formation assay.

HEp-2 cells were seeded on to 96-well plates at 1 × 10⁴ cells/well and cultured overnight, followed by washing with PBS. The cells were infected with RSV A2 at an MOI of 1 in maintenance medium. Each test compound was added to the wells at 24 hpi. The infected cells were incubated at 37°C in the presence of 5% CO₂ for 3 days and then examined microscopically for syncytium formation.

Selection and genotypic characterization of the resistant RSV strains.

HEp-2 cells were seeded on to 12-well plates and cultured until confluence. The cells were washed with PBS and infected with RSV A2 at an MOI of 0.05 in the presence of 0.3 nM TP0591816. The infected cells were incubated at 37°C in the presence of 5% CO₂ and harvested when a CPE was observed. The harvested cells were in a 10-fold volume of fresh medium containing either the same or a 3-fold higher concentration of TP0591816 and then transferred to newly cultured cells. The variants were passaged until the concentration reached 72.9 nM; control viruses were also passaged in parallel, but without the addition of TP0591816. The RNA of the wild-type and resistant viruses was reverse transcribed and amplified by PCR with primer pairs specific for the F gene. The PCR products were sequenced, and the mutation in the resistant virus was identified by comparing with the sequence of the wild-type virus (17).

Mouse model of RSV infection.

All of the in vivo experimental protocols used in this study were approved a priori by the Institutional Animal Care and Use Committee of Taisho Pharmaceutical Co., Ltd., and were in accordance with the Guidelines for Proper Conduct of Animal Experiments (Science Council of Japan, 2006). Female BALB/c mice (8 weeks of age) purchased from Charles River Laboratories Japan, Inc. (Kanagawa, Japan), were used. TP0591816 was dissolved in 10% hydroxypropyl-β-cyclodextrin and administered subcutaneously to the mice at the doses of 1, 10, and 100 mg/kg at 1 h prior to the virus inoculation. The mice were then anesthetized by intramuscular injection of ketamine (62.5 mg/kg) and xylazine (12.5 mg/kg) and inoculated intranasally with 1 × 10⁷ PFU of RSV A2 in 100 μl of culture medium. On day 4 after the RSV challenge, the mice were euthanized, and the lungs were harvested and homogenized. The virus titers in the lung homogenates were determined by plaque assay as described above. The statistical significance of the difference versus the vehicle group was analyzed by Steel’s test.

PAMPA.

The permeability of TP0591816 was evaluated using the PAMPA Evolution instrument (Pion, Inc., Woburn, MA). The permeation of TP0591816 across an artificial membrane after a 4-h incubation at room temperature was quantified using a UV plate reader. The P_app at pH 6.2 was calculated using PAMPA Evolution software (Pion, Inc.).

In vitro metabolic stability study.

The metabolic stability of TP0591816 was determined using pooled human liver microsomes (XenoTech, LLC, Lenexa, KS). TP0591816 was incubated with a reaction mixture containing 0.25 mg/ml human liver microsomes and an NADPH generation system (1.3 mmol/liter β-NADP⁺, 3.3 mmol/liter glucose-6-phosphate, 3.3 mmol/liter MgCl₂, and 0.4 U/ml glucose-6-phosphate dehydrogenase), in the presence of 100 mmol/liter potassium sodium phosphate buffer (pH 7.4). Reactions were initiated by the addition of the NADPH regenerating system. After incubation for 5, 10, 15, and 30 min at 37°C, reactions were terminated by the addition of acetonitrile/methanol (9:1, vol/vol), and the resultant mixtures were centrifuged at 3,974 × g for 10 min at 4°C. An aliquot of the supernatant was injected into a liquid chromatography-tandem mass spectrometry (LC-MS/MS) system. CL_int was calculated as described by Obach (43). The f_u,mic was estimated using the lipophilicity relationship algorithms described by Hallifax and Houston (44).

Plasma protein binding.

The plasma protein binding of TP0591816 in humans was evaluated by the equilibrium dialysis method. Equilibrium dialysis was conducted on a 96-well equilibrium dialysis device (HTDialysis, LLC, Gales Ferry, CT) with a 12- to 14-kDa-cutoff dialysis membrane. Blank plasma was spiked with TP0591816 dissolved in dimethyl sulfoxide (DMSO) at a final concentration of 1 μg/ml. The plasma samples were equilibrated with phosphate buffer (pH 7.4) at 37°C in 5% CO₂ for 4 h. After the dialysis, the concentration of TP0591816 in the plasma (C_p) and phosphate buffer (C_u) were determined by LC-MS/MS. The f_u,p was calculated as f_u,P = C_u/C_p.

Pharmacokinetic evaluations in the monkey.

The pharmacokinetics and urinary excretion of TP0591816 in male cynomolgus monkeys were evaluated following intravenous administration of the drug at the dose of 0.25 mg/kg in a cocktail of compounds. TP0591816 was dissolved in 20% 2-hydroxypropyl-β-cyclodextrin and administered intravenously into the saphenous vein of fasting cynomolgus monkeys. Blood samples were obtained from the cephalic vein of the animals prior to the dosing, and at 0.0833, 0.25, 0.5, 1, 2, 4, 6, 8, and 24 h after the dosing; the samples were centrifuged to separate plasma. Urine was collected for 24 h after the dosing.

The concentration of TP0591816 in each plasma and urine sample was determined by liquid chromatography-tandem mass spectrometry (LC-MS/MS). The plasma concentration-time profiles of TP0591816 were analyzed by a noncompartmental analysis method using the pharmacokinetic analysis software Phoenix WinNonlin v8.0 (Certara, Princeton, NJ), and the systemic clearance (CL) and the apparent steady-state volume of distribution (Vd_ss) were calculated. The fraction excreted in urine was calculated by dividing the amount excreted in the urine by the administered dose. The CL_R was calculated as the systemic clearance multiplied by the fraction excreted in the urine.

PBPK modeling.

Physiologically based pharmacokinetic (PBPK) modeling was performed using the Simcyp software v17 release 1 (Simcyp, Sheffield, UK) to predict the plasma concentration-time profile of TP0591816 in humans. The Simcyp model for TP0591816 was developed using the physicochemical properties, in vitro data, and predicted pharmacokinetic parameters. The Vd_ss (liter/kg) was predicted from simple allometric scaling of the data from monkeys using equation 1 (45).

$${\text{Vd}}_{\text{ss,human}}=0.79\times {\text{Vd}}_{\text{ss,monkey}}$$ (1)

The hepatic clearance was predicted from the intrinsic clearance in human liver microsomes. The renal clearance (CL_R [liter/h/kg]) was predicted by simple allometric scaling of the data from monkeys using equation 2 (46).

$${\text{CL}}_{\text{R,human}}=0.407\times {\text{CL}}_{\text{R,monkey}}$$ (2)

Bioavailability was predicted using the PAMPA permeability and intrinsic clearance in human liver microsomes based on the assumption that TP0591816 was mainly metabolized by CYP3A4. The plasma concentration-time profile of TP0591816 after oral administration of a 200-mg dose was simulated for 100 subjects (10 trials × 10 subjects).

Determination of the effective clinical dose of TP0591816 for the treatment of RSV infection.

The effective clinical doses of TP0591816 for the treatment of infections caused by RSV A2 and D486N, F488L, or T400I mutants were predicted. The target trough plasma concentration for each virus strain was calculated as the paEC₉₅ value using equation 3.

$$\text{target concentration}=\frac{\text{effective concentration for 95\% inhibition}}{\text{fraction unbound in plasma}}$$ (3)

Plasma concentration of TP0591816 at 12 h after oral administration of 200 mg was simulated by PBPK modeling using Simcyp, as described previously, and the effective dose (mg/day) was predicted from equation 4, based on the assumption of linear PK.

$$\text{dose}=\frac{\text{target plasma concentration}}{\text{mean simulated plasma concentration of TP0591816 after oral administration of 200 mg}}\times 200$$ (4)