Bis-benzylisoquinoline alkaloids inhibit African swine fever virus internalization and replication by impairing late endosomal/lysosomal function

ABSTRACT

African swine fever (ASF), caused by the African swine fever virus (ASFV), is a highly infectious disease afflicting domestic pigs and wild boars. It exhibits an alarming acute infection fatality rate of up to 100%. Regrettably, no commercial vaccines or specific drugs for combating this disease are currently available. This study evaluated the anti-ASFV activities in porcine alveolar macrophages, 3D4/21 cells, and PK-15 cells of four bis-benzylisoquinoline alkaloids (BBAs): cepharanthine (CEP), tetrandrine, fangchinoline, and iso-tetrandrine. Furthermore, we demonstrated that CEP, which exhibited the highest selectivity index (SI = 81.31), alkalized late endosomes/lysosomes, hindered ASFV endosomal transport, disrupted virus uncoating signals, and thereby inhibited ASFV internalization. Additionally, CEP disrupted ASFV DNA synthesis, leading to the inhibition of viral replication. Moreover, berbamine was labeled with NBD to synthesize a fluorescent probe to study the cellular location of these BBAs. By co-staining with Lyso-Tracker and lysosome-associated membrane protein 1, we demonstrated that BBAs target the endolysosomal compartments for the first time. Our data together indicated that BBAs are a class of natural products with significant inhibitory effects against ASFV infection. These findings suggest their potential efficacy as agents for the prevention and control of ASF, offering valuable references for the identification of potential drug targets.

IMPORTANCE

The urgency and severity of African swine fever (ASF) underscore the critical need for effective interventions against this highly infectious disease, which poses a grave threat to domestic pigs and wild boars. Our study reveals the potent anti-African swine fever virus (ASFV) efficacy of bis-benzylisoquinoline alkaloids (BBAs), particularly evident in the absence of progeny virus production under a 5 µM concentration treatment. The structural similarity among cepharanthine, tetrandrine, fangchinoline, and iso-tetrandrine, coupled with their analogous inhibitory stages and comparable selectivity indexes, strongly suggests a shared antiviral mechanism within this drug category. Further investigation revealed that BBAs localize to lysosomes and inhibit the internalization and replication of ASFV by disrupting the endosomal/lysosomal function. These collective results have profound implications for ASF prevention and control, suggesting the potential of the investigated agents as prophylactic and therapeutic measures. Furthermore, our study offers crucial insights into identifying drug targets and laying the groundwork for innovative interventions.

INTRODUCTION

African swine fever (ASF) is a highly infectious disease that affects domestic pigs and wild boars. It is caused by the African swine fever virus (ASFV) and can be fatal, with a fatality rate of up to 100%. Being the sole member of the Asfarviridae family and the only identified DNA arbovirus, the ASFV has a linear double-stranded DNA genome spanning approximately 170–193 kb, encoding 151–167 open reading frames (1). ASF was first reported in Kenya in 1921 and has since spread across Asia, Africa, Europe, and America (2, 3). The initial ASF outbreak in China occurred in 2018 (4, 5), resulting in considerable economic losses exceeding $130 billion (6). Unfortunately, there are currently no vaccines or specific drugs available to combat this disease. Key challenges in achieving successful ASF vaccine development include the intricate multi-layered structure and the constrained comprehension of virus infection, replication, and immune mechanisms (7). While significant efforts are being made to develop effective vaccines against ASFV (8–10), problems such as achieving cross-protection, ensuring safety, and the absence of suitable animal models persist (11). Hence, it is crucial to simultaneously explore antiviral agents as an alternative strategy for comprehensive disease control measures.

Over the past few decades, compounds with in vitro anti-ASFV activity can be categorized into six classes based on specific targets and mechanisms: nucleoside analogs, interferons, antibiotics, small interfering RNA, CRISPR/Cas9, and natural products from the plant (12). Among them, natural products offer advantages as antiviral drugs due to the diverse array of bioactive components, biocompatibility, low risk of drug resistance, and potential as a source for new drug discovery (13). Recently, a range of plant-derived compounds, including genistein, genkwanin, emodin, rhapontigenin, dihydromyricetin, toosendanin, and luteolin, have been identified for their anti-ASFV activity in vitro (14–19). Regrettably, these studies have not succeeded in clarifying the specific cellular targets and mechanisms that contribute to the anti-ASFV properties of the drugs. These results suggest that employing natural products as anti-ASFV drugs holds promise. Further exploration is required to identify the drug targets of these natural products and understand their antiviral mechanisms.

Bis-benzylisoquinoline alkaloids (BBAs) are a class of natural alkaloids widely distributed in the plant kingdom, characterized by diverse structures and various physiological activities, showcasing versatile structural types and multiple biological functions in areas, including anti-inflammation (20), antioxidant activity (21), anti-tumor responses (22), antibacterial infections (23), parasitic infections (24), and viral infections. In recent years, multiple reports have underscored that BBAs, encompassing cepharanthine (CEP), tetrandrine (TET), fangchinoline (FAN), and isotetrandrine (ISO), selectively target various stages of the envelope virus. For instance, CEP inhibits the entry stage of Hantaan virus and SARS-CoV-2 (25–27), while targeting the replication stage in Herpes simplex virus and hepatitis B virus (28, 29). Additionally, CEP simultaneously hampers the entry and replication of human immunodeficiency virus through distinct mechanisms (30, 31). Similarly, TET disrupts the transport of Ebola virus and SARS-CoV-2 from early endosomes to late endosomes through unique mechanisms (32, 33). Conversely, FAN impedes the replication of porcine epidemic diarrhea virus by blocking cellular autophagic flux (34). Hence, we hypothesize that BBAs are broad-spectrum antiviral drugs specifically designed to target various stages of enveloped viruses, including ASFV. Our previous research findings indicated that berbamine (BBM), a BBA, exhibited in vitro antiviral activity against ASFV (35). Nevertheless, the precise stages of the ASFV life cycle regulated by BBAs and the underlying mechanisms remain unclear.

In this study, we investigated the anti-ASFV activities of four BBAs, CEP, TET, FAN, and ISO, which have structures highly similar to BBM. We focused on CEP to better understand how BBA hinders the internalization of ASFV by inhibiting late endosome/lysosome acidification, which in turn suppresses viral uncoating. In addition, it has been observed that CEP serves to impair the DNA replication process of the ASFV genome, thereby effectively preventing the formation of viral factory structures. These findings indicate that BBAs, which include CEP, hold considerable promise as a means of both preventing and treating ASFV infection within the pig farming industry.

RESULTS

CEP, TET, FAN, and ISO inhibited ASFV infection in PAMs with high selectivity index

The chemical structures of CEP, TET, FAN, and ISO are shown in Fig. 1A. The four compounds, CEP, TET, FAN, and ISO, have similar structures. All four compounds are rigid macrocycles containing two tetrahydroisoquinolines, which are linked with a diphenyl ether moiety similarly, especially for TET, FAN, and ISO. All aromatic rings in the four compounds are substituted with methoxy groups (for TET, FAN, and ISO) or methylenedioxy groups (for CEP). Since porcine alveolar macrophages (PAMs) are the major target cells of ASFV infection (36), we first tested the cytotoxicity of CEP, TET, FAN, and ISO in PAMs post-incubation for 24 h by 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl tetrazolium bromide (MTT) assay, and half-maximal cytotoxic concentrations (CC₅₀) was calculated by GraphPad Prism 8.0. The results indicate that CEP, TET, FAN, and ISO have CC₅₀ values of 39.98 ± 3.19 µM, 36.02 ± 2.67 µM, 45.27 ± 1.80 µM, and 14.37 ± 2.33 µM, respectively (Fig. 1B).

Fig 1.

Cellular toxicity and anti-ASFV activity of CEP, TET, FAN, and ISO in PAMs. (A) Chemical structures of CEP, TET, FAN, and ISO. (B) Cellular toxicity of CEP, TET, FAN, and ISO was examined in PAMs at 24 h post-incubation using MTT assay, and CC₅₀ was calculated using GraphPad Prism 8.0. (C) CEP, TET, FAN, and ISO inhibited the ASFV CP204L gene and B646L gene transcription in PAMs. (D) CEP, TET, FAN, and ISO inhibited the ASFV p72 expression in PAMs. (E) CEP, TET, FAN, and ISO inhibited the ASFV titer in PAMs. Each datum represents the results of three independent experiments (mean ± SD, n = 3). The data were compared using Student’s t-tests; ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; and ****P < 0.0001.

Next, the antiviral effects of CEP, TET, FAN, and ISO against ASFV in PAMs were examined by reverse transcription-quantitative PCR (RT-qPCR), Western blotting, and endpoint dilution assay, respectively. As shown in Fig. 1C, the transcription level of the CP204L gene (encoding ASFV early protein p30) and B646L gene (encoding ASFV late protein p72) was significantly inhibited by CEP, TET, FAN, and ISO in a dose-dependent manner at 72 h post-infection (hpi). Half-maximal inhibitory concentration (IC₅₀) values of the four drugs against the infections by ASFV were determined by RT-qPCR, and the selectivity index (SI) ranged from 69.96 to 81.31 (Table 1). While the variance compared to other BBAs is not substantial, CEP exhibited the highest SI, indicating a preferable efficacy and safety. We further examined the antiviral effect of CEP, TET, FAN, and ISO on the ASFV p72 protein expression level. As shown in Fig. 1D, at concentrations ranging from 1 to 5 µM, four drugs demonstrated a significant, dose-dependent inhibition of ASFV p72 levels in PAMs. Specifically, when treated with these drugs at a concentration of 5 µM, a complete suppression of p72 expression was observed at 48 hpi compared to the control treated with DMSO. The anti-ASFV activity of CEP was evaluated under infection conditions at 24, 36, and 48 hpi. Both RT-qPCR and Western blotting results demonstrated that CEP exhibited inhibitory effects on ASFV infection at different time points (Fig. S1). Consistently, four drugs reduced ASFV titer, determined by indirect immunofluorescence assay (IFA)-based endpoint dilution assay, in a dose-dependent manner in PAMs at 72 hpi. Virus titer was non-detectable under 5 µM of drug treatment (Fig. 1E).

TABLE 1.

Inhibitory activity of CEP, TET, FAN, and ISO against ASFV in PAMs

Drug	IC50 (μM）	SI
CEP	0.49 ± 0.03	81.31
TET	0.46 ± 0.04	79.08
FAN	0.62 ± 0.11	73.61
ISO	0.21 ± 0.12	69.96

These results demonstrated that CEP, TET, FAN, and ISO, as a class of BBAs, exhibited a similar anti-ASFV activity in PAMs with limited cytotoxicity.

CEP, TET, FAN, and ISO inhibited ASFV infection in 3D4/21 and PK-15 cells

Since monocyte-macrophages are the primary target cells for ASFV (37), the assessment of anti-ASFV activity was primarily conducted on PAMs. However, in vitro evaluation of the antiviral efficacy of drugs needs validation across various cell types. 3D4/21 cells and PK-15 cells are immortalized cell lines derived from porcine alveolar macrophages and porcine kidney cells, respectively. We observed impaired replication and proliferation ability of ASFV in 3D4/21 and PK-15 cells (Fig. S2), consistent with other studies (38). Therefore, we employed 3D4/21 and PK-15 cells to assess the antiviral efficacy of CEP, TET, FAN, and ISO against ASFV. An MTT assay was performed to evaluate the cytotoxicity of CEP, TET, FAN, and ISO in these two porcine cell lines and demonstrated comparable cytotoxicity and CC₅₀ in 3D4/21 cells (Fig. 2A) and PK-15 cells (Fig. 2B).

Fig 2.

Cellular toxicity and anti-ASFV activity of CEP, TET, FAN, and ISO in 3D4/21 cells and PK-15 cells. (A) The CC₅₀ of CEP, TET, FAN, and ISO in 3D4/21 cells. (B) The CC₅₀ of CEP, TET, FAN, and ISO in PK-15 cells. (C) CEP, TET, FAN, and ISO inhibited ASFV B646L gene transcription in 3D4/21 cells. (D) CEP, TET, FAN, and ISO inhibited ASFV B646L gene transcription in PK-15 cells. Each datum represents the results of three independent experiments (mean ± SD, n = 3). The data were compared using Student’s t-tests; ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; and ****P < 0.0001.

To investigate the antiviral activity of CEP, TET, FAN, and ISO in PK-15 and 3D4/21 cells, cells were infected with the ASFV (multiplicity of infection, MOI = 0.1) for 1.5 h and then the virus solution was replaced with 10% fetal bovine serum (FBS) Dulbecco’s modified Eagle medium (DMEM) (for PK-15 cells) or 10% FBS RPMI-1640 (for 3D4/21 cells) containing 5 µM drugs and incubated until 24 hpi. The RT-qPCR revealed that CEP, TET, FAN, and ISO inhibited the ASFV B646L gene transcription level in 3D4/21 cells at concentrations ranging from 0.4 to 6.4 µM (Fig. 2C) and that in PK-15 cells at concentrations ranging from 0.3 to 5 µM (Fig. 2D). Importantly, this effect was concentration-dependent, suggesting that the anti-ASFV activity of these drugs was not cell-type dependent.

CEP, TET, FAN, and ISO inhibited ASFV infection in different treatment modes

We conducted the drug pre-treatment assay and the time-of-addition assay to explore the specific stage(s) of the ASFV infection during which CEP, TET, FAN, and ISO exhibited the antiviral effects. For drug pre-treatment, 5 µM of drugs was pre-treated for 4 h at 37°C and then infected with ASFV at an MOI of 0.1, and ASFV B646L gene transcription level was evaluated by RT-qPCR at 24 hpi. Results, as shown in Fig. 3A, show that pre-treatment with CEP, TET, FAN, and ISO led to significant inhibition of the transcription level of the ASFV B646L gene.

Fig 3.

CEP, TET, FAN, and ISO inhibited ASFV infection in pre-, co-, and post-treatment modes in PAMs. (A) Pre-treatment with CEP, TET, FAN, and ISO inhibited ASFV B646L gene transcription. (B) The schematic diagram of the time-of-addition assay. (C) The ASFV B646L gene transcription level in the different experimental groups was evaluated with an RT-qPCR. (D) The progeny viruses in the different experimental groups were titrated. Each datum represents the results of three independent experiments (mean ± SD, n = 3). The data were compared using Student’s t-tests; ND, none detected; **P < 0.01; ***P < 0.001; and ****P < 0.0001.

In Fig. 3B, the pattern diagram of the time-of-addition assay is presented. The experiment involved incubating PAMs with 5 µM of CEP, TET, FAN, or ISO in four different groups: the control group (PAMs were infected with ASFV for 1.5 h without drug treatment), the full-time group [PAMs were treated throughout the ASFV life cycle (−2–24 hpi)], the during-time group [PAMs were treated only at the ASFV entry stage (0–1.5 hpi)], and the post-time group [PAMs were treated only at the post-entry stage (1.5–24 hpi)]. The RT-qPCR demonstrated that the ASFV B646L transcription levels decreased dramatically in all drug-treated groups (Fig. 3C). The progeny viruses were titrated with an IFA, and the virus production of all drug-treated groups was reduced significantly compared with that of the control group (Fig. 3D). Notably, whether the during-time group or post-time group, the antiviral efficacy was much lower compared to the full-time group, indicating that the significant anti-ASFV effects of BBAs resulted from a comprehensive action targeting various stages of viral infection.

Hence, CEP, TET, FAN, and ISO impacted various stages of ASFV infection, encompassing both entry and post-entry phases, contributing to their antiviral efficacy.

CEP blocked the ASFV transport pathway by impairing endolysosomal function

Subsequently, we focused on CEP as it had the highest selectivity index, with the aim of investigating its antiviral mechanisms. To examine the impact of CEP on the ASFV entry stage, we performed Western blotting to assess p72 protein levels during the binding and internalization stages. In Fig. 4A, we observed that CEP treatment during the ASFV binding stage did not significantly reduce p72 protein expression compared to the ASFV-infected group without CEP treatment. However, in the internalization group, weak p72 protein bands were detected. The relative expression level of p72, normalized to the mock group, was analyzed using ImageJ software. The quantitative results further corroborate that CEP exerts an inhibitory effect on p72 protein expression during the internalization stage of ASFV rather than the binding stage (Fig. 4B). Once the virus particles are internalized, they undergo a process of disassembly that is driven by low pH. This ultimately leads to a membrane fusion event between the inner envelope and the late endosomal membrane, which releases the genome-containing core into the cytoplasm (39). Following this, p72, the major capsid protein of ASFV (40), is degraded (36). However, CEP appears to inhibit p72 protein degradation after the internalization of ASFV. Furthermore, we utilized quantitative PCR to detect the levels of viral nucleic acid during the viral adsorption and internalization stages. However, the results indicated that CEP treatment during the adsorption and internalization stages did not significantly affect the levels of ASFV viral nucleic acid, even when the drug concentration was increased to 10 µM (Fig. S3). This suggests that CEP may have a potential inhibitory effect on the virus uncoating process.

Fig 4.

CEP impaired lysosomal function and blocked the ASFV transport pathway. (A) CEP inhibited ASFV capsid p72 protein degradation during ASFV internalization stage in PAMs. (B) The relative p72 expression level normalized to the mock group was analyzed by ImageJ software. (C) 3D4/21 cells were seeded in the confocal dish and exposed to CEP (10 µM) for 4 h at 37°C. Staining in 200 nM Lyso-Tracker at 37°C for 1 h. Red fluorescent vesicles represented acidic organelles. Scale bar, 30 µm. (D) The fluorescence intensity of Lyso-Tracker red in 40 cells per group was semi-quantitatively analyzed using ImageJ software. (E) CEP-induced increasing diameter of lysosome and undigested lysosomal contents accumulation was observed by transmission electron microscopy at the ultrastructural level. The area of the cellular lysosome (Lyso) is marked with dashed lines. Scale bar at left panel, 1 µm; scale bar at right panel, 500 nm. (F) Quantification of the normal lysosomes. Data are expressed as percentages of normal lysosomes per cell. (G) PAMs were treated with or without CEP and chloroquine for 4 h, followed by LC3B immunoblot analysis. (H) 3D4/21 cells were pre-treated with CEP (10 µM) for 6 h at 37°C and then the FITC-conjugated dextran was pulsed for 1.5 h at 4°C. After another incubation at 37°C for 0, 30, and 60 min, late endosomes were labeled with an antibody against Rab7A followed by an Alexa 594-conjugated secondary antibody. Co-localization of FITC-Dextran and RAB7A was observed by laser scanning microscope. Scale bar, 5 µm. (I) PAMs were exposed to CEP (10 µM) or Baf A1 (100 nM, endosomal transport inhibitor) for 6 h at 37°C and infected with ASFV (MOI = 3) for 1.5 h at 4°C. After fixation, ASFV particles were labeled with an antibody against protein p30 followed by an Alexa 488-conjugated secondary antibody, and Rab7A followed by an Alexa 594-conjugated secondary antibody. Co-localization of ASFV p30 and RAB7A was observed by laser scanning microscope. Scale bar, 5 µm. Each datum represents the results of three independent experiments (mean ± SD, n = 3). The data were compared using Student’s t-tests; ****P＜0.0001.

The acidic environment of late endosomes and lysosomes provides the chemical signal that triggers ASFV uncoating (39). As CEP inhibits the ASFV uncoating process, we used the Lyso-Tracker probe to label acidic organelles, including late endosomes and lysosomes, to investigate CEP’s effect on these organelles. The Lyso-Tracker probe is a dye that exhibits fluorescence, with the intensity being inversely related to the pH values. Figure 4C shows a significant decrease in the intensity of red fluorescence in the CEP group compared to the DMSO group. Moreover, in the Baf A1 group (bafilomycin A1, an inhibitor of vacuolar-type H-ATPase), no red fluorescence signal was observed. A semi-quantitative analysis was conducted using ImageJ software, which involved 40 cells per group. The analysis confirmed that the average fluorescence intensity was notably lower in the CEP and Baf A1 groups than in the DMSO group (Fig. 4D). These data indicated that CEP, similar to the endolysosomal acidifying inhibitor Baf A1, elevated the pH of acidic organelles in porcine alveolar macrophages. Notably, cellular lysosomes in the CEP group exhibited morphological changes and enlargement. In addition, Lyso-Tracker probe detection for TET, FAN, and ISO yielded comparable results (Fig. S4), indicating that the BBAs we studied are potent inhibitors of endolysosomal systems.

To further explore the impact of CEP on lysosomal morphology and function, we collected macrophages that were incubated with 10 µM CEP for 24 h. We then examined the lysosomal morphology of these cells using transmission electron microscopy (TEM). As shown in Fig. 4E and F, CEP treatment resulted in an abnormally increased diameter of lysosomes compared to DMSO-treated PAMs, and these lysosomes contained numerous undigested cellular contents, with the number of normal lysosomes significantly reduced by ~40%. In addition, CEP pretreatment led to a dose-dependent accumulation of LC3-II (Fig. 4G), indicating that CEP impairs lysosomal function, resembling the effect of the autophagy inhibitor chloroquine (41).

We further hypothesized that CEP may disrupt endosomal transport pathways. To investigate this, we initially examined CEP’s effect on FITC-Dextran, a fluid-phase marker, and its co-localization with the late endosomal/lysosomal marker RAB7A. As shown in Fig. 4H, at 0 min, both the DMSO and CEP groups showed non-co-localized FITC-Dextran with RAB7A, as most of the FITC-Dextran was still attached to the cell surface. At 30 min, co-localization of FITC-Dextran and RAB7A was observed in both the DMSO group and the CEP group, indicating that FITC-Dextran is undergoing intracellular transport into late endosomes. However, at 60 min, in the DMSO group, FITC-Dextran almost completely co-localized with RAB7A, whereas in the CEP group, there remained instances where FITC-Dextran did not co-localize with RAB7A. In addition, the colocalization analysis of FITC-Dextran and RAB7A revealed that the lack of colocalization between FITC and RAB7A persists after 60 min of CEP treatment, indicating the blockage of FITC-Dextran transport within endosomes by CEP (Fig. S5). This suggested that after treatment with CEP, the co-localization of FITC-Dextran with the RAB7A decreased, indicating that CEP inhibits the process of FITC-Dextran transport from early endosomes to late endosomes, potentially impeding intracellular transport in porcine alveolar macrophages.

To investigate the inhibitory effect of CEP on the endosomal transport of ASFV in PAMs, immunofluorescence staining was employed to analyze the co-localization of ASFV p30 protein and RAB7A under different durations of CEP treatment. The endosomal transport inhibitor Baf A1 was added as a positive control. As shown in Fig. 4I, at 0 min, co-localization of the viral p30 protein and RAB7A was observed, with p30 predominantly localized near the cell membrane, suggesting viral adsorption to the cell surface. At 30 min, the co-localization of p30 with RAB7A increased gradually in the DMSO group, Baf A1 group, and CEP group, suggesting a gradual progression of the virus into late endosomes. At 60 min, co-localization of viral p30 protein and RAB7 was observed in the DMSO group, with p30 distributed and clustered in the perinuclear cytoplasmic region. However, in the Baf A1 group and CEP group, the distribution of p30 occurred in the cytoplasmic region away from the perinuclear area. These results suggested that CEP, similar to the intracellular transport inhibitor Baf A1, hinders the process of ASFV transport from endosomes to the perinuclear cytoplasmic region where the viral factory is located.

In conclusion, CEP alkalinized the late endosomes/lysosomes in porcine alveolar macrophages, hindering ASFV intracellular transport and disrupting the chemical signal for viral uncoating, thereby inhibiting the later stage of the ASFV entry process.

CEP inhibited ASFV genomic DNA replication and prevented virus factory formation

To examine the impact of CEP on ASFV post-entry stage, we performed the drug addition after ASFV infection assay. PAMs infected with ASFV were treated with CEP at different time points: 5, 8, 11, 14, and 17 hpi (Fig. 5A). RT-qPCR results revealed significant suppression of ASFV B646L gene transcription by CEP at 5, 8, and 11 hpi, with no significant difference at 14 and 17 hpi compared to the control group (Fig. 5B). Western blotting results also demonstrated inhibition of ASFV p72 protein expression by CEP at 5, 8, and 11 hpi (Fig. 5C). Conversely, at 14 and 17 hpi, p72 protein expression remained similar to the control group, aligning with the RT-qPCR findings. These results suggested that CEP inhibits the replication phase of ASFV, especially in the early stages of replication.

Fig 5.

CEP inhibited ASFV replication and viral factory formation. (A) Schematic diagram of drug addition after ASFV infection assay. PAMs infected with ASFV were treated with CEP at different times post-infection. At 24 hpi, the cells were harvested for determining ASFV B646L mRNA using (B) RT-PCR, and viral p72 detection using (C) Western blotting. (D) Immunolocalization of BrdU incorporation in ASFV-infected PAMs. Cells seeded on glass slides were infected with ASFV. After 1.5 h adsorption, the viral inoculum was replaced with fresh medium, and cells were exposed to a single 30-min pulse of BrdU (at different times post-infection), fixed, permeabilized, and followed by a step of depurination with HCl to detect viral replication sites. ASFV particles were labeled with an antibody against protein p30 followed by an Alexa 488-conjugated secondary antibody, and BrdU followed by an Alexa 594-conjugated secondary antibody. Scale bar, 5 µm. (E) Electron microscopy analysis of ASF viral factory in infected PAMs after CEP treatment. PAMs were infected with ASFV at an MOI of 1 in the presence or absence of 10 µM of CEP and fixed at 24 hpi. In the absence of CEP, viral factories (VF, marked with dashed lines) contained envelope precursors and abundant immature and mature icosahedral particles (red arrowheads). In the presence of CEP, no viral factories were observed but only single-membrane structures of damaged virus particles (red arrowheads) and an occurrence of the lipid droplet (LD) aggregation. Scale bar at left panel, 1 µm; scale bar at right panel, 500 nm. Each datum represents the results of three independent experiments (mean ± SD, n = 3). The data were compared using Student’s t-tests; ns, not significant; *P < 0.05; ***P < 0.001; and ****P < 0.0001.

BrdU, a thymidine analog that can replace thymidine in replicating DNA molecules, was used to label ASFV genome DNA synthesis (42). CEP treatment was applied at different time points (3, 5, 8, and 11 hpi) following ASFV infection, with uniform BrdU labeling at 24 hpi. Immunofluorescence staining and confocal imaging of BrdU and ASFV early protein p30 were conducted. In the 3 hpi group, no specific fluorescence of ASFV p30 with BrdU was observed, indicating an inhibitory effect of CEP on ASFV early protein p30 expression and early genome DNA replication. In the 5, 8, and 11 hpi groups, ASFV p30 protein expression was detected in conjunction with BrdU but with reduced fluorescence intensity and distribution range compared to the control group (Fig. 5D). Additionally, the earlier the CEP treatment, the lower the expression of ASFV p30 and BrdU fluorescence signal intensity, suggesting the inhibition of ASFV genome DNA synthesis, particularly during the early stages of replication by CEP treatment.

We next analyzed whether CEP treatment affected the ASFV morphogenesis that takes place at perinuclear viral factories. PAMs were incubated or not incubated with CEP after virus adsorption. Viral factories were then examined at 24 hpi by TEM. As shown in Fig. 5E, cytoplasmic factories formed in the absence of CEP displayed typical viral structures, including envelope precursors and immature and mature icosahedral particles. However, in contrast, CEP-treated infected PAMs showed the absence of clustered mature or immature virus particles near the nucleus. Instead, single-membrane structures of damaged virus particles were observed in the cytoplasm and lysosomes. Additionally, we also observed the aggregation of lipid droplets (LDs). This could be attributed to the inhibitory effect of CEP on ASFV replication, resulting in the absence of the necessary conditions for virus morphogenesis.

In summary, these findings suggested that CEP inhibits the early replication stage of ASFV by suppressing ASFV genome DNA synthesis. This inhibition disrupted the formation of viral factories, which contributed to the anti-ASFV effect.

Subcellular localization of BBM-NBD in the lysosome

Fluorescent probes are powerful tools for exploring the subcellular localization of small ligand molecules, offering high efficiency in both temporal and spatial resolution (43). However, attaching a fluorescent group to CEP is challenging due to its protected phenolic hydroxyl groups. Therefore, BBM, as an analog of CEP, was chosen for the fluorescent-labeled probe, as it is much easier for structural modification. Fluorescence-labeled berbamine was synthesized as illustrated in Fig. 6A through a copper-catalyzed click reaction between alkyne-bearing compound BBM-1 and azide-labeled NBD (NBD-N₃) concisely. An MTT assay was performed to evaluate the cytotoxicity of BBM-NBD in PAMs and demonstrated CC₅₀ as 11.76 µM (Fig. 6B). The antiviral effects of BBM and BBM-NBD were further assessed at a concentration of 5 µM. Results indicated that both compounds exhibited inhibitory effects on the transcription of the ASFV B646L gene, with comparable levels of inhibition (Fig. 6C), suggesting that the conjugation of the NBD with BBM did not cause a significant change in the antiviral efficacy of the drug. We then applied BBM-NBD to perform fluorescence imaging in living cells using fluorescence microscopy. As seen in Fig. 6D, the group with BBM-NBD exhibited green fluorescence colocalized with red fluorescence from Lyso-Tracker, suggesting that BBM-NBD targets acidic organelles in living cells. Moreover, the fluorescence intensity of Lyso-Tracker decreased, and the fluorescence morphology enlarged under treatment with BBM and BBM-NBD, suggesting that both BBM and BBM-NBD alter the lysosomal acidity and morphology of lysosomes, akin to other BBAs. The sustained co-localization of BBM-NBD with Lyso-Tracker was also investigated (Fig. S6); co-localization with Lyso-Tracker was observed as early as 1 h of pre-incubation with BBM-NBD and remained consistent up to 3 h, indicating that BBM-NBD entered lysosomes, where it became trapped, continually impacting lysosomal morphology and function. To further confirm the co-localization of BBM-NBD with lysosomes, cells incubated with BBM-NBD were fixed, and immunofluorescence staining was performed for co-localization analysis with lysosome-associated membrane protein 1 (LAMP1). As depicted in Fig. 6E, BBM-NBD exhibited co-localization with LAMP1, providing additional evidence of the co-localization of BBM-NBD with lysosomes. Additionally, under low magnification, no significant impact of BBM-NBD on the protein levels of LAMP1 was observed (Fig. S7). We also investigated the correlation between BBM-NBD and mitochondria to confirm that BBM-NBD selectively locates in endolysosomes or lysosomes. Results showed that there was no colocalization between BBM-NBD and Mito-Tracker red (Fig. S8). As reported by Cao et al. (44), the modified NBD fluorescent group can serve as a live-cell indicator for lysosomal pH. Pre-alkalization of lysosomes using NH₄Cl demonstrated that BBM-NBD maintained its lysosomal localization (Fig. S9), indicating that its targeting is independent of lysosomal pH and suggesting a direct interaction with lysosomal proteins.

Fig 6.

The drug activity and lysosomal subcellular location of BBM-NBD. (A) Synthesis of BBM-NBD. (B) The CC₅₀ of BBM-NBD in PAMs. (C) BBM and BBM-NBD inhibited ASFV B646L gene transcription. ns, not significant. (D) Co-localization of BBM-NBD and Lyso-Tracker. Cells were incubated with BBM-NBD for 3 h and Lyso-Tracker for 1 h, followed by live-cell imaging to observe their co-localization. A co-localization analysis of NBD and Lyso-Tracker fluorescence was measured using ImageJ software. Scale bar, 30 µm. (E) Co-localization of BBM-NBD and LAMP1. Cells were incubated with BBM-NBD for 3 h; after fixation, cells were incubated with an antibody against LAMP1 followed by an Alexa 594-conjugated secondary antibody. A co-localization of BBM-NBD and LAMP1 was observed using a laser scanning microscope. A co-localization analysis of NBD and LAMP1 was measured using ImageJ software. Scale bar, 5 µm. Each datum represents the results of three independent experiments (mean ± SD, n = 3). The data were compared using Student’s t-tests; ns, not significant.

These results indicated the potential for direct interaction between BBM-NBD and proteins within lysosomes. Furthermore, BBM-NBD demonstrated comparable antiviral activity against ASFV and regulatory effects on lysosomal functions with other BBAs, suggesting a consistent shared drug target within lysosomes.

DISCUSSION

The large viral genome and intricate infection mechanisms of ASFV significantly impede vaccine development progress. Therefore, the urgent need to explore alternative approaches for preventing ASFV is evident. Antiviral drugs have demonstrated significant efficacy, and the history of ASFV antiviral drug development spans several decades. Continuing to delve into ASFV antiviral drugs, deciphering their mechanisms of action, and identifying new drug targets hold crucial practical significance for the scientific control of ASFV. Given the highly similar chemical structures shared between CEP, TET, FAN, and ISO with BBM, alongside the documented antiviral activity of BBAs against a range of enveloped viruses (25–34, 45, 46), and our prior findings demonstrating BBM’s potent inhibition of ASFV infection, focusing on CEP, TET, FAN, and ISO as research subjects serves to reinforce the broader antiviral potential of BBAs as natural compounds against ASFV. This not only enhances their research significance but also elevates their reference value within the field.

Here, we demonstrated the antiviral activity of the BBAs, CEP, TET, FAN, and ISO, which are capable of inhibiting ASFV internalization and DNA replication. We established that CEP, TET, FAN, and ISO display dose-dependent antiviral activity at non-cytotoxic concentrations in PAMs, immortalized porcine alveolar macrophages (3D4/21), and porcine kidney cell line (PK-15) against ASFV. Remarkably, none of the four BBAs exhibited detectable infectious viral particles in PAMs treated at a 5 µM concentration. As natural products, these drugs demonstrate notable anti-ASFV capabilities among reported antiviral drugs for ASFV. For instance, the toosendanin-treated ASFV-infected PAMs still detected ASFV titers of approximately 5.5 log₁₀TCID₅₀/mL even at the highest concentration of 3 µM (18). Considering ASFV-caused lethality in pigs with an intramuscular injection dose as low as 1 HAD₅₀ (47), the strong inhibition of infectious viral particles at safe concentrations in PAMs by BBAs suggested significant anti-ASFV potential in vivo. Furthermore, we utilized the time-of-addition assay and found that BBAs inhibited both the entry and post-entry stages of ASFV. Given the high structural similarity among CEP, TET, FAN, and ISO, along with their analogous inhibitory stages against the virus and comparable selectivity indexes, it strongly suggests a shared antiviral mechanism within this drug category. Therefore, we focused our investigation on CEP, which exhibited the highest selectivity index against ASFV, to understand the impact of BBAs on both the entry and post-entry stages of the virus.

A schematic summary of the inhibitory cascade of CEP on ASFV internalization and replication is shown in Fig. 7. The entry of ASFV comprises both the virus adsorption on the cell surface and the subsequent internalization through endocytosis (40). We evaluated the effects of CEP during these two distinct stages. It is a common practice to extend sample collection times from 24 to 96 hpi in current antiviral drug studies targeting the ASFV entry stage (15, 48, 49). However, considering that CEP may influence specific cellular processes and potentially impact post-entry stages of ASFV infection, we opted to promptly collect samples for analysis after ASFV completed adsorption (incubation at 4°C for 1.5 h) and internalization (incubation at 4°C for 1.5 h and transfer to 37°C for 1 h) to avoid potential misinterpretation of drug efficacy during the virus entry stage. We excluded the inhibitory effect on the ASFV adsorption stage and instead identified CEP inhibition of the ASFV capsid protein degradation during internalization. Furthermore, CEP inhibits the intracellular transport and uncoating stages of ASFV, which do not significantly impact viral nucleic acid levels, and quantitative PCR results revealed that CEP treatment does not affect viral nucleic acid levels during the adsorption or internalization stages. CEP altered lysosome morphology and disrupted functionality, likely due to its alkalization of porcine alveolar macrophage lysosomes, leading to lysosomal membrane rupture or expansion. The alkalinization effect on late endosomes/lysosomes by BBAs is likely a significant factor in their broad-spectrum targeting of envelope viruses, given the acidic environment crucial for membrane fusion in these viruses. It is worth mentioning that a recent study indicated that TET inhibited ASFV entry by blocking the PI3K/Akt pathway (50). Although the author had not elaborated on the relationship between the PI3K/Akt pathway and ASFV entry, our speculation is that BBAs may disrupt ASFV endolysosomal transport by modulating the PI3K-AKT signaling pathway. This is supported by the fact that the activation of the PI3K/AKT pathway protects L-type calcium ion channels, while inhibitors targeting the related molecule PIKfyve disrupt endolysosomal transport, thereby impeding the virus entry into the host cell cytoplasm (51). Therefore, our results, along with those of other research team members, collectively confirmed that BBAs target the internalization stage of ASFV.

Fig 7.

Schematic overview of the inhibitory cascade of CEP on ASFV internalization and replication. CEP inhibited ASFV internalization by alkalizing late endosomes/lysosomes and inhibited ASFV replication by disrupting viral DNA synthesis. Created with BioRender.com.

The replication cycle of ASFV, from initial attachment to the cell surface to the release of progeny viruses, spans approximately 24 h. Upon entering the cell, ASFV expresses the early protein ASFV p30 within 4 hpi, triggering viral DNA replication; between 6 and 8 hpi, there is extensive replication of the viral genome, forming viral factories; around 16 hpi, late protein expression occurs, signifying the completion of viral DNA replication (36). Therefore, PAMs were treated with CEP at 5, 8, 11, 14, and 17 hpi to determine the specific CEP inhibition stages against ASFV replication. However, the downregulation of ASFV B646L transcription and p72 protein expression was not sufficient to directly demonstrate the impairment of the viral genomic DNA replication by CEP treatment. Employing BrdU labeling assay, newly synthesized ASFV genomic DNA was evaluated. The inhibitory effect of the expression level of ASFV p30 protein and the intensity of labeled BrdU weakened with the post-infection treatment of CEP, providing further evidence for the impairment of CEP on ASFV genomic DNA synthesis, particularly prominent during the early stages of replication. Furthermore, the typical viral factory in ASFV-infected PAMs following CEP treatment was unobservable using transmission electron microscopy, possibly due to CEP inhibiting ASFV genomic DNA replication and, consequently, lacking the prerequisites for viral morphology. It is intriguing to note that CEP induced the accumulation of LDs in macrophages (Fig. 5E). Given that BBAs, described as autophagy inhibitors (52), are likely to disrupt lysosomal lipid degradation function by inhibiting autophagy, as reported by Miyamae et al. (53), TET induces lipid accumulation in a hepatic stellate cell line through the blockage of autophagy. Lipids function as an efficient energy storage form, enabling viruses to utilize these energy reservoirs for the execution of various processes throughout their life cycle, such as protein expression and nucleic acid synthesis (54). Hence, we speculated that CEP was likely to inhibit autophagy by disrupting lysosomal function, resulting in the accumulation of LDs, and hindering the provision of adequate energy for ASFV genomic DNA synthesis.

Drug structure analysis reveals that BBM, with its phenolic hydroxyl groups, is more suitable for fluorescence labeling compared to CEP, and the efficacy of BBM-NBD, coupled with NBD fluorophore, aligns with other BBAs, supported by its antiviral activity against ASFV and lysosomal function similar to CEP, suggesting consistent efficacy in utilizing different BBAs for drug modification. After coupling BBM with the NBD fluorophore to obtain BBM-NBD, we demonstrated for the first time the lysosomal localization of BBAs for the first time. Although the mechanism by which BBM-NBD enters lysosomes was not explicitly demonstrated, the results suggested its retention within lysosomes, exerting a persistent impact on lysosomal morphology and function. Considering the lysosomal subcellular localization of BBAs and the notable impact on lysosomal morphology and function, the drug targets are assumed to be within the lysosomes. Various BBAs, demonstrated inhibitors of cellular calcium ion channels (55), are hypothesized to target the calcium ion channel protein localized in lysosomes, given the critical role of ion concentration homeostasis within these organelles.

Furthermore, a more in-depth exploration of the antiviral targets of BBAs against ASFV, coupled with a comprehensive understanding of the antiviral mechanisms, not only provides additional theoretical support for the clinical application of these drugs but also contributes to a better elucidation of the interplay between ASFV and its host. Screening in silico significantly expedites the drug development process. Utilizing ligand-based reverse virtual screening, multiple ligand-protein interaction relationships were identified (56). Therefore, the integration of reverse virtual screening with functional annotation of screened proteins has the potential to accelerate the target selection process for BBAs against ASFV. Additionally, employing omics approaches to identify proteins with ligand-protein interaction capabilities holds promise for discovering potential drug targets. Our results indicated that the NBD labeling of BBM achieved a high labeling efficiency without affecting the anti-ASFV activity. Consequently, biotin labeling of BBM for affinity purification emerges as a highly viable approach for interactome proteomics screening. However, it is important to note that these methods rely on drug-protein interactions, and proteins identified as interactors may not necessarily exhibit functional activities indicative of pharmacological relevance. On the other hand, it is crucial to evaluate the in vivo efficacy of antiviral drugs. However, we had not conducted in vivo experiments to assess the antiviral efficacy of BBAs against ASFV yet. BBAs have been applied in clinical research, and current reports indicate minimal adverse reactions and no significant safety issues associated with their use. Notably, Dong et al. (57) reported that at a safe dose of 11.1 mg/kg of body weight, CEP reduced the viral load and tissue lesions in the intestines of porcine epidemic diarrhea virus (PEDV)-infected piglets, offering a crucial reference for evaluating the efficacy against ASFV in pigs. In summary, future investigations into the anti-ASFV properties of BBAs should focus on screening drug targets, elucidating specific anti-ASFV mechanisms, and evaluating antiviral effects within the porcine organism.

In conclusion, we have demonstrated the antiviral activity of BBAs, revealed the inhibitory effect on the internalization and replication of ASFV, and, for the first time, demonstrated the lysosomal distribution of BBA. The present findings have significant implications for the prevention and control of ASF. Specifically, we suggest that the agents under investigation display potential efficacy as prophylactic and therapeutic measures against ASF. Our study offers valuable insights for the identification of potential drug targets, which could be harnessed for the development of novel interventions.

MATERIALS AND METHODS

Cell culture and virus

Porcine alveolar macrophages were isolated from 4-week-old specific pathogen-free pigs and cultured in Roswell Park Memorial Institute 1640 Medium (RPMI-1640; Gibco) supplemented with 10% fetal bovine serum (ExCell). Porcine kidney-15 cells (PK-15; ATCC CCL-33) and immortalized porcine alveolar macrophage 3D4/21 (ATCC CRL-2843) cells were cultured in DMEM (Gibco) and RPMI-1640 medium, respectively, supplemented with 10% fetal bovine serum. The ASFV strain GZ201801 (GenBank: MT496893.1) was obtained from the Research Center for African Swine Fever Prevention and Control, South China Agricultural University (Guangzhou, China).

Antibodies and reagents

ASFV p72 mouse monoclonal antibody was obtained from Zoonogen (Cat#M100068). ASFV p30 mouse monoclonal antibody was generously provided by Professor Daxin Peng of Yangzhou University. β-actin mouse monoclonal antibody (Cat#AF0003), GAPDH mouse monoclonal antibody (Cat#AF0006), HRP-conjugated Goat Anti-Mouse IgG (Cat#A0216), and HRP-conjugated Goat Anti-Rabbit IgG (Cat#A0208) were purchased from Beyotime Biotechnology. RAB7A rabbit polyclonal antibody (Cat#GTX130847) was purchased from GeneTex. LC3B rabbit polyclonal antibody (Cat#NB100-2220) was purchased from NOVUSBio. The BrdU rabbit polyclonal antibody (Cat#A20304) was purchased from Abclone. LAMP1 rabbit monoclonal antibody (Cat#9091) was purchased from Cell Signaling Technology. The goat anti-mouse Alexa Fluor 488 conjugate antibody (Cat#HS231-01) was purchased from TransGen Biotech. The goat anti-rabbit Alexa Fluor 594 (Cat#A-11012) was purchased from Thermo Fisher Scientific. FITC-Dextran (Cat#009-090-050) was purchased from Jackson Immuno.

Cytotoxicity assay

The cytotoxicity of BBAs was evaluated by using an MTT assay. Briefly, PAMs, PK-15, and 3D4/21 plated in 96-well plates were treated with increasing concentrations (from 1 to 100 µM) of CEP (Selleckchem, Cat#S4238), TET (Selleckchem, Cat#S2403), FAN (Selleckchem, Cat#S3606) and ISO (MedChemExpress, Cat#HY-N6045) for 24 h at 37°C in 5% CO₂. Then, the culture medium was removed and replaced with 100 µL MTT (0.5 mg/mL) and incubated at 37°C for 4 h. After the removal of the supernatant, 150 µL DMSO was added in each well to dissolve the formazan crystals for 10 min. Optical density (OD) was measured at 490 nm using a microplate reader. GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA) was used to calculate half-maximal cytotoxic concentrations (CC₅₀).

Indirect immunofluorescence assay

Cells were washed with PBS, followed by fixation with 4% paraformaldehyde for 20 min. Subsequently, 0.25% (vol/vol) Triton X-100 was added to each well for 30 min at room temperature (RT) to permeabilize the cell membrane. After three PBS rinses, cells underwent blocking with PBS containing 5% bovine serum albumin (BSA) at RT for 1 h. The ASFV p30 mouse monoclonal antibody diluted 1/500 was incubated at 4°C overnight, and goat anti-mouse Alexa Fluor 488 conjugate antibody diluted 1/1,000 was applied at RT for 1 h. The cell nucleus was stained with 4′,6-diamidino-2-phenylindole (DAPI). Image acquisition was performed using a Nikon Eclipse Ti-U fluorescence microscope.

Western blotting assay

Cells were lysed in RIPA lysis buffer containing 1 mM phenylmethylsulfonylfluoride at 4°C. The supernatant was harvested by centrifugation at 10,000 g for 20 min at 4°C. Equal amounts of cell lysates were loaded onto sodium dodecyl sulfate-polyacrylamide gels and then transferred to a PVDF membrane. The membranes were blocked by incubation with 5% skim milk powder for 1 h at RT. Then, ASFV p72 mouse monoclonal antibody diluted 1/500 was incubated at 4°C overnight and HRP-conjugated Goat Anti-Mouse IgG diluted 1/1,000 for 1 h at RT. The results were analyzed using Odyssey Sa (LI-COR, Lincoln, NE, USA).

Reverse transcription-quantitative PCR

RNA was extracted from samples using a total RNA rapid extraction kit (Fastagen, Shanghai, China) and reverse transcripted as complementary DNA (cDNA) using a PrimeScript RT reagent kit with gDNA Eraser (Takara Bio). Gene expression in cDNA samples was measured by ChamQ Universal SYBR qPCR Master Mix according to the manufacturer’s instructions (Vazyme Biotech). The sequences of primers used for RT-qPCR are listed in Table 2. Quantitative real-time PCR was performed on the CFX96 real-time PCR detection system (Bio-Rad). The RNA relative expression of each target gene was normalized to GAPDH expression and then calculated using the 2^-△△CT method.

TABLE 2.

Sequences of primers used for RT-qPCR

Gene	Sense (5′−3′)	Antisense (5′−3′)
B646L	CCCAGGRGATAAAATGACTG	CACTRGTTCCCTCCACCGATA
CP204L	GAGGAGACGGAATCCTCAGC	GCAAGCATATACAGCTTGGAGT
GAPDH	GAAGGTCGGAGTGAACGGATTT	TGGGTGGAATCATACTGGAACA

TCID₅₀ assay

The PAMs were seeded in 96-well plates and incubated at 37°C for 4 h. After discarding the supernatant, cells were infected with 10-fold serially diluted samples (test group) or RPMI 1640 medium (blank control group) for 24 h. Both the test and blank control groups comprised eight replicate wells each. Virus detection was performed using the IFA method, and TCID₅₀ was calculated by employing the Reed and Muench method.

Antiviral activity assay

The antiviral activity assay was performed to compare drugs’ in vitro ASFV-inhibiting capacities. PAMs, 3D4/21, or PK-15 cell monolayers grown in 24-well plates were pre-incubated by drugs with different concentrations for 2 h and infected with ASFV (MOI = 0.1) for 1.5 h at 37°C. Supernatants were removed, and a fresh medium containing different concentrations of each drug was then added. Cells and supernatants were then collected at 72 hpi, and the virus titer was determined by an endpoint dilution assay and viral RNA level by RT-qPCR. ASFV p72 protein level was determined by Western blotting at 48 hpi.

Drug pre-treatment assay

3 × 10⁵ PAMs were seeded in 24-well plates and incubated with 5 µM of drugs for 4 h at 37°C, followed by three time washes with PBS, and then infected with ASFV at an MOI of 0.1 for 1.5 h at 37°C. Then, the virus–drug mixture was removed, and the cells were washed three times with PBS and incubated with complete medium for another 24 h. Total RNA was extracted, and the transcription level of the B646L gene was detected by RT-qPCR.

Time-of-addition assay

3 × 10⁵ PAMs were seeded in 24-well plates and then infected with the ASFV at an MOI of 0.1 for 1.5 h at 37°C. Drugs were co-treated (during-time), post-treated (post-time), or treated all over the time (full-time) during the ASFV infection. For the during-time group, cells were simultaneously incubated with ASFV and 5 µM of drugs for 1.5 h at 37°C. After that, the virus–drug mixture was removed, and the cells were washed three times with PBS prior to the complete medium being added. For the post-time group, cells were first infected with ASFV for 1.5 h at 37°C. After that, the virus was removed, and the cells were washed three times with PBS prior to the complete medium containing 5 µM of drugs being added. For the full-time group, cells were initially incubated with a 5 µM concentration of drugs for 2 h. Subsequently, the solution was discarded, and the cells were washed three times with PBS. Then, the cells were concurrently exposed to ASFV and 5 µM of drugs for 1.5 h at 37°C. Following this, the virus was removed, and the cells were washed three times with PBS before being treated with a complete medium containing 5 µM of drugs. At 24 hpi, samples were collected to determine virus titer by an endpoint dilution assay and ASFV B646L gene mRNA level by RT-qPCR.

Live cell imaging

3D4/21 cells were grown in confocal dishes to 50% confluency and treated with corresponding drugs. At the appropriate time point, cell supernatants were discarded and replaced with a complete culture medium containing a Lyso-Tracker Red fluorescent probe at a concentration of 200 nM for an additional 1 h at 37°C. After that, the fluorescent probe dye was removed and washed with PBS prior to replacing it with a complete culture medium. The live cell imaging for all experimental groups was completed within 1 h using an Olympus FV10i confocal laser scanning microscope.

ASFV binding and internalization analysis

For the ASFV binding phase, PAMs were infected with ASFV at an MOI of 3 in the presence of 5 µM of CEP at 4°C for 1.5 h to facilitate viral adsorption. Subsequently, the cells were washed thrice with pre-chilled PBS to remove unadsorbed virus particles. For the ASFV internalization phase, PAMs were infected with ASFV at an MOI of 3 in the absence of 5 µM of CEP at 4°C for 1.5 h and washed by pre-chilled PBS for three times, but the infection was extended for one additional hour in the presence of 5 µM of CEP at 37°C to allow virus internalization. The cells were then washed three times with pre-warmed PBS. Subsequently, 0.5 mg/mL of protease K was added to the cells and incubated at 37°C for 1 min. The cells were washed three times with PBS to eliminate any virus particles that were bound but not internalized. At each time point, Western blotting was performed to detect ASFV major capsid protein p72.

Fluid-phase uptake assay

For the FITC-dextran uptake assay, 3D4/21 cells (1 × 10⁵ cells per well) in 24-well plates were initially exposed to 10 µM of CEP for 6 h at 37°C. Following this treatment, the cells underwent two washes with pre-chilled PBS and were subsequently incubated with 250 µg/mL of FITC-conjugated dextran for 1.5 h at 4°C. The medium was then replaced with pre-warmed complete medium and continued to incubate at 37°C for 0, 30, and 60 min. At each time point, dextran uptake was imaged using an Olympus FV10i confocal laser scanning microscope.

For ASFV uptake assay, PAMs (3 × 10⁵ cells per well) in 24-well plates were initially exposed to 10 µM of CEP or 100 nM of Baf A1 for 6 h at 37°C. Following this treatment, the cells underwent two washes with pre-chilled PBS and were subsequently infected with the ASFV at an MOI of 3 for 1.5 h at 4°C. The virus was then discarded and continued to incubate with the pre-warmed complete medium at 37°C for 0, 30, and 60 min. At each time point, ASFV uptake was imaged using an Olympus FV10i confocal laser scanning microscope.

Transmission electron microscopy

PAMs were treated with fixing solution [2% glutaraldehyde and 2% paraformaldehyde in 0.1 M phosphate buffer (PB), pH = 7.0] and stored at 4°C. The fixed samples were washed with PB four times, post-fixed with 1% osmium tetroxide (OsO₄), followed by dehydration with gradient ethanol. The samples were transferred in acetone, embedded in Spurr’s resin (SPI, USA), and polymerized at 70°C for 48 h. The sample blocks were cut into 70 nm sections by using an ultramicrotome (UC7, Leica). Sections were collected with copper grids and stained with uranyl acetate and lead citrate. The stained grids were inspected by a Thermo Fisher Talos L120C electron microscope and imaged by a Ceta2 camera.

BrdU labeling assay

For indirect immunofluorescence for viral DNA labeling in PAMs, PAM cells were seeded into coverslips with 3 × 10⁵ cells for 24 h. After that, PAMs were infected with ASFV at an MOI of 0.5. At different times post-infection, 5-bromo-2′-deoxyuridine (BrdU, 150 µM, Sigma-Aldrich) was incubated to label the newly synthesized viral DNA. After a 30-min exposure to BrdU, cells were washed twice with PBS, thereafter fixed in 4% paraformaldehyde for 10 min at RT, washed with PBS solution, and permeabilized with PBS/Triton X-100 (0.3%, vol/vol) for 5 min. After two additional washes with PBS, cells were incubated with hydrogen chloride solution (1 N HCl, 10 min, 37°C) to remove purines from DNA molecules and to expose incorporated BrdU epitopes of newly synthesized viral genomes. Cells were then blocked with PBS/BSA (PBS containing 5% bovine serum albumin) for 1 h. BrdU incorporation was detected by indirect immunofluorescence using a monoclonal anti-BrdU rabbit antibody (1:500) overnight at 4°C and a 12-h incubation with an ASFV p30 (1:500) at 4°C. A second antibody incubation was performed for 12 h at 4°C using Alexa Fluor 594 anti-rabbit (1:50). Washing solutions were performed with PBS/Tween20 (0.05%). DAPI was used to stain both viral and cellular DNA. Finally, coverslips were mounted in a Prolong Gold antifade mounting medium. Images were collected on an Olympus FV10i confocal laser scanning microscope.

Synthesis of BBM-NBD

To prepare BBM-NBD, a solution of BBM-1 (0.1 mmol, 1.0 eq) and NBD-N₃ (0.1 mmol, 1.0 eq) was created in a mixture of MeOH, DCM, and H₂O (5 mL/2 mL/1 mL). CuSO₄ (0.15 mmol, 1.5 eq) and sodium ascorbate (VitCNa, 0.3 mmol, 3.0 eq) were added, and the solution was stirred at room temperature for 24 h. Afterward, the mixture was extracted with DCM, and the organic layer was collected. The collected organic layer was then purified using preparative HPLC to yield BBM-NBD as a yellow solid. HRMS (ESI) m/z: [M + H]+ calcd for C55H62O10N9, 1,008.4614; found, 1,008.4596.

Statistical analysis

All data were analyzed using GraphPad Prism 8.0 software (GraphPad Software, San Diego, CA, USA) and are presented as the mean ± standard error of the mean of at least three independent experiments. Statistical comparisons between groups were performed using Student’s t-tests. Two-tailed P values were determined (*P < 0.05; **P < 0.01; ***P < 0.001; and ****P < 0.0001).

DATA AVAILABILITY

The data underlying this article will be shared upon reasonable request to the corresponding author.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jvi.00327-24.

Supplemental figures. jvi.00327-24-s0001.docx.

Figures S1 to S9.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.