Introduction

It has been 3 years since the coronavirus disease-19 (COVID-19) pandemic was declared in March 2020. Currently, there are many vaccine products available to prevent severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) infection. Nine vaccine products have acquired Emergency Use Authorization (EUA) from the US Food and Drug Administration (FDA) as of December 31, 2022 [1]. Still, the effective antiviral drugs to help eliminate the viruses in the patient’s body are limited in number. As of December 31, 2022, there were only three antiviral drugs that have received the FDA EUA for the treatment of COVID-19 patients: molnupiravir, Paxlovid (ritonavir-boosted nirmatrelvir), and remdesivir. The first mentioned is a nucleotide analog that disrupts the virus replication in the host cells, while the other two work by inhibiting the viral enzymes [1, 2]. Therefore, the exploration of new effective antiviral drugs against SARS-CoV-2 is obviously needed.

Indeed, the development of novel drugs is always challenging since it races against time as the outbreak progresses to over [2]. The drug candidates need to pass many assessments before they can be used for humans. In addition, conventional drug screening procedures require high biosafety level laboratories, at least BSL-3 [3, 4]. The limited number of adequate research facilities is a serious bottleneck in the discovery of new drugs, especially for lower-middle-income countries. For those reasons, various alternative high-throughput platforms—using synthetic peptide [5], recombinant virus [6], virus-like particle (VLP) [7], and engineered bacteria [8]—have been developed to screen potential drug candidates in more effective and efficient ways.

Using a synthetic biology approach, a novel target-based drug screening platform, called a dimer-based screening system (DBSS), has been developed for identifying antimicrobial drug candidates from diverse bioactive compounds and drug repurposing for various pathogens, such as Mycobacterium tuberculosis [9, 10], human immunodeficiency virus (HIV) [11,12,13,14], and hepatitis B virus (HBV) [15]. Inspired by Furuta et al. [16] and Okada et al. [17], DBSS assesses the capability of drug candidates to inhibit the dimerization of bacterial or viral proteins by using a genetically engineered Escherichia coli. Since the platform did not directly involve the target pathogen, it could be done in low biosafety level laboratories. Adopting the concept, Fibriani et al. [18] developed a DBSS targeting the C-terminal domain (CTD) of the SARS-CoV-2 nucleocapsid protein as a screening system for COVID-19 drug candidates (Fig. 1).

Fig. 1
figure 1

Dimer-based screening system (DBSS) for COVID-19 antiviral drug screening that applied in this study. The system screened the antiviral drug candidates that have the capability to inhibit dimerization of the C-terminal domain (CTD) of SARS-CoV-2 nucleocapsid. In the presence of viral protein inhibitors, the genetically engineered E. coli will emit a high fluorescence signal. The illustration was created with BioRender

Full size image

Porphyrin is one of the natural bioactive compounds that has demonstrated antiviral activities against numerous enveloped viruses, including HBV, HIV, dengue virus, Lassa virus, and also influenza A virus [19, 37]. Compared to synthetic drugs, natural-based medicines have milder adverse effects and a lower risk in develo** resistance for long-term uses. However, common organic compounds have poor solubility in water, limiting their delivery and bioavailability. To overcome those constraints, different drug delivery systems have been developed, such as micelles, liposomes, microspheres, and nanoparticles [37, 38]. As demonstrated by Ragab et al. [39], encapsulating chrysin, a natural flavone with anticancer activity, within a chitosan nanoparticle enhances its bioactivity in vitro. Therefore, this study also investigated the biological activity improvement of nanoparticles, specifically carbon dots, derived from porphyrin in disrupting SARS-CoV-2 infection.

Based on the in silico analysis, porphyrin displayed a potential to inhibit the dimerization of SARS-CoV-2 nucleocapsid. Porphyrin could spontaneously bind to the SARS-CoV-2 N-CTD (binding affinity energy of − 8.6 kcal/mol). The value is comparable to several COVID-19 antiviral candidates targeting the same viral protein, including silmitasertib (− 7.89 kcal/mol), fedratinib (− 8.2 kcal/mol), nintedanib (− 8.4 kcal/mol), dovitinib (− 8.6 kcal/mol), and rapamycin (− 8.9 kcal/mol), even higher than TMCB (− 7.05 kcal/mol), lopinavir (− 6.58 kcal/mol), sapanisertib (− 6.14 kcal/mol), chloroquine (− 5.62 kcal/mol), Arbidol (− 5.32 kcal/mol), oseltamivir (− 5.08 kcal/mol), ribavirin (− 4.86 kcal/mol), favipiravir (− 4.44 kcal/mol), hydroxychloroquine (− 4.32 kcal/mol), and remdesivir (− 3.46 kcal/mol) [40,4). As shown in Table 1, treatment of 4 μg/mL por-CDs for 72 h was able to reduce CPEs in the infected cells by more than 30% with the viral titer up to 4000 PFU. This antiviral capacity is comparable to other porphyrin derivatives, protoporphyrin IX and verteporfin, reported by Gu et al. [44].

Complementing the proof-of-concept demonstrated in this study, further pharmacological evaluation is still required to gain a more comprehensive understanding of the mechanism of por-CDs in disrupting SARS-CoV-2 infection. Exploration of other mechanisms of action of the antiviral drug candidate is imperative, as shown by a similar experiment by Marín-Palma et al. [58]. Furthermore, in vivo assessment is also mandatory, as they may result in different levels of toxicity and antiviral activity [59].

To the best of our knowledge, our study is the first to utilize carbon dots to increase the performance of porphyrin as a SARS-CoV-2 antiviral drug. However, the other applications of por-CDs could be explored further. It has been known that carbon dots possessed a fluorescence activity, thus making them capable of being used in many biomedical applications, including diagnostics, biosensors, photoacoustic imaging, therapeutics, and the simultaneous therapy/imaging applications called theranostics [25, 26, 60,61,62]. As shown in the previous studies, due to their great penetration capacity into cells, carbon dots increased the antimicrobial efficiency of porphyrin [63] and also improved its anticancer activity in photodynamic and photothermal therapy [64, 65]. Following a similar approach, future prospective studies would be explorations of various natural compound-derived carbon-dots that have the potential to be developed as antiviral, antibacterial, or even anticancer drugs.

Conclusions

This study discovered the potential of porphyrin-derived C-dots to be developed as safe and effective antiviral drugs for COVID-19. Using the DBSS targeting the CTD of SARS-CoV-2 nucleocapsid, porphyrin demonstrated a dimerization inhibitory activity against the N-CTD at a concentration range of 4–8 μg/mL; however, por-CDs were able to prevent the viral antigen dimerization in a broader concentration range, which was 4–10 μg/mL. The improved performances of por-CDs were further assessed in Vero E6 cells, where the por-CDs were shown to possess a lower cytotoxicity than porphyrin. Exposuring por-CDs at a concentration range of 2–10 μg/mL for 72 h only reduced the cell viability by as much as 26.88%, which surpassed the international standards for drug cytotoxicity, indicating that the antiviral drug candidate was non-toxic. For comparison, a significant viability reduction was observed in the Vero E6 cells exposed to porphyrin above the concentration of 6 μg/mL. Por-CDs also demonstrated better performance in disrupting SARS-CoV-2 infection in vitro compared to porphyrin. At the lowest concentration of 4 μg/mL, por-CDs significantly suppressed CPEs (75% reduction) in infected cells. Meanwhile, in the cells treated with porphyrin, a similar output was generated at the concentration of 6 μg/mL, suggesting an improvement of antiviral capacity after transforming porphyrin into carbon dots particles. Nevertheless, since it was still a proof-of-concept experiment, further assessments are required to ensure their safety and effectiveness in treating COVID-19 in humans. Overall, this study demonstrated the biological activity improvements of bioactive compounds in the form of C-dots. In addition, this study validated DBSS as a reliable novel drug screening platform that can be used under limited access to high -containment laboratory facilities, so that the similar approach can be applied to other natural compounds and promising drug candidates to treat other diseases.

Materials and methods

Cells and virus

Genetically engineered Escherichia coli BL21 (DE3) expressing a modified SARS-CoV-2 nucleocapsid as the dimer-based screening system (DBSS) was established by Fibriani et al. [18]. The bacteria culture was maintained in Luria–Bertani (LB) agar medium containing 200 μg/mL ampicillin and refreshed every 2 weeks. The bacterial cell stock was preserved at − 80 °C. Maintenance and experiments involving the engineered E. coli cells were conducted at School of Life Sciences and Technology, Institut Teknologi Bandung, Indonesia.

Vero E6 cell line was maintained in Dulbecco’s modified eagle medium (DMEM, Sigma-Aldrich, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS, Gibco, USA) and 1% penicillin/streptomycin (Sigma-Aldrich, USA). Every 2–4 days, the confluent cell monolayers were harvested by trypsinization and seeded into a new vessel. The cells were used for cytotoxicity and antiviral activity assays once they reached a passage number of 3.

SARS-CoV-2 (GISAID Accession ID: EPI_ISL_4004658) was isolated from Bogor, Indonesia, in May 2020. The virus was propagated in the Vero E6 cell line and the viral titer was determined using 50% tissue culture infectious dose (TCID50) assay. The TCID50/mL value was converted into plaque-forming unit (PFU)/mL by dividing it by 0.7 [66]. The viral culture was preserved at − 80 °C. All experiments involving viruses were conducted in a certified BSL-3 biocontainment facility at the Indonesian National Research and Innovation Agency (BRIN), Indonesia, and the experiment protocols have been approved by the Biosafety and Ethics Committees of BRIN.

Compound preparation

Porphyrin (C20H14N4; CAS No. 101–60-0) was purchased from Muse Chemicals (Cat No. M071749, USA) in the form of powder. Porphyrin was solubilized in 100% dimethyl sulfoxide (DMSO) and preserved at − 20 °C, protected from light. Meanwhile, porphyrin-derived C-dots (por-CDs) were prepared through a solvothermal method as follows. Precursor solution was made by stirring 1 M citric acid (C6H8O7; CAS No. 77–92-9) and 5 M urea (CH4N2O; CAS No. 57–13-6) in distilled water for 15 min at room temperature. Separately, 0.03 M porphyrin was diluted in DMSO. Then, the porphyrin and precursor solution were transferred into the Teflon inner autoclave and heated for 5 h at 160 °C. The solution was centrifuged, filtered by RC filter 0.22 μm, and freeze-dried for 3 days. In preparation for assays, the freeze-dried por-CDs were solubilized in sterile double-distilled water. The freeze-dried por-CDs were preserved at room temperature and protected from light, while the por-CDs solution was preserved at − 20 °C.

For the viral protein inhibition assay (DBSS), all tested compounds were prepared in their appropriate solvents to stock concentrations of 100 and 500 μg/mL. For the cytotoxicity assay, all tested compounds were diluted in complete DMEM supplemented with 10% FBS to final concentrations of 2, 4, 6, 8, and 10 μg/mL. Meanwhile, for the antiviral activity assay, complete DMEM supplemented with 2% FBS was used rather than 10%.

In the DBSS procedure, the final concentration of DMSO was 5% (v/v). For up to 6 h, exposure to 5% DMSO only gave moderate effects on the growth of E. coli cells [67]. On the other hand, since DMSO concentration of more than 1% is considered toxic for animal cells [68], the final concentration of DMSO for all assays using Vero E6 cells was 1% (v/v).

In silico analysis

Protein and ligand

The crystal structure of the C-terminal domain (CTD) of SARS-CoV-2 nucleocapsid phosphoprotein (PDB ID: 6ZCO) was retrieved from the Protein Database (https://www.rcsb.org/), deposited by Zinzula et al. [69]. The molecular structure of porphyrin (PubChem ID: 66868) was retrieved from PubChem (https://pubchem.ncbi.nlm.nih.gov/). All files were formatted to.pdbqt for molecular docking using AutoDock Tools 1.5.6 [70].

Molecular docking

The interaction of the tested compounds and the target viral protein was modeled using a molecular docking approach. Porphyrin was docked to the SARS-CoV-2 N-CTD using AutoDock Vina [71]. Grid box was constructed with dimensions of 36 × 38 × 34 Å and coordinates [2.482 Å, − 5.594 Å, − 1.035 Å] in a spacing of 1.00 Å. The percentage of interaction (%interaction), which is the percentage of dimerization residues of CTD of SARS-CoV-2 nucleocapsid that interacted with the ligand, was calculated as follows.

$$\mathrm{\%Interaction }= \frac{\mathrm{Number\;of\;residues\;of\;the\;protein\;interacted\;with\;the\;ligands }}{\mathrm{Total\;number\;of\;residues\;of\;the\;protein}} \times 100\%$$

In addition, the docking results were visualized using BIOVIA Discovery Studio (BIOVIA, USA).

Viral protein inhibition assay using dimer-based screening system

The viral protein inhibition assay using the dimer-based screening system (DBSS) was described by Fibriani et al. [18]. Briefly, the assay for porphyrin and por-CDs was performed as follows. A single colony of the engineered E. coli BL21 (DE3) was inoculated into 5-mL Luria–Bertani (LB) broth containing 200 μg/mL ampicillin and activated by 8-h incubation at 37 °C with 150 rpm shaking, followed by overnight incubation. The activated bacterial culture was transferred to fresh LB broth with an inoculum of 10% (v/v) and incubated at 37 °C with shaking until it reached optical densities (OD600) of 0.4. After the addition of 0.4 μM IPTG, the bacterial culture was transferred to 96-well plates (95 μL/well). Then, the tested compounds were added (5 μL/well), followed by 4-h incubation at 37 °C with shaking. As vehicle control (0 μg/mL porphyrin treatment), cells were treated with 5% DMSO. After the incubation, the cell culture optical density was read at 600 nm using GloMaxⓇ Explorer (Promega, USA), followed by fluorescence intensity measurement (excitation at 475 nm and emission at 500–550 nm). Relative fluorescence intensity is defined as the fluorescence intensity relative to the culture biomass at OD600. The optical density and fluorescence intensity values of each sample were normalized by blank, then the relative fluorescence intensity was calculated as follows.

$$\mathrm{Relative\;fluorescence\;intensity }= \frac{\mathrm{normalized\;fluorescence\;intensity}}{\mathrm{normalized\;cell\;culture\;optical\;density}}$$

Cytotoxicity assay on Vero E6 cells

The cytotoxicity of porphyrin and por-CDs on Vero E6 cells was evaluated using the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assay, which quantifies the viable cells after treatment of the tested compounds. Vero E6 cells were seeded in 96-well plates (100 μL/well) at a density of 2 × 104 cells/well and incubated at 37 °C with 5% CO2. After incubation for 24 h, the culture medium was discarded, followed by the addition of the diluted compounds to cell monolayers (100 μL/well). For 0 μg/mL porphyrin treatment (vehicle control), cells were treated with 1% DMSO-contained medium. The assay was performed with three replicates (n = 3). The plates were incubated for 72 h at 37 °C with 5% CO2. At the end of the treatment, the supernatant was discarded, then the cells were washed with phosphate buffer saline (PBS). MTT (0.5 mg/mL) was added 100 μL/well and incubated for 3 h, at 37 °C with 5% CO2, in dark condition. Finally, 100% DMSO was added (100 μL/well) to dissolve the formazan crystals formed. After 10-min incubation at room temperature, the absorbance was read at 570 nm using Varioskan™ LUX Multimode Microplate Spectrophotometer (ThermoScientific, USA). The absorbance value of each sample was normalized by blank before the cell viability was calculated as follows.

$$\mathrm{Cell\;viability }(\mathrm{\%}) = \frac{\mathrm{normalized\;absorbance\;of\;samples}}{\mathrm{normalized\;absorbance\;of\;untreated\;cells}} \times 100\mathrm{\%}$$

The tested compounds at certain concentrations which resulted in cell viability of ≥ 70% were considered non-toxic [33]. The 50% cytotoxicity concentration (CC50) values of porphyrin and por-CDs were obtained by linear regression calculation.

Antiviral activity assay against SARS-CoV-2 on Vero E6 cells

In order to examine the antiviral activity of porphyrin and por-CDs in vitro against SARS-CoV-2, we challenged the virus-infected Vero E6 cells with the tested compounds. The antiviral activity was determined based on the reduction of the cytopathic effects (CPEs) on the treated infected cells, compared with the untreated infected cells [72, 73].

The confluent Vero E6 cell monolayers (2 × 104 cells/well) were then infected with SARS-CoV-2 at MOI (multiplicity of infection) of 0.1, equal to 2000 PFU/well, in 100 μL DMEM supplemented with 2% FBS and incubated for 1 h at 37 °C. To allow maximum viral adsorption, the plate was shaken every 15 min. After the incubation, the inoculum was removed. The diluted compounds were added to the infected cells (100 μL/well), followed by incubation for 72 h at 37 °C with 5% CO2. The untreated (0 μg/mL compounds) non-infected cells were used as the cell viability control (minimal CPEs), while the untreated infected cells were used as the cell infection control (maximum CPEs). For the vehicle control of porphyrin treatment, 1% DMSO-contained medium was used. The assay was performed with six replicates (n = 6). Meanwhile, the cells were fixed with 4% formaldehyde and stained with 0.5% crystal violet (CV) solution. CPE on each well was observed under inverted microscope before and after CV staining.

Furthermore, we explored the maximum viral titer that could be inhibited by the tested compounds at the optimum non-toxic concentrations. The assay was performed as the previous step, except the Vero E6 cells were infected with varying titers of SARS-CoV-2: 32,000, 16,000, 8000, 4000, and 2000 PFU/well, corresponding to MOI of 1.6, 0.8, 0.4, 0.2, and 0.1. The assay was performed with twelve replicates (n = 12).

Data and statistical analysis

All data processing and statistical analysis were performed using Microsoft Excel (Microsoft, USA) and SPSS Statistics (IBM, USA). Data were presented as mean ± standard deviation. Statistical differences were tested using Student’s t-test for parametric data or chi-square test for non-parametric data, where p ≤ 0.05 was considered significantly different.

For the DBSS, the normalized fluorescence values for control and treatment groups were analyzed for homogeneity using Levene’s test. Data that had been confirmed as homogeneous were analyzed for significance using one-way analysis of variance (ANOVA), followed by post hoc analysis using Tukey’s test, where p ≤ 0.05 was considered significantly different.