Antiviral Activity of Micafungin and Its Derivatives against SARS-CoV-2 RNA Replication

Echinocandin antifungal drugs, including micafungin, anidulafungin, and caspofungin, have been recently reported to exhibit antiviral effects against various viruses such as flavivirus, alphavirus, and coronavirus. In this study, we focused on micafungin and its derivatives and analyzed their antiviral activities against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The micafungin derivatives Mi-2 and Mi-5 showed higher antiviral activity than micafungin, with 50% maximal inhibitory concentration (IC50) of 5.25 and 6.51 µM, respectively (3.8 to 4.7-fold stronger than micafungin) and 50% cytotoxic concentration (CC50) of >64 µM in VeroE6/TMPRSS2 cells. This high anti-SARS-CoV-2 activity was also conserved in human lung epithelial cell-derived Calu-3 cells. Micafungin, Mi-2, and Mi-5 were suggested to inhibit the intracellular virus replication process; additionally, these compounds were active against SARS-CoV-2 variants, including Delta (AY.122, hCoV-19/Japan/TY11-927/2021), Omicron (BA.1.18, hCoV-19/Japan/TY38-873/2021), a variant resistant to remdesivir (R10/E796G C799F), and a variant resistant to casirivimab/imdevimab antibody cocktail (E406W); thus, our results provide basic evidence for the potential use of micafungin derivatives for developing antiviral agents.


Introduction
Coronavirus disease 2019 (COVID- 19), which is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), remains a global public health emergency with more than 648 million infections and 6.6 million deaths as of December 2022 (https: //covid19.who.int/, accessed on 10 December 2022). In the past 20 years, SARS-CoV, Middle East respiratory syndrome coronavirus, and SARS-CoV-2 have emerged as novel human coronaviruses, which indicates the potential threat of this viral group. There are currently several US Food and Drug Administration-approved anti-COVID-19 agents, including antiviral agents such as remdesivir (RDV), molnupiravir, and nirmatrelvir. RDV and molnupiravir are nucleoside analogues that target viral polymerases and show broadspectrum antiviral effects; they were developed as drug candidates against viruses including ebola virus, hepatitis C viruses, and respiratory syncytia virus, before the COVID-19 Viruses 2023, 15, 452 2 of 8 pandemic [1][2][3][4][5]. Continuous development and studies on these antiviral agents enabled the rapid and successful application of these agents for treatment of COVID-19 within two years of the pandemic; thus, the identification and analysis of new classes of antiviral agents are important for not only developing a drug against the current infectious disease but also for developing a rapid response for managing new emerging infectious diseases, which should be stressed given the probable outbreaks of emerging viral infections in the future.
The time-of-addition analysis was performed by adding compounds at three different time points to VeroE6/TMPRSS2 cells that were inoculated with SARS-CoV-2 Wuhan strain at an MOI of 0.1 (see Figure 1A): (a, whole life cycle, gray) present during the 1 h virus inoculation and maintained throughout the 24 h infection period; (b, entry, blue) treated during the 1 h virus inoculation and for an additional 2 h and then removed; (c, post-entry, green) added at 2 h after virus inoculation and treated for the remaining 22 h until harvest. Remdesivir and cepharanthine were used as positive controls for a replication and an entry inhibitor, respectively, as reported [28,29].
The time-of-addition analysis was performed by adding compounds at three different time points to VeroE6/TMPRSS2 cells that were inoculated with SARS-CoV-2 Wuhan strain at an MOI of 0.1 (see Figure 1A): (a, whole life cycle, gray) present during the 1 h virus inoculation and maintained throughout the 24 h infection period; (b, entry, blue) treated during the 1 h virus inoculation and for an additional 2 h and then removed; (c, post-entry, green) added at 2 h after virus inoculation and treated for the remaining 22 h until harvest. Remdesivir and cepharanthine were used as positive controls for a replication and an entry inhibitor, respectively, as reported [28,29]. The Calu-3 cell infection assay was performed by inoculating the Wuhan strain for 3 h and incubating for an additional 21 h after which the extracellular viral RNA concentration was measured and 10 µM RDV was used as a positive control in the assay (see Figure  S2C).

Quantification of Viral RNA
Viral RNA in the culture supernatant was extracted using a MagMAX Viral/Pathogen II Nucleic Acid Isolation kit (Thermo Fisher Scientific) and quantified via real-time RT-PCR analysis using a one-step quantitative RT-PCR kit (Thunderbird Probe One-step qRT-PCR kit; Toyobo, Osaka, Japan) using the SARS-CoV-2-specific primers (Forward: 5′-ACAGGTACGTTAATAGTTAATAGCGT-3′, Reverse: 5′- The Calu-3 cell infection assay was performed by inoculating the Wuhan strain for 3 h and incubating for an additional 21 h after which the extracellular viral RNA concentration was measured and 10 µM RDV was used as a positive control in the assay (see Figure S2C).

Cell Viability
Cell viability was measured via a cytotoxicity assay using cell-counting kit-8 (Dojindo laboratories, Kumamoto, Japan) according to the manufacturer's instructions (see Figure S2B,D). This examination evaluates the cellular dehydrogenase activities by treating tetrazolium salt, WST-8, which is reduced under dehydrogenase to produce a yellow formazan dye. VeroE6/TMPRSS2 and Calu-3 cells cultured for 24 h with the tested compounds were evaluated by this assay.
2.6. Calculation of IC 50 , IC 90 , IC 99 , and CC 50 Inhibitory concentrations of 50, 90, and 99% maximum as well as 50% maximal cytotoxic concentration (IC 50 , IC 90 , IC 99 , and CC 50 ) of each compound were determined from a regression line [Y = AX + B] of two values that sandwich 50, 90, or 99% inhibition at compound concentration (X) and inhibition value (Y). X (IC 50 , IC 90 , IC 99 , and CC 50 values) was calculated when Y in the regression line was substituted with 50, 90, or 99.

MCFG and Its Derivatives Show Antiviral Activity against SARS-CoV-2 Infections
We used MCFG, anidulafungin, and caspofungin as approved echinocandin antifungal agents (see Figure S1) to evaluate anti-SARS-CoV-2 activity in a cell culture infection model [29]. Figure S2A,B show the viral RNA levels and cell viabilities, respectively, upon treatment with different concentrations of compounds. The IC 50 , IC 90, IC 99 , and CC 50 were calculated and are shown in Figure S2. MCFG and anidulafungin, among the three approved echinocandin antifungal agents, reduced the viral RNA levels in a dose-dependent manner (IC 50 = 26.1 and 7.09 µM, respectively), while caspofungin did not show any antiviral effect up to a concentration of 64 µM (see Figure S2A). The cell viability assay showed a dose-dependent cytotoxicity of anidulafungin (CC 50 = 24.6 µM) (see Figure S2B). Since anidulafungin had a narrow window between anti-SARS-CoV-2 activity and cytotoxicity, we focused on MCFG and synthesized derivatives. We synthesized six MCFG derivatives, Mi-1, Mi-2, Mi-3, Mi-4, Mi-5, and Mi-6 by changing the side chains of MCFG (see Materials and Methods, and Figure S1). In the SARS-CoV-2 infection assay, Mi-1, Mi-2, Mi-3, Mi-4, and Mi-5 reduced the viral RNA concentration without apparent cytotoxicity, while Mi-6 had no effect on viral RNA levels up to 64 µM (see Figure S2A,B). Among them, Mi-2 and Mi-5 showed lower IC 50 s than that of MCFG (IC 50 = 5.25 and 6.51 µM, respectively; 3.8 to 4.7-fold lower than MCFG), but no remarkable reduction in cell viability (see Figure S2A,B); further, in Calu-3 cells, Mi-2 and Mi-5 showed stronger antiviral activity than MCFG (see Figure S2C, IC 50 = 55.3, 10.1, and 5.71 µM for MCFG, Mi-2, and Mi-5, respectively), but with a detectable toxicity for Mi-5 (see Figure S2D, CC 50 = 48.5 µM) (see Table 1); thus, we produced new derivatives of MCFG that have higher antiviral potencies.

MCFG and Its Derivatives Inhibit SARS-CoV-2 Replication
To determine the stages in the SARS-CoV-2 life cycle inhibited by MCFG, Mi-2, and Mi-5, we performed a time-of-addition assay in which antiviral activities were evaluated after treatment with the compounds at three different times (see Figure 1A, a-c). In this assay, RDV, a reported SARS-CoV-2 replication inhibitor [32], was used as a positive control and showed no inhibitory effect when added at the virus entry phase (see Figure 1B, lane 5), but showed remarkable antiviral activity at the post-entry phase (see Figure 1B, lane 6). We also confirmed the mode of action of CEP, a SARS-CoV-2 entry inhibitor that inhibits SARS-CoV-2 particle binding to cells [29], based on considerable inhibition of the virus entry phase (see Figure 1B, lane 8) as well as a lower viral activity at post-entry phase. This is likely to be obtained by inhibition of multiple rounds of viral re-infection (see Figure 1B, lane 9) [29]. In this assay, Mi-2 and Mi-5 clearly reduced viral RNA levels when added during the whole life cycle and the post-entry phase, but exhibited no inhibitory effect on the virus entry phase, similar to the effects of RDV (see Figure 1B, lanes 10-18). These data suggest that MCFG, Mi-2, and Mi-5 target virus replication, rather than the process for viral entry.

Discussion
Echinocandin antifungal drugs have been reported to show antiviral activities against multiple viruses. Kim et al. reported that MCFG inhibits Enterovirus 71 (EV71) infection in the first study describing the antiviral activity of an echinocandin compound [10]. This study suggested that MCFG targets virus replication and reported an estimated IC 50 of 5-8 µM to a luciferase carrying EV71 replicon in Vero cells and weaker antiviral activity against coxsackievirus group B type 3-infected Hela cells and human rhinovirusinfected H1HeLa cells [10]. Recently, Ho et al. reported that MCFG inhibits the infection of Chikungunya virus (CHIKV) using U2OS or BHK-21 cells [11], Dengue virus serotype 2 (DENV-2) in Vero cells [12], and Zika virus (ZIKV) in Vero cells [14] (IC 50 = 20.63, 10.23, and 7.35 µM, respectively). They also reported that anidulafungin inhibits DENV-2 [12] and ZIKV infections [14] (IC 50 = 3.24 and 2.08 µM, respectively) and that caspofungin inhibits DENV-2 [12] and ZIKV [14] infections in the parallel infection assays (IC 50 = 20.78 and 75.75 µM, respectively). MCFG and anidulafungin have also been reported to inhibit SARS-CoV-2 infection in Vero cells with IC 50 = 5.5-12.9 µM [15] and 4.64 µM [16], respectively, consistent with the anti-SARS-CoV-2 activities shown in the present study (see Figure S2A and Table 1). Regarding the target of antiviral action of these echinocandin drugs, MCFG inhibits the replication of CHIKV [11] and the entry of DENV-2 [12], while anidulafungin inhibits ZIKV entry [14], which is mainly demonstrated in time-ofaddition assays; also, MCFG was reported to inhibit the helicase enzymatic activity of DENV-2 in vitro (IC 50 = 4.98 µM) [13], in addition to its anti-DENV entry activity [12], speculating that MCFG may potentially have multiple targets for antiviral activity. In a structural computer modeling of SARS-CoV-2 proteins, MCFG has been predicted to interact with 3CLpro [33,34] and was shown to inhibit 3CLpro enzymatic activity in vitro (IC 50 = 47.63 µM) [33]. It should be clarified whether 3CLpro is an actual target for MCFG, Mi-2, and Mi-5 in the future. Additionally, anidulafungin inhibits angiotensin converting enzyme 2-spike protein interaction in SARS-CoV-2 pseudovirus entry assay [35]; thus, MCFG and the related echinocandins may have multiple targets as broad-spectrum antiviral agents. Given the possible reduced efficacy of antiviral activity against emerging SARS-CoV-2 variants, development of additional anti-SARS-CoV-2 drugs with new mode of action is demanded. Identification of the target molecule of MCFG and its derivatives would be needed in the future.
In this study, we summarized the structure-activity relationship of the MCFG derivatives in Figure 2. Comparison of MCFG (IC 50 = 26.1 µM), Mi-5 (IC 50 = 6.51 µM), and Mi-6 (IC 50 > 64 µM), which share the same basic skeleton, R2, and R3 with different structure in R1, indicated the importance of the R1 structure for anti-SARS-CoV-2 activity. Since Mi-2, Mi-5, Mi-4, and Mi-1 had higher anti-SARS-CoV-2 activities than that of MCFG, replacement of the R1 moiety of MCFG with structures containing two six-membered rings (cyclohexane, piperidine, or piperazine) might increase its activity. By comparing Mi-2 and Mi-4 that have identical R2 and similar R1, the SO 3 H group in R3 seems preferable for high anti-SARS-CoV-2 activity. Further derivative analysis based on the aforementioned structure-activity relationship is expected to yield more potent antiviral compounds. The pharmacokinetics of MCFG for human adults indicate that the Cmax and half-life are, reportedly, 7.2 mg/L (5.67 µM) and 11-17 h upon administration at 100 mg daily, respectively; 100 mg/d is the approved dose for anti-fungal treatment [7]. The exposure concentration of MCFG in serum in treated adults is lower than the anti-SARS-CoV-2 IC50 calculated in this study, but that in the human lung has not been well determined. In addition, the pharmacokinetics profiles of the MCFG derivatives are needed to predict the antiviral potency in an in vivo setting. Further development of MCFG derivatives may be useful not only for developing anti-SARS-CoV-2 drugs but also as broad-spectrum antiviral drugs, which are helpful for developing a rapid response to probable new virus pandemics in the future.
Supplementary Materials: The following supporting information can be downloaded at: The pharmacokinetics of MCFG for human adults indicate that the C max and half-life are, reportedly, 7.2 mg/L (5.67 µM) and 11-17 h upon administration at 100 mg daily, respectively; 100 mg/d is the approved dose for anti-fungal treatment [7]. The exposure concentration of MCFG in serum in treated adults is lower than the anti-SARS-CoV-2 IC 50 calculated in this study, but that in the human lung has not been well determined. In addition, the pharmacokinetics profiles of the MCFG derivatives are needed to predict the antiviral potency in an in vivo setting. Further development of MCFG derivatives may be useful not only for developing anti-SARS-CoV-2 drugs but also as broad-spectrum antiviral drugs, which are helpful for developing a rapid response to probable new virus pandemics in the future.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.