Novel Tetrahydroisoquinoline-Based Heterocyclic Compounds Efficiently Inhibit SARS-CoV-2 Infection In Vitro

The ongoing COVID-19 pandemic has caused over six million deaths and huge economic burdens worldwide. Antivirals against its causative agent, SARS-CoV-2, are in urgent demand. Previously, we reported that heterocylic compounds, i.e., chloroquine (CQ) and hydroxychloroquine (HCQ), are potent in inhibiting SARS-CoV-2 replication in vitro. In this study, we discussed the syntheses of two novel heterocylic compounds: tert-butyl rel-4-(((3R,4S)-3-(1H-indol-3-yl)-1-oxo-2-propyl-1,2,3,4-tetrahydroisoquinolin-4-yl)methyl)piperazine-1-carboxylate (trans-1) and rel-(3R,4S)-3-(1H-indol-3-yl)-4-(piperazin-1-ylmethyl)-2-propyl-3,4-dihydroisoquinolin-1(2H)-one (trans-2), which effectively suppressed authentic SARS-CoV-2 replication in Vero E6 cells. Compound trans-1 showed higher anti-SARS-CoV-2 activity than trans-2, with a half maximal effective concentration (EC50) of 3.15 μM and a selective index (SI) exceeding 63.49, which demonstrated comparable potency to CQ or HCQ. Additional anti-SARS-CoV-2 tests on Calu-3 human lung cells showed that trans-1 efficiently inhibited viral replication (EC50 = 2.78 μM; SI: > 71.94) and performed better than CQ (EC50 = 44.90 μM; SI = 2.94). The time of an addition assay showed that the action mechanism of trans-1 differed from that of CQ, as it mainly inhibited the post-entry viral replication in both Vero E6 and Calu-3 cells. In addition, the differences between the antiviral mechanisms of these novel compounds and CQ were discussed.


Introduction
The COVID-19 pandemic caused by SARS-CoV-2 has resulted in more than 650 million human cases with an approximately 1% mortality (WHO data; 5 January 2023) [1]. Moreover, virus spillover to wild animals from people and transmission from animals such as pets to humans have been observed [2,3]; therefore, SARS-CoV-2 is expected to persist for a long period. As such, effective antivirals against SARS-CoV-2 are essential to human health care. Although there are a few approved small-molecule drugs for the treatment of COVID-19, such as Remdesivir [4], Paxlovid [5], and Molnupiravir [6], additional alternatives against different antiviral targets are needed in the wake of continuously emerging variants of SARS-CoV-2, especially for those with resistance to current drugs.
Heterocyclic compounds are a class of organic cyclic compounds with at least one hetero atom; most heteroatoms are nitrogen, sulfur, and oxygen [7]. These ring structures serve as the framework of many biological molecules such as DNA, RNA, hormones, and vitamins, making them indispensable motifs for drug discovery. Numerous FDAapproved drugs contain these heterocyclic structures, many of which are potent for the treatment of viral disease. For example, berberine inhibits the viral replication of herpes and Chikungunya viruses [8,9], as well as the entry of hepatitis C virus [10]. Moreover, it is active toward influenza virus in vitro and in vivo [11]. Isoquinoline alkaloids tetrandrine, fangchinoline, and cepharanthine hinder the expression of spike and nucleocapsid proteins in coronavirus OC43 in human lung cells [12]; palmatine inhibits the replication of West Nile, Zika, and Dengue viruses [13][14][15]; sanguinarine has antiviral effects against HIV protease and herpes simplex virus [16].
We previously reported that the anti-malarial drugs, i.e., chloroquine (CQ) and hydroxychloroquine (HCQ), are potent to inhibit SARS-CoV-2 replication in vitro [17,18]. One possible hypothesis for the inhibitory effect of CQ and HCQ is that the basicity of heterocycle in these compounds influences the acidity of intracellular organelles and hence membrane fusion [19] and the release of viral genetic materials into the cells. Heterocyclic compounds can potentially be modified to yield more optimized or varying bio-activities. For example, quinoline ring structure-based 1-oxo-2,3-disubstituted tetrahydroisoquinoline-4-carboxamides exhibit antiparasitic properties, particularly against Plasmodium falciparum, which is resistant to chloroquine treatment [20].

Cells and Viruses
African green monkey kidney Vero E6 epithelial cells were obtained from American Type Culture Collection (ATCC, NO. 1586) and maintained in an EMEM culture medium with 10% FBS. Human lung epithelial cells-Calu-3-were obtained from ATCC (HTB-55) and maintained in DMEM supplemented with 10% FBS. A clinically isolated SARS-CoV-2 strain WIV-04 [21] and the delta variant strain B.1.617.2 (IVCAS-6.7585) were obtained from the National Virus Resource Center (Wuhan, China) and were used for antiviral assessment following the approved standard operation procedures of Biosafety Level 3 (BSL-3) laboratory at Wuhan Institute of Virology, Chinese Academy of Sciences.

Chemical Synthesis of Novel Heterocyclic Compounds
All solvents used in the present study were of HPLC grade and are commercially available. We synthesized the starting tosylate in previous studies [22]; 1-Boc-piperazine and trifluoroacetic acid are commercially available and were used as supplied. The melting points (m.p.) of the compounds were determined on Boetius PHMK 0.5 apparatus and were uncorrected. NMR spectra were obtained using a Bruker Avance III 500 HD NMR spectrometer operating at 500.13 MHz for 1 H and 125.76 MHz for 13 C NMR. The chemical shifts are presented in ppm (δ) using tetramethylsilane (TMS) as an internal standard. Liquid chromatography mass spectrometry analysis (LC-MS) was carried out on a Q Exactive ® hybrid quadrupole-Orbitrap ® mass spectrometer (ThermoScientific Co, Waltham, MA, USA) equipped with a HESI ® (heated electrospray ionization) module, a Tur-boFlow ® Ultra High Performance Liquid Chromatography (UHPLC) system (Thermo-Scientific Co, Waltham, MA, USA), and an HTC PAL ® autosampler (CTC Analytics, Zwingen, Switzerland). The synthetic procedures for trans-1 and trans-2 were as follows.
2.2.1. Synthesis of tert-butyl rel-4-(((3R,4S)-3-(1H-indol-3-yl)-1-oxo-2-propyl-1,2,3,4tetrahydroisoquinolin-4-yl)methyl)piperazine-1-carboxylate (trans-1) A mixture of tosylate (0.489 g, 1 mmol) and 1-Boc-piperazine (0.559 g, 3 mmol) in dry toluene (5 mL) was refluxed, until the starting tosylate reacted completely (TLC). The reaction mixture was cooled down to 25 • C, and crystals were formed, filtered and discarded. The filtrate was concentrated, and another portion of crystals was collected. The solvent from the filtrate containing the desired product was evaporated under reduced pressure, and the residual oil was crystallized in ethyl acetate (2 mL) and light petroleum (initially 6 mL). An additional 3 mL of light petroleum was added upon crystal formation. Crystals were collected by filtration and dried, thus yielding 0.  Compound trans-1 (0.402 g, 8 mmol) was dissolved in trifluoroacetic acid (2 mL) and sonicated for 15 min. The reaction mixture was concentrated under reduced pressure, and the oily residue was triturated with 10% Na 2 CO 3 solution (6 mL) upon which crystals were formed. The solid was collected by filtration to yield 0.123 g of white crystals. The filtrate was extracted with ethyl acetate (3 × 10 mL), dried with Na 2 SO 4 and evaporated to obtain an additional 0.068 g of white crystals for a combined yield of 0.191 g (60%

Antiviral Assessment
Compounds were dissolved in dimethyl sulfoxide (DMSO) and diluted to a working concentration with a cell culture medium. Cells were incubated with compounds following infection with SARS-CoV-2 (MOI = 0.05) for 24 h or 48 h for Vero E6 and Calu-3 cells, respectively. The antiviral effect was measured using two previously reported approaches [17]: (1) quantification of the cell supernatant to assess progeny virus yield; and (2) immunofluorescence staining of the viral nucleocapsid protein (NP) to illustrate the infected cells. For quantitative real-time RT-PCR (qRT-PCR), the viral RNA was extracted by a commercial kit (Takara Bio, Beijing, China, #9766) and reverse-transcribed with a PrimeScript RT Reagent Kit (Takara Bio, Beijing, China, #RR047B). qRT-PCR was performed on StepOne Plus Real-time PCR system (Applied Biosystem) with TB Green Premix Ex Taq II (Takara Bio, Beijing, China, #RR820A). The primers used for qRT-PCR were RBD-qF: 5 -CAATGGTTTAACAGGCACAGG-3 and RBD-qR:5 -CTCAAGTGTCTGTGGATCACG-3 .

Cytotoxicity of Tested Compounds
To evaluate the cytotoxicity of compounds, a series of diluted concentrations of compounds were incubated with Vero E6 or Calu-3 cells in a 96-well plate (1 × 10 4 cells/well) for 24 h, following cell viability assessment with a cell count kit-8 (Beyotime, Shanghai, China, #C0039) according to the manufacturer's instructions. The OD 450 values of the compound treated cells (OD compound ), DMSO-treated cells (OD DMSO ), and a medium without cells (OD blank ) were obtained for calculating the normalized cytotoxicity under each concentration, which was expressed as: % cytotoxicity = 100% − (OD compound − OD blank ) / (OD DMSO − OD blank ) × 100%. The CC 50 for each compound on each cell type was calculated using Graphpad Prism 8.0 software.

Time of the Drug Addition Assay
To determine the point where trans-1 inhibited viral replication, a time of an additional assay was performed using three treatments: "Full-time", "Entry", and "Post-entry". For the "Entry" treatment, cells were pretreated with trans-1 (20 µM) for 1 h and then infected with SARS-CoV-2 (MOI = 0.05). After incubation for 1 h, compound trans-1 containing a medium was removed and replaced in a fresh medium after washing once with PBS, and cells were cultured for an additional 24 or 48 h for Vero E6 and Calu-3 cells, respectively. For the "Post-entry" treatment, the cells were incubated with viruses for 1 h, after which the cells were washed with PBS and trans-1 (20 µM) containing a medium was added. For the "Full-time" treatment, cells were treated with trans-1 before, during, and after virus incubation. Samples were collected in three ways: 1) the cell supernatant was collected for virus production detection; 2) infected cells were fixed with 4% w/v Paraformaldehyde (PFA) for immunofluorescence assay; and 3) infected cells were lysed with 1 × SDS-PAGE loading buffer (50 mM Tris-HCl, 2% w/v SDS, 0.1% w/v bromophenol blue, 10% v/v glycerol, and 1% v/v β-mercaptoethanol) for Western blot analysis.

Synthesis of Novel Tetrahydroisoquinoline-Based Heterocyclic Compounds
Based on the results of the biological activity of the previously synthesized 1,2,3,4tetrahydroisoquinoline (THIQ) derivatives and piperidinones, additional heterocyclic moieties to the THIQ core structure were selected. The synthesis of trans-1 was based on the known in the literature reaction scheme ( Figure 1A) [22]. Compound trans-2 was obtained from isoquinoline trans-1 after the removal of the Boc-protective group in trifluoroacetic acid ( Figure 1B).

Synthesis of Novel Tetrahydroisoquinoline-Based Heterocyclic Compounds
Based on the results of the biological activity of the previously synthesiz tetrahydroisoquinoline (THIQ) derivatives and piperidinones, additional he moieties to the THIQ core structure were selected. The synthesis of trans-1 was the known in the literature reaction scheme ( Figure 1A) [22]. Compound trans tained from isoquinoline trans-1 after the removal of the Boc-protective group in acetic acid ( Figure 1B).

Anti-SARS-CoV-2 Activity of Novel Heterocyclic Compounds
The anti-SARS-CoV-2 potentials of the two THIC derivatives trans-1 and t the seven additional heterocyclic compounds previously synthesized by us tested on Vero E6 cells. Compounds trans-3−9 shared the same main heterocycli as trans-1 and trans-2, and their general structures are displayed in Figure S7

Anti-SARS-CoV-2 Activity of Novel Heterocyclic Compounds
The anti-SARS-CoV-2 potentials of the two THIQ derivatives trans-1 and trans-2 and the seven additional heterocyclic compounds previously synthesized by us [23] were tested on Vero E6 cells. Compounds trans-3-9 shared the same main heterocyclic moieties as trans-1 and trans-2, and their general structures are displayed in Figure S7 (Supporting Information). Two dosages of each compound (20 µM and 5 µM) were subjected to Vero E6 cell culture, following viral infection with an MOI of 0.05. Chloroquine (CQ) (10 µM) was used as an inhibitory positive control. The supernatant of cell culture was harvested at 24 h p.i. for measuring progeny virus production by qRT-PCR. As shown in Figure 2, trans-1 (both 20 µM and 5 µM) and trans-2 (20 µM) with the indicated concentration significantly decreased the production of viral progeny more than 10-fold (Figure 2A). The immunofluorescence staining of the viral NP also suggested that trans-1 and trans-2 with the indicated concentration inhibited the replication of SARS-CoV-2 ( Figure 2B).

Compound Trans-1 Efficiently Inhibits SARS-CoV-2 Replication in Human
CQ was found to have limited potency in inhibiting viral infection cells that express co-receptor TMPRSS2, such as Calu-3, which facilitates ing and mediates viral entry via plasma membrane fusion instead of endo wondered whether the heterocyclic compound developed in this study, most promising candidate trans-1, could inhibit SARS-CoV-2 infection cells. As shown in Figure 4A,B, surprisingly, trans-1 (10 μM) inhibited SA variant (B.1.617.2) infection of Calu-3, which was not the case for CQ ( Fig  EC50 values for trans-1 and CQ in Calu-3 were 2.78 μM and 44.9 μM, resp 4C). Therefore, trans-1, a novel heterocyclic compound, may have a unique anism that effectively blocks SARS-CoV-2 infection in human lung cells.

Compound Trans-1 Efficiently Inhibits SARS-CoV-2 Replication in Human Lung Cells
CQ was found to have limited potency in inhibiting viral infection of human lung cells that express co-receptor TMPRSS2, such as Calu-3, which facilitates S protein priming and mediates viral entry via plasma membrane fusion instead of endocytosis [24]. We wondered whether the heterocyclic compound developed in this study, particularly the most promising candidate trans-1, could inhibit SARS-CoV-2 infection in human lung cells. As shown in Figure 4A,B, surprisingly, trans-1 (10 µM) inhibited SARS-CoV-2 Delta variant (B.1.617.2) infection of Calu-3, which was not the case for CQ ( Figure 4A,B). The EC 50 values for trans-1 and CQ in Calu-3 were 2.78 µM and 44.9 µM, respectively ( Figure 4C). Therefore, trans-1, a novel heterocyclic compound, may have a unique antiviral mechanism that effectively blocks SARS-CoV-2 infection in human lung cells. The supernatant was quantified for the production of viral progeny, and the inhibition rate was normalized to that of the DMSO vehicle control group. Combined with the cytotoxicity data of each compound, the dose−response curves of the inhibition rate and cytotoxicity were generated using GraphPad Prism 8 software.

Mode of Action of Trans-1
As CQ inhibits virus endocytosis and, as mentioned above, trans-1 probably has a different antiviral mechanism, we tested whether trans-1 also inhibits viral entry. A time of addition assay was performed. As shown in Figure 5A, trans-1 inhibited the production of viral progeny by almost 100% in both the "Full-time" and "Post-entry" treatment groups, and it only inhibited the production of viral progeny by ~20% and ~10% in the "Entry" treatment on Vero E6 and Calu-3 cells, respectively. The NP expression detected by Western blot ( Figure 5B) and immunofluorescence ( Figure 5C) also confirmed the inhibitory effect in the "Full-time" and "Post-entry" stages. Therefore, trans-1 inhibits SARS-CoV-2 replication mainly at the post-entry stage, which differs from CQ. The supernatant was quantified for the production of viral progeny, and the inhibition rate was normalized to that of the DMSO vehicle control group. Combined with the cytotoxicity data of each compound, the dose−response curves of the inhibition rate and cytotoxicity were generated using GraphPad Prism 8 software.

Mode of Action of Trans-1
As CQ inhibits virus endocytosis and, as mentioned above, trans-1 probably has a different antiviral mechanism, we tested whether trans-1 also inhibits viral entry. A time of addition assay was performed. As shown in Figure 5A, trans-1 inhibited the production of viral progeny by almost 100% in both the "Full-time" and "Post-entry" treatment groups, and it only inhibited the production of viral progeny by~20% and~10% in the "Entry" treatment on Vero E6 and Calu-3 cells, respectively. The NP expression detected by Western blot ( Figure 5B) and immunofluorescence ( Figure 5C) also confirmed the inhibitory effect in the "Full-time" and "Post-entry" stages. Therefore, trans-1 inhibits SARS-CoV-2 replication mainly at the post-entry stage, which differs from CQ.

Discussion
Piperazine, alkylpiperazine, and phenylpiperazine are important heterocyclic building blocks extensively evaluated in the development of various biologically active agents including anticonvulsants, antidepressants, antimalarials, anti-HIV compounds [25], and carbonic anhydrase inhibitors [26]. We previously reported that the heterocyclic drug CQ [17], HCQ [18], and other anti-malaria drugs such as arteannuin B [27] exhibit potency in the inhibition of SARS-CoV-2 infection in vitro. In this study, we developed two novel tetrahydroisoquinoline-based heterocyclic compounds (Figure 1 and Supporting Information) and tested their anti-SARS-CoV-2 activities (Figure 2). Both trans-1 and trans-2 effectively inhibited viral infection, with EC50 values of 3.15 μM and 12.02 μM, respectively ( Figure 3A,C). Compounds trans-1 and trans-2 had similar foundations, and the removal of the Boc-protection of amino group in trans-2 was the only difference. Therefore, considering that trans-2 was less biologically active against SARS-CoV-2, the N-Boc-piperazine group in the 4 th place appeared to be crucial for the anti-viral activity. This protection group also seemed to reduce compound cytotoxicity to cells as the CC50 of trans-2 was 67.78 μM ( Figure 3C), whereas low cytotoxicity was observed for trans-1 even at 200 μM ( Figures 3A and 4C).
Alkaloids, CQ, and HCQ could elevate the pH of endosome, thus inhibiting viral entry which demands low endosomal pH [28]. The cleavage of S1/S2 of spike protein is essential for SARS-CoV-2 entry and could be mediated by endosomal cathepsin B/L or

Discussion
Piperazine, alkylpiperazine, and phenylpiperazine are important heterocyclic building blocks extensively evaluated in the development of various biologically active agents including anticonvulsants, antidepressants, antimalarials, anti-HIV compounds [25], and carbonic anhydrase inhibitors [26]. We previously reported that the heterocyclic drug CQ [17], HCQ [18], and other anti-malaria drugs such as arteannuin B [27] exhibit potency in the inhibition of SARS-CoV-2 infection in vitro. In this study, we developed two novel tetrahydroisoquinoline-based heterocyclic compounds (Figure 1 and Supporting Information) and tested their anti-SARS-CoV-2 activities (Figure 2). Both trans-1 and trans-2 effectively inhibited viral infection, with EC 50 values of 3.15 µM and 12.02 µM, respectively ( Figure 3A,C). Compounds trans-1 and trans-2 had similar foundations, and the removal of the Boc-protection of amino group in trans-2 was the only difference. Therefore, considering that trans-2 was less biologically active against SARS-CoV-2, the N-Boc-piperazine group in the 4th place appeared to be crucial for the anti-viral activity. This protection group also seemed to reduce compound cytotoxicity to cells as the CC 50 of trans-2 was 67.78 µM ( Figure 3C), whereas low cytotoxicity was observed for trans-1 even at 200 µM ( Figures 3A and 4C).
Alkaloids, CQ, and HCQ could elevate the pH of endosome, thus inhibiting viral entry which demands low endosomal pH [28]. The cleavage of S1/S2 of spike protein is essential for SARS-CoV-2 entry and could be mediated by endosomal cathepsin B/L or TMPRSS2 located on the cell surface [29]. Because the activity of endosomal cathepsin relies on acid pH, CQ effectively suppresses the endosomal entry of SARS-CoV-2 in TMPRSS2 negative cells [18]. However, for TMPRSS2-expressing cells such as respiratory epithelial cells and lung cells (such as Calu-3), CQ showed limited antiviral potency [24]. This may explain the limited clinical anti-SARS-CoV-2 efficacy of CQ and HCQ [30]. Because trans-1 is the most promising compound we have obtained so far and our starting strategy to synthesize novel heterocyclic compounds was based on CQ's antiviral activity, we attempted to dissect the working mechanism of trans-1. First, we tested the anti-SARS-CoV-2 activities of trans-1, trans-2, and CQ on Calu-3 cells. As shown in Figure 4, only trans-1 showed potency in the inhibition of viral infection under 10 µM, and the EC 50 value of trans-1 was considerably higher than that of CQ (2.78 µM versus 44.9 µM). This result was unexpected and suggested that trans-1 might possess a unique mechanism for the inhibition of viral replication. Therefore, a time of addition assay was performed to reveal whether trans-1 functions at the viral entry stage in the same way CQ does. The results showed that trans-1 inhibited SARS-CoV-2 replication mainly at the post-entry stage in both Vero E6 and Calu-3 cells ( Figure 5). Compound trans-1 probably targets the host cell for antiviral activity, but this target might not be associated with pH increase in endosome as CQ does. Of course, we do not exclude the possibility that trans-1 may directly target viral machines to inhibit virus replication in host cells. Taken together, due to different working mechanisms, trans-1 is promising in that it potently inhibits SARS-CoV-2 infection in human lung cells, while CQ and HCQ fail to.
In this study, trans-1 was shown with potency to inhibit WIV-04 and Delta SARS-CoV-2 strains. If trans-1 targets the host cell biological events, it would be potent in the inhibition of most variants of concern, including ongoing endemic Omicron strains. Although the in vitro anti-SARS-CoV-2 activity of trans-1 is not as good as current clinically used therapeutics such as Molnupiravir and Paxlovid, the in vivo antiviral efficacy is a more important parameter and it remains to be determined. Thus, additional preclinical data are needed to evaluate whether trans-1 is a promising candidate for the treatment of COVID-19 and the consequent mitigation of the pandemic, including the detailed anti-viral mechanism of trans-1 and the in vivo potency in the inhibition of SARS-CoV-2 infection of proper animal models. In addition, based on the structural formulations of trans-1 and trans-2, more analogs have been synthesized, and their anti-coronavirus activity as well as cytotoxicity are being tested. Therefore, a comprehensive view of the relationship between the structure and the bio-activity of these THIQ analogs would offer insights into a more optimized design of heterocyclic compounds with a broad spectrum of anti-coronavirus activity in the future.