Letter to the Editor in response to the articles 'Lianhuaqingwen exerts anti-viral and anti-inflammatory activities against novel coronavirus (SARS-CoV-2)' and 'Liu Shen capsule shows antiviral and anti-inflammatory abilities against novel coronavirus SARS-CoV-2 via suppression of NF-κB signaling pathway.'

Dear Editor

We congratulate Li Runfeng et al. and Qinhai Ma et al. on their in vitro analyses [1,2] wherein SARS-CoV-2 replication, virus morphology, and the number of virus particles in Vero E6 cells as well as mRNA levels of pro-inflammatory cytokines (TNF-α, IL-6, IL-1β, IL-8, CCL-2/MCP-1, and CXCL-10/IP-10) in Huh-7 cells were directly inhibited or reduced by crude extracts of Lianhuaqingwen (LH, 150–600 μg mL−1) and Liu Shen (LS, 0.25–2 μg mL−1). This was further supported by the clinical analysis of LH efficacy against COVID-19 [3]. These in vivo/vitro observations prove to be meaningful as COVID-19, owing to its fast-spreading nature, has become a global threat with no or scarcely available antivirals or vaccine. Moreover, LH and LS crude extracts while being inconsistent with their circulation profiles, when directly administered, produce potential interference in virus replication, morphology, and mRNA levels in SARS-CoV-2-treated cells, similar to that observed in in vitro studies on more than 38,597 herbal crude extracts (PubMed, 2020/5/4). However, the use of herbal products such as LH and LS is not preferred because of their unclear representation of absorbed bioactive compounds (ABCs). Furthermore, their unabsorbed compounds, concentrations of absorbed compounds, excipients, metal ions, and impurities in crude extract may lead to false responses ex vivo after administration. The precise content of the above-mentioned herbal ABCs has remained elusive [4]. Methods such as the bioethnopharmaceutical analytical pharmacology strategy we propose here in which three ABCs almost replicate multi-herbal cardioprotection within 3 months [5] are indeed useful. Nonetheless, generally, the experimental time for in vivo profiling, including that for the above, is longer than that for in vitro profiling, which is unsuitable for rapid screening of antivirals and overcoming the challenge. Herein, we developed a rapid strategy for eliminating false ex vivo responses to LH in H2O2-treated Huh-7 cells by using LH post-dose supernatant of deproteinized serum (LPSDS) as the herbal LH proxy for ex vivo samples. H2O2, a well-known stressor of inflammation and oxidative stress, is an indispensable mediator and inducer of oxidative damage via reactive oxygen species (ROS) generation and inflammatory cytokine-induced damage following coronavirus attack [6,7]. H2O2-treated endothelium cells induced relevant time- and dose-specific not dependent changes in ROS levels (unpublished data), whereas Huh-7 cells treated for a fixed time with a fixed dose of H2O2 induced a relatively stable increase in ROS levels (Fig. 1 C). After oral administration of LH to rats, the proxy LPSDS, wherein proteins in serum are removed by using a hot water bath, was consistent with the amount circulating in ex vivo systems without any interference. This approach has been successfully applied to Chaihu-Shugan-San (CSS, seven herbs)-induced neuroprotection, vascular protection, and prokinetics where 10 ABCs [Meranzin hydrate (PubChem CID: 5070783); Albiflorin (PubChem CID: 24868421); Ferulic acid (PubChem CID: 445858); Naringin (PubChem CID: 442428); Hesperidin (PubChem CID: 10621); Glycyrrhizic acid (PubChem CID: 14982); Saikosaponin A (PubChem CID: 167928); Nobiletin (PubChem CID: 72344); Tangeretin (PubChem CID: 68077); Carbenoxolone (PubChem CID: 636403)] play a crucial role via in vivo anti-inflammatory and docking experiments (unpublished). Safety assessment should be performed after clinical translational application of serum with or without protein, thereinto such as sera antibodies IgG, IgA, and IgM, as herbal ex vivo proxy relevant to COVID-19 patients. The early sera antibodies produced during infection neutralize or mediate clearance of virus via binding virus particles and forming immune complexes. The immune complexes promote inflammation and exacerbate symptoms via interactions between antibody Fc domains and FcγRIIIa pathways [8], whereby we assessed the comparative effects of blank deproteinized serum (BDS) or blank serum (BS) containing-IgG on inflammatory cytokines.

Fig. 1

Quantification of IL-6, TNF-α, ROS and apoptosis rate in Huh-7 cells. Quantification of IL-6 (A), TNF-α (B), ROS (C) and apoptosis (D) in each group: blank, model, 75 μg/mL LH(L-LPSDS), 150 μg/mL LH(M-LPSDS), 300 μg/mL LH, 600 μg/mL LH(H-LPSDS), BS, BDS, and LMS. LH, Lianhuaqingwen; BS, blank serum; BDS, blank deproteinized serum; LMS, Lianhuaqingwen medicated serum; H-/M-/L-LPSDS, High-/Middle-/Low-dose Lianhuaqingwen post-dose supernatant of deproteinized serum. Pound sign or asterisk denotes a significant difference between two groups in the repeated-measures ANOVA (P＜0.0001, vs. the blank group; ★★★★P＜0.0001, vs. the model group; ★★P＜0.01, vs. the model group; ★P＜0.05, vs. the model group). Data are expressed as mean ± SD.

We congratulate Li Runfeng et al. and Qinhai Ma et al. on their in vitro analyses [1,2] wherein SARS-CoV-2 replication, virus morphology, and the number of virus particles in Vero E6 cells as well as mRNA levels of pro-inflammatory cytokines (TNF-α, IL-6, IL-1β, IL-8, CCL-2/MCP-1, and CXCL-10/IP-10) in Huh-7 cells were directly inhibited or reduced by crude extracts of Lianhuaqingwen (LH, 150–600 μg mL−1) and Liu Shen (LS, 0.25–2 μg mL−1). This was further supported by the clinical analysis of LH efficacy against COVID-19 [3]. These in vivo/vitro observations prove to be meaningful as COVID-19, owing to its fast-spreading nature, has become a global threat with no or scarcely available antivirals or vaccine. Moreover, LH and LS crude extracts while being inconsistent with their circulation profiles, when directly administered, produce potential interference in virus replication, morphology, and mRNA levels in SARS-CoV-2-treated cells, similar to that observed in in vitro studies on more than 38,597 herbal crude extracts (PubMed, 2020/5/4). However, the use of herbal products such as LH and LS is not preferred because of their unclear representation of absorbed bioactive compounds (ABCs). Furthermore, their unabsorbed compounds, concentrations of absorbed compounds, excipients, metal ions, and impurities in crude extract may lead to false responses ex vivo after administration. The precise content of the above-mentioned herbal ABCs has remained elusive [4]. Methods such as the bioethnopharmaceutical analytical pharmacology strategy we propose here in which three ABCs almost replicate multi-herbal cardioprotection within 3 months [5] are indeed useful. Nonetheless, generally, the experimental time for in vivo profiling, including that for the above, is longer than that for in vitro profiling, which is unsuitable for rapid screening of antivirals and overcoming the challenge. Herein, we developed a rapid strategy for eliminating false ex vivo responses to LH in H2O2-treated Huh-7 cells by using LH post-dose supernatant of deproteinized serum (LPSDS) as the herbal LH proxy for ex vivo samples. H2O2, a well-known stressor of inflammation and oxidative stress, is an indispensable mediator and inducer of oxidative damage via reactive oxygen species (ROS) generation and inflammatory cytokine-induced damage following coronavirus attack [6,7]. H2O2-treated endothelium cells induced relevant time- and dose-specific not dependent changes in ROS levels (unpublished data), whereas Huh-7 cells treated for a fixed time with a fixed dose of H2O2 induced a relatively stable increase in ROS levels (Fig. 1C). After oral administration of LH to rats, the proxy LPSDS, wherein proteins in serum are removed by using a hot water bath, was consistent with the amount circulating in ex vivo systems without any interference. This approach has been successfully applied to Chaihu-Shugan-San (CSS, seven herbs)-induced neuroprotection, vascular protection, and prokinetics where 10 ABCs [Meranzin hydrate (PubChem CID: 5070783); Albiflorin (PubChem CID: 24868421); Ferulic acid (PubChem CID: 445858); Naringin (PubChem CID: 442428); Hesperidin (PubChem CID: 10621); Glycyrrhizic acid (PubChem CID: 14982); Saikosaponin A (PubChem CID: 167928); Nobiletin (PubChem CID: 72344); Tangeretin (PubChem CID: 68077); Carbenoxolone (PubChem CID: 636403)] play a crucial role via in vivo anti-inflammatory and docking experiments (unpublished). Safety assessment should be performed after clinical translational application of serum with or without protein, thereinto such as sera antibodies IgG, IgA, and IgM, as herbal ex vivo proxy relevant to COVID-19 patients. The early sera antibodies produced during infection neutralize or mediate clearance of virus via binding virus particles and forming immune complexes. The immune complexes promote inflammation and exacerbate symptoms via interactions between antibody Fc domains and FcγRIIIa pathways [8], whereby we assessed the comparative effects of blank deproteinized serum (BDS) or blank serum (BS) containing-IgG on inflammatory cytokines.

Thus, using the above design, we conducted simulations of relevant experiments in Huh-7 cells. In H2O2-treated Huh-7 cells, IL-6, TNF-α, and ROS were selected to mimic the virus-induced inflammatory response (different etiology with similar inflammation), and apoptosis rate following oxidative stress was measured to evaluate cellular function. The virus and H2O2 increased TNF-α levels by ∼4.0- [1] and ∼7.0-fold (vs. blank, in Fig. 1B), respectively. We compared crude LH (Lot No. B2001019 [1]; its nine chromatographic peaks were detected) extracts with LPSDS by using these indices for their consistency. The present LH ex vivo dose (150–600 μg mL−1 [1]) was almost identical to ∼5%–19 % (bioavailability) of conversion in vivo dose following calculation via coefficient 6.7 [5]{Pinkel, 1958 #3} for a rat (0.1541 g kg−1), which is conversion dose from patient dose (0.0230 g kg−1) [3]. The theoretical bioavailability in rats is 0%–100 %, i.e. blank dose to complete absorption, with the latter observed near sites of intravascular injection. Among the dose-dependent (150–600 μg mL−1) ex vivo responses to LH, 600 μg mL−1 significantly reduced mRNA levels (pg mL−1) of IL-6 by 44.12 % [60.53 ± 0.54 vs. 108.34 ± 1.39 (model); p＜0.0001] (Fig. 1A), TNF-α by 61.30 % [16.73 ± 0.40 vs. 43.23 ± 0.50 (model); p＜0.0001] (Fig. 1B), as well as ROS level by 54.42 % [8611.63 ± 454.62 vs. 18893.20 ± 147.31 (model); p＜0.0001] (Fig. 1C). It also inhibited apoptosis by 57.71 % [8.47 ± 0.39 vs. 20.03 ± 0.41 (model); p＜0.0001] (Fig. 1D). Unexpectedly, the reduction of the selected indices by LH (600 μg mL−1) was lesser than that by its high-dose LPSDS, i.e. H-LPSDS (IL-6, 44.12 % vs. 64.60 %; TNF-α, 61.30 % vs. 79.10 %; ROS, 54.42 % vs. 82.03 %; and apoptosis, 57.71 % vs. 83.76 %. Middle- and low-doses LPSDS (M- and l-LPSDS) were shown in Fig. 1A-D), which differs from the results of experiments on herbal extracts similar to LH [9]. Compared with H-, M- and L-LPSDS, LH has relatively unclear apoptosis-inducing substances, while as in vitro agent, BS (with volume-escalating limit) and BDS showed milder effects on the selected indices. Apoptosis as an originally designed cellular survival functional index unexpectedly docked just published pathogenesis [10] and the present pharmacology relevant to coronavirus patients in Fig. 1D. Although ex vivo responses to BDS and BS were mild (Fig. 1A-D), it is necessary to eliminate this minor interference because of potential endogenous anti- or pro-inflammation [8]. Here, we hypothesize that LH if applied to COVID-19-treated Huh-7 cells in its proxy form (LPSDS) will exhibit stronger attenuation of virus replication and suppression of inflammation. The precise antiviral compounds of LH elucidated in the above-mentioned CSS research could further support this hypothesis.

Thus, using the above design, we conducted simulations of relevant experiments in Huh-7 cells. In H2O2-treated Huh-7 cells, IL-6, TNF-α, and ROS were selected to mimic the virus-induced inflammatory response (different etiology with similar inflammation), and apoptosis rate following oxidative stress was measured to evaluate cellular function. The virus and H2O2 increased TNF-α levels by ∼4.0- [1] and ∼7.0-fold (vs. blank, in Fig. 1B), respectively. We compared crude LH (Lot No. B2001019 [1]; its nine chromatographic peaks were detected) extracts with LPSDS by using these indices for their consistency. The present LH ex vivo dose (150–600 μg mL−1 [1]) was almost identical to ∼5%–19 % (bioavailability) of conversion in vivo dose following calculation via coefficient 6.7 [5]{Pinkel, 1958 #3} for a rat (0.1541 g kg−1), which is conversion dose from patient dose (0.0230 g kg−1) [3]. The theoretical bioavailability in rats is 0%–100 %, i.e. blank dose to complete absorption, with the latter observed near sites of intravascular injection. Among the dose-dependent (150–600 μg mL−1) ex vivo responses to LH, 600 μg mL−1 significantly reduced mRNA levels (pg mL−1) of IL-6 by 44.12 % [60.53 ± 0.54 vs. 108.34 ± 1.39 (model); p＜0.0001] (Fig. 1A), TNF-α by 61.30 % [16.73 ± 0.40 vs. 43.23 ± 0.50 (model); p＜0.0001] (Fig. 1B), as well as ROS level by 54.42 % [8611.63 ± 454.62 vs. 18893.20 ± 147.31 (model); p＜0.0001] (Fig. 1C). It also inhibited apoptosis by 57.71 % [8.47 ± 0.39 vs. 20.03 ± 0.41 (model); p＜0.0001] (Fig. 1D). Unexpectedly, the reduction of the selected indices by LH (600 μg mL−1) was lesser than that by its high-dose LPSDS, i.e. H-LPSDS (IL-6, 44.12 % vs. 64.60 %; TNF-α, 61.30 % vs. 79.10 %; ROS, 54.42 % vs. 82.03 %; and apoptosis, 57.71 % vs. 83.76 %. Middle- and low-doses LPSDS (M- and L-LPSDS) were shown in Fig. 1A-D), which differs from the results of experiments on herbal extracts similar to LH [9]. Compared with H-, M- and L-LPSDS, LH has relatively unclear apoptosis-inducing substances, while as in vitro agent, BS (with volume-escalating limit) and BDS showed milder effects on the selected indices. Apoptosis as an originally designed cellular survival functional index unexpectedly docked just published pathogenesis [10] and the present pharmacology relevant to coronavirus patients in Fig. 1D. Although ex vivo responses to BDS and BS were mild (Fig. 1A-D), it is necessary to eliminate this minor interference because of potential endogenous anti- or pro-inflammation [8]. Here, we hypothesize that LH if applied to COVID-19-treated Huh-7 cells in its proxy form (LPSDS) will exhibit stronger attenuation of virus replication and suppression of inflammation. The precise antiviral compounds of LH elucidated in the above-mentioned CSS research could further support this hypothesis.

Financial support

This work was supported by the National Natural Science foundation of China (grant numbers: 81973589 and 81373855) and the Priority Academic Program Development of Jiangsu Higher Education Institutions (integration of Chinese and western medicine).

Authors’ contributions

Xi Huang as the correspondence author designed this work, provided funding, and wrote the letter to the editor; Qiulong Zhao, Liang Xia, and Shaoqi Shi conducted experiments, data analyses, and submitted manuscript. All authors reviewed and approved the final version.

Declaration of Competing Interest

The authors have no conflict of interest to declare with respect to this manuscript.