Abstract

   COVID-19 is an acute respiratory disease caused by SARS-CoV-2 that has
   spawned a worldwide pandemic. ADAM17 is a sheddase associated with the
   modulation of the receptor ACE2 of SARS-CoV-2. Studies have revealed
   that malignant phenotypes of several cancer types are closely relevant
   to highly expressed ADAM17. However, ADAM17 regulation in SARS-CoV-2
   invasion and its role on small molecules are unclear. Here, we
   evaluated the ADAM17 inhibitory effects of cordycepin (CD),
   thymoquinone (TQ), and N6, N6-dimethyladenosine (m^6[2]A), on cancer
   cells and predicted the anti-COVID-19 potential of the three compounds
   and their underlying signaling pathways by network pharmacology. It was
   found that CD, TQ, and m^6[2]A repressed the ADAM17 expression upon
   different cancer cells remarkably. Moreover, CD inhibited GFP-positive
   syncytia formation significantly, suggesting its potential against
   SARS-CoV-2. Pharmacological analysis by constructing CD-, TQ-, and
   m^6[2]A-based drug-target COVID-19 networks further indicated that
   ADAM17 is a potential target for anti-COVID-19 therapy with these
   compounds, and the mechanism might be relevant to viral infection and
   transmembrane receptors-mediated signal transduction. These findings
   imply that ADAM17 is of potentially medical significance for cancer
   patients infected with SARS-CoV-2, which provides potential new targets
   and insights for developing innovative drugs against COVID-19.

   Keywords: ADAM17, cancer, SARS-CoV-2, cordycepin (CD), thymoquinone
   (TQ), N6, N6-Dimethyladenosine (m^6[2]A), network-pharmacological
   analysis

1. Introduction

   Since December 2019, SARS-CoV-2 has caused the global spread of
   COVID-19 (Corona Virus Disease 2019), which seriously affects people’s
   healthy lives [[46]1]. SARS-CoV-2 is an enveloped virus that has caused
   approximately 300 million infections and more than 5 million deaths
   worldwide. Multiple receptors that may be associated with SARS-CoV-2
   have been identified, such as transmembrane serine protease 2
   (TMPRSS2), transmembrane serine protease 4 (TMPRSS4),
   angiotensin-converting enzyme 2 (ACE2), aminopeptidase N (APN),
   cathepsin L (CTSL), heparan sulfate proteoglycans (HSPGs), furin,
   neuropilin-1 (NRP1), heat shock protein A5 (HSPA5), O-acetylated sialic
   acid (O-Ac-Sia), etc. [[47]2,[48]3,[49]4,[50]5,[51]6,[52]7]. Studies
   have shown that ACE2 is expressed broadly on the membrane of epithelial
   cells in the apical region, and especially on the cilium [[53]8].
   Coronavirus host cell entry is attributed to ACE2 receptors on the
   alveolar surface [[54]9,[55]10]. Specifically, the S protein is a very
   important surface protein of SARS-CoV-2 and exists as a trimer with
   more than 300 amino acids in each monomer forming a receptor-binding
   domain that binds tightly to ACE2 and facilitates viral entry of lung
   cells [[56]11]. ACE2 was found to potentiate cell invasion by the
   COVID-19-spike (S) protein spseudovirus, but no similar phenomenon was
   observed in ACE2 mutant cells [[57]12]. Therefore, it seems that
   cells/tissues expressing ACE2 are susceptible to invasion by this
   virus. Conversely, in ACE2 mutant mouse models, peak coronavirus
   injection aggravated acute lung injury and even lung failure, whereas
   recombinant ACE2 treatment dramatically attenuates acute lung injury
   [[58]13,[59]14]. On this basis, the researchers proposed the dubious
   hypothesis that an increase in plasma membrane-bound ACE2 may lead to
   higher infection rates, although the critical function of ACE2 within
   COVID-19 infection remains unknown.

   A disintegrin and metalloproteinase 17 (ADAM17; OMIM: 603639), also
   known as TNFα converting enzyme (TACE), CD156B, NISBD1, and CSVP are
   mainly present in two forms: precursor and activated ADAM17 [[60]15].
   ADAM17 facilitates EGF receptor-, TNF-α receptor-, and IL-6
   receptor-mediated signal transduction by cleavage/hydrolysis of the
   ectodomains of more than 80 membrane proteins (e.g., pro-inflammatory
   cytokines, adhesion molecules, receptors, and growth factors) and
   regulates multiple pathophysiological processes, including tumor,
   inflammation, immunity, growth, and metastasis [[61]16,[62]17,[63]18].
   Furthermore, ADAM17 activity is also regulated by posttranslational
   modifications, pro-domain removal, conformational changes, and
   glycosylation [[64]19]. The proteins iRhom1 and iRhom2 are required for
   ADAM17 trafficking and enzyme activation [[65]20,[66]21]. ADAM17 is a
   newly identified SARS-CoV-2-related protein. It is highly expressed in
   nasopharyngeal, bronchial, and lung tissues [[67]22]. As a type I
   transmembrane protein and sheddase, ADAM17 probably contributes to the
   ectodomain shedding of ACE2 [[68]23], leading to the formation of
   soluble ACE2 and the induction of SARS-CoV-2 virus fusion with host
   cells [[69]24,[70]25]. Evidence for ADAM17-mediated infection of
   SARS-CoV-2 in human lung cells suggests that ADAM17 triggers SARS-CoV-2
   invasion by cleaving the S protein of the virus [[71]26]. Suppression
   of ADAM17 by reducing the shedding of ACE2 may reduce viral load and
   protect host cells from viral infection [[72]27]. In addition to
   affecting membrane protein shedding, ADAM17 also induces local tumor
   metastasis and invasion through degradation of the cellular basement
   membrane and extracellular matrix. The fact that ADAM17 has been
   reported to be expressed in diverse malignancies, but minimally in
   normal cells [[73]28], implies its role in tumors. In this regard,
   ADAM17 inhibition might alleviate the severe infections that occur in
   patients with malignant tumors. Previously, it was found that ADAM17
   expression was higher in tumors than in normal tissues, and that it
   reduced the prognosis of patients with malignant tumors [[74]22].
   Interestingly, ADAM17 was also associated with immune cell infiltration
   and immunomodulators in tumor and paired normal tissues [[75]22]. These
   results implicate the importance of ADAM17 in COVID-19-infected cancer
   patients and provide a new thread for developing anti-COVID-19 drugs.

   Chinese herbal medicines and their active ingredients are characterized
   by multi-targets, multi-pathways, and low toxicity, and are widely used
   in antiviral and anti-tumor therapy. Cordycepin (CD), N6,
   N6-dimethyladenosine (m^6[2]A), and thymoquinone (TQ) are nucleoside
   analogues with a wide range of pharmacological functions. CD is a
   natural ingredient originally found in traditional Chinese medicine
   (TCM) as a medicinal mushroom [[76]29,[77]30]. Our previous studies
   showed that CD inhibited the malignant progression of drug-resistant
   non-small cell lung cancer through regulation of the AMPK signaling
   pathway [[78]29] and repressed triple-negative breast cancer (TNBC)
   cell migration and invasion by the down regulation of
   epithelial-to-mesenchymal transition inducible transcription factors
   (EMT-TFs) [[79]31]. TQ is a biologically active phytochemical component
   in the seed oil of Panicola nigra [[80]32]. Numerous studies have shown
   that TQ can protect against cardiovascular diseases [[81]33] and viral
   infections [[82]34]. Our results suggested that TQ or its analogs
   synergistically enhanced the antitumor effects of chemotherapeutic
   agents [[83]35] and repressed the metastasis and invasion of distinct
   tumor cells [[84]36,[85]37]. m^6[2]A is a modified ribonucleoside in
   rRNA and an endogenous A3 adenosine receptor ligand [[86]38,[87]39]. It
   is also a marine-derived AKT inhibitor [[88]40], and its analogues
   showed enhanced antitumor effects [[89]41]. CD, TQ, and m^6[2]A showed
   anticancer roles both in cells and animals. However, the regulatory
   effects of CD, TQ, and m^6[2]A on ADAM17 expression and their potential
   anti-COVID-19 mechanisms remain unclear.

   In this study, we evaluated the ADAM17 inhibitory effects of CD, TQ,
   and m^6[2]A on cancer cells and constructed a visual
   drug-target-disease network to predict prospective targets and
   mechanisms of the anti-COVID-19 drugs using network pharmacology.

2. Results

2.1. The Mutation and Methylation of ADAM17 in Pan-Cancer

   Previously, we studied the expression of ADAM17 in cancer patients
   and/or normal subjects using bioinformatics analysis. Hereby, we
   further explored epigenetic modification and gene mutation and their
   impact on prognosis. Oncogenic mutations may lead to malignant
   phenotypes of tumors, recurrence, and drug resistance. Through the
   analysis of mutations in 32 types of cancers, we revealed that ADAM17
   expression exhibited the highest mutation frequency, accounting for
   10.53% of 57 cases of Uterine Carcinosarcomas, followed by Uterine
   Corpus Endometrial Carcinoma, accounting for 6.62% of 529 cases,
   whereas Kidney Renal Clear Cell Carcinoma had the lowest mutation rate,
   with 0.39% of 511 cases ([90]Figure 1A). ADAM17 mutations were not
   observed in Adrenocortical Carcinoma, Thyroid Carcinoma, Acute Myeloid
   Leukemia, Cholangiocarcinoma, and Uveal Melanoma. The detailed
   landscape of ADAM17 mutations appears to be distributed across the
   whole AMAD17 region, with missense being the dominant mutation type. We
   further explored the survival rates, but there was no significant
   difference between the two groups, although the median monthly
   survivals (overall, progression-free, disease-specific, and disease
   free) were longer in the altered group than in the unchanged group
   (Data not shown).

Figure 1.

   [91]Figure 1
   [92]Open in a new tab

   The mutation and methylation of ADAM17 in pan-cancer. (A) Summary of
   ADAM17 mutations across pan-cancerous tissues. Distinct colors show
   differing mutation types. (B) The hot spots of ADAM17 mutations across
   pan-cancer tissues. (C) The methylation of ADAM17 in BLCA. (D) The
   methylation of ADAM17 in READ. (E) The methylation of ADAM17 in THCA.
   (F) The methylation of ADAM17 in BRCA. (G) The methylation of ADAM17 in
   KIRC. (H). The methylation of ADAM17 in KIRP. (I) The methylation of
   ADAM17 in LUSC. (J) The methylation of ADAM17 in PAAD. (K) The
   methylation of ADAM17 in PRAD. (L) The methylation of ADAM17 in SARC.
   (M) The methylation of ADAM17 in SKCM. See [93]Figure 2 for the full
   name of the cancer types. The p < 0.05 was considered with statistical
   significance.

   DNA methylation regulates gene expression. The methylation profiles of
   23 types of cancers were analyzed by the DNMIVD database. We found that
   the methylation status of the ADAM17 promoter in BLCA, THCA, and READ
   tumor tissues was significantly lower than correlated normal tissues
   ([94]Figure 1C–E). The methylation status of the ADAM17 promoter was
   markedly higher in BRCA, KIRC, KIRP, LUSC, PAAD, PRAD, SARC, and SKCM
   tumor tissues than in paired normal tissues ([95]Figure 1F–M). The
   hypomethylation of ADAM17 in PAAD tumor tissues appeared to be
   positively correlated with the higher expression shown in our previous
   report, implying that hypomethylation of the ADAM17 promotor may be
   responsible for its increased expression in PAAD tissues. However,
   promoter methylation in other cancer types may not be the only
   mechanism regulating ADAM17 overexpression.

2.2. Treatment with m^6[2]A, TQ, and CD Represses ADAM17 Expression in
Distinct Cancer Cells

   CD, m^6[2]A, and TQ are nucleoside analogues that have broad
   pharmacological effects, such as anti-tumor, anti-virus, etc. In this
   regard, we analyzed the effects of CD, m^6[2]A, and TQ on ADAM17 levels
   in different cancer cells. Interestingly, CD, m^6[2]A, and TQ
   dose-dependently reduced ADAM17 protein levels in different cancer
   cells ([96]Figure 2A–I, upper panels), but mRNA levels were not reduced
   ([97]Figure 2A–I, bottom panels), implying the potential of the three
   compounds in improving viral susceptibility in patients with malignant
   tumors.

Figure 2.

   [98]Figure 2
   [99]Open in a new tab

   CD, TQ, and m^6[2]A inhibit ADAM17 expression in distinct cancer cells.
   (A) CD-treated lung cancer cell line H460 was measured for alterations
   in ADAM17 protein (upper panel) and mRNA (lower panel) levels. (B)
   CD-treated breast cancer cell line BT549 was measured for alterations
   in ADAM17 protein (upper panel) and mRNA (lower panel) levels. (C)
   CD-treated breast cancer cell line MDA-MB-231 was measured for
   alterations in ADAM17 protein (upper panel) and mRNA (lower panel)
   levels. (D) CD-treated breast cancer cell line MCF7 was measured for
   alterations in ADAM17 protein (upper panel) and mRNA (lower panel)
   levels. (E) m^6[2]A-treated lung cancer cell line H1975 was measured
   for alterations in ADAM17 protein (upper panel) and mRNA (lower panel)
   levels. (F) m^6[2]A-treated breast cancer cell line BT549 was measured
   for alterations in ADAM17 protein (upper panel) and mRNA (lower panel)
   levels. (G) m^6[2]A-treated breast cancer cell line MDA-MB-231 was
   measured for alterations in ADAM17 protein (upper panel) and mRNA
   (lower panel) levels. (H) TQ-treated breast cancer cell line MCF7 was
   measured for alterations in ADAM17 protein (upper panel) and mRNA
   (lower panel) levels. (I) TQ-treated lung cancer cell line H460 was
   measured for alterations in ADAM17 protein (upper panel) and mRNA
   (lower panel) levels.

2.3. Treatment with m^6[2]A or CD Inhibits ADAM17 Translation but Prevents
Its Degradation

   The stability of the ADAM17 protein was detected in CHX-treated BT549
   cells in the presence/ absence of m^6[2]A. In CHX-treated cells, ADAM17
   protein had a half-life of less than 2.3 h, whereas the addition of
   m^6[2]A significantly extended the half-life of ADAM17 protein to 4 h
   ([100]Figure 3A,B). Further, quantitative analysis showed that the
   combination of m^6[2]A and CHX markedly decreased the degradation rate
   of ADAM17 protein by nearly 50% compared to CHX treatment alone
   ([101]Figure 3C). Such results were then observed in cells treated with
   CD or CHX alone and their combination ([102]Figure 3D–F). Overall,
   these results indicate that m^6[2]A or CD treatment inhibits ADAM17
   translation, but prevents its degradation.

Figure 3.

   [103]Figure 3
   [104]Open in a new tab

   Treatment with m^6[2]A or CD inhibits the translation, but prevents the
   degradation of ADAM17 in BT549 cancer cells. (A) Stability of the
   ADAM17 protein was detected in CHX-treated cells in the presence or
   absence of m^6[2]A. (B) The quantitative results of A. (C) Quantitated
   the ADAM17 protein levels with m^6[2]A, but without CHX treatment. (D)
   Stability of the ADAM17 protein was detected in CHX-treated cells in
   the presence or absence of CD. (E) The quantitative results of (D). (F)
   Quantitated the ADAM17 protein levels with CD, but without CHX
   treatment. 40 µg/mL was used for the final concentration of CHX.

2.4. Network Pharmacology Analyzes Potential Targets of m^6[2]A, TQ, and CD

   Based on the databases, we collected a total of 100, 100, and 82
   potential targets for m^6[2]A, TQ, and CD, respectively. The
   interactions of these targets for different compounds analyzed in the
   STRING database were visualized with Cytoscape software (version
   3.6.1). In [105]Figure 4A, [106]Figure 5A, and [107]Figure 6A, we found
   that ADAM17 is a potential target for m^6[2]A and TQ, but not CD. To
   further explore the relationship between the three compounds and their
   targets, KEGG analysis was performed. A total of 134 pathways of
   m^6[2]A, 75 pathways of TQ, and 71 pathways of CD were enriched by KEGG
   analysis (adjusted p < 0.01). Based on the GeneRatio values and
   p-value, the top 20 pathways of these compounds were selected, e.g.,
   pathways in cancer, human cytomegalovirus infection, lipid and
   atherosclerosis, chemical carcinogenesis-receptor activation,
   neuroactive ligand-receptor interaction, purine metabolism, calcium
   signaling pathway, glucagon signaling pathway, and ABC transporters,
   and were shown in [108]Figure 4B, [109]Figure 5B, and [110]Figure 6B.

Figure 4.

   [111]Figure 4
   [112]Open in a new tab

   The “m^6[2]A −targets” network built by Cytoscape 3.6.1. PPI network
   (A) and KEGG pathway enrichment analysis (B) of m^6[2]A. The nodes
   represent proteins, and the edges refer to interactions between
   proteins. The size of the circles and their distinct colors represent
   degrees.

Figure 5.

   [113]Figure 5
   [114]Open in a new tab

   The “TQ−targets” network built by Cytoscape 3.6.1. PPI network (A) and
   KEGG pathway enrichment analysis (B) of TQ. The nodes represent
   proteins, and the edges refer to interactions between proteins. The
   size of the circles and their distinct colors represent degrees.

Figure 6.

   [115]Figure 6
   [116]Open in a new tab

   The “CD−targets” network built by Cytoscape 3.6.1. PPI network (A) and
   KEGG pathway enrichment analysis (B) of CD. The nodes represent
   proteins, and the edges refer to interactions between proteins. The
   size of the circles and their distinct colors represent degrees.

2.5. Construction of the Compounds-Targets-COVID-19 Network

   To elucidate the mechanism of the compounds against COVID-19, we
   collected the common differentially expressed genes of the three
   constructed compound-target networks and 1145 COVID-19-associated
   target gene networks retrieved from the databases. Protein-protein
   interactions were obtained by uploading the above described candidate
   targets to the STRING database and assigning p-values less than 1, 10,
   and 16. We found that ADAM17 is also one of the larger nodes,
   indicating that ADAM17 may play a very important role in the
   anti-COVID-19 effect of the three compounds ([117]Figure 7A,
   [118]Figure 8A, and [119]Figure 9A). We further conducted GO
   enrichment/KEGG analysis to investigate the potential mechanism of the
   three compounds in COVID-19, using the DAVID database.

Figure 7.

   [120]Figure 7
   [121]Open in a new tab

   Establishment of an ingredient-target-disease network for m^6[2]A in
   COVID-19. (A) The Venn diagram exhibited a total of 24 m^6[2]A targets
   for COVID-19 were extracted. ADAM17 was one of the key targets.
   Enrichment analysis of pathways for the 24 hub genes. (B) The top 3
   terms of BP, CC, and MF enrichment. (C) Enrichment analysis of pathways
   for the 24 DEGs.

Figure 8.

   [122]Figure 8
   [123]Open in a new tab

   Establishment of an ingredient-target-disease network for TQ in
   COVID-19. (A) The Venn diagram exhibited a total of 25 TQ targets for
   COVID-19 were extracted. ADAM17 was one of the key targets. Enrichment
   analysis of pathways for the 25 hub genes. (B) The top 3 terms of BP,
   CC, and MF enrichment. (C) Enrichment analysis of pathways for the 25
   DEGs.

Figure 9.

   [124]Figure 9
   [125]Open in a new tab

   Establishment of an ingredient-target-disease network for CD in
   COVID-19. (A). The Venn diagram exhibited a total of 16 CD targets for
   COVID-19 were extracted. Enrichment analysis of pathways for the 16 hub
   genes. (B) The top 3 terms of BP, CC, and MF enrichment. (C) Enrichment
   analysis of pathways for the 16 DEGs.

   For GO-term analysis, key terms of BP enrichment include: (1) m^6[2]A:
   peptidyl-serine phosphorylation, cellular response to the drug,
   positive regulation of cellular protein localization; (2) TQ: response
   to lipopolysaccharide, multicellular organismal homeostasis, response
   to molecule of bacterial origin; (3) CD: peptidyl-tyrosine
   modification, protein autophosphorylation, peptidyl-tyrosine
   phosphorylation ([126]Figure 7B, [127]Figure 8B, and [128]Figure 9B).
   MF analysis was mainly enriched in (1) m^6[2]A: ubiquitin-like protein
   ligase binding, endopeptidase activity, ubiquitin protein ligase
   binding; (2) TQ: steroid hormone receptor activity, transcription
   factor activity, nuclear receptor activity, direct ligand regulated
   sequence-specific DNA binding; (3) CD: transmembrane receptor protein
   kinase activity, non-membrane spanning protein tyrosine kinase
   activity, protein tyrosine kinase activity ([129]Figure 7B, [130]Figure
   8B, and [131]Figure 9B). CC terms revealed that: (1) m^6[2]A was
   associated with membrane region, membrane microdomain, and membrane
   raft; (2) TQ was associated with the early phagosome, transcription
   factor complex, and endolysosome; (3) CD was associated with
   transferring phosphorus-containing groups, transferase complex, protein
   kinase complex, and serine/threonine protein kinase complex
   ([132]Figure 7B, [133]Figure 8B, and [134]Figure 9B). For KEGG
   analysis, multiple candidate genes have interacted with
   COVID-19-related pathways. Among them, virus infection and protein
   phosphorylation were considered 2 of the top 12 enrichment pathways
   associated with these compounds ([135]Figure 7C, [136]Figure 8C, and
   [137]Figure 9C). Additionally, the anti-COVID-19 mechanism may also be
   related to ubiquitin-like protein ligase binding, membrane microdomain,
   membrane raft, direct ligand-regulated sequence-specific DNA binding,
   endolysosome, and transmembrane receptor protein kinase activity.

2.6. CD Is Able to Inhibit Syncytia Formation

   Emerging evidence indicates that a large number of multinucleated cells
   characteristic of syncytial pathology are present in patients with
   COVID-19 [[138]42], which is a pathological hallmark of SARS-CoV-2
   infection. Syncytium formation is required for the participation of the
   SARS-CoV-2 Spike protein and the host cell ACE2 [[139]26]. Here,
   SARS-CoV-2-Spike plasmids with GFP fluorescence were transfected into
   293T-ACE2 cells and co-incubated with untransfected 293T-ACE2 cells for
   24 h, followed by the addition of 20 μM CD for another 24 h. We
   observed many larger syncytia with GFP green fluorescence in control
   cells, indicating SARS-CoV-2 cell invasion ([140]Supplementary Figure
   S1A). However, the area of fluorescence of GFP-positive syncytia was
   significantly reduced upon treatment with CD ([141]Supplementary Figure
   S1A). Further, quantitative analysis revealed that the mean
   fluorescence area of CD-treated syncytia was reduced to 44.34% ± 25.54
   compared to the control group ([142]Supplementary Figure S1B).

3. Discussions

   COVID-19 has become a global public health problem. It is of great help
   to understand the expression of potential SARS-CoV-2 receptors (e.g.,
   ACE2 and ADAM17) in host cells/tissues to reduce the replication and
   transmission of the virus and the severity of COVID-19. Studies have
   shown that ACE2 is involved in the viral invasion of host cells,
   leading to infection by binding to the B domain of the COVID-19 virus’s
   S protein [[143]43,[144]44,[145]45]. Similar to ACE2, ADAM17 is another
   newly identified virus recognition receptor [[146]46], and its
   distributed and expressed levels may mirror the susceptibility of the
   virus, its replication, and its invasion. Inhibition of ADAM17
   expression protects the body from COVID-19 infection [[147]47].
   Increasing evidence indicates that ADAM17 is expressed in multiple
   malignancies at higher levels than in paired normal tissues
   [[148]28,[149]48], suggesting its specificity in tumors. Moreover,
   invasion and metastasis of distinct malignancies can be attributed to
   ADAM17-mediated degradation of the cellular basement membrane and
   extracellular matrix [[150]48]. However, the underlying role of ADAM17
   in cancer patients infected with COVID-19 remains unknown.

   In a previously published paper [[151]22], we explored ADAM17 protein
   or its mRNA expression in pan-cancer and adjacent normal tissues. It
   was found that ADAM17 was significantly higher in multiple human tumor
   tissues than in adjacent tissues, and that cancer patients with high
   ADAM17 expression had a poor clinical prognosis. Gene mutations favor
   the progression, recurrence, and chemo-resistance of malignant tumors,
   and DNA methylation can provoke structural changes in chromosomes that
   lead to tumorigenesis by turning off tumor suppressor genes. Actually,
   point mutations of the ADAM17 catalytic domain have been identified in
   tumor samples from cancer patients, which are associated with
   tumor-related dysfunction [[152]49]. Meanwhile, protein
   posttranslational modifications have been found to be closely relevant
   to the regulation of ADMA17 activation [[153]50,[154]51]. We speculate
   that both may probably be related to tumor progression. In this study,
   ADAM17 was found to be expressed in 57 cases of uterine carcinosarcoma
   with the highest mutation frequency (10.53%), while it was barely
   expressed in 511 cases of kidney renal clear cell carcinoma with a
   mutation rate of 0.39%. By analyzing the methylation profiles, the
   methylation status of the ADAM17 promoter in BLCA, READ, and THCA tumor
   tissues was distinctly lower than in adjacent normal tissues, whereas
   its methylation status was higher in BRCA, KIRC, KIRP, LUSC, PAAD,
   PRAD, SARC, and SKCM tumor tissues. Our previous study indicated that
   hypomethylation of ADAM17 in PAAD tumor tissues was positively
   correlated with high ADAM17 expression, indicating that hypomethylation
   of the ADAM17 promoter may be responsible for its increased expression
   in PAAD tissues. However, promoter methylation in other tumor types may
   not be the only mechanism regulating ADAM17 overexpression. In
   addition, the catalytic domain of ADAM17 is required for the cleavage
   of the substrates [[155]50]. The cleavage and release of some
   substrates, including ACE2, directly drive viral invasion [[156]11].
   Hence, it is not difficult to speculate about the effect of
   site-specific mutations or post-translational modifications of the
   catalytic domain on the shedding activity of ADAM17 and viral invasion.

   Alterations in ADAM17 expression affect susceptibility to viral
   infection and the severity of COVID-19, implying the importance of
   targeting ADAM17 in patients with malignant tumors infected with
   COVID-19 [[157]47]. CD is a nucleoside derivative extracted from
   Cordyceps Sinensis with a wide-range of biological activities,
   including antiviral replication, anticancer, anti-inflammatory,
   antidepressant, hepatic, neuroprotective, etc.
   [[158]29,[159]52,[160]53]. Recent studies have shown that CD is able to
   conjugate to the S protein and Mpro protein of SARS-CoV-2. In vitro
   SARS-CoV-2 invasion assay supports the antiviral effect of CD
   [[161]54]. In this study, CD suppressed ADAM17 expression, especially
   in lung cancer, which supports the possibility of developing
   anti-SARS-CoV-2 drugs and suggests a role for CD in anti-SARS-CoV-2
   therapy in cancer patients by inhibiting ADAM17 expression. m^6[2]A,
   another nucleoside derivative, exhibited an inhibitory effect on CTSL
   protein expression. TQ, a major component of N. Sativa, has been
   reported to potentially suppress the development of COVID-19 by binding
   to TMPRSS2 [[162]32,[163]55]. TQ exerts a suppressive effect on the
   malignant progression of cancer cells [[164]36,[165]56]. Chase assays
   confirmed that treatment with m^6[2]A or CD increased the stability of
   ADAM17 protein, but significantly decreased its total protein levels,
   suggesting that m^6[2]A/CD treatment alone inhibits ADAM17 translation,
   but prevented its degradation. The different mechanisms may be
   connected with the action of m^6[2]A. Collectively, our results showed
   that CD, m^6[2]A, and TQ remarkably reduced ADAM17 protein levels in
   distinct tumor cells. A large number of multinucleated cells
   characteristic of syncytial pathology are present in COVID-19 patients
   [[166]42], which is a pathological hallmark of SARS-CoV-2 infection.
   Syncytium formation is required for the participation of SARS-CoV-2
   Spike protein and ACE2 protein of the host cells [[167]26]. Our results
   showed many large syncytia with GFP green fluorescence in control
   cells, whereas the area of fluorescence of GFP-positive syncytia was
   significantly reduced upon addition of CD. The above results suggest a
   potential for these drugs against SARS-CoV-2, although further studies
   are needed.

   Network pharmacology is an approach based on systematic biological
   theory to predict underlying molecular mechanisms through
   high-throughput virtual computing and database retrieval, as well as
   the establishment of the drug (compound)-target-disease-pathway
   network. Recently, a large number of studies have employed this
   approach to explain the complex interrelationships between drugs and
   diseases. Herein, the relationship between anti-COVID-19 treatment and
   m^6[2]A/TQ/CD treatment was fully explored by network pharmacology.
   These data suggested that 100, 100, and 82 targets were respectively
   associated with m^6[2]A, TQ, and CD on COVID-19, in which (GAPDH, MMP9,
   SRC, EGFR, EZH2, MAPK1, PRMT1), (IL-6, PTGS2, PPARG, SLC6A4, PPARA,
   ACHE, NR3C1), and (ADK, MTR, AKT1, MTHFR, MTRR, MUT, MAT1A, DCK) were
   considered as core targets, respectively. It has been reported that SRC
   [[168]57], EGFR [[169]58], EZH2, IL-6 [[170]59], PTGS2 [[171]60], PPARG
   [[172]61], SLC6A4 [[173]62], PPARA [[174]63], ACHE [[175]64], NR3C1
   [[176]65], ADK [[177]66], AKT1 [[178]67], and MTHFR [[179]68] are
   closely related to the pathogenesis of COVID-19. Interestingly, we
   found that ADAM17 is a potential target for m^6[2]A and TQ, rather than
   CD, which may be related to the number of articles published in these
   years. Further, 134 m^6[2]A-related pathways, 75 TQ-related pathways,
   and 71 CD-related pathways were enriched by KEGG enrichment analysis,
   among which 20 pathways were significantly enriched, including pathways
   in cancer, human cytomegalovirus infection, lipid and atherosclerosis,
   chemical carcinogenesis-receptor activation, purine metabolism, calcium
   signaling pathway, neuroactive ligand-receptor interaction, glucagon
   signaling pathway, ABC transporters, etc. In recent years, evidence has
   shown that COVID-19 is implicated in neuroactive ligand-receptor
   interactions [[180]69], and calcium signaling pathways conduce to the
   shedding of ACE2 catalytic outer domains during COVID-19 infection
   [[181]70]. Purine metabolism is found to be associated with COVID-19
   infection [[182]71]. According to the findings of the network
   pharmacology analysis, we identified the regulatory mechanisms of
   m^6[2]A, TQ, and CD in COVID-19 disease involving multiple pathways and
   multiple targets. Protein-protein interaction results indicated that
   ADAM17 might play critical roles in the anti-COVID-19 activity of these
   three compounds. Further studies revealed that multiple candidate genes
   interact with COVID-19-related pathways, with the viral infection and
   protein phosphorylation identified as 2 of the top 12 enrichment
   pathways associated with these compounds. In addition,
   sequence-specific DNA-binding endolysosome and transmembrane receptor
   protein kinase activity directly regulated by ligands in the
   ubiquitin-like protein ligase membrane microdomain may also be the
   mechanism of its resistance to COVID-19. Possible explanations for the
   mechanism may be as follows: ADAM17 is involved to some extent in the
   potential antiviral effects of small molecule compounds of traditional
   Chinese medicine. ADAM17 also is a shedding enzyme that helps shed and
   activate nearly 100 substrates and, thus, possesses a wide range of
   biological effects. Some core targets (such as EGFR, IL-6, ACE2, etc.)
   in the “drug-target” network are the direct substrates of ADAM17.
   Although it does not play a primary role in the network, ADAM17
   probably mediates the shedding and activation of substrates EGFR
   [[183]72], IL-6 [[184]73,[185]74,[186]75], and ACE2 [[187]76] and then
   causes inflammation and immune responses, which further demonstrates
   the multi-target and versatility of the active molecules of traditional
   Chinese medicine.

4. Materials and Methods

4.1. Online Databases

   ADAM17 promoter DNA methylation analysis was conducted with the DNA
   Methylation Interactive Visualization Database (DNMIVD) [[188]77]. The
   ADAM17 mutation was analyzed by the cBioPortal for Cancer Genomics
   [[189]4,[190]78].

4.2. Cell Culture and Drug Treatments

   Human lung adenocarcinoma cells (H1975 and H460) and breast cancer
   cells (BT549, MDA-MB-231, and MCF7) were supplied by ATCC (American
   Type Culture Collection) (VA, USA). BT549 cells and H1975 were cultured
   in RPMI1640 medium containing 10% FBS (fetal bovine serum) and 1%
   penicillin/streptomycin (PS). The remaining 3 cells were maintained in
   DMEM mediums with 10% FBS and 1% PS. All the cells were incubated at 37
   °C in a 5% CO[2] incubator. For the following in vitro experiments,
   cells were seeded into 12-well plates and treated with distinct
   concentrations of cordycepin (CD, 10–40 μM), thymoquinone (TQ, 5–20
   μM), and N6, N6-Dimethyladenosine (m^6[2]A, 10–40 μM) for 24 h. The
   whole protein lysates were gathered and analyzed by Western blotting.
   Total RNA was extracted and reverse transcription was conducted.

4.3. Western Blotting and Chase Assays

   Western blotting was performed as described previously [[191]79]. The
   PVDF membranes with quantified protein samples were blocked with 5%
   non-fat milk and incubated with a primary antibody to ADAM17 (cat #:
   HPA051575, Sigma-Aldrich), β-actin, or tubulin in 2% fat-free milk at 4
   °C and blotted with a secondary antibody at ambient temperature.
   Protein bands were exposed and photographed using the Syngene G: BOX
   Imaging System (Cambridge, UK). Chase assays for ADMA17 protein
   stability were performed with indicated cycloheximide (CHX) and m^6[2]A
   or CD in BT549 cancer cells.

4.4. Semi-Quantitative RT-PCR

   For the detection of ADAM17 mRNA, the total RNA of different tumor
   cells was extracted, as per instructions from the manufacturer, using a
   Total RNA Extraction Kit. Synthesized cDNAs were subjected to
   semi-quantitative RT-PCR with forward primer 5′-cccaccagagactcgagaag-3′
   and the reverse primer 5′-caaccacgtgtccagtgaag-3′ [[192]80]. The
   product size is 279 bp. ACTB was used as the internal control.

4.5. Establishment of Small Molecule Compounds-Target Interaction

   TQ, m^6[2]A, and CD were searched for in the Traditional Chinese
   Medicine Systems Pharmacology (TCMSP, [193]http://tcmspw.com/tcmsp.php,
   accessed on 15 January 2022), Herbal Ingredients’ Targets Database
   (HIT, [194]http://lifecenter.biosino.org/hit/, accessed on 15 January
   2022), Traditional Chinese medicine integrative database for herb
   molecular mechanism analysis (TCMID,
   [195]http://www.megabionet.org/tcmid/, accessed on 15 January 2022),
   TCM@Taiwan ([196]http://tcm.cmu.edu.tw, accessed on 15 January 2022),
   CancerHSP ([197]http://ibts.hkbu.edu.hk/LSP/CancerHSP.php, accessed on
   15 January 2022), Naturally occurring Plant based Anticancerous
   Compound-Activity-Target DataBase (NPACT,
   [198]http://crdd.osdd.net/raghava/npact/, accessed on 15 January 2022),
   Natural Product Activity and Species Source Database (NPASS,
   [199]http://bidd2.nus.edu.sg/NPASS, accessed on 15 January 2022) system
   network-pharmacological databases. Drug likeness (DL) of greater than
   0.18 and oral bioavailability (OB) of greater than 30% were employed as
   criteria for screening potential compounds. Cytoscape software (version
   3.6.1) for the construction and visualization of the interactions
   between the three compounds and the target was used.

4.6. Gene Ontology and Pathway Enrichment Analysis

   COVID-19-related gene expression data were obtained from the
   GeneCards^®: The Human Gene Database ([200]https://www.genecards.org/,
   accessed on 12 September 2021) and Gene Expression Omnibus Database
   ([201]https://www.ncbi.nlm.nih.gov/geo/, accessed on 12 September
   2021). The genes that correlate with SARS-CoV-2 infection and negative
   control in the gene expression profile of this dataset were analyzed
   using the GEN2R online tool. A total of 1145 differentially expressed
   genes (DEGs) were obtained based on p ≤ 0.05 and fold change (FC) ≥
   1.5. Protein-protein interaction (PPI) analyses were performed using
   the STRING database ([202]https://cn.string-db.org/, accessed on 16
   January 2022) to obtain common genes. The Kyoto Encyclopedia of Genes
   (KEGG) and Gene Ontology (GO) analyses were carried out by the DAVID
   database ([203]http://david.abcc.ncifcrf.gov/, accessed on 16 January
   2022). Go annotations were used to analyze the following three terms:
   cellular composition (CC), biological processes (BP), and molecular
   function (MF).

4.7. Syncytia Formation Assay

   Syncytia formation is the hallmark event in SARS-CoV-2 cellular
   infections [[204]26]. In this study, SARS-CoV-2-Spike plasmids with GFP
   fluorescence pCDH-CMV-HnCoV-S-EF1-copGFP purchased from Shanghai
   HedgehogBio Science and Technology Ltd. were transfected into 293T-ACE2
   cells and co-incubated with untransfected 293T-ACE2 cells for 24 h,
   followed by the addition of 20 μM CD for another 24 h. Syncytia
   formation in each group was observed and analyzed using a ZOE
   Fluorescent Cell Imager (Bio-Rad, Hercules, CA, USA).

5. Conclusions

   In conclusion, hypomethylation of ADAM17 was positively correlated with
   high ADAM17 expression in PAAD tumor tissues, suggesting that
   hypomethylation of the ADAM17 promoter may be responsible for the
   increased ADAM17 expression in PAAD tissues. In addition, ADAM17
   protein levels were significantly reduced in lung and breast cancer
   cells treated with nucleoside compounds, such as CD, TQ, and m^6[2]A.
   CD significantly decreased the area of fluorescence of GFP-positive
   syncytia compared to the control group. Further, a network pharmacology
   approach was employed to establish a drug-disease-target network and
   analyze the potential molecular pathways by GO and KEGG enrichment
   analysis. Overall, this research not only insinuates the medical
   significance of ADAM17 for COVID-19 cancer patients, but it also sheds
   potential light on the treatment of COVID-19.

Abbreviations

   SARS-CoV-2 Severe acute respiratory syndrome coronavirus 2
   COVID-19 Coronavirus Disease 2019
   ADAM17 A disintegrin and metalloproteinase domain 17
   ATCC American Type Culture Collection
   CD Cordycepin
   TQ Thymoquinone
   m^6[2]A N6, N6-Dimethyladenosine
   CHX Cycloheximide
   TCMSP Traditional Chinese Medicine Systems Pharmacology
   TCMID Traditional Chinese medicine integrative database
   HIT Herbal Ingredients’ Targets Database
   NPACT Naturally occurring Plant based Anticancerous
   Compound-Activity-Target DataBase
   NPASS Natural Product Activity and Species Source Database
   DL Drug likeness
   OB Oral bioavailability
   DEGs Differentially expressed genes
   KEGG Kyoto Encyclopedia of Genes
   GO Gene ontology
   PPI Protein-protein interaction
   CC Cellular composition
   BP Biological processes
   MF Molecular function
   [205]Open in a new tab

Supplementary Materials

   The following supporting information can be downloaded at:
   [206]https://www.mdpi.com/article/10.3390/molecules27249044/s1, Figure
   S1: CD significantly inhibits syncytia formation.
   [207]Click here for additional data file.^ (406.9KB, zip)

Author Contributions

   J.F. (Junjiang Fu) conceived and coordinated the study. J.H., S.L.,
   Z.L., Q.T., T.L., C.W., X.L., Z.M., K.W., J.C. and J.F. (Jiewen Fu)
   carried out the experiments and analyzed and interpreted data. K.W.
   edited the manuscript. J.F. (Junjiang Fu) wrote and edited the
   manuscript. K.W. and J.F. (Junjiang Fu) revised the manuscript. All
   authors have read and agreed to the published version of the
   manuscript.

Institutional Review Board Statement

   Not applicable.

Informed Consent Statement

   Not applicable.

Data Availability Statement

   Not applicable.

Conflicts of Interest

   The authors declare no conflict of interest.

Funding Statement

   This work was financially supported by the Foundation of Southwest
   Medical University (grant nos. 2021ZKMS004, 2021ZKQN109), the
   Foundation of Science and Technology Department of Sichuan Province
   (grant no. 2022NSFSC0737), in part by the National Natural Science
   Foundation of China (grant nos. 81672887, 30371493, 82073263), and the
   Research Foundation of Luzhou City (grant no. 2021-SYF-37).

Footnotes

   Publisher’s Note: MDPI stays neutral with regard to jurisdictional
   claims in published maps and institutional affiliations.

References