Impact of TMPRSS2 Expression, Mutation Prognostics, and Small Molecule (CD, AD, TQ, and TQFL12) Inhibition on Pan-Cancer Tumors and Susceptibility to SARS-CoV-2

As a cellular protease, transmembrane serine protease 2 (TMPRSS2) plays roles in various physiological and pathological processes, including cancer and viral entry, such as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Herein, we conducted expression, mutation, and prognostic analyses for the TMPRSS2 gene in pan-cancers as well as in COVID-19-infected lung tissues. The results indicate that TMPRSS2 expression was highest in prostate cancer. A high expression of TMPRSS2 was significantly associated with a short overall survival in breast invasive carcinoma (BRCA), sarcoma (SARC), and uveal melanoma (UVM), while a low expression of TMPRSS2 was significantly associated with a short overall survival in lung adenocarcinoma (LUAD), demonstrating TMPRSS2 roles in cancer patient susceptibility and severity. Additionally, TMPRSS2 expression in COVID-19-infected lung tissues was significantly reduced compared to healthy lung tissues, indicating that a low TMPRSS2 expression may result in COVID-19 severity and death. Importantly, TMPRSS2 mutation frequency was significantly higher in prostate adenocarcinoma (PRAD), and the mutant TMPRSS2 pan-cancer group was significantly associated with long overall, progression-free, disease-specific, and disease-free survival rates compared to the wild-type (WT) TMPRSS2 pan-cancer group, demonstrating loss of functional roles due to mutation. Cancer cell lines were treated with small molecules, including cordycepin (CD), adenosine (AD), thymoquinone (TQ), and TQFL12, to mediate TMPRSS2 expression. Notably, CD, AD, TQ, and TQFL12 inhibited TMPRSS2 expression in cancer cell lines, including the PC3 prostate cancer cell line, implying a therapeutic role for preventing COVID-19 in cancer patients. Together, these findings are the first to demonstrate that small molecules, such as CD, AD, TQ, and TQFL12, inhibit TMPRSS2 expression, providing novel therapeutic strategies for preventing COVID-19 and cancers.


Introduction
Transmembrane serine protease 2 (TMPRSS2, OMIM: 602060) encodes a 492 amino acid protein with a molecular weight of 53,859 Da, which contains a functional serine protease domain [1]. TMPRSS2 has been reported to be involved in various pathological and physiological processes [2][3][4]. Recurrent fusions at the TMPRSS2 5 UTR (untranslated region) to ETV1 (ETS variant transcription factor 1) or ERG (ETS-related gene) lead to outlier expression and drive progression of prostate cancer [5].
Serine proteases proteolytically cleave and activate viral S proteins (spike glycoproteins), thereby facilitating fusion of the virus with the cell membrane. Two independent mechanisms are involved in human severe acute respiratory syndrome coronavirus (SARS-CoV) infection entering host cells: (1) proteolytic cleavage of angiotensin converting enzyme 2 (ACE2), which promotes viral entry; and (2) cleavage of coronavirus S protein, which activates itself. Proteolytic cleavage and activation of the S protein is required for infection of human coronavirus 229E (HCoV-229E) and human coronavirus EMC (HCoV-EMC), as well as the F0 fusion glycoprotein of Sendai virus (SeV), human metapneumovirus (HMPV), and human parainfluenza viruses (HPIVs) 1, 2, 3, 4a, and 4b.
Herein, we conducted expression, mutation, and prognostic analyses for the TMPRSS2 gene in pan-cancers as well as in COVID-19-infected lung tissues. Cancer cell lines were treated with the small molecules CD, AD, TQ, and TQFL12 to inhibit TMPRSS2 expression in cancer cell lines.

TMPRSS2 Expressions and Prognostics in Pan-Cancers
Previous bioinformatics analysis of TMPRSS2 expression profiles in 32 different tumor tissues has indicated that TMPRSS2 is significantly upregulated in six cancer tissues, including prostate adenocarcinoma (PRAD), and significantly downregulated in six cancer tissues compared to corresponding normal tissues [20]. We further analyzed TMPRSS2 mRNA expression in pan-cancers and found that TMPRSS2 mRNA expression was highest in prostate cancer with 234.4 fragments per kilobase of exon model per million mapped fragments (FPKM) ( Figure 1A). TMPRSS2 protein expression was highest in prostate cancer followed by urothelial cancer, renal cancer, and pancreatic cancer, and TMPRSS2 was expressed at low levels in lung cancer; TMPRSS2 expression was undetected in the remaining cancer tissue types, including breast cancer ( Figure 1B). Moreover, immunohistochemistry (IHC) showed weak to moderate membranous and/or granular cytoplasmic immunoreactivity in lung and breast cancer tissues ( Figure 1C-H). expressed at low levels in lung cancer; TMPRSS2 expression was undetected in the remaining cancer tissue types, including breast cancer ( Figure 1B). Moreover, immunohistochemistry (IHC) showed weak to moderate membranous and/or granular cytoplasmic immunoreactivity in lung and breast cancer tissues ( Figure 1C-H). We next conducted survival analysis in the pan-cancers. A high expression of TMPRSS2 was significantly associated with a short overall survival in breast invasive carcinoma (BRCA), sarcoma (SARC), and uveal melanoma (UVM), as well as with a long overall survival in lung adenocarcinoma (LUAD) (Figure 2A-D). We next conducted survival analysis in the pan-cancers. A high expression of TM-PRSS2 was significantly associated with a short overall survival in breast invasive carcinoma (BRCA), sarcoma (SARC), and uveal melanoma (UVM), as well as with a long overall survival in lung adenocarcinoma (LUAD) (Figure 2A maining cancer tissue types, including breast cancer ( Figure 1B). Moreover, immunohistochemistry (IHC) showed weak to moderate membranous and/or granular cytoplasmic immunoreactivity in lung and breast cancer tissues ( Figure 1C-H). We next conducted survival analysis in the pan-cancers. A high expression of TMPRSS2 was significantly associated with a short overall survival in breast invasive carcinoma (BRCA), sarcoma (SARC), and uveal melanoma (UVM), as well as with a long overall survival in lung adenocarcinoma (LUAD) (Figure 2A

TMPRSS2 Mutations and Prognostics in Pan-Cancers
Mutation analysis of the pan-cancers demonstrated that the mutation frequency was significantly high in PRAD (42.71%) and low in uterine corpus endometrial carcinoma (UCES) (4.73%), and 7 of the 32 types of cancer had no mutations ( Figure 3A). The plots for the TMPRSS2 mutation types are shown in Figure 3B, and an overview of mutations in TMPRSS2 is shown in Figure 3C. Different types of TMPRSS2 mutations were found, and missense mutations were the dominant mutation type.

TMPRSS2 Mutations and Prognostics in Pan-Cancers
Mutation analysis of the pan-cancers demonstrated that the mutation frequency was significantly high in PRAD (42.71%) and low in uterine corpus endometrial carcinoma (UCES) (4.73%), and 7 of the 32 types of cancer had no mutations ( Figure 3A). The plots for the TMPRSS2 mutation types are shown in Figure 3B, and an overview of mutations in TMPRSS2 is shown in Figure 3C. Different types of TMPRSS2 mutations were found, and missense mutations were the dominant mutation type. Further survival analysis was conducted with and without TMPRSS2 mutations in the pan-cancers, which demonstrated that the mutated TMPRSS2 group was significantly associated with long overall ( Figure 4A), progression-free ( Figure 4B), disease-specific ( Figure 4C), and disease-free ( Figure 4D) survival rates compared to the wild-type (WT) TMPRSS2 group (Table 1). Thus, these findings indicate that TMPRSS2 mutations may be a prognostic marker for long survival rates in pan-cancers. Further survival analysis was conducted with and without TMPRSS2 mutations in the pan-cancers, which demonstrated that the mutated TMPRSS2 group was significantly associated with long overall ( Figure 4A), progression-free ( Figure 4B), disease-specific ( Figure 4C), and disease-free ( Figure 4D) survival rates compared to the wild-type (WT) TMPRSS2 group (Table 1). Thus, these findings indicate that TMPRSS2 mutations may be a prognostic marker for long survival rates in pan-cancers.

CD, AD, TQ, and TQFL12 Inhibit TMPRSS2 Expression in Cancer Cell Lines
Molecular docking analysis has predicted that small molecules have functional inhibitory effects on TMPRSS2 [21]. Thus, we investigated whether small molecules (CD, AD, TQ, and TQFL12) regulate TMPRSS2 expression. The results indicate that CD inhibited TMPRSS2 expression in H460 ( Figure 5A Figure 5E) and 22RV1 cells ( Figure 5F) at the protein level in a dose-dependent manner, and TQFL12 inhibited TMPRSS2 expression in 22RV1 ( Figure 5G) and PC3 cells ( Figure 5H) at the protein level in a dose-dependent manner. Further, AD inhibited TMPRSS2 expression in H460 ( Figure 5I) and 22RV1 cells ( Figure 5J) at the protein level in a dose-dependent manner. However, the small molecules did not significantly change TMPRSS2 mRNA levels, except for CD treatment of 22RV1 cells.   Figure 5F) at the protein level in a dose-dependent manner, and TQFL12 inhibited TMPRSS2 expression in 22RV1 ( Figure 5G) and PC3 cells ( Figure 5H) at the protein level in a dose-dependent manner. Further, AD inhibited TMPRSS2 expression in H460 ( Figure 5I) and 22RV1 cells ( Figure 5J) at the protein level in a dose-dependent manner. However, the small molecules did not significantly change TMPRSS2 mRNA levels, except for CD treatment of 22RV1 cells.

Treatment with CD Inhibits the Translation and Promotes the Degradation of TMPRSS2 Protein
We next investigated the protein stability of TMPRSS2 protein using cycloheximide (CHX) treatment in the presence or absence of CD treatment in 22RV1 cancer cells. The results showed that CD treatment increased the protein stability of TMPRSS2 compared to the control with an increase in the half-life from~3 h to >8 h ( Figure 6A,B). To further verify whether CD decreases the protein level of TMPRSS2, we quantitated TMPRSS2 protein levels and found that CD treatment decreased the TMPRSS2 protein levels by more than 30% ( Figure 6C). Overall, these results indicate that CD treatment inhibits the translation and promotes the degradation of TMPRSS2 protein.

Treatment with CD Inhibits the Translation and Promotes the Degradation of TMPRSS2 Protein
We next investigated the protein stability of TMPRSS2 protein using cycloheximide (CHX) treatment in the presence or absence of CD treatment in 22RV1 cancer cells. The results showed that CD treatment increased the protein stability of TMPRSS2 compared to the control with an increase in the half-life from ~3 h to >8 h ( Figure 6A,B). To further verify whether CD decreases the protein level of TMPRSS2, we quantitated TMPRSS2 protein levels and found that CD treatment decreased the TMPRSS2 protein levels by more than 30% ( Figure 6C). Overall, these results indicate that CD treatment inhibits the translation and promotes the degradation of TMPRSS2 protein.

Treatment with CD Inhibits the Translation and Promotes the Degradation of TMPRSS2 Protein
We next investigated the protein stability of TMPRSS2 protein using cycloheximide (CHX) treatment in the presence or absence of CD treatment in 22RV1 cancer cells. The results showed that CD treatment increased the protein stability of TMPRSS2 compared to the control with an increase in the half-life from ~3 h to >8 h ( Figure 6A,B). To further verify whether CD decreases the protein level of TMPRSS2, we quantitated TMPRSS2 protein levels and found that CD treatment decreased the TMPRSS2 protein levels by more than 30% ( Figure 6C). Overall, these results indicate that CD treatment inhibits the translation and promotes the degradation of TMPRSS2 protein. Please note that the amount of TMPRSS2 protein in CD+ treatments remained lower over the entire period than that in CD−. (C) Quantitative results' comparison of TMPRSS2 protein levels without and with CD treatments but without CHX treatments in Figure 6A. Left column shows without CD treatments (CD−), while right column shows with CD treatments (CD+). The final concentration of CHX was 40 µg/mL. h, hour (s) of CHX treatment. In the left panel of Figure 5A, all the lanes were added CD for 1 h prior adding CHX treatments.

TMPRSS2 Expression in COVID-19-Infected Lungs and Control Lungs
Because SARS-CoV infection downregulates TMPRSS2 expression in cultured cells [22], we investigated the changes in TMPRSS2 expression in COVID-19-infected lungs. By analyzing the single-cell RNA-sequencing dataset (GSE171524) of COVID-19-infected lungs and control lungs [23], the expression levels of TMPRSS2 in COVID-19-infected lungs were significantly reduced compared to the control lungs ( Figure 7A). By further analyzing different cell types, we found that the expression levels of TMPRSS2 in SARS-CoV-2-infected lungs were reduced in epithelial cells but significantly increased in myeloid cells compared to control lungs ( Figure 7B). Consistent with these results in Figure 7A, analysis of the GSE152075 dataset indicates that nasopharyngeal swabs of SARS-CoV-2-infected patients show significantly lower TMPRSS2 expression in virus-infected individuals compared to healthy individuals [24]. Moreover, the expression levels of TMPRSS2 were very high in epithelial cells of both SARS-CoV-2-infected lungs and control lungs ( Figure 7B). + CD treatment. Please note that the amount of TMPRSS2 protein in CD + treatments remained lower over the entire period than that in CD −. (C) Quantitative results' comparison of TMPRSS2 protein levels without and with CD treatments but without CHX treatments in Figure 6A. Left column shows without CD treatments (CD-), while right column shows with CD treatments (CD+). The final concentration of CHX was 40 µg/mL. h, hour (s) of CHX treatment. In the left panel of Figure 5A, all the lanes were added CD for 1 h prior adding CHX treatments.

TMPRSS2 Expression in COVID-19-Infected Lungs and Control Lungs
Because SARS-CoV infection downregulates TMPRSS2 expression in cultured cells [22], we investigated the changes in TMPRSS2 expression in COVID-19-infected lungs. By analyzing the single-cell RNA-sequencing dataset (GSE171524) of COVID-19-infected lungs and control lungs [23], the expression levels of TMPRSS2 in COVID-19-infected lungs were significantly reduced compared to the control lungs ( Figure 7A). By further analyzing different cell types, we found that the expression levels of TMPRSS2 in SARS-CoV-2-infected lungs were reduced in epithelial cells but significantly increased in myeloid cells compared to control lungs ( Figure 7B). Consistent with these results in Figure 7A, analysis of the GSE152075 dataset indicates that nasopharyngeal swabs of SARS-CoV-2-infected patients show significantly lower TMPRSS2 expression in virus-infected individuals compared to healthy individuals [24]. Moreover, the expression levels of TMPRSS2 were very high in epithelial cells of both SARS-CoV-2-infected lungs and control lungs ( Figure 7B).

Discussion
Highly expressed viral entry factors may play critical roles in the invasion of SARS-CoV-2 [25][26][27][28]. A TMPRSS2-expressing cell line has been shown to have high susceptibility for SARS-CoV-2 invasion [29,30]. In the present study, we conducted expression, mutation, and prognostic analyses for the TMPRSS2 gene in pan-cancers and in COVID-19infected lung tissues. TMPRSS2 expression was highest in prostate cancer followed by urothelial cancer, renal cancer, and pancreatic cancer. In addition, TMPRSS2 expression was low in lung cancer, and was not detected in other cancer tissues, including breast cancer. IHC revealed weak to moderate membranous and/or granular cytoplasmic immunoreactivity in lung and breast cancer tissues, which was supported by a previous study

Discussion
Highly expressed viral entry factors may play critical roles in the invasion of SARS-CoV-2 [25][26][27][28]. A TMPRSS2-expressing cell line has been shown to have high susceptibility for SARS-CoV-2 invasion [29,30]. In the present study, we conducted expression, mutation, and prognostic analyses for the TMPRSS2 gene in pan-cancers and in COVID-19-infected lung tissues. TMPRSS2 expression was highest in prostate cancer followed by urothelial cancer, renal cancer, and pancreatic cancer. In addition, TMPRSS2 expression was low in lung cancer, and was not detected in other cancer tissues, including breast cancer. IHC revealed weak to moderate membranous and/or granular cytoplasmic immunoreactivity in lung and breast cancer tissues, which was supported by a previous study reporting that prostate cancer patients have a high risk for SARS-CoV-2 infection compared to non-cancer patients [31]. A high expression of TMPRSS2 was significantly associated with a short overall survival in BRCA, SARC, and UVM, while a low expression of TMPRSS2 was significantly associated with a short overall survival in LUAD. These results demonstrate a role of TMPRSS2 in SARS-CoV-2 invasion, cancer susceptibility, and cancer severity in patients with PRCA, BRCA, SARC, and UVM. We further analyzed TMPRSS2 expression using a dataset containing COVID-19-infected lungs and control lungs, and we found that the levels of TMPRSS2 in COVID-19-infected lungs were significantly reduced compared to the control lungs. Considering that a low expression of TMPRSS2 is significantly associated with a short overall survival in LUAD, low TMPRSS2 expression may result in severity and death of LUAD cancer patients infected with COVID-19 or SARS-CoV-2.
Mutation analysis of the pan-cancers revealed that the TMPRSS2 mutation frequency was significantly higher in PRAD and that the mutated TMPRSS2 group was significantly associated with long overall, progression-free, disease-specific, and disease-free survival compared to the WT TMPRSS2 group, demonstrating a loss of function roles for TMPRSS2 mutations as prognostic markers in pan-cancers. However, it remains unknown whether TMPRSS2 mutation affects COVID-19 severity, thereby indicating the need for additional studies to understand the causation roles of TMPRSS2 mutation in COVID-19.
Targeting SARS-CoV-2 entry factors, including TMPRSS2, may be a therapeutic strategy against COVID-19 [32][33][34]. Molecular docking analysis has predicted that small molecules have a functional inhibitory effect on TMPRSS2 [21]. Many dietary flavonoids show promising multitarget activities against SARS-CoV-2 [35]. A small-molecule compound, N-0385, has recently been reported to act as a pan-SARS-CoV-2 prophylactic and therapeutic agent with TMPRSS2 inhibitory activity [36]. Thus, the present study investigated whether small molecules (CD, AD, TQ, and TQFL12) regulate TMPRSS2 expression. We found that CD inhibited TMPRSS2 expression in H460, MCF7, MDA-MB-231, and PC3 cells. Moreover, we found that TQ inhibited TMPRSS2 expression in MCF7 and 22RV1 cells and that TQFL12 inhibited TMPRSS2 expression in PC3 and 22RV1 cells. Further, AD inhibited TMPRSS2 expression in H460 and 22RV1 cells. Notably, AD is a natural nucleotide from an intermediate product of metabolism in the human body, and has been reported to play roles in COVID-19 pathogenesis and therapeutic opportunities [37]. Moreover, CD treatment inhibited the translation and promoted the degradation of TMPRSS2. CD, TQ, and TQFL12 have anti-cancer suppressive roles both in vitro and in vivo [38][39][40][41]. Natural product CD is a derivative (analog) from AD, while TQFL12 is a novel synthetic derivative from TQ [38,42]. Taken together, these results imply that CD, AD, TQ, and TQFL12 may have therapeutic roles in preventing COVID-19 and cancers.

Reagents, Antibodies, and Cell Lines
CD and TQ have been previously described [48], and TQFL12 is a new synthetic TQderivative [38]. AD was purchased from Macklin Inc. (Shangai, China, A6218-25 g). The TMPRSS2 antibody for Western blotting and immunohistochemistry (IHC) was purchased from Sigma-Aldrich (cat #: HPA035787, Burlington, MA, USA). The indicated cancer cell lines and culture conditions have been previously described [47,48].

Immunohistochemistry (IHC)
The IHC protocol has been previously described [47,48]. In brief, formalin-fixed paraffin-embedded tissues from Chinese breast cancer and lung cancer patients were subjected to antibody staining with the TMPRSS2 antibody (1:100 dilution) for IHC. Informed consent forms were obtained for the cancer patient tissues [47].

Western Blotting
The PC3 and 22RV1 prostate cancer cell lines, the H460 lung cancer cell line, and the MCF7 breast cancer cell line were utilized in the present study. Western blotting for TMPRSS2 was performed using cells treated with or without CD, AD (0, 10, 20, and 40 µm), TQ, or TQFL12 (0, 5, 10, and 20 µm) for 24 h. β-actin antibodies were used as the internal controls. Assays for cycloheximide (CHX)-based protein stability were performed as previously described [49]. The 22RV1 cells were treated with and without CD and with indicated CHX, followed by Western blotting. All experiments were repeated three times.

Semi-Quantitative RT-PCR for TMPRSS2
The semi-quantitative RT-PCR assays were conducted using the above-described treated cancer cells. The following primers for RT-PCR were used: RT-TMPRSS2-L, 5caccaccagctattggacct-3 ; and RT-TMPRSS2-R, 5 -acacgccatcacaccagtta-3 . The PCR product size was 390 bp. The ACTB and GDPDH genes were used as the internal controls. All experiments were repeated three times.

Conclusions
The TMPRSS2 gene is highly expressed in cancer tissues, specifically in PRAD tumors, implying susceptibility to SARS-CoV-2 and severity for COVID-19. TMPRSS2 mutations may be a prognostic marker for long survival rates in pan-cancers. This is the first study to demonstrate that the small molecules CD, AD, TQ, and TQFL12 inhibit TMPRSS2 expression, which may have therapeutic roles via targeting TMPRSS2 to prevent SARS-CoV-2 invasion in cancer patients during the COVID-19 pandemic.  Institutional Review Board Statement: The present study was approved by the Ethics Committee of Southwest Medical University. Informed consent was obtained from patients.

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study. Data Availability Statement: Not applicable.