Biological Evaluation of Alkyl Triphenylphosphonium Ostruthin Derivatives as Potential Anti-Inflammatory Agents Targeting the Nuclear Factor B Signaling Pathway in Human Lung Adenocarcinoma A549 Cells

Ostruthin (6-geranyl-7-hydroxycoumarin) is one of the constituents isolated from Paramignya trimera and has been classified as a simple coumarin. We recently reported the synthesis of alkyl triphenylphosphonium (TPP) derivatives from ostruthin and evaluated their anticancer activities. In the present study, we demonstrated that alkyl TPP ostruthin derivatives inhibited the up-regulation of cell-surface intercellular adhesion molecule-1 (ICAM-1) in human lung adenocarcinoma A549 cells stimulated with tumor necrosis factor-α (TNF-α) without affecting cell viability, while ostruthin itself exerted cytotoxicity against A549 cells. The heptyl TPP ostruthin derivative (termed OS8) attenuated the up-regulation of ICAM-1 mRNA expression at concentrations higher than 40 μM in TNF-α-stimulated A549 cells. OS8 inhibited TNF-α-induced nuclear factor κB (NF-κB)-responsive luciferase reporter activity at concentrations higher than 40 μM, but did not affect the translocation of the NF-κB subunit RelA in response to the TNF-α stimulation at concentrations up to 100 μM. A chromatin immunoprecipitation assay showed that OS8 at 100 μM prevented the binding of RelA to the ICAM-1 promoter. We also showed that OS8 at 100 μM inhibited the TNF-α-induced phosphorylation of RelA at Ser 536. Moreover, the TNF-α-induced phosphorylation of an inhibitor of NF-κB α and extracellular signal-regulated kinase was reduced by OS8. These results indicate that OS8 has potential as an anti-inflammatory agent that targets the NF-κB signaling pathway.


Introduction
Cell adhesion molecules, such as intercellular adhesion molecule-1 (ICAM-1; also known as CD54), play an essential role in the recruitment of leukocytes in blood vessels and their migration to local inflammation sites [1,2]. The up-regulation of ICAM-1 on endothelial cells is stimulated by proinflammatory cytokines, such as tumor necrosis BioChem 2021, 1 108 factor-α (TNF-α) and interleukin-1 [3]. These proinflammatory cytokines primarily induce the nuclear factor κB (NF-κB) signaling pathway [4,5]. In response to TNF-α, the NF-κB subunit RelA (also known as p65) is translocated from the cytoplasm to the nucleus and then binds to the κB sites of target genes [6,7]. The NF-κB signaling pathway contributes to immune responses, inflammation, and survival by promoting the expression of various intrinsic genes [8].
We recently reported the synthesis and biological evaluation o phenylphosphonium (TPP) ostruthin derivatives to enhance the pharmacoki ties of ostruthin [17]. The modification of small molecules with TPP catio shown to promote mitochondrial delivery [18]. Ostruthin exhibited cytoto against human pancreatic cancer PANC-1 cells, human cervical cancer He human hepatic cancer HepG2 cells [17]. Ostruthin TPP derivatives with diff of alkyl linkers were found to exert stronger or weaker cytotoxic activities ag 1, HeLa, and HepG2 cells [17]. Collectively, these findings clearly showed t lent linking of ostruthin with a TPP moiety and alkyl linkers influenced its tivity.
Regarding anti-inflammatory activity, ostruthin has been shown to sup opolysaccharide-induced expression of inducible NO synthase and cycloox mouse microglia BV-2 cells [15]. Ostruthin was recently identified in an ext danum ostruthium (L.) Koch as an active constituent that inhibits TNF-α-ind reporter activity in human embryonic kidney 293 cells [16]. In the present stu uated the anti-inflammatory activities of ostruthin and its TPP derivatives w lengths of alkyl linkers in human lung adenocarcinoma A549 cells (Figure 1 obtained showed that the alkyl TPP derivatives of ostruthin exhibited anti-i activity without affecting cell viability, while ostruthin itself showed cytotox  We recently reported the synthesis and biological evaluation of alkyl triphenylphosphonium (TPP) ostruthin derivatives to enhance the pharmacokinetic properties of ostruthin [17]. The modification of small molecules with TPP cations has been shown to promote mitochondrial delivery [18]. Ostruthin exhibited cytotoxic activities against human pancreatic cancer PANC-1 cells, human cervical cancer HeLa cells, and human hepatic cancer HepG2 cells [17]. Ostruthin TPP derivatives with different lengths of alkyl linkers were found to exert stronger or weaker cytotoxic activities against PANC-1, HeLa, and HepG2 cells [17]. Collectively, these findings clearly showed that the covalent linking of ostruthin with a TPP moiety and alkyl linkers influenced its cytotoxic activity.
Regarding anti-inflammatory activity, ostruthin has been shown to suppress the lipopolysaccharide-induced expression of inducible NO synthase and cyclooxygenase-2 in mouse microglia BV-2 cells [15]. Ostruthin was recently identified in an extract of Peucedanum ostruthium (L.) Koch as an active constituent that inhibits TNF-α-induced NF-κB reporter activity in human embryonic kidney 293 cells [16]. In the present study, we evaluated the anti-inflammatory activities of ostruthin and its TPP derivatives with different lengths of alkyl linkers in human lung adenocarcinoma A549 cells (Figure 1). The results obtained showed that the alkyl TPP derivatives of ostruthin exhibited antiinflammatory activity without affecting cell viability, while ostruthin itself showed cytotoxic activity.

Cells
Human lung adenocarcinoma A549 cells (JCRB0076) and human fibrosarcoma HT-1080 cells (JCRB9113) were originally obtained from the National Institutes of Biomedical Innovation, Health and Nutrition JCRB Cell Bank (Osaka, Japan). A549 cells and HT-1080 cells were maintained in RPMI 1640 medium supplemented with heat-inactivated fetal calf serum (Sigma-Aldrich, St. Louis, MO, USA) and penicillin-streptomycin mixed solution (Nacalai Tesque, Kyoto, Japan), as described in our previous studies [19,20].

Reporter Assay
The ICAM-1 promoter (−1604 to +40) was inserted into the KpnI and HindIII sites of the pGL4.22 [luc2CP/Puro] vector (Promega, Madison, WI, USA). The internal HindIII site of the ICAM-1 promoter was mutated by the replacement of −273 T to A. The NF-κBresponsive firefly luciferase reporter plasmid was previously described [22]. A549 cells were transiently transfected with firefly luciferase reporter plasmids, together with the cytomegalovirus promoter-driven Renilla luciferase reporter plasmid by using HilyMax transfection reagent (Dojindo, Kumamoto, Japan), and used in the reporter assay as previously described [23]. Relative light units were measured with Lumitester C-110 (Kikkoman Biochemifa, Tokyo, Japan).

Flow Cytometry
The amount of cell-surface ICAM-1 was assessed by flow cytometry as previously described [24]. A mouse anti-human ICAM-1 antibody (clone 15.2) and mouse IgG1 isotype control antibody (MOPC-21; BioLegend, San Diego, CA, USA) together with a phycoerythrin-labeled anti-mouse IgG antibody (Jackson ImmunoResearch) were used. Cells were stained serially with primary and secondary antibodies, and washed twice with PBS containing 2% BSA and 2 mM EDTA. Stained cells were measured by FACSCalibur (BD Biosciences, San Jose, CA, USA), followed by data analysis using FlowJo software (Tomy Digital Biology, Tokyo, Japan).

Western Blotting
Nuclear and cytoplasmic fractions were prepared as described in our previous study [28]. Cells were washed with PBS and then lyzed with Triton X-100 lysis buffer (1% Triton X-100, 50 mM Tris-HCl (pH 7.4) and 2 mM dithiothreitol) containing cOmplete TM (Sigma-Aldrich) as protease inhibitor cocktail. Cell lysates were centrifuged (15,300× g, 5 min) to separate pellets and supernatants as cytoplasmic fractions. Pellets were washed with 1% Triton X-100 lysis buffer, then solubilized with sonication and collected as nuclear factions. Proteins were separated by SDS-PAGE and transferred to ClearTrans ® nitrocellulose membrane , and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (6C5; Santa Cruz Biotechnology), and peroxidase-conjugated goat anti-mouse IgG(H + L) antibody and peroxidase-conjugated goat anti-rabbit IgG(H + L) antibody (Jackson ImmunoResearch) as secondary antibodies were used for Western blotting. The blots for target proteins were reprobed with the γ1-actin protein, lamin A/C protein, and GAPDH protein. Protein bands were analyzed by ImageQuant LAS 4000 mini (GE Healthcare Japan, Tokyo, Japan). The expression of γ1-actin was similar to that of β-actin under our experimental conditions.

ChIP Assay
A chromatin immunoprecipitation (ChIP) assay was performed as described in our previous studies [25,29]. Cells were fixed with 1% formaldehyde and washed with PBS. The fixed cells were suspended serially in three lysis buffers, followed by sonication to shear chromatin DNA. The anti-RelA antibody (C-20) and Dynabeads TM protein A (Thermo Fisher Scientific) were used for immunoprecipitation. Beads were washed with different buffers, including elution buffer. Immunoprecipitated and input DNA were treated with ribonuclease A and then proteinase K, followed by purification with the QIAquick PCR purification kit (QIAGEN, Hilden, Germany). Operated primer pairs for the amplification of the ICAM-1 promoter regions (−286 to −90 and −625 to −446) [30] were used for real-time PCR. Primer-specific standard curves were used for the quantitation of initial DNA levels.

Statistical Analysis
Data were evaluated by a one-way ANOVA and Tukey's test for multiple comparisons.

Biological Evaluation of Ostruthin and Its Alkyl TPP Derivatives on the Viability of A549 Cells
Ostruthin was purified from the CHCl 3 extract of P. trimera [11]. Four TPP derivatives were synthesized from ostruthin as a scaffold ( Figure 1) [17]. As the first step, MTT assay was used to examine the effects of the test compounds on the viability of A549 cells. The MTT assay revealed that ostruthin inhibited MTT-reducing activity at IC 50 values of 83.2 ± 2.5 and 82.8 ± 0.9 µM (n = 3, calculated from three independent experiments) in the absence and presence of TNF-α, respectively ( Figure 2). In contrast, the four alkyl TPP derivatives of ostruthin at concentrations up to 100 µM did not significantly decrease MTT-reducing activity ( Figure 2).

Biological Evaluation of Ostruthin and Its Alkyl TPP Derivatives on the Viability of A549 Cells
Ostruthin was purified from the CHCl3 extract of P. trimera [11]. Four TPP derivatives were synthesized from ostruthin as a scaffold ( Figure 1) [17]. As the first step, MTT assay was used to examine the effects of the test compounds on the viability of A549 cells. The MTT assay revealed that ostruthin inhibited MTT-reducing activity at IC50 values of 83.2 ± 2.5 and 82.8 ± 0.9 µM (n = 3, calculated from three independent experiments) in the absence and presence of TNF-α, respectively ( Figure 2). In contrast, the four alkyl TPP derivatives of ostruthin at concentrations up to 100 μM did not significantly decrease MTTreducing activity ( Figure 2).
Crystal violet staining was also used to assess cell viability because it mainly stains adherent cells. Ostruthin decreased the number of crystal violet-stained cells at IC50 values of 63.4 ± 3.1 and 68.6 ± 4.4 μM (n = 3) in the absence and presence of TNF-α, respectively ( Figure S1). The number of crystal violet-stained cells was not affected by OS5 or OS8, but was slightly reduced by OS6 and OS7 ( Figure S1). These results confirmed that OS5, OS6, OS7, and OS8 did not markedly affect the viability of A549 cells at concentrations up to 100 µM.  Crystal violet staining was also used to assess cell viability because it mainly stains adherent cells. Ostruthin decreased the number of crystal violet-stained cells at IC 50 values of 63.4 ± 3.1 and 68.6 ± 4.4 µM (n = 3) in the absence and presence of TNF-α, respectively ( Figure S1). The number of crystal violet-stained cells was not affected by OS5 or OS8, but was slightly reduced by OS6 and OS7 ( Figure S1). These results confirmed that OS5, OS6, OS7, and OS8 did not markedly affect the viability of A549 cells at concentrations up to 100 µM.

Biological Evaluation of Ostruthin and Its Alkyl TPP Derivatives on the TNF-α-Induced Expression of Cell-Surface ICAM-1 in A549 Cells
In A549 cells, the TNF-α stimulation up-regulates the expression of cell-surface ICAM-1 within 6 h at sufficient levels, and this may be quantitatively measured using a cell-ELISA [19]. The treatment with TNF-α significantly increased the cell-surface expression of ICAM-1 ( Figure 3). The expression level of cell-surface ICAM-1 was reduced by all of the test compounds in a dose-dependent manner ( Figure 3). Among the test compounds examined, ostruthin exerted the strongest inhibitory effects and attenuated the up-regulation of cell-surface ICAM-1 induced by TNF-α at an IC 50 value of 55.3 ± 2.5 µM (n = 3) and completely at concentrations higher than 70 µM ( Figure 3). OS5, OS6, OS7, and OS8 interfered with cell-surface ICAM-1 expression at IC 50 values of >100, 88.6 ± 5.9, 65.1 ± 14.9, and 68.6 ± 8.3 µM (n = 3), respectively ( Figure 3). These results suggested that the strong inhibition of cell-surface ICAM-1 expression by ostruthin was at least partly due to reductions in cell viability and attached cells. In contrast, the alkyl TPP derivatives of ostruthin exerted inhibitory effects against the TNF-α-dependent up-regulation of cell-surface ICAM-1 without significantly affecting cell viability. The flow cytometric assay showed that OS8 inhibited the TNF-α-induced expression of cell-surface ICAM-1 in a dose-dependent manner with an IC 50 value of 62.0 ± 3.4 µM (n = 3) ( Figure S2), which confirmed cell-ELISA-based results (Figure 3).

Biological Evaluation of Ostruthin and Its Alkyl TPP Derivatives on the TNF-α-Induced Expression of Cell-Surface ICAM-1 in A549 Cells
In A549 cells, the TNF-α stimulation up-regulates the expression of cell-surface ICAM-1 within 6 h at sufficient levels, and this may be quantitatively measured using a cell-ELISA [19]. The treatment with TNF-α significantly increased the cell-surface expression of ICAM-1 ( Figure 3). The expression level of cell-surface ICAM-1 was reduced by all of the test compounds in a dose-dependent manner ( Figure 3). Among the test compounds examined, ostruthin exerted the strongest inhibitory effects and attenuated the up-regulation of cell-surface ICAM-1 induced by TNF-α at an IC50 value of 55.3 ± 2.5 µM (n = 3) and completely at concentrations higher than 70 μM ( Figure 3). OS5, OS6, OS7, and OS8 interfered with cell-surface ICAM-1 expression at IC50 values of >100, 88.6 ± 5.9, 65.1 ± 14.9, and 68.6 ± 8.3 µM (n = 3), respectively ( Figure 3). These results suggested that the strong inhibition of cell-surface ICAM-1 expression by ostruthin was at least partly due to reductions in cell viability and attached cells. In contrast, the alkyl TPP derivatives of ostruthin exerted inhibitory effects against the TNF-α-dependent up-regulation of cell-surface ICAM-1 without significantly affecting cell viability. The flow cytometric assay showed that OS8 inhibited the TNF-α-induced expression of cell-surface ICAM-1 in a dose-dependent manner with an IC50 value of 62.0 ± 3.4 µM (n = 3) ( Figure S2), which confirmed cell-ELISA-based results ( Figure 3).

Biological Evaluation of Ostruthin and Its Alkyl TPP Derivatives on TNF-α-Induced ICAM-1 Promoter-Driven Luciferase Activity in A549 Cells
We investigated the effects of ostruthin alkyl TPP derivatives on ICAM-1 promoterdriven luciferase reporter activity induced by TNF-α. A 2.5-h incubation with TNF-α increased luciferase activity by approximately 3-fold from that in unstimulated cells ( Figure 4A-E). All test compounds reduced TNF-α-induced luciferase reporter activity ( Figure 4A-E). Among the alkyl TPP derivatives of ostruthin, OS8 significantly inhibited ICAM-1 promoter-luciferase reporter activity at concentrations higher than 40 µM ( Figure 4E). Therefore, OS8 was considered to be the most potent anti-inflammatory agent without any influence on cell viability. any influence on cell viability.
We further examined the effects of OS8 on TNF-α-induced ICAM-1 transcription using quantitative PCR. In response to TNF-α, ICAM-1 mRNA levels were approximately 400-fold higher than those in untreated cells ( Figure 4F). OS8 inhibited TNF-α-induced ICAM-1 mRNA expression in a dose-dependent manner and strongly at concentrations of 70 and 100 µM ( Figure 4F). These results demonstrated that OS8 inhibited TNF-α-induced ICAM-1 transcription.

OS8 Prevented TNF-α-Induced ICAM-1 Expression and NF-κB-Responsive Luciferase Activity in HT-1080 Cells
Human fibrosarcoma HT-1080 cells were employed to generalize the biological activity of OS8. The MTT assay showed that OS8 slightly decreased the viability of HT-1080 We further examined the effects of OS8 on TNF-α-induced ICAM-1 transcription using quantitative PCR. In response to TNF-α, ICAM-1 mRNA levels were approximately 400-fold higher than those in untreated cells ( Figure 4F). OS8 inhibited TNF-α-induced ICAM-1 mRNA expression in a dose-dependent manner and strongly at concentrations of 70 and 100 µM ( Figure 4F). These results demonstrated that OS8 inhibited TNF-α-induced ICAM-1 transcription.
The translocation of the NF-κB subunit RelA from the cytoplasm to the nucleus was observed in A549 cells within 30 min of the TNF-α stimulation [28,31]. The TNF-α stimulation induced an approximately 2-fold increase in nuclear RelA levels ( Figure 6B-D). OS8 did not decrease the amount of nuclear RelA at concentrations up to 100 µM ( Figure 6B,C). These results revealed that OS8 did not affect TNF-α-induced nuclear RelA translocation.
These results revealed that OS8 did not affect TNF-α-induced nuclear RelA translocation.
Based on the above results, we speculated that OS8 may prevent the binding of RelA to the ICAM-1 promoter. Multiple NF-κB-responsive sites were previously reported to be located in the ICAM-1 promoter [30,32]. A ChIP assay was performed to investigate the direct binding of RelA to these NF-κB-responsive sites. The binding of RelA to the ICAM-1 promoter was markedly increased by the TNF-α stimulation ( Figure 6E,F). OS8 at 100 µM significantly suppressed the increase in RelA binding ( Figure 6E,F). These results indicated that OS8 prevented the binding of RelA to the ICAM-1 promoter, which accounted for the reduction in ICAM-1 mRNA expression. Figure 6. OS8 did not affect the nuclear translocation of RelA, but inhibited NF-κB-responsive luciferase activity and RelA binding to the ICAM-1 promoter in TNF-α-stimulated A549 cells. (A) A549 cells were transfected with NF-κB-responsive and control cytomegalovirus promoter-driven luciferase reporter plasmids. Transfected A549 cells were treated as described in Figure 5D. Luciferase activity (fold) represents the mean ± S.E. of three independent experiments. *** p < 0.001. (B-D) A549 cells were treated as described in Figure 4F, except for an incubation without (−) or with (+) TNF-α for 30 min. Nuclear and cytoplasmic lysates were analyzed by Western blotting. Representative blots from three independent experiments are shown (B). The protein level of RelA in the nucleus (C) and cytoplasm (D) was standardized by that of lamin A/C and GAPDH, respectively. The RelA protein (%) represents the mean ± S.E. of three independent experiments. * p < 0.05. (E,F) A549 cells were treated as described in Figure 6B-D. A ChIP assay was performed with an anti-RelA antibody. IP DNA (% input) for ICAM-1 promoter regions at −625 to −446 (E) and −286 to −90 (F) represents the mean ± S.E. of three independent experiments. * p < 0.05 and ** p < 0.01. Figure 6. OS8 did not affect the nuclear translocation of RelA, but inhibited NF-κB-responsive luciferase activity and RelA binding to the ICAM-1 promoter in TNF-α-stimulated A549 cells. (A) A549 cells were transfected with NF-κB-responsive and control cytomegalovirus promoter-driven luciferase reporter plasmids. Transfected A549 cells were treated as described in Figure 5D. Luciferase activity (fold) represents the mean ± S.E. of three independent experiments. *** p < 0.001. (B-D) A549 cells were treated as described in Figure 4F, except for an incubation without (−) or with (+) TNF-α for 30 min. Nuclear and cytoplasmic lysates were analyzed by Western blotting. Representative blots from three independent experiments are shown (B). The protein level of RelA in the nucleus (C) and cytoplasm (D) was standardized by that of lamin A/C and GAPDH, respectively. The RelA protein (%) represents the mean ± S.E. of three independent experiments. * p < 0.05. (E,F) A549 cells were treated as described in Figure 6B-D. A ChIP assay was performed with an anti-RelA antibody. IP DNA (% input) for ICAM-1 promoter regions at −625 to −446 (E) and −286 to −90 (F) represents the mean ± S.E. of three independent experiments. * p < 0.05 and ** p < 0.01. Based on the above results, we speculated that OS8 may prevent the binding of RelA to the ICAM-1 promoter. Multiple NF-κB-responsive sites were previously reported to be located in the ICAM-1 promoter [30,32]. A ChIP assay was performed to investigate the direct binding of RelA to these NF-κB-responsive sites. The binding of RelA to the ICAM-1 promoter was markedly increased by the TNF-α stimulation ( Figure 6E,F). OS8 at 100 µM significantly suppressed the increase in RelA binding ( Figure 6E,F). These results indicated that OS8 prevented the binding of RelA to the ICAM-1 promoter, which accounted for the reduction in ICAM-1 mRNA expression.

OS8 Inhibited the Phosphorylation of IκBα in TNF-α-Stimulated A549 Cells
To elucidate the mechanisms by which OS8 inhibits the binding of RelA to the ICAM-1 promoter, we examined the phosphorylation of components of the NF-κB signaling pathway. In A549 cells, the results of the time-course experiment showed that the TNF-α stimulation induced the phosphorylation of IκBα within 5 min, which was followed by the degradation of IκBα within 15 min ( Figure 7A-C). OS8 at 100 µM markedly diminished TNF-α-induced IκBα phosphorylation ( Figure 7A,B). In contrast, TNF-α-induced IκBα degradation was delayed, but proceeded for 30 min in the presence of OS8 ( Figure 7A,C). This result is consistent with OS8 not markedly affecting the nuclear translocation of RelA 30 min after a stimulation with TNF-α ( Figure 6B,C). The results of the dose-dependent experiment showed that OS8 at 100 µM suppressed the degradation of IκBα to some extent 15 min after the TNF-α stimulation ( Figure 7D,E).

OS8 Inhibited the Phosphorylation of IκBα in TNF-α-Stimulated A549 Cells
To elucidate the mechanisms by which OS8 inhibits the binding of RelA to the ICAM-1 promoter, we examined the phosphorylation of components of the NF-κB signaling pathway. In A549 cells, the results of the time-course experiment showed that the TNF-α stimulation induced the phosphorylation of IκBα within 5 min, which was followed by the degradation of IκBα within 15 min ( Figure 7A-C). OS8 at 100 µM markedly diminished TNF-α-induced IκBα phosphorylation ( Figure 7A,B). In contrast, TNF-α-induced IκBα degradation was delayed, but proceeded for 30 min in the presence of OS8 ( Figure  7A,C). This result is consistent with OS8 not markedly affecting the nuclear translocation of RelA 30 min after a stimulation with TNF-α ( Figure 6B,C). The results of the dose-dependent experiment showed that OS8 at 100 µM suppressed the degradation of IκBα to some extent 15 min after the TNF-α stimulation ( Figure 7D,E). cells were preincubated with OS8 for 1 h and then incubated without (−) or with (+) TNF-α (2.5 ng/mL) for the indicated times in the presence or absence of OS8 (100 µM). Cytoplasmic lysates were analyzed by Western blotting. Representative blots from three independent experiments are shown (A). The protein levels of phospho-IκBα and total IκBα were standardized by that of γ1-actin (B,C). * p < 0.05, ** p < 0.01, and *** p < 0.001. (D,E) A549 cells were treated as described in Figure 4F, except for an incubation without (−) or with (+) TNF-α for 15 min. Cytoplasmic lysates were analyzed by Western blotting. Representative blots from three independent experiments are shown (D). The protein level of IκBα was standardized by that of β-actin. ** p < 0.01 and *** p < 0.001. . The protein levels of phospho-IκBα and total IκBα were standardized by that of γ1-actin (B,C). * p < 0.05, ** p < 0.01, and *** p < 0.001. (D,E) A549 cells were treated as described in Figure 4F, except for an incubation without (−) or with (+) TNF-α for 15 min. Cytoplasmic lysates were analyzed by Western blotting. Representative blots from three independent experiments are shown (D). The protein level of IκBα was standardized by that of β-actin. ** p < 0.01 and *** p < 0.001.

Discussion
Ostruthin inhibited TNF-α-induced ICAM-1 expression as anti-inflammatory activity in A549 cells, but also exhibited cytotoxic activity in the same cell line. However, in the present study, the synthetic alkyl TPP derivatives of ostruthin inhibited the TNF-α-induced expression of ICAM-1 without affecting the viability of A549 cells. We selected OS8 as the most potent derivative and showed that it inhibited TNF-α-induced ICAM-1 mRNA expression and NF-κB-dependent reporter activity. In TNF-α-stimulated A549 cells, OS8 did not affect RelA nuclear translocation, but prevented the subsequent binding of RelA to the ICAM-1 promoter. The present results also demonstrated that OS8 inhibited the TNF-α-induced phosphorylation of IκBα, RelA, and ERK. Based on similarities in the chemical structures of the derivatives, other compounds (OS5, OS6, and OS7) may inhibit TNF-α-induced ICAM-1 expression by a similar mechanism to that of OS8.
We showed that ostruthin markedly reduced TNF-α-induced ICAM-1 expression in A549 cells. A previous study reported that ostruthin prevented the lipopolysaccharideinduced expression of inducible NO synthase and cyclooxygenase-2 in mouse BV-2 microglial cell line [15]. Ostruthin was recently shown to inhibit TNF-α-induced NF-κB reporter activity in human embryonic kidney 293 cells [16]. Consistent with this finding, daphnetin (7,8-dihydroxycoumarin), a coumarin derivative closely related to ostruthin, was reported to block the translocation of NF-κB subunits in concanavalin A-stimulated mouse T cells [36]. It also decreased the level of phosphorylated RelA in high glucosestimulated human glomerular mesangial cells [37]. These findings suggest the potential of ostruthin as an agent that targets the NF-κB signaling pathway. However, ostruthin exhibited cytotoxicity against A549 cells at concentrations that decreased ICAM-1 expression. Among the OS derivatives, OS8 inhibited TNF-α-induced ICAM-1 expression at an IC 50 value of 68.6 ± 8.3 µM, which was a slightly higher concentration than that of ostruthin (55.3 ± 2.5 µM). In contrast, OS8 did not affect the viability of A549 cells at concentrations up to 100 µM. Therefore, the alkyl TPP modification of ostruthin reduced its cytotoxic activity and improved its potential for anti-inflammatory activity.
We previously demonstrated that ostruthin exhibited cytotoxic activity against three human cancer cell lines, i.e., pancreatic PANC-1 cells, cervical HeLa cells, and hepatic HepG2 cells [17]. We also showed that ostruthin reduced the viability of human lung adenocarcinoma A549 cells. Consistent with these findings, ostruthin exhibited cytotoxic activities against human pancreatic cancer cell lines (BcPc3, CFPAC1, PANC-1, and PSN-1) and rat neural PC12 cells [14,38,39]. Ostruthin also reduced the proliferation of normal cells, i.e., rat vascular smooth muscle cells and human pancreatic ductal epithelial cells [40][41][42]. Although sensitivity to ostruthin may vary in different types of cells and/or be influenced by experimental conditions (e.g., incubation time), these findings collectively indicated that ostruthin exhibits cytotoxic activities against cancerous and normal cells. OS7 and OS8 exhibited stronger cytotoxicity against PANC-1 cells and HeLa cells than ostruthin, but weaker cytotoxic activity against HepG2 cells [17]. In the present study, all TPP derivatives (OS5, OS6, OS7, and OS8) exhibited weaker cytotoxic activities against A549 cells. These results suggest that ostruthin and ostruthin derivatives target multiple proteins that are essential for cell survival and proliferation, thereby interfering with different target proteins in a cell context-dependent manner. Moreover, the selectivity of ostruthin and its derivatives among target proteins in each cell line is altered by the modification of a TPP moiety and/or alkyl linkers because TPP modifications are utilized conceptionally for mitochondrial delivery [18].
In addition to cell-based biological activities, previous studies reported that ostruthin directly inhibited the enzymatic activities of ethoxyresorufin O-dealkylase and pentoxyresorufin O-dealkylase (cytochrome P450 family members) [43], acetylcholine esterase [44], and α-glucosidase [11,45]. Furthermore, ostruthin has been identified as an activator of TWIK-related K + channels [46]. These biological activities of ostruthin do not appear to be directly linked to the inhibition of the NF-κB signaling pathway or cytotoxic activity. Nevertheless, these findings suggest that ostruthin and its derivatives have the ability to target multiple proteins in the cell.
Regarding its mechanism of action, we revealed that OS8 did not affect the nuclear translocation of RelA, but inhibited the binding of RelA to the ICAM-1 promoter in A549 cells. Moreover, we showed that OS8 inhibited the TNF-α-induced phosphorylation of IκBα, RelA, and ERK. OS8 delayed TNF-α-induced IκBα degradation, but did not markedly affect the nuclear translocation of RelA at later time points. Previous studies demonstrated that the phosphorylation of Ser 536 is mediated by multiple protein kinases, such as IκB kinase β, and is involved in transcriptional activation, nuclear localization, and protein stability [33][34][35]. Therefore, the inhibition of RelA phosphorylation by OS8 appears to be responsible for the inhibition of TNF-α-induced NF-κB activation, including RelA binding to the ICAM-1 promoter. Moreover, the activation of ERK was required for the TNF-α-induced RelA phosphorylation at Ser 536 in mouse epidermal JB cells [47]. It is also possible that the inhibition of TNF-α-induced ERK activation by OS8 is involved in the inhibition of RelA phosphorylation. Further experiments are needed to elucidate the molecular mechanisms by which OS8 inhibits the TNF-α-induced phosphorylation of IκBα, RelA, and ERK.

Conclusions
Ostruthin exhibited anti-inflammatory and cytotoxic activities in A549 cells. The present results demonstrated that the alkyl TPP modification reduced the cytotoxic activity of ostruthin while maintaining its anti-inflammatory activity. OS8 inhibited TNF-α-induced ICAM-1 expression without affecting cell viability. OS8 prevented the NF-κB signaling pathway and inhibited the binding of the NF-κB subunit RelA to the ICAM-1 promoter. OS8 inhibited the TNF-α-induced phosphorylation of IκBα, RelA, and ERK. In conclusion, the present results revealed that OS8 has potential as an anti-inflammatory agent that targets the NF-κB signaling pathway.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/biochem1020010/s1, Figure S1: Ostruthin decreased cell viability, whereas OS5, OS6, OS7, and OS8 did not, Figure S2: OS8 inhibited TNF-α-induced cell surface ICAM-1 expression in a dose-dependent manner. Funding: This work was partly supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant number 19H02885 (to T.K.) and Grants-in-Aid from the JSPS Core-to-Core program, B Asia-Africa Science Platforms.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

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