L‐kynurenine and quinolinic acid inhibited markers of cell survival in B16 F10 melanoma cells in vitro

Melanoma is an aggressive malignancy and remains a major cause of skin cancer mortality, highlighting the need for new treatment strategies. Recent findings revealed that L‐kynurenine and quinolinic acid induce cytotoxicity and morphological changes in B16 F10 melanoma cells in vitro. This paper highlights the effects of L‐kynurenine and quinolinic acid at previously determined half‐maximal inhibitory concentrations on cell cycle progression, cell death and extracellular signal‐regulated protein kinase inhibition. Melanoma, B16 F10 and murine macrophages, RAW 264.7 cells were used in this study, as both cell lines express all the enzymes associated with the kynurenine pathway. Post exposure to the compounds at half‐maximal inhibitory concentrations, transmission electron microscopy was used to assess intracellular morphological changes. Flow cytometry was used to analyse cell cycle progression and quantify apoptosis via the dual staining of Annexin V and propidium iodide and cell survival via extracellular signal‐regulated protein kinase. L‐kynurenine and quinolinic acid at half‐maximal inhibitory concentrations induced intracellular morphological changes representative of cell death. Flow cytometry revealed alterations in cell cycle distribution, increased apoptosis and significantly inhibition of cell survival. L‐kynurenine and quinolinic acid are exogenous kynurenine compounds which inhibited cell survival through extracellular signal‐regulated protein kinase inhibition, induced cell cycle alterations and induced apoptosis in B16 F10 melanoma cells.

involves a multistep process whereby tumour cells invade the surrounding tissue, migrate into the blood or lymphatic circulatory system in a process known as intravasation and extravasate at the secondary site, where they adhere and proliferate (van Zijl et al., 2011).As such, the typical hallmarks of cancer include evading growth suppressors, resisting cell death, enabling replicative immortality and sustained proliferative signalling (Hanahan & Weinberg, 2011).The uncontrolled cell proliferation in cancer is promoted by impaired cell cycle regulation and deregulated cell death (Zhivotovsky & Orrenius, 2010).It is well known that deregulated cell death is often promoted by hyperactivated oncogenic signalling pathways in cancer (Sever & Brugge, 2015).One of these pathways includes the mitogen-activated protein kinase (MAP-K) pathway, which comprises four subpathways, including the extracellular signal-regulated kinase 1/2 (ERK1/2) cascade.Furthermore, ERK1/2 phosphorylation is associated with cell differentiation, survival and proliferation (Nkandeu et al., 2022).
Although decades of research contributed to the advancement of knowledge regarding new treatments to combat cancer and alleviate critical side effects, metastatic melanoma has evaded all attempts at treatment (Davis et al., 2019).Recently, it has been reported that kynurenine compounds, namely L-kynurenine (L-kyn) and quinolinic acid (Quin) induced cytotoxicity and subcellular morphological changes in B16 F10 melanoma cells at half-maximal inhibitory concentration (IC 50 ) after 48 h in vitro (Basson et al., 2023b).
The kynurenine pathway allows for NAD + production, which is a cofactor and may support the energy requirements of cancer cells to sustain uncontrolled proliferation (Moffett & Namboodiri, 2003;Navas & Carnero, 2021).Overall, kynurenine metabolites have been reported to exert effects on a multitude of physiological processes while demonstrating implications in carcinogenesis and metastasis.
Notably, L-kyn, the central metabolite in the kynurenine pathway, has previously been identified as a potent cytotoxic compound against B16 F10 melanoma cells, surpassing the effects of Quin, an end metabolite in the same pathway (Basson et al., 2023b).However, activated downstream signalling pathways, such as aryl hydrocarbon receptors (Ahr) activation resulting from L-kyn (an AhR agonist), previously induced contrasting biological effects in different cancers, yielding pro-and anticancerous effects (Bungsu et al., 2021;Nkandeu et al., 2022).Given the contradictory impact of kynurenine metabolites on cancer progression, there is a critical need to elucidate the specific roles of these metabolites in the intricate pathways of melanoma carcinogenesis and metastasis.In light of these considerations, the aim of this study is to investigate the effects of L-kyn and Quin at IC 50 on cell cycle progression, cell death (apoptosis and necrosis) and ERK1/2 inhibition.By investigating the potential nuanced effects of kynurenine metabolites, this research aims to provide insights to guide targeted treatments for melanoma.

| Cell lines
The melanoma (B16 F10) cell line is characterised as a standard cell line for melanoma research and was previously used to determine the IC 50 values, which were used in this study (Basson et al., 2023b).The B16 F10 cell line was purchased from the American Type Culture Collection B16 F10 (ATCC ® CRL-6475™) and used between passages 4-12.The RAW 264.7 cell line is a noncancerous, macrophage-like cell line and was used as the control cell line, as macrophages are known to express all the enzymes associated with the kynurenine pathway (Guillemin et al., 2003).
This cell line was purchased from CELLONEX and used between passages 6-20.

| Sample preparation
The selection of concentrations for this study was informed by a preceding study wherein the IC 50 values of these metabolites were determined through the employment of a crystal violet assay, spanning a concentration range extending from 0 to 5 mM.As such, all compounds were prepared to IC 50 concentrations as previously determined, where L-kyn had an IC 50 of 1.74 mM and Quin acid 8.23 mM at 48 h in B16 F10 melanoma cells (Basson et al., 2023b).
Briefly, L-kyn, Quin and nocodazole (NOC) (a microtubule inhibitor used as a positive control) were dissolved in ≤0.1% volume per volume (v/v) dimethyl sulfoxide and diluted in 0.1 M phosphate-buffered saline (PBS).Positive control cells were exposed to 1.30 mM NOC diluted in ddH 2 O.A vehicle control sample of complete culture medium (CCM) and PBS in a 1:1 ratio was included as PBS serves as the diluent for L- kyn and Quin.Similarly, a CCM:ddH 2 O vehicle control was included, as ddH 2 O acts as the diluent for NOC.This approach aligns with the methodology of a prior study (Basson et al., 2023b).

| General culture maintenance
Cells were cultured in sterile culture flasks in CCM, containing Dulbecco's modified essential medium, 10% foetal calf serum and 1% antibiotics (amphotericin/penicillin/streptomycin) in a humidified atmosphere at 37°C with 5% CO 2 in a Forma Scientific waterjacketed incubator.To detach and plate cells, B16 F10 cells were trypsinised with TrypLE™ Express whereas RAW 264.7 cells were scraped and resuspended in 1 mL of CCM and counted.

| Transmission electron microscopy
A volume of 4.5 mL at a concentration of 20 × 10 4 B16 F10 and RAW 264.7 cells/mL was plated in a T25 cm 2 tissue culture flask (9 × 10 5 cells/ flask).B16 F10 and RAW 264.7 cells were treated at stated IC 50 values (Section 2.2).After 48 h the cells were harvested, and cell suspensions were washed with 0.1 M PBS.All washes included centrifugation at 3000 rpm.The cells were then fixed with 2.5% glutaraldehyde/ formaldehyde solution for 1 h, then washed three times with 0.1 M PBS for 10 min.One millilitre of osmium tetroxide solution (1%) was then added for 1 h to the cells, followed by washing three times with 0.1 M PBS for 10 min each and then cells were consecutively dehydrated with 30%, 50%, 70%, 90% and three times with 100% ethanol (10 min each).
After dehydration, the cell pellet was suspended in propylene oxide for 10 min and then suspended in a 2:1 mixture of propylene oxide and epoxy resin for 1h in an angled rotator.The samples were then suspended in a 1:2 mixture of propylene oxide and epoxy resin overnight in an angled rotator.The pellet was then set in a mould with 100% epoxy resin in a microcentrifuge tube and placed in an oven at 65°C for 36 h.
The moulds were then trimmed and sectioned into ultrathin sections with glass and then with a diamond knife, using a microtome.The samples were then stained in urinyl acetate for 5 min and then washed three times in ddH 2 O.Samples were then stained with lead citrate for 2 min, followed by three times washes in ddH 2 O.A paper filter was then used to blot samples and dried before viewing them with a transmission electron microscope (JOEL JEM-2100F, 200 kV FE (Field Emission) microscope (Peabody).Different sections per sample were imaged.

| Flow cytometry
A volume of 9 mL at a concentration of 70 × 10 4 B16 F10 and RAW 264.7 cells/mL was plated in a T75 cm 2 tissue culture flask (6.3 × 10 6 cells/flask).After allowing for attachment, cells were treated as previously described at stated IC 50 values.Cells were harvested, centrifuged at 3000 rpm for 5 min and washed twice with icecold PBS.Cells were then fixed in ice-cold 70% methanol (v/v) dropwise while gently vortex mixing and thereafter stored at −20°C until further analysis.On the day of analysis, cells were washed once in 0.5% bovine serum albumin (BSA)/PBS buffer (for flow cytometry, all washes included centrifugation at 3000 rpm for 3 min) before adding the stain or antibody.
Additionally, cells were vortexed before analysing them on the Flow Check™ Pro (Beckman Coulter) attached to an air-cooled argon laser.
For all flow cytometry analyses, unstained control cells treated with PBS were used to set positive gates for the cells treated with the kynurenine exogenous compounds.Analysed data was derived from a minimum of 15,000 cells using Kaluza C data analysis software (Version 1.1.00003.20057;Beckman Coulter).Debris and doublets were gated out and, therefore, excluded from further analyses.All flow cytometry data is representative of at least two experimental repeats (n = 2) (mean ± standard error of mean [SEM]) and presented as a composite univariate histogram.

| Statistics
All quantitative data was represented as mean ± SEM.Flow cytometry experiments were conducted twice and tested for significant differences using an analysis of variance, with Tukey post hoc analysis where p ≤ .05 was considered significant.The statistical programme GraphPad Prism was used for statistical analyses.For all statistical analyses, kynurenine exogenous compounds were compared to the control cells treated with 0.1 M PBS, whereas NOC-treated cells were compared to the control cells treated with ddH 2 O.

| Transmission electron microscopy
A summary of the intracellular morphological changes is included in the Supporting Information files.

| Cell cycle progression
Flow cytometry histograms and a table summary for the B16 F10 and RAW 264.7 cell lines are included in the supplementary files.For the B16 F10 cell line, the ddH 2 O-treated control cells did not display any significant effects compared to the PBS-treated control cells.L-kyn exposure induced an accumulation of cells in the sub-G1 phase (39.68 ± 0.63) as well as a decrease in cells occupying the G1 (49.40 ± 0.71), S (6.92 ± 0.02) and G2/M (3.87 ± 0.40) phase (Figure 2a).
The treatment of B16 F10 cells with Quin resulted in an accumulation of cells in the sub-G1 phase (67.91 ± 3.63) as well as a decrease in cells occupying the G1 (18.58 ± 3.14) and G2/M (0.39 ± 0.06) phase (Figure 2a).NOC exposure led to a decrease in cells occupying the G1 phase (40.29 ± 1.51) as well as an increase in cells in the S (15.52 ± 0.08) and G2/M (33.65 ± 2.91) phase (Figure 2a).
In RAW 264.7 cells, the ddH 2 O-treated control cells did not display any significant effects compared to the PBS-treated control cells.L-kyn displayed a significant increase in cells occupying the sub-G1 phase (19.64 ± 2.72) and displayed a decreased cell population in the G1 (56.53 ± 1.60) (Figure 2b).Quin did not cause significant changes in the cell cycle distribution (Figure 2b).NOC exposure induced a decrease in cells occupying the G1 phase (28.24 ± 0.88) and the S phase (3.45 ± 0.25) and a significant increase in the G2/M (63.82 ± 2.36) phase (Figure 2b).

| Cell death
For the B16 F10 cell line, the ddH 2 O-treated control cells did not display any significant effects compared to the PBS-treated control cells.L-kyn exposure induced a decrease in the viable cell population (9.39 ± 1.20) as well as an increase in the necrotic population (10.73 ± 2.85), the early apoptotic population (32.30 ± 5.79) and late apoptotic cell population (47.58 ± 4.14).The treatment of B16 F10 cells with Quin resulted in a decrease in the viable cell population (68.83 ± 0.09) as well as an increase in the early apoptotic cell population (29.48 ± 0.51) (Figure 4a).NOC exposure led to a decrease in viable cells (66.73 ± 0.02) as well as an increase in cells in the early apoptotic stage (Figure 4a). the treatment of RAW 264.7 cells with Quin resulted in a slight increase in the viable cell population (83.07 ± 0.36) and the necrotic cell population (15.88 ± 0.06), as well as a decrease in the early apoptotic cell population (0.43 ± 0.78) (Figure 4b).NOC exposure led to a decrease in viable cells (61.51 ± 0.05) as well as an increase in cells in the late apoptotic stage (7.22 ± 0.46) and early apoptotic stage (15.95 ± 1.01) (Figure 4b).

| DISCUSSION
In a prior investigation, the authors investigated the cytotoxicity and morphological changes induced by L-kyn, Quin and kynurenic acid (KA) on B16 F10, RAW 264.7 and HaCat cells at 24, 48 and 72 h, spanning concentrations from 0 to 5 mM.The current study tested L-kyn and Quin at these previously determined 48-h IC 50 values on B16 F10 and RAW 264.7 cells (Basson et al., 2023b), which both express all the kynurenine enzymes (Guillemin et al., 2003).This current study found that (1) L-kyn and Quin each displayed different effects in terms of intracellular morphological changes, cell cycle progression, cell death and ERK1/2 inhibition and (2) that both L-kyn and Quin displayed different effects in different cell lines (B16 F10 compared to RAW 264.7 cells).These factors may be explained by the fact that different metabolites in the kynurenine pathway act as active substances, where each metabolite exerts different effects on different cell populations despite its common origin (Austin & Rendina, 2015;Moffett & Namboodiri, 2003).Therefore, L-kyn, which is an Ahr agonist (Walczak et al., 2020a), may exert different effects than Quin, which is a Nmethyl-D-aspartate type glutamate receptors agonist (Nkandeu et al., 2022).
First, TEM was used to observe intracellular morphological changes induced by kynurenine compounds on B16 F10 and RAW 264.7 cells, as TEM is the gold standard to confirm apoptosis (Elmore, 2007).Both L-kyn and Quin induced swelling of the Golgi apparatus in B16 F10 cells but not in RAW 264.7 cells.A previous study by Bottone et al. indicated that morphological changes of the Golgi apparatus, including fragmentation, distension and swelling, can occur under apoptotic conditions (Bottone et al., 2013).Furthermore, L-kyn and Quin induced the formation of myelin figures in B16 F10 cells, but not in RAW 264.7 cells.A previous study suggested that these myelin figures may be derived from cell membranes that coil into whorls during reversible cell injury, but become more prominent in irreversibly injured cells (Miller & Zachary, 2017).A previous study investigating the cytotoxic oxysterols (7KC, 7β-OH and 5β, 6β-epoxide), concluded that myelin figure formation occurred independently of caspase activation (Vejux et al., 2007).Wilkonson elucidated that the emergence of whorls may be a consequence of autophagy induced by endoplasmic reticulum stress (Wilkinson, 2019).Previous evidence indicated that Quin induced striatal neuron death, through both apoptotic and autophagic mechanisms, induced by nuclear factor kappa B and p53 (Lugo-Huitrón et al., 2013).Furthermore, this study demonstrated that Quin induced an increased amount of lysosomes and vacuoles in both cell lines, which has previously been linked to cell death (Appelqvist et al., 2013), such as autophagy (Salazar et al., 2011).These considerations suggested that some relationships may exist between kynurenine compounds, cell death and myelin figure formation and that future studies should investigate the effects of kynurenine compounds on different types of cell death and the possible crosstalk between apoptosis and autophagy.
The cell cycle is regarded as an important determinant of cancer progression, as cell cycle checkpoint disruption is implicated in cancer progression, leading to uncontrolled proliferation and resistance to apoptotic cell death (Visconti et al., 2016).A previous study demonstrated the apoptotic effects of L-kyn on natural killer cells, demonstrating its effect in immune escape.Notably, the observed apoptotic effects did not manifest through cell cycle arrest, despite a marginal elevation of cells in the G0/G1 phase (Song et al., 2011).Data from this study indicated that treatment of B16 F10 cells with L-kyn   and Quin resulted in a significant increase in cells occupying the sub-G1 phase of the cell cycle, indicating the initiation of apoptosis as apoptotic cells have lower DNA sustainability (Darzynkiewicz et al., 1992).In addition, a decreased amount of cells resided in the G2/M phase, indicating that cells did not enter the mitotic phase due to irreparable DNA damage (Hayashi & Karlseder, 2013).These changes were accompanied by a significant decrease in the G1 phase, indicating a reduction in cell growth.However, these changes were less significant in the RAW 264.7 cells.This may indicate that one of the mechanisms by which L-kyn and Quin may act to inhibit the proliferation of B16 F10 melanoma cells is through the inhibition of cell cycle progression.
Another important marker of cell proliferation is ERK1/2, which belongs to the MAP-K family (Guo et al., 2020) and is upregulated in up to 80% of melanomas (Nkandeu et al., 2022).Our results indicate that both L-kyn and Quin significantly inhibited ERK1/2 activation in B16 F10 cells and that L-kyn inhibited ERK1/2 activation in B16 F10 melanoma cells to a greater extent than Quin.This is in alignment with a recent study, indicating that L-kyn (an Ahr agonist) was the most potent cytotoxic kynurenine metabolite towards B16 F10 melanoma cells (Basson et al., 2023b).These results are supported by a previous study where KA (another Ahr agonist in the kynurenine pathway) activated GPR35, leading to the inhibition of phosphorylation of various signalling proteins, such as Akt, ERK and p38 mitogen-activated protein kinase (p38).However, the specific study did not investigate the effects of L-kyn and Quin on these signalling pathways.Furthermore, the study suggested that KA may alter proliferation, cell cycle, cell survival and migration through the disruption of the phosphoinositide 3-kinase/ protein kinase B PI3K/Akt pathway as well as differentiation, proliferation or apoptosis through the disruption of the ERK pathway (Walczak et al., 2020b).
In the current study, however, ERK1/2 was not activated in RAW 264.7 cells.Therefore, the authors suggest that the level of expression and activation of these receptors may differ between B16 F10 and RAW 264.7 cell lines, thus giving rise to the differences in the mechanism of the effects observed.In addition, a previous study reported that ERK1/2 activation leads to the phosphorylation and subsequent activation of numerous signalling molecules in the MAP-K pathway, which regulate cell processes, such as cell cycle progression, differentiation and survival (Sugiura et al., 2021).
Therefore, the inhibition of ERK1/2 is an important antitumour strategy for melanoma (Zhuang, 2005).Ongoing activation of the ERK MAPK-K pathway can transform normal cells into tumour cells, where the inhibition of the ERK MAP-K pathway can inhibit tumour growth (Sugiura et al., 2021).In addition to the correlation between ERK1/2 and tumour growth, ERK1/2 is one of the most important role players in tumour metastasis (Zhuang, 2005), which may indicate the potential antimetastatic properties of L-kyn and Quin.Huang et al. found that the inhibition of the ERK/MAP-K pathway in lymphoma cells suppressed cell proliferation and promoted apoptosis (Huang et al., 2019).
In support of the cell cycle, morphological and ERK1/2 activation results, it was found that L-kyn reduced viable cells and induced cell death, namely necrosis, early and late apoptosis in B16 F10 cells, but only early and late apoptosis in RAW 264.7 cells.Finding that L-kyn induces both early and late apoptosis can be explained by the fact that the loss of the membrane integrity is a gradual process.The cell membrane of early apoptotic cells is thus still intact and impermeable to PI.If these apoptotic cells are not successfully removed by macrophage-mediated phagocytosis, cells will progress to the late apoptotic phase, where the cell membrane integrity is compromised and thus permeable to PI (Kari et al., 2022).The controversy regarding AhR activation in tumour apoptosis is reflected in the literature, as one study indicated that L-kyn suppressed apoptosis in P20E breast cancer cells (Bekki et al., 2015), while another found that the exogenous treatment of A375 melanoma cells with L-kyn induced necrosis, but not apoptosis (Walczak et al., 2020a).In the current study, Quin also caused a reduction in viable B16 F10 melanoma cells, while inducing early apoptosis.However, early apoptosis was suppressed in Quin-treated RAW 264.7 cells and instead led to an increase in necrosis.Necrosis is a significant contributor to inflammation due to the release of intracellular substances, which may be detrimental to surrounding cells (Yang et al., 2015).
Therefore, this warrants further research to elaborate on the effects of Quin on other noncancerous cell lines as it may be potentially toxic to surrounding cells.
In conclusion, this study provides a better understanding of the in vitro therapeutic effects of L-kyn and Quin against B16 F10 melanoma cells and confirms that both of these kynurenine metabolites exert a different mechanism of action, which is cell line dependent.Furthermore, kynurenine compounds inhibited B16 F10 cell proliferation and survival through ERK1/2 activation inhibition, cell cycle alterations, cell death and intracellular morphological changes, with less pronounced effects in the noncancerous RAW 264.7 macrophage cell line.However, these findings need to be substantiated with additional cell lines.
2.5.3 | Apoptosis detection with Annexin V Flow cytometry was utilised to quantify apoptosis with the dual staining of Annexin V conjugated to a fluorochrome, fluorescein isothiocyanate (FITC) and PI, using the Invitrogen™ ApoDETECT Annexin V-FITC Kit from Thermo Fisher Scientific.Cells were resuspended in 1 mL of 1× binding buffer.Cells in suspension (190 μL) were transferred to microcentrifuge tubes and 10 μL Annexin V-FITC was added.Samples were incubated for 10 min in the dark at room temperature.After 10 min, samples were vortexed, and the supernatant was discarded.Cells were resuspended in 190 μL of 1× binding buffer, and 10 μL of PI was added.Dot plots for PI (FL3 log) versus dot plots for Annexin V-FITC (FL1 log) were created for B16 F10 and RAW 264.7 cells and can be found in the supplementary files.The data were grouped into four quadrants, each representing a different type of cell death.Cell death was analysed as follows: viable cells in the lower left quadrant, where FITC (−) and PI (−); necrotic cells in the lower right quadrant, where FITC (−) and PI (+); early apoptotic cells in the upper left quadrant where FITC (+) and PI (−) and late apoptotic cells in the upper right quadrant where FITC (+) and PI (+).
figures were present and Golgi complexes swollen.NOC-treated RAW 264.7 cells (Figure1s,t) presented with membrane blebbing, an increased amount of lysosomes and vacuoles, swollen mitochondria and swollen smooth endoplasmic reticulum.

Figure 2a ,
Figure 2a,b demonstrate the quantitative representation of the percentage gated B16 F10 and RAW 264.7 cells in the different phases of the cell cycle.

F
I G U R E 2 Quantitative representation of the percentage gated B16 F10 (a) and RAW 264.7 (b) cells in the different phases of the cell cycle (sub-G1, G1, S and G2/M) after 48 h for control cells treated with phosphate-buffered saline (PBS), control cells treated with ddH 2 O, L-kyntreated cells at 1.74 mM, Quin-treated cells at 8.23 mM and positive control cells treated with nocodazole at 1.30 mM.The bar graphs represent the average of at least two experimental repeats, with the standard error of mean indicated by the error bars.*p ≤ .05;**p ≤ .01;***p ≤ .001indicates a significant difference compared to the control treated with PBS or ddH 2 O.

For
the RAW 264.7 cell line, the ddH 2 O-treated control cells displayed a slightly lower percentage of cells in the early apoptotic phase (74.68 ± 0.18) (p = .02),as well as a slight increase in necrotic cells (p = .05)in comparison to the PBS-treated control cells (77.10 ± 0.43).L- kyn exposure induced a decrease in the viable population (30.51 ± 1.23) as well as an increase in the late apoptotic population (19.04 ± 1.86) and early apoptotic population (38.01 ± 0.28) (Figure 4b).Interestingly, F I G U R E 3 Quantitative representation of extracellular signal-regulated kinase 1/2 (ERK1/2) activation in B16 F10 cells and RAW 264.7 cells after 48 h for control cells treated with phosphate-buffered saline (PBS), control cells treated with ddH 2 O, L-kyn-treated cells at 1.74 mM, Quin acid-treated cells at 8.23 mM, positive control cells treated with nocodazole at 1.30 mM.The bar graphs represent the average of at least two experimental repeats, with the standard error of mean indicated by the error bars.**p ≤ .01;****p ≤ .0001indicates a significant difference compared to the control cells treated with PBS or ddH 2 O. F I G U R E 4 Quantitative representation of cell death in B16 F10 (a) and RAW 264.7 (b) cells after 48 h for control cells treated with phosphate-buffered saline (PBS), control cells treated with ddH 2 O, L-kyn-treated cells at 1.74 mM, Quin-treated cells at 8.23 mM and positive control cells treated with nocodazole at 1.30 mM.The bar graphs represent the average of at least two experimental repeats, with the standard error of mean indicated by the error bars.*p ≤ .05;**p ≤ .01;****p ≤ .0001indicates a significant difference compared to the control treated with PBS or ddH 2