Hydrogen sulfide acts as a double-edged sword in human hepatocellular carcinoma cells through EGFR/ERK/MMP-2 and PTEN/AKT signaling pathways

Hydrogen sulfide (H2S) is involved in cancer biological processes. However, there are several controversies concerning the role of H2S in cancer development and progression. In this study, we found that the growth and migration of hepatocellular carcinoma (HCC) cells were enhanced by 10–100 μM NaHS and dose-dependently inhibited by 600–1000 μM NaHS. The apoptotic levels were reduced by 25–100 μM NaHS but increased by 400–1000 μM NaHS in HCC cells. After treatment with 25–50 μM NaHS, the protein levels of p-EGFR, p-ERK, MMP-2, and p-AKT were increased, whereas the levels of PTEN and the ratio of BAX/BCL-2 were down-regulated. Administration of 800–1000 μM NaHS showed opposite effects on these protein levels in HCC cells. However, H2S showed no effects on the growth, migration, apoptosis, and the protein levels of the EGFR/ERK/MMP-2 and PTEN/AKT signaling pathways in L02 cells. Furthermore, 25–100 μM NaHS promoted HCC tumor growth and blood vessel formation, while 800–1000 μM NaHS inhibited angiogenesis and tumor growth with no obvious systemic toxicity. These results indicate that H2S acts as a double-edged sword in HCC cells through EGFR/ERK/MMP-2 and PTEN/AKT signaling pathways. Novel H2S donors could be designed and applied for further antitumor research.

TdT-mediated dUTP-biotin nick end labeling (TUNEL) assay. TUNEL   (G-I) The percentages of viable cells were determined using MTT assay and the cell viability of every cell line without NaHS treatment was normalized as 100% and considered to be the control group. Data are presented as mean ± SEM of three independent experiments; *P < 0.05, **P < 0.01 compared with the control group.
Scientific RepoRts | 7: 5134 | DOI: 10.1038/s41598-017-05457-z Measurement of H 2 S levels. The concentrations of H 2 S in both cells and culture supernatant were determined using enzyme-linked immunosorbent assay (ELISA) kits according to the manufacturer's instructions (LanpaiBio, Shanghai, China). Briefly, the cells and culture supernatant were collected to test the levels of H 2 S. Then, the standard controls were diluted, in which the concentration of H 2 S was 15 ng/L, 30 ng/L, 60 ng/L, 120 ng/L, and 180 ng/L, respectively. The samples were diluted and incubated 0.5 h at 37 °C. After H 2 S was bound and the plates were washed, the conjugate reagent was added to the well and incubated 0.5 h at 37 °C. After washing, the colour-developing agents were added to each well and incubated 15 min at 37 °C. The optical density of each well was measured with a microplate reader (Bio-Rad, CA, USA) at 450 nm. A standard curve was generated by plotting the logarithm of the mean absorbance for each standard versus the logarithm of the known The human normal hepatocyte cell line L02 and HCC cell lines SMMC-7721 and Huh-7 were seeded into 6-well plates at 2 × 10 5 cells/ well. After incubation with NaHS at 0 (Control), 10, 25, 50, 100, 200, 400, 600, 800, and 1000 μM for 24 h, the effect of H 2 S on cell migration was measured by wound-healing assay; original magnification 40x. (D-F) The migration rates of L02, SMMC-7721, and Huh-7 cells were calculated by the formula shown above. Data are presented as mean ± SEM of three independent experiments; *P < 0.05, **P < 0.01 compared with the control group.
H 2 S concentration. The value for the blank was subtracted from both the samples and the standard controls. The experiments were repeated three times.
parallel to the skin surface and W is the dimension perpendicular to L and parallel to the surface 19 . The mice were sacrificed at 24 h after the last administration. The tumors were excised and weighted to evaluate the inhibition rate (IR). The IR of tumor growth was calculated as IR (%) = [(A − B)/A] × 100, where A is the average tumor weight of the control group, and B is that of the treatment group 18 . Hematoxylin and eosin (HE) staining. A necropsy examination was performed immediately after sacrifice. Tumor samples and major organs, including heart, liver, spleen, lung, kidney, and brain, were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned at 5 μm thickness, and processed according to the HE staining protocol. The stained tissues were observed using an Olympus BX51 microscope (Olympus, Tokyo, Japan). Immunohistochemistry (IHC) and evaluation. CD34 is considered an ideal biomarker for vascular endothelial cells, and its immunostaining density is represented by the tumor microvessel density (MVD) 20 . To determine the tumor MVD, tumor tissue sections were stained by IHC using CD34 antibody (1:100, Santa Cruz, CA, USA). The sections were scanned under a light microscope with low-power magnification (100x) to identify areas with the angiogenic vessels. Stained vessels with a clearly defined lumen or well-defined linear vessel shape were counted in six high-power (400x) fields from the representative tumor zone, and the mean value was regarded as MVD.
Mice were observed and weighed daily during the experimental period. The blood was collected from the heart before sacrifice at the end of the experiment. Total number of white blood cell (WBC) was measured using an automated analyzer (Beckman Coulter, Inc., Fullerton, CA, USA). The heart, liver, spleen, lung, kidney, and brain were removed and weighed immediately after sacrifice, and the relative organ weights (g/100 g body weight) were calculated.
Statistical analysis. Data are presented as means ± standard error of the mean (SEM). The differences between multiple groups were analyzed by one-way analysis of variance (ANOVA) using SPSS 17.0 software, followed by Tukey's test. A P value of less than 0.05 was considered to be statistically significant.

Results
The levels of H 2 S in human HCC cells were higher than those in L02 cells. As shown in Fig. 1A-C, the protein levels of CSE and CBS in both SMMC-7721 and Huh-7 cells were significantly higher than those in L02 cells. While the protein levels of 3-MST in SMMC-7721 and Huh-7 cells were lower than those in L02 cells ( Fig. 1A and D). Furthermore, the levels of H 2 S in SMMC-7721 and Huh-7cells, as well as in the supernatant were notably higher than those in L02 cells and the supernatant (Fig. 1E and F). These results indicated that H 2 S could be involved in the development and progression of human HCC cells.

Dual effects of H 2 S on the growth and migration of HCC cells. H 2 S had no obvious effects on the
proliferation and viability of human normal hepatocytes ( Fig. 2A,D and G). The proliferation and viability of SMMC-7721 and Huh-7 cells were significantly enhanced by 10-100 μM NaHS and dose-dependently inhibited by 600-1000 μM NaHS (Fig. 2B,C,E,F,H and I). H 2 S did not affect the migration of L02 cells (Fig. 3A and D). The migration distance of HCC cells was significantly increased by 10-100 μM NaHS but decreased by 400-1000 μM NaHS relative to the control group (Fig. 3B,C,E and F). These results suggested that H 2 S plays an important role in the regulation of liver cancer cell growth and migration. Dual effects of H 2 S on apoptosis in HCC cells. TUNEL staining was performed to evaluate the apoptosis of L02, SMMC-7721, and Huh-7 cells with NaHS treatment. In L02 cells, there was no obvious change between each group (Fig. 4A and D). The numbers of apoptotic cells were significantly reduced by 25-100 μM NaHS but dose-dependently increased by 400-1000 μM NaHS in both SMMC-7721 and Huh-7 cells (Fig. 4B,C,E and F). Fig. 5A-D, H 2 S showed no obvious effect on the protein expression of each group in L02 cells. EGFR phosphorylation results in ERK-mediated signaling transduction, which supports cell growth and facilitates cellular motility 21,22 . Treatment with 25-50 μM NaHS dose-dependently increased EGFR and ERK phosphorylation in HCC cells with no obvious changes in total expression. However, 800-1000 μM NaHS reduced EGFR and ERK phosphorylation levels (Fig. 5A,E,F,H and I). Because MMP-2 is involved in cancer cell migration and invasion, which could be regulated by ERK1/2 signaling 23, 24 , MMP-2 expression in HCC cells was further examined. In HCC cells, 25-100 μM NaHS significantly increased MMP-2 protein expression, whereas 800-1000 μM NaHS decreased expression (Fig. 5A,G and J).  (Fig. 6A-D). Increasing evidence indicates that AKT activation is positively correlated with cancer progression 24,25 . We found that 25-50 μM NaHS dose-dependently up-regulated p-AKT without changing the total AKT level in HCC cells. However, 800-1000 μM NaHS reduced AKT phosphorylation (Fig. 6A,F and I). PTEN is a key negative regulator of the PI3K/AKT pathway and is frequently inactivated in human tumors 26 . Treatment with 25-100 μM NaHS reduced PTEN expression, whereas 800-1000 μM NaHS increased the level of PTEN protein (Fig. 5A,E and H). The ratio between BAX and BCL-2 proteins is an important factor in apoptosis regulation 27,28 . An increased BAX/BCL-2 ratio is sufficient to directly activate mitochondrial apoptosis in mammalian cells 27,29 . Treatment with 10-100 μM NaHS decreased the BAX/BCL-2 ratio, whereas 600-1000 μM NaHS increased it in HCC cells (Fig. 6A,G and J). H 2 S modulated the growth and angiogenesis of HCC xenograft tumors in nude mice. SMMC-7721 and Huh-7 cells have been widely used to establish xenograft tumor models [30][31][32] . We therefore investigated the effect of H 2 S on HCC xenograft growth in BALB/c nude mice. Treatment with 25-100 μM NaHS promoted the growth of HCC xenograft tumors in nude mice; however, 800-1000 μM NaHS exhibited significant anti-HCC effects (Fig. 7A-F). Treatment with 25-100 μM NaHS significantly increased the MVD, whereas 800-1000 μM NaHS dose-dependently decreased the MVD (Fig. 8A-D).

H 2 S mediated the EGFR/ERK/MMP-2 pathway in HCC cells. As shown in
H 2 S showed no obvious toxicity in nude mice bearing HCC xenografts. The body weights of mice were monitored every day throughout the experiment. Initially, there were no significant differences between the groups. However, at the end of the study, mice administered 25-100 μM NaHS had significantly lower body weights than the control group ( Fig. 9A and B). Since loss of body weight is a common feature of cancer development 33 , 25-100 μM NaHS might significantly accelerate cancer progression. Although the body weights in each group increased slowly during the experiment, the increase for mice administered 50 μM NaHS was the lowest (Fig. 9C and D). There were no significant differences in the total WBC count ( Fig. 9E and F) and relative weights of heart, liver, spleen, lung, kidney, and brain (Tables 1 and 2). Furthermore, there were no obvious morphological differences of heart, liver, spleen, lung, kidney, and brain in each group (Fig. 10A and B).

Discussion
Increasing evidence suggests that H 2 S plays important roles in many physiological and pathophysiological processes in mammals 34 . However, the role of H 2 S in cancer development and progression is controversial. We proposed that treatment with relatively low levels of exogenous H 2 S could promote cancer cell growth, whereas relatively high concentrations of H 2 S might exhibit anticancer effects 14 . To test our hypothesis, normal human showed that the levels of H 2 S in human HCC cells and the supernatant were significantly higher than those in L02 cells and the supernatant, suggesting that H 2 S is involved in the development and progression of human HCC cells. In addition, we found that H 2 S showed no obvious effects on the proliferation, viability, and migration of human normal hepatocytes. Treatment with 10-100 μM NaHS significantly promoted the growth and migration of HCC cells, whereas 600-1000 μM NaHS exhibited opposite effects in a dose-dependent manner. These results showed that H 2 S could play an important role in the proliferation, viability, and migration of human HCC cells.
To elucidate the molecular mechanisms by which H 2 S affects the biological behaviors of HCC cells, we measured the expression levels of several growth factors and proteins in HCC cells. EGFR, a member of the receptor tyrosine kinase family, plays a pivotal role in several cellular functions and is considered an attractive target for cancer therapy 24 . EGFR activation is involved in cancer cell proliferation, migration, and invasion, and it is overexpressed in various cancers including HCC 35,36 . EGFR phosphorylation can activate several downstream signaling pathways including PI3K/AKT and MAPK/MEK/ERK 16,35 . MMP-2 is positively regulated by the ERK pathway and regulates degradation of extracellular matrix components, which plays a key role in cancer metastasis 24,37 . A recent study found that 500 μM NaHS markedly increased MMP-2 expression in human PLC/ PRF/5 hepatoma cells 38 . Our results herein demonstrated that 25-50 μM NaHS increased EGFR and ERK1/2 phosphorylation, whereas they were decreased by 800-1000 μM NaHS. Furthermore, 25-100 μM NaHS significantly increased MMP-2 protein expression, whereas 800-1000 μM NaHS decreased its expression in HCC Figure 10. Effect of H 2 S on the structures of the heart, liver, spleen, lung, kidney, and brain in mice. Histopathology of the above-mentioned organs of the mice bearing SMMC-7721 (A) and Huh-7 (B) xenografts after treatment with NaHS at the indicated concentrations for 14 days. No significant morphological damage was observed. HE staining; original magnification 100x.
cells. The inconsistent results may be attributable to the different physiological properties among HCC cell lines. Furthermore, H 2 S showed no obvious effect on the protein expression of each group in L02 cells. These results suggested that H 2 S regulates the growth and migration of HCC cells through the EGFR/ERK/MMP-2 signaling pathway.
Apoptosis, an intrinsic cell-suicide program, is critical for normal development and maintenance of tissue homeostasis in multicellular organisms 14,39 . Herein, we examined apoptosis and the expression levels of major apoptosis-related proteins. Our results demonstrated that 25-100 μM NaHS significantly reduced apoptosis; however, 400-1000 μM NaHS dose-dependently increased apoptosis in HCC cells. There was no obvious change between each group in L02 cells. In addition, treatment with 400-1000 μM NaHS decreased the migration of HCC cells, suggesting that high concentrations of NaHS could induce apoptosis. The PI3K/AKT pathway plays a critical role in the proliferation, migration, metabolism, and apoptosis of cancer cells 23,35,40 . PTEN, a tumor suppressor protein, possesses alkaline phosphatase and protein phosphatase activities and can block PI3K/AKT signaling 26,41 . The decreased expression of PTEN may result in the activation of the AKT/ERK pathways, leading to the proliferation and migration of HCC cells. However, the increased level of PTEN could induce apoptosis of HCC cells. We examined the p-AKT and PTEN expression levels in HCC cells after treatment with H 2 S and found that 25-50 μM NaHS up-regulated AKT phosphorylation, and 25-100 µM NaHS decreased PTEN expression; however, 800-1000 μM NaHS dose-dependently reduced AKT phosphorylation and increased PTEN expression. BAX and BCL-2 are the downstream apoptosis regulators of the PI3K/AKT pathway, and the BAX/BCL-2 ratio is widely used as an important factor in the regulation of apoptosis 29,30 . We found that 10-100 μM NaHS decreased the BAX/BCL-2 ratio, whereas 600-1000 μM NaHS dose-dependently increased the BAX/BCL-2 ratio in HCC cells. However, there was no obvious change of the protein level of each group in L02 cells. These results together indicated that PTEN/PI3K/AKT signaling might represent another mechanism underlying the effects of H 2 S on HCC cell growth and migration.
Consistent with the in vitro findings, 25-100 μM NaHS significantly promoted xenograft tumor growth in nude mice; however, 800-1000 μM NaHS exhibited dose-dependent anti-HCC effects. Angiogenesis plays a key role in solid tumor growth, invasion, and metastasis and is an attractive target for tumor therapy 14,18,39 . To determine the effect of H 2 S on angiogenesis, we observed the MVD in dissected tumor tissues by anti-CD34 immunostaining and found that it was significantly increased with 25-100 μM NaHS treatment, suggesting that H 2 S had a pro-angiogenic effect in agreement with previous findings 15 . Intriguingly, 800-1000 μM NaHS dose-dependently reduced the MVD, consistent with the effect of H 2 S on HCC xenograft tumor growth in nude mice. These results demonstrated that H 2 S could regulate tumor growth by mediating angiogenesis. Furthermore, there were no significant differences in the relative organ weights and morphologies of heart, liver, spleen, lung, kidney, and brain or in total WBC count among the groups, indicating no obvious systemic toxicity. Thus, treatment with relatively high concentrations of an H 2 S donor may effectively exert antitumor effects.
In conclusion, our data indicate that the effect of H 2 S on HCC cells is a double-edged sword mediated by EGFR/ERK/MMP-2 and PTEN/AKT signaling pathways. Novel slow-releasing H 2 S donors and H 2 S-releasing hybrid drugs could be designed and applied for further antitumor research.