Pitavastatin inhibits exogenous free fatty acid (FFA)-enhanced proliferation and migration but promotes apoptosis in vitro

Chronic HFD intake can promote multiple types of tumor progression. However, the role of HFD and lipid-lowering drug, pitavastatin, on lung cancer progression is still unclear. Therefore, we used the mixture solution of palmitate acid (PA) and oleic acid (OA) as exogenous FFAs to mimic a high-fat environment. First, we determined the effect of pitavastatin on proliferation. We treated A549 and NCI-H520 cells with pitavastatin and then determined cell viability by the MTT method. As shown in Fig. 1A, B, pitavastatin reduced cell viability in dose- and time-dependent manners, whereas the cell viability of A549 and NCI-H520 cells was increased by FFAs (Fig. 1C, D). To further explore the role of pitavastatin in FFA-enhanced cell proliferation, we treated the cells with FFAs or pitavastatin plus FFAs and found that pitavastatin largely reduced FFA-enhanced cell proliferation (Fig. 1F). Moreover, the results of cell cycle analysis conducted by flow cytometry indicated that FFAs increased the percentage of cells in S phase, which was attenuated by pitavastatin treatment (Fig. 1E). These data suggest that exogenous lipids promote proliferative capacity by inducing the transition from G1 to S phase. Additionally, the wound healing test indicated that FFAs significantly promoted cell migration, which was completely blocked by pitavastatin in both A549 and NCI-H520 cells (Fig. 1G).

Fig. 1

Pitavastatin inhibits FFA-enhanced cell proliferation and migration but enhances apoptosis in lung cancer cell lines. A–D A549 and NCI-H520 cells were treated with indicated concentration of pitavastatin (A) or FFAs (C) for 24 h, or 5 µM pitavastatin (B) or 150 µM FFAs (D) at the indicated time. Cell viability was determined using MTT assay. E-K A549 and NCI-H520 cells are treated with 150 µM FFAs or 5 µM pitavastatin plus 150 µM FFAs for 24 h (others) or 48 h (G); FACS (E), MTT (F), scratch (G), and Annexin V-FITC/PI staining (H) assays were used to evaluate cell cycle, viability, migration, and apoptosis, respectively. Protein expression of CDH1, PCNA, vimentin, Bcl-2, and BAX was determined by Western blot with density quantitative analysis (I). Cellular TG (J) and FFA (K) levels were measured by indicated assay kits. Mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001 vs control group; #p < 0.05; ##p < 0.01; ###p < 0.001 vs FFA-treated group; A–D and F: n = 6; E, G–K: n = 3; Pita, pitavastatin

In order to explore the effect of FFAs and pitavastatin on cell apoptosis, we treated cells and measured the proportion of cell apoptosis by Annexin V-FITC/PI staining assay kit. Our results showed that FFAs significantly reduced A549 and NCI-H520 cell apoptosis at both early and late stages, while pitavastatin reversed FFA-inhibited cell apoptosis (Fig. 1H). To unravel the underlying mechanisms, we determined expression of proteins involved in proliferation (proliferating cell nuclear antigen, PCNA), migration and invasion [cadherin 1 (CDH1) and vimentin], and apoptosis [B-cell lymphoma/leukenfia-2 (Bcl-2) and Bcl-2-associated X (BAX)] in A549 and NCI-H520 cells. We found that the protein expression of PCNA, vimentin, and Bcl-2 was increased, but that of CDH1 and BAX was decreased by FFAs, in both A549 and NCI-H520 cells. However, these changes were reversed by pitavastatin administration (Fig. 1I). Taken together, the above results suggest that pitavastatin attenuated FFA-enhanced cell viability, and the effects were associated with the regulation of proteins involved in cell proliferation, migration, invasion, and apoptosis.

Mounting evidence indicates that the excess uptake and production of FFAs in most solid malignancies is increased. The FFAs provide nutrients and energy for tumor growth and migration. Therefore, we determined the cellular TG, FFA, and cholesterol content. Our results showed that the exogenous FFAs increased TG, FFA, and cholesterol content, which was almost reversed by pitavastatin (Fig. 1J, K and Fig. S1), indicating that pitavastatin-inhibited cell viability may be correlated to the reducing lipid accumulation in cells.

Pitavastatin inhibits HFD-exacerbated NSCLC progression in C57BL/6J mice

To investigate the role of pitavastatin and HFD on tumor growth in mice, Lewis lung cancer–bearing C57BL/6J mice were fed a normal chow (NC) or HFD in the presence or absence of pitavastatin (Fig. 2A). After euthanasia, mouse tumor tissues were collected. Compared to NC group, we found that the tumor volume and weight were increased in the HFD group mice. Importantly, pitavastatin administration decreased tumor volume and weight in both NC and HFD feeding conditions (Fig. 2B–D). In addition, HFD exacerbated metastasis in lung tissues, which was also attenuated by pitavastatin treatment (Fig. 2E). At the molecular level, tumor tissues of the HFD group mice exhibited high expression of Ki-67 and vimentin, but low expression of BAX. However, pitavastatin treatment significantly increased BAX, while inhibited Ki-67 and vimentin expressions in both feeding conditions (Fig. 2F). Taken together, our results show that HFD promotes tumor growth, which can be largely reversed by the administration of pitavastatin.

Fig. 2

Pitavastatin inhibits the progression of NSCLC in C57BL/6J mice. A Experimental design: C57BL/6J mice in four groups received the following treatment: normal chow (NC) groups (5 mice/group), fed normal food and injected with NS or pitavastatin solution (2 mg/kg/day) for 5 weeks; HFD groups (7 mice/group), fed HFD and injected with NS or pitavastatin solution for 5 weeks. At the third week, mice received s.c. injection of LLC1 cells (106 cells/100 µL/mouse). At the end of experiment, blood, tumor, and lung tissues were collected. B Tumor size was determined once every 3 days. C, D Tumor samples were photographed (C) and weighted (D). E Paraffin sections of lung tissues were performed H&E staining, and the lung carcinogenesis area was quantified by ImageJ software. F Tumor paraffin sections were performed IHC staining to detect the expression of Ki-67, BAX, and vimentin, and mean density (MD) was quantified by ImageJ software. Mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001 (n ≥ 5); Pita, pitavastatin

Epidemiological evidence suggests that HFD alters lipid composition in the tumor microenvironment, which affects the activation of oncogenes and promotes tumor progression (Peck and Schulze 2019). Subsequently, we measured lipid levels in serum and tumor tissues. We found that HFD elevated serum T-CHO, LDL-C, HDL-C, TG, and FFA levels (Table S2). Consistent with the tumor inhibition effects, pitavastatin also reduced HFD-induced T-CHO, TG, and FFAs levels in serum. Furthermore, we indicated that both TG and FFA levels in orthotopic tumors were enhanced by HFD, and these changes were nearly reversed by the administration of pitavastatin. Therefore, our in vitro and in vivo results reveal that pitavastatin significantly inhibits tumor progression by regulating lipid metabolism.

Pitavastatin inhibits lung cancer progression by reducing CD36 expression

CD36 is a receptor of fatty acids that can promote the development of cardiovascular diseases and cancer (Wang and Li 2019). To evaluate whether CD36 is involved in HFD-enhanced or pitavastatin-reduced cancer, we determined CD36 expression in tumor samples or cells. As shown in Fig. 3A, we found that CD36 levels in tumors of the HFD groups were higher than those in the tumors of the NC groups, while pitavastatin inhibited CD36 expression in those two conditions. Consistently, the in vitro results also found that CD36 expression was enhanced by FFAs in A549 and NCI-H520 cells but reduced by pitavastatin treatment (Fig. 3B). In order to explore whether CD36 is related to the development of NSCLC, we collected serum from 14 NSCLC patients and 24 healthy human volunteers. Our results found that soluble CD36 (sCD36) levels in plasma were higher in NSCLC patients than in healthy ones (Fig. 3C). More importantly, we showed that CD36 protein levels in tumor tissues of NSCLC patients were much higher than those in adjacent non-cancerous tissues (Fig. 3D). To sum up, the above results suggest that CD36 may play an important role in NSCLC progression.

Fig. 3

CD36 expression is positively associated with lung cancer development. A Tumor paraffin sections collected from Fig. 2A were conducted IHC staining to detect CD36 expression with quantified MD by ImageJ software; mean ± SEM; *p < 0.05; ***p < 0.001 (n = 5). B A549 and NCI-H520 cells were treated with 150 µM FFAs or 5 µM pitavastatin plus FFAs for 24 h; protein expression of CD36 was determined by Western blot with density quantitative analysis. Mean ± SEM; **p < 0.01; ***p < 0.001 vs control group; #p < 0.05; ###p < 0.001 vs FFA-treated group (n = 3); Pita, pitavastatin. C Plasma sCD36 levels were detected by the CD36 ELISA kit. Mean ± SEM; ***p < 0.001. D Expression of CD36 in cancer tissues (n = 5) and adjacent non-cancerous tissues (n = 2) was detected by IHC staining, and MD was quantified by ImageJ software. ANCT, adjacent non-cancerous tissues

To further investigate the critical role of CD36 in NSCLC, we transfected A549 cells with the pCMV-CD36 plasmid for overexpression or NCI-H520 cells with the CasRX-CD36 plasmid to knockdown CD36. First, we verified the transfection efficacy of plasmids, and CD36 levels were enhanced ~ 2.5- or ~ 0.4-fold by transfection with the pCMV-CD36 or CasRX-CD36 plasmid, respectively (Figure S2). CD36 overexpression enhanced cell viability (Fig. 4A, upper panel), proliferation (Fig. 4B), and migration (Fig. 4D, left and right panels) while reducing the ratio of apoptotic cells (Fig. 4E, left and right panels). In contrast, cell viability (Fig. 4A, lower panel), proliferation (Fig. 4C), and migration (Fig. 4D, middle and right panels) were reduced in CD36-knockdown cells, but cell apoptosis was enhanced (Fig. 4E, middle and right panels). More importantly, at the molecular level, cells overexpressing CD36 showed enhanced PCNA, vimentin, and Bcl-2 expressions but reduced CDH1 and BAX levels (Fig. 4F, left and right panels). Consistent with cell phenotype results, proliferation-, migration- and apoptosis-related protein expressions were also regulated by CD36 inhibition, which was opposite to observations in the overexpressed cells (Fig. 4F, middle and right panels). Moreover, FFA-enhanced cell viability was largely reduced (Fig. 4G), while the reduction effects of FFA on cell apoptosis were enhanced (Fig. 4H) in CD36-knockdown NCI-H520 cells. Consistently, FFA-regulated vimentin, PCNA, and BAX were also attenuated in CD36-knockdown NCI-H520 cells (Figure S3). Taken together, these results indicate that CD36 levels are positively associated with cancer development.

Fig. 4

CD36 expression is positively correlated with cell proliferation and migration in vitro. A–F A549 cells were transfected with pCMV or pCMV-CD36 plasmid, and NCI-H520 cells were transfected with CasRX or CasRX-CD36 plasmid for 12 h in serum-free medium and then cultured in complete medium for another 24 h. Cells were collected for determination of cell viability (A), cell cycle (B, C), migration (D), and apoptosis (E) by MTT assay, FACS, wound healing test, and Annexin V-FITC/PI staining, respectively. Protein expression of CD36, CDH1, PCNA, vimentin, Bcl-2, and BAX was determined by Western blot with density quantitative analysis (F). G, H NCI-H520 cells were transfected with CasRX or CasRX-CD36 plasmid for 12 h and then cultured in complete medium for 24 h, followed by FFAs (150 µM) treatment for 24 h. Cells were collected for determination of cell viability (G) and apoptosis (H). Mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001; A, G: n = 6; B–F, H: n = 3

The results in Fig. 3A, B showed that pitavastatin reduced HFD or FFA-induced CD36 expression in tumor tissues or cells, indicating that pitavastatin-reduced cancer development may be regulated by CD36. Therefore, we treated CD36-knockdown or overexpression cells with or without pitavastatin. Our results showed that CD36 overexpression enhanced cell viability (Fig. 5A, upper panel), proliferation (Fig. 5B), and migration (Fig. 5D, left and right panels) but reduced cell apoptosis (Fig. 5E, upper panel), which was largely attenuated by pitavastatin in A549 cells. In contrast to CD36 overexpression, both CD36 knockdown and pitavastatin treatment inhibited cell viability (Fig. 5A, lower panel), proliferation (Fig. 5C), and migration (Fig. 5D, middle and right panels) but increased cell apoptosis (Fig. 5E, lower panel) in NCI-H520 cells. Consistent with the phenotypic results in cells, pitavastatin also reversed CD36 overexpression–regulated cell proliferation-, migration-, and apoptosis-related protein expression in A549 cells (Fig. 5F). However, the regulatory effect of pitavastatin in related protein expression was attenuated in CD36-knockdown cells (Fig. 5G). Therefore, we demonstrate that CD36 is crucial to the development of NSCLC and that pitavastatin-mediated regulation of cancer processes depends on CD36 expression, at least in part.

Fig. 5

The effects of pitavastatin on cell proliferation, migration, and apoptosis are related to CD36 expression. A549 cells were transfected with pCMV or pCMV-CD36 plasmid, and NCI-H520 cells were transfected with CasRX or CasRX-CD36 plasmid for 12 h and then cultured in complete medium for 24 h, followed by received pitavastatin (5 µM) treatment for 24 h. Cells were collected for determination of cell viability (A), cell cycle (B, C), migration (D), and apoptosis (E). Protein expression of CD36, PCNA, Bcl-2, and BAX was determined by Western blot with density quantitative analysis (F, G). Mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001; A: n = 6; B–F: n = 3; Pita, pitavastatin

The inhibitory effects of pitavastatin on HFD-enhanced tumor progression are impaired in CD36−/− mice

The above in vitro results indicated that pitavastatin-reduced cell proliferation almost depends on CD36 reduction. Therefore, we used global CD36-deficient (CD36−/−) mice for determining the protective role of pitavastatin in lung cancer. C57BL/6J and CD36−/− mice were fed with HFD and received normal saline (NS) or pitavastatin injection for 5 weeks. At the third week, mice were injected with LLC1 cells (Fig. 6A). Compared to C57BL/6 J mice, tumor progression was largely impaired in CD36−/− mice. In addition, pitavastatin administration reduced tumor size and weight in both C57BL/6J and CD36−/− mice. However, the protective effects of pitavastatin in CD36−/− mice were weaker than those in C57BL/6 J mice (Fig. 6B–D). In addition, H&E staining showed that tumor metastasis was suppressed in the lung tissues of CD36−/− mice (Fig. 6E). Furthermore, the results of IHC staining of tumor sections showed that the expression of Ki-67 and vimentin in CD36−/− mice was much lower than that in C57BL/6J mice, while BAX expression was higher in CD36−/− mice (Fig. 6F). Consistent with the tumor reduction effects, pitavastatin also regulated related protein expression but exhibited fewer effects in CD36−/− mice. Our data suggest that CD36 deficiency greatly reduces lung cancer progression and that the protective effect of pitavastatin on lung cancer is impaired in CD36−/− mice.

Fig. 6

The reduction effects of pitavastatin on tumor progression were impaired in CD36−/− mice. A Experimental design: C57BL/6J and CD36−/− mice (9 mice/group) in four groups fed with HFD and injected with NS or pitavastatin solution (2 mg/kg/day) for 5 weeks. At the third week, mice received s.c. injection of LLC1 cells (106 cells/100 µL/mouse). At the end of experiment, blood, tumor, and lung tissues were collected. B Tumor size was determined once every 3 days. C, D Tumor tissues were photographed (C) and weighted (D). E, F Paraffin sections of lung tissues were performed H&E staining, and lung carcinogenesis area was quantified by ImageJ software (E), or conduced IHC staining to detect Ki-67, BAX, and vimentin expression with MD quantified by ImageJ software (F). Mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001; B, D: n = 8; E, F: n = 5; Pita, pitavastatin

We indicated that pitavastatin inhibited cell proliferation and cancer development through regulating lipid profiles (Fig. 1J, K and Table S2). To explore whether the impaired protective role of pitavastatin in CD36−/− mouse is also related to lipid levels, we determined lipid profiles in serum and tumor tissues. As shown in Table S3, we found that lipid levels in CD36−/− mouse serum were much lower than those in WT mouse serum. In addition, the reduction effects of pitavastatin in T-CHO, TG, and FFA levels were largely impaired in CD36−/− mice. Taken together, these results indicate that pitavastatin-mediated reduction in tumor progression is partly dependent on CD36 expression and lipid levels.

Pitavastatin-attenuated tumor progression is regulated by the CD36/AKT/mTOR pathway

The AKT-mTOR pathway is activated in varieties of tumors and performs an important function in regulating cell growth, promoting cell invasion and metastasis, and enhancing neo-angiogenesis (LoRusso 2016). To explore whether the AKT-mTOR pathway is involved in pitavastatin-reduced or CD36-induced cell proliferation, we determined the activation of AKT and mTOR in cells and tumor tissues. Both phosphorylated AKT (p-AKT) and p-mTOR levels were enhanced by exogenous FFA treatment while being significantly reduced by pitavastatin treatment in A549 and NCI-H520 cells (Fig. 7A, B and Fig. S4A and B). In addition, we found that CD36 overexpression also activated AKT and mTOR (Fig. 7C and Fig. S4C). In contrast, reduction of CD36 expression inhibited p-AKT and p-mTOR levels (Fig. 7D and Fig. S4D). Either reduced or enhanced p-AKT and p-mTOR by CD36 inhibition or overexpression can also be further regulated by pitavastatin. Compared to normal or overexpression cells, the reduction effects of pitavastatin on the levels of p-AKT and p-mTOR were lower in CD36-knockdown cells. Notably, compared with control cells, the upregulated levels of p-AKT and p-mTOR by exogenous FFAs in CD36-knockdown NCI-H520 cells were obviously impaired (Fig. 7E and Fig. S4E). Moreover, in vivo results also indicated that pitavastatin reduced p-AKT and p-mTOR levels in tumor tissues of both NC and HFD conditions (Fig. 7F, G). Consistent with in vitro results (Fig. 7D), we found that the reduction effect of pitavastatin on p-AKT and p-mTOR expressions in tumor tissues of CD36−/− mice was less than that in WT mice (Fig. 7G).

Fig. 7

The reduction effects of pitavastatin on tumor progression are regulated by CD36/AKT/mTOR pathway. A–E A549 (A) and NCI-H520 (B) cells were treated with 150 µM FFAs or 5 µM pitavastatin plus FFAs for 24 h. A549 cells were transfected with pCMV or pCMV-CD36 plasmid for 12 h (C); NCI-H520 cells were transfected with CasRX or CasRX-CD36 plasmid (D, E) for 12 h and then cultured in complete medium for 24 h, followed by treatment with 5 µM pitavastatin (C, D) or 150 µM FFAs (E) for 24 h. Protein expression of p-AKT, AKT, p-mTOR, and mTOR was detected by Western blot. F, G Tumor paraffin sections collected from Fig. 2A (F) or Fig. 6A (G) were performed IHC staining to detect the expression of p-AKT and p-mTOR with MD quantified by ImageJ software. H–L A549 cells were transfected with pCMV or pCMV-CD36 plasmid for 12 h and then cultured in complete medium for 24 h, followed by received LY294002 (10 µM) treatment for 24 h. Cells were collected for determination of cell viability (H) and apoptosis (I, J). Protein expression of CD36, vimentin, PCNA, BAX, p-AKT, AKT, p-mTOR, and mTOR was determined by Western blot (K, L). Mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001; A–E, I–L: n = 3; F, G: n = 5; H: n = 6; Pita, pitavastatin; LY, LY294002

To further confirm the association between CD36 and the AKT pathway, we treated A549 cells with LY294002 (an AKT inhibitor) and found that LY294002 significantly inhibited cell viability (Fig. 7H) but enhanced apoptosis (Fig. 7I, J), with greater degree in CD36-overexpressed A549 cells than in control cells. In addition, LY294002 also regulated related protein levels, and the activity of AKT and mTOR (Fig. 7K, L and Fig. S4F and G). Therefore, our results suggest that pitavastatin attenuates HFD-accelerated NSCLC development contributed by reducing CD36-AKT-mTOR pathway, at least in part.