DLPC induces ferroptosis in cancer cells

Phosphatidylcholine (PC) is the most abundant phospholipid in mammalian cells. It accounts for approximately 50% of all phospholipids, represents the main component of cellular and subcellular membranes [1], and is also found as a metabolite in serum. PC contributes to regulating cellular functions, and its level may be altered under certain physiological and pathological conditions; thus, it may serve as a marker for disease progression [2]. PC is generally studied as a whole. However, PC is a mixture of many species that may have distinct functions [3]. Therefore, it is worthwhile to study the individual species of PC.

Previous studies revealed that PC is involved in cancer development. The levels of fatty acids and complex lipids, including PC, are elevated in cancer cells, and increased PC biosynthesis contributes to the malignant transformation of intestinal epithelial cells [4]. Ingram et al. [5] found that drug-resistant prostate cancer cell lines have higher levels of PC than parental control cells. These findings suggest that PC promotes cancer development, but other groups have reported opposite effects. For example, Yu et al. [6] reported that PCYT1A (the rate-limiting enzyme for PC synthesis) suppresses cancer cell proliferation and migration by inhibiting the mTORC1 pathway in lung adenocarcinoma. Moreover, the expression of PCYT1A is associated with better survival in patients with urothelial cancer of the bladder [7]. Given these contradictory findings, further work is needed to clarify the roles of PC and individual PC species in cancer cells.

The mouse colon cancer cell line MC38 originated from the colon tumor tissue of C57BL/6 mice that were long-term exposed to the carcinogen 1,2-dimethylhydrazine (DMH). A murine colon carcinoma model generated by subcutaneously inoculating MC38 cells in mice is one of the most commonly used powerful models for the study of tumor biology and anti-tumor drug development, with the advantages of a high tumor formation rate and tumor metastasis similar to that of human colon carcinomas [8]. Thus, we first tested whether PC affects cell growth of MC38 cells. The cells were cultured for 24 h with various doses of PC, and the number of viable cells was measured using a Cell Counting Kit-8. The results revealed that PC dose-dependently decreased cell viability, and the 1.5 mM dose group exhibited the most significant effect ( Supplementary Figure S1A).

PC is a mixture of many components. Soybean is one of the rich sources for PC, and the main component of PC is 1, 2-dilinoleoylsn-glycero-3-phosphocholine (DLPC) (40%–52% in PC). The structure of PC comprises a backbone of glycerol, a lipid head group and two acyl chains linked to the three hydroxyl groups of glycerol. The diversity of PC components mainly comes from the highly diverse acyl-chain composition [9]. Generally, each of the two acyl chains ranges from 16 to 22-carbon-long, and 0 to 6 unsaturated bonds are distributed on the two acyl chains. This complex combination generates many PC components. Moreover, unlike proteins and nucleic acids, obtaining pure individual PC components relies on purification or commercial synthesis, which are very expensive. Another problem is that, due to technical limitations, only some PC components can be commercially synthesized. These factors make it infeasible to systematically screen all the PC components. Therefore, we choose to first test the effect of DLPC, the main component of PC. DLPC is an 18:2/18:2 PC that has a glycerol ester skeleton, one phosphate group, one cholinergic group, two 18-carbon-long acyl-chain tails and two unsaturated bonds on each chain.

To explore the impact of the length of the acyl chain and the degree of unsaturation on the effect of PC, several commercially available PC components structurally similar to DLPC were also tested, as shown in Supplementary Figure S1B. Three of the selected components were PC (18:0/18:2), PC (18:1/18:1) and PC (18:0/18:1), which have the same length of acyl chain as DLPC, but the number and distribution of the unsaturated bonds on the two acyl chains are slightly different from those of DLPC. The other two selected components were PC (16:0/16:0) and PC (16:0/18:1), which have different length of acyl chain and degrees of unsaturation from DLPC. Among them, only DLPC significantly inhibited MC38 cell growth ( Supplementary Figure S1B). This finding suggested that DLPC is an active component in the inhibitory effect of PC on MC38 cancer cells. Notably, the ability of DLPC to inhibit MC38 cell viability was greater than that of PC at the same concentration. Interestingly, the effect of DLPC was not observed in cells treated with structurally similar PCs having fatty acid chains of the same length but different number of unsaturated bonds or with PCs having acyl chains of different lengths. This suggests that the length of the acyl chains and the number of unsaturated bonds are relevant to the ability of DLPC to decrease MC38 cell viability. Our results show that individual species of PC might have distinct functions.

As shown in Figure 1A, the ability of DLPC to inhibit MC38 cell viability was dose-dependent. A strong colony formation ability is a characteristic of cancer cells and reflects their proliferation capacity. Here, we found that DLPC treatment significantly decreased the ability of MC38 cancer cells to form colonies ( Figure 1B). Notably, 0.1 mM DLPC significantly decreased the colony number of MC38 cells but did not obviously decrease cell viability. This finding suggested that 0.1 mM DLPC is sufficient to impair the proliferative capacity of MC38 cells but not to kill them.

Figure 1 .

Different types of cells have different sensitivity to DLPC

(A) Cell viability of MC38 cells treated with DLPC. (B) Colony formation of MC38 cells treated with DLPC. Left panel: images of colony formation. Right panel: statistics of colony numbers. (C–H) Cell viability of colon cancer cell lines (C), liver cancer cell lines (D), osteosarcoma cell lines (E), melanoma cell lines (F), hematological malignancy cell lines (G), and primary mouse splenocytes and hepatocytes (H) treated with DLPC for 24 h. All cultured cells were treated when they reached 60%–70% confluence via logarithmic growth. The value of each group was normalized to that of its corresponding solvent-only control group (0 mM), which was designated 100%. All the experiments were independently repeated at least three times. The results are presented as the mean±SD. * P<0.05, ** P<0.01, *** P<0.001, **** P<0.0001. All P values were calculated using two-tailed Student’s t-test.

We next assessed whether DLPC had similar effects on other types of cells. Both colon cancer and liver cancer are digestive system tumors with high morbidity. Melanoma is a type of tumor that originates from the epithelium and has a high incidence rate in Western countries. The B16F10 melanoma cell line is commonly used in syngeneic murine models and is widely used in the study of tumor biology and anti-tumor drug development. Hematological tumors are characteristically different from solid tumors. These frequently studied tumor types were included in our experiments. The response patterns of three human colon cancer cell lines, HT-29, RKO, and SW480, were similar to those of mouse MC38 colon cancer cells, with 0.5 mM DLPC triggering the most prominent inhibitory effect ( Figure 1C). DLPC also inhibited the viability of two liver cancer cell lines, human SMMC-7721 cells and mouse Hepa1-6 cells ( Figure 1D), and dramatically inhibited the viability of two human osteosarcoma cell lines, MG63 cells and U2OS cells ( Figure 1E). Two melanoma cell lines, human A875 cells and mouse B16F10 cells, were less sensitive to DLPC ( Figure 1F). For hematological tumors, DLPC did not affect the viability of the human T-lymphocytic leukemia cell line, Jurkat, or the multiple myeloma cell line, MM.1S ( Figure 1G). These results suggest that the sensitivity of different types of cells to DLPC treatment varies. The same types of cancer cells show similar sensitivities to DLPC treatment. However, the DLPC sensitivity is somewhat heterogeneous within each tested cell type.

We also measured the effects of DLPC on normal cells. As myelosuppression and hepatotoxicity are two common and serious side effects of anti-cancer drugs, we investigated the effect of DLPC on normal splenocytes and hepatocytes. Primary mouse splenocytes and hepatocytes were freshly isolated and cultured for several days in vitro and then treated with or without DLPC. Notably, primary splenocytes and hepatocytes showed no growth inhibition when treated with DLPC ( Figure 1H). In fact, DLPC was observed to increase hepatocyte viability ( Figure 1H).

The above results showed that not all cells are sensitive to the inhibitory effect of DLPC. We speculate that the cell type-specific sensitivity to DLPC might reflect differences in cellular membrane features, functions or tissues of origin. We found that osteosarcoma cells were the most sensitive to DLPC, suggesting that osteosarcoma might be further investigated as a candidate target for DLPC treatment. Liver cancer cells were less sensitive to DLPC in our experimental setting. Surprisingly, in normal hepatocytes, DLPC increased cell viability. This may reflect the unique features of the liver cell lineage. Consistent with this finding, previous studies found that DLPC protects hepatocytes from ethanol-induced apoptosis [10] and that PCs can reduce portal pressure and improve liver function in CCl4-induced liver injury [11]. The hematological cells tested in the present study, including normal splenocytes and malignant cells, consistently resisted the inhibitory effect of DLPC. This suggests that hematological malignancy is not a suitable target for DLPC treatment. However, the resistance of normal splenocytes and hepatocytes to DLPC suggests that DLPC may have relatively mild side effects as a potential therapeutic agent.

MC38 cells were used to further investigate the mechanism underlying the DLPC-induced decrease in cell viability. Corresponding to the cell viability results, the cell morphology changed as the dose of DLPC increased. Most of the cells lost their intact shape and shrank in cultures treated with 0.5 mM DLPC ( Figure 2A). Propidium Iodide (PI) staining was used to further verify cell death. Consistent with the cell morphology findings, the number of PI-stained cells increased with the increase in the concentration of DLPC, with the 0.5 mM DLPC-treated group exhibiting the most dead cells ( Figure 2B).

Figure 2 .

DLPC induces ferroptosis in MC38 cells

(A) Morphology of cells treated with DLPC for 24 h under light microscopy. Scale bar: 100 μm. (B) PI staining images of cells treated with DLPC for 24 h under fluorescence microscopy. Scale bar: 100 μm. (C) Cell viability of cells treated for 24 h with DLPC and inhibitors of ferroptosis or apoptosis. The value of each group was normalized to that of its corresponding solvent-only control group (Control), which was designated 100%. (D) Cell viability of cells treated for 24 h with DLPC and an iron ion-chelating agent. The value of each group was normalized to that of its corresponding solvent-only control group (Control), which was designated 100%. (E) Flow cytometry results obtained for BODIPY-C11-stained cells treated with DLPC and a ferroptosis inhibitor (Lip-1) for 4 h. (F) Morphology of organelles in cells treated with DLPC for 4 h under electron microscopy. Yellow arrows indicate the mitochondria. Scale bar: 2 μm. All cultured cells were treated when they reached 60%–70% confluence via logarithmic growth. All the experiments were independently repeated at least three times. The results are presented as the mean±SD. * P<0.05, ** P<0.01, *** P<0.001. ns indicates no statistical difference. All P values were calculated using two-tailed Student’s t-test.

Ferroptosis is an iron-dependent form of non-apoptotic cell death that has attracted significant attention in recent years for its potential as an alternative strategy for cancer treatment. Mechanistically, ferroptosis is triggered by excessive iron-dependent peroxidation of polyunsaturated fatty acid (PUFA)-containing phospholipids in cell membranes. Given that DLPC is a PUFA, we hypothesized that DLPC may induce cancer cell death by triggering ferroptosis. Thus, we assessed the features of apoptosis and ferroptosis, the two most common types of cell death. The results revealed that ferroptosis was the only parameter that changed significantly upon DLPC treatment. Consistent with this, the ferroptosis inhibitors Fer-1 and Lip-1 dramatically abolished the ability of DLPC to inhibit MC38 cancer cell growth, whereas the apoptosis inhibitor Z-VAD did not ( Figure 2C). The iron chelator desferrioxamine (DFO) also significantly attenuated the ability of DLPC to inhibit MC38 cancer cell growth ( Figure 2D). These data suggested that DLPC might induce ferroptosis in MC38 cells. Ferroptosis is associated with increased lipid peroxidation in cell membranes. We thus stained cells with BODIPY-C11, which is a fluorescent probe that detects reactive oxygen species (ROS) in cells via the detection of an oxidation-induced shift in the fluorescence emission peak from ~590 to ~510 nm. Flow cytometric analysis revealed that DLPC treatment of MC38 cells increased the fluorescence intensity at ~510 nm, and this change was attenuated by the ferroptosis inhibitor Lip-1 ( Figure 2E). As ferroptosis triggers characteristic morphological changes in organelles, we used electron microscopy to observe the morphology of organelles in MC38 cells treated with and without DLPC. Compared to those of control cells, the mitochondria of DLPC-treated cells were smaller in size and exhibited increased membrane density and fewer mitochondrial cristae, which are characteristic of mitochondria in cells undergoing ferroptosis ( Figure 2F). These observations supported the notion that DLPC induces MC38 cancer cell death by inducing ferroptosis.

In summary, we herein show that DLPC, an 18:2/18:2 species of PC, decreases cancer cell viability by inducing ferroptosis without affecting normal cells. To our knowledge, this is the first report revealing the impact of DLPC on cancer cells and the underlying molecular mechanism involved. Based on our present results, we suggest that DLPC might be a potential anti-cancer drug candidate.

Supplementary Data

Supplementary data is available at Acta Biochimica et Biophysica Sinica online.

COMPETING INTERESTS

The authors declare that they have no conflict of interest.

Funding Statement

This work was supported by the grant from the National Key R&D Program of China (No. 2022YFA0806002).