Skip to main content
NCBI home page
EMBO Reports logoLink to EMBO Reports
. 2019 Aug 5;20(10):e48115. doi: 10.15252/embr.201948115

Lin28 enhances de novo fatty acid synthesis to promote cancer progression via SREBP‐1

Yang Zhang 1,2,†, Chenchen Li 2,†, Chuanzhen Hu 3, Qian Wu 4, Yongping Cai 5, Songge Xing 1,2, Hui Lu 2, Lin Wang 2, De Huang 2, Linchong Sun 6, Tingting Li 2, Xiaoping He 2, Xiuying Zhong 6, Junfeng Wang 7, Ping Gao 1,6, Zachary J Smith 3, Weidong Jia 1,✉, Huafeng Zhang 1,2,✉
PMCID: PMC6776893  PMID: 31379107

Abstract

Lin28 plays an important role in promoting tumor development, whereas its exact functions and underlying mechanisms are largely unknown. Here, we show that both human homologs of Lin28 accelerate de novo fatty acid synthesis and promote the conversion from saturated to unsaturated fatty acids via the regulation of SREBP‐1. By directly binding to the mRNAs of both SREBP‐1 and SCAP, Lin28A/B enhance the translation and maturation of SREBP‐1, and protect cancer cells from lipotoxicity. Lin28A/B‐stimulated tumor growth is abrogated by SREBP‐1 inhibition and by the impairment of the RNA binding properties of Lin28A/B, respectively. Collectively, our findings uncover that post‐transcriptional regulation by Lin28A/B enhances de novo fatty acid synthesis and metabolic conversion of saturated and unsaturated fatty acids via SREBP‐1, which is critical for cancer progression.

Keywords: de novo fatty acid synthesis, Lin28, lipotoxicity, saturated and unsaturated fatty acids, SREBP cleavage‐activating protein, SREBP‐1

Subject Categories: Cancer, Metabolism

Introduction

Lin28 is highly correlated with a wide range of human malignancies, including neuroblastoma, hepatocellular carcinoma, melanoma, and Wilms’ tumor, and is associated with poor clinical outcome via the repression of let‐7 microRNAs 1, 2. While the Lin28/let‐7 axis is traditionally known to drive multiple tumor development in murine models 3, 4, 5, 6, 7, 8, 9, the major functions of Lin28A and its homologue Lin28B are not solely dependent on let‐7. Lin28 (including Lin28A and Lin28B) also serves as RNA binding proteins with both a cold‐shock domain (CSD) and a cys‐cys‐his‐cys (CCHC) domain. Lin28 can directly bind mRNAs at GGAGA motifs and enhance mRNA translation, thereby regulating mRNA metabolism, cell cycle, cell growth, glycolysis, and oxidative phosphorylation (OxPhos)‐related protein expression 10, 11, 12, 13, 14, 15. These studies have provided extensive insights into the roles of Lin28 independent of let‐7. Regarding metabolism regulation, Lin28 has been reported to regulate glucose metabolism by PI3K‐AKT‐mTOR signaling 14 and enhance OxPhos metabolism to stimulate tissue repair 15. Lin28 is also reported to regulate stem cell metabolism and pluripotency 16. Of note, our group recently demonstrated that the Lin28/let‐7 axis regulates the Warburg effect via PDK1 under normoxic condition in cancer cells 7. Nevertheless, the illustration of the interplay between the oncogenic effects of Lin28 and the metabolic pathway is just getting started and it is yet to be explored whether Lin28 plays other critical roles in regulating metabolism during tumor progression.

Metabolic reprogramming has been defined as a core hallmark of cancer 17. The cancer cell metabolism is now extended far beyond the original observation of aerobic glycolysis, the so‐called Warburg effect 18, 19. Besides typical changes in glucose metabolism, cancer cells were also observed to prefer de novo fatty acid synthesis to maintain rapid cell growth and cell proliferation 20, 21. Several reactions are involved in converting carbons from nutrient to fatty acids. ATP‐citrate lyase (ACLY), acetyl‐CoA carboxylase (ACC), fatty acid synthase (FASN), and stearoyl‐CoA desaturase (SCD) are the key enzymes involved in generating fatty acids from glucose and reductive glutamine metabolism in cancer cells 22, 23, 24. Likewise, acetate, as a carbon source, contributes to producing acetyl‐CoA for de novo fatty acid synthesis in certain tumors including liver cancers 25. The crucial transcriptional regulator of lipid synthesis, sterol regulatory element‐binding protein 1 (SREBP‐1), which extensively targets fatty acid synthesis genes including ACLY, ACC, FASN, and SCD, is synthesized as an inactive precursor. SREBP cleavage‐activating protein (SCAP) binds to the SREBP‐1 precursors to form a complex, which is embedded to the endoplasmic reticulum (ER). When sterol is deficient in cells, SCAP escorts the SREBP‐1 precursors to the Golgi where they are activated by a two‐step cleavage 26, 27. Inhibition of SREBP‐1 induces ER stress through loss of fatty acid desaturation in human glioblastoma cells 28, 29. These studies underline the importance of fatty acid synthesis in cancer cell biology, whereas the underlying regulatory mechanisms for fatty acid synthesis in cancer cells are still largely unknown.

Our group has previously documented that the Lin28/let‐7 axis regulates the Warburg effect via PDK1 7. It is intriguing to note that, by staining the tumor samples generated from liver cancer cells with forced expression of Lin28, we observed a significant increase in lipid accumulation, thus prompting us to hypothesize that Lin28 may regulate lipid metabolism during cancer progression. Here, we provide ample evidence to reveal that Lin28A and Lin28B promote de novo fatty acid synthesis in cancer cells. We observed that both Lin28A and Lin28B bind to mRNAs of SREBP‐1 and SCAP to enhance the translation and maturation of SREBP‐1, a master lipid synthesis regulator that increases multiple triglyceride species and fatty acids levels and promotes the conversion of saturated fatty acids to unsaturated ones. Furthermore, lack of Lin28 induces ER stress via de novo lipogenic disorders and, importantly, the dysfunction of Lin28 as an RBP abrogates the lipid accumulation and cancer progression. Collectively, our results establish that Lin28 enhances de novo fatty acid synthesis and cancer progression via a previously unappreciated mechanism of SREBP‐1 regulation.

Results

Lin28A/B enhance lipid accumulation in cancer cells

Our previous studies have demonstrated that the Lin28/let‐7 axis facilitates the Warburg effect to promote cancer progression 7. However, little is known about Lin28 in regulation of lipid metabolism in cancer cells. It is intriguing to note that, by staining the mouse tumor samples generated from PLC or Hep3B cells overexpressing Lin28A/B with oil red O, we observed a significant increase in lipid accumulation (Fig 1A). On the other hand, suppression of Lin28A or Lin28B by shRNAs decreased lipid accumulation in mouse xenograft derived from Hep3B cells (Appendix Fig S1A), suggesting that Lin28A and Lin28B are involved in the regulation of lipid metabolism during liver cancer progression. Consistently, in cultured PLC cells, forced expression of Lin28A or Lin28B led to increased cellular lipid accumulation as well as elevated cellular triglyceride (TG) levels (Fig 1B and C). Meanwhile, knockdown of Lin28A or Lin28B suppressed cellular lipid accumulation as well as cellular TG levels in PLC cells (Fig 1D and E). Similar results were observed in HepG2 and Hep3B cells (Appendix Fig S1B), further confirming that Lin28A and Lin28B regulate lipid metabolism in liver cancer cells.

Figure 1. Lin28A/B enhance lipid accumulation in cancer cells.

Figure 1

Open in a new tab
  1. Neutral lipids were measured in the mouse tumor samples generated from PLC or Hep3B cells overexpressing Lin28A or Lin28B by oil red O staining. Scale bars, 100 μm. Bar graphs depicted the results from the analysis of image of oil red O staining by Photoshop. Data were presented as mean ± SD. Data are representative of five independent experiments. *P < 0.05 as compared to EV group; Student's t‐test.
  2. Cellular neutral lipids were measured in PLC cells overexpressing Lin28A or Lin28B by Nile red staining. Scale bars, 50 μm.
  3. Cellular TG was measured in PLC cells overexpressing Lin28A or Lin28B by Biochemical Triglyceride Determination Kit. The values were normalized to cellular protein. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to EV group; Student's t‐test.
  4. Cellular neutral lipids were measured in PLC cells expressing NTC, shLin28A, or shLin28B by Nile red staining. Scale bars, 50 μm.
  5. Cellular TG was measured in PLC cells expressing NTC, shLin28A, or shLin28B by Biochemical Triglyceride Determination Kit. The values were normalized to cellular protein. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to NTC group; Student's t‐test.
  6. Lipids in PLC cells expressing NTC, shLin28A, or shLin28B were measured by LC‐MS. The heatmap highlighted relative changes for each type of lipid metabolite (left panel) and detailed type of TG (right panel) with their individual average ion counts normalized to NTC. Data are representative of four independent experiments. Lipid species abbreviations are available in Appendix Table S3.
  7. Total cellular fatty acids were determined from PLC cells expressing NTC, shLin28A, or shLin28B by GC‐MS. Average ion counts of metabolites were normalized to NTC. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding NTC group; Student's t‐test.

Next, liquid chromatography–mass spectrometry (LC‐MS) was performed to detect the changes in global lipid species focusing mainly on glycerides and phosphatides. LC‐MS data revealed that suppression of Lin28A/B markedly decreased multiple types of lipids such as triglyceride (TG), phosphatidylethanolamine (PE), sphingomyelin (SM), lysophosphatidylcholine (LPC), phosphatidylcholine (PC), and phosphatidylinositol (PI). Among them, TG species exhibited the most extensive decrease (Fig 1F, Appendix Fig S1C and D, and Appendix Tables S1 and S2), which was consistent with the cellular TG measurement data by colorimetric assay (Fig 1C and E). Further analysis by gas chromatography–mass spectrometry (GC‐MS) showed that cellular long‐chain fatty acids, such as palmitic acid (C16:0), palmitoleic acid (C16:1), stearic acid (C18:0), and oleic acid (C18:1), were markedly reduced by shLin28A or shLin28B in PLC cells (Fig 1G), indicating that Lin28A and Lin28B promote lipid accumulation probably by regulating fatty acid metabolism.

Lin28A/B promote de novo fatty acid synthesis via SREBP‐1

To explore how Lin28A and Lin28B regulate fatty acid metabolism, we detected the major enzymes involved in de novo fatty acid synthesis, fatty acid β‐oxidation, fatty acid uptake, and glucose metabolism. Interestingly, Western blot analysis revealed that de novo fatty acid synthetic enzymes FASN, SCD1, ACC1, and ACLY were markedly increased by Lin28A/B in PLC cells (Fig 2A). Consistently, suppression of Lin28A or Lin28B by shRNAs reduced the protein levels of the fatty acid synthetic enzymes in PLC cells (Fig 2B). To explore whether de novo fatty acid synthesis contributes to the change in cellular fatty acids induced by Lin28A or Lin28B, we carried out metabolic flux analysis using U‐13C2 acetate or U‐13C6 glucose. As a result, 13C‐incorporated palmitic acid (C16:0), palmitoleic acid (C16:1), stearic acid (C18:0), and oleic acid (C18:1) derived from U‐13C2 acetate or U‐13C6 glucose carbons were significantly decreased in PLC cells expressing shLin28A or shLin28B (Fig 2C and Appendix Fig S2A–C), demonstrating that Lin28A/B enhance de novo fatty acid synthesis.

Figure 2. Lin28A/B promote de novo fatty acid synthesis via SREBP‐1.

Figure 2

Open in a new tab
  • A
    Protein levels of metabolic enzymes were determined by Western blot in PLC cells overexpressing Lin28A or Lin28B. β‐Actin served as loading control.
  • B
    Protein levels of metabolic enzymes in de novo fatty acid synthesis were determined by Western blot in PLC cells expressing NTC, shLin28A, or shLin28B. β‐Actin served as loading control.
  • C
    The proportion of labeled fatty acids were determined in PLC cells expressing NTC, shLin28A, or shLin28B incubated with U‐13C2 acetate by GC‐MS. Average ion counts of metabolites were normalized to NTC. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding NTC group; Student's t‐test.
  • D
    Protein levels of transcription factor in de novo fatty acid synthesis and SCAP were determined by Western blot in PLC cells overexpressing Lin28A or Lin28B. β‐Actin served as loading control.
  • E
    Protein levels of SREBP‐1 and SCAP were determined by Western blot in PLC cells expressing NTC, shLin28A, or shLin28B. β‐Actin served as loading control.
  • F
    mRNA levels of SREBP‐1, SCAP, and metabolic enzymes were determined by qRT–PCR in PLC cells overexpressing Lin28A or Lin28B. ACSL6 served as negative control. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding EV group; Student's t‐test.
  • G, H
    PLC cells overexpressing Lin28A or Lin28B were further infected with viruses expressing NTC, shSREBP‐1 (G), or shSCAP (H). Protein levels of SREBP‐1, SCAP, and metabolic enzymes in de novo fatty acid synthesis were determined by Western blot. β‐Actin served as loading control.
  • I
    PLC cells overexpressing Lin28A or Lin28B were further infected with viruses expressing NTC, shSREBP‐1, or shSCAP. Cellular TG was measured. The values were normalized to cellular protein. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared between indicated groups. NS, not significant; Student's t‐test.
  • J
    PLC cells overexpressing Lin28A or Lin28B were further infected with viruses expressing NTC or shSREBP‐1. Cellular neutral lipids were measured by Nile red staining. Scale bars, 50 μm.

Source data are available online for this figure.

To address how Lin28A and Lin28B enhance the expression of fatty acid synthetic enzymes, we detected the master de novo fatty acid synthesis regulators including sterol regulatory element‐binding transcription factor 1 (SREBP‐1), SREBP cleavage‐activating protein (SCAP, which is the direct cleavage factor for SREBP family), carbohydrate‐responsive element‐binding protein (ChREBP), and liver X receptor alpha (LXRα) 26, 27, 30, 31. Interestingly, overexpression of Lin28A or Lin28B resulted in a remarkable increase in SREBP‐1 precursors, mature SREBP‐1, and SCAP protein in PLC cells (Fig 2D). Consistently, suppression of Lin28A or Lin28B by shRNAs reduced the protein levels of SREBP‐1 precursors, mature SREBP‐1, and SCAP in PLC cells (Fig 2E). Similar results were observed in cultured HepG2 and Hep3B cells as well as in the xenograft tumors derived from Hep3B cells (Appendix Fig S2D–G). Furthermore, qRT–PCR experiments showed that Lin28A/B had no effect on SREBP‐1 or SCAP mRNA levels, but markedly increased mRNA levels of de novo fatty acid synthetic enzymes which are the downstream targets of SREBP‐1 26 (Fig 2F), indicating that SREBP‐1 and SCAP are regulated by Lin28A/B at the post‐transcriptional level.

Consistent with a previous report that SCAP escorts the SREBP‐1 precursors to the Golgi where SREBP‐1 precursors are converted to mature SREBP‐1 by a two‐step cleavage 27, we also observed that knockdown of SCAP decreased the mature SREBP‐1 levels in PLC cells (Appendix Fig S2G). Furthermore, our data confirmed that suppression of SREBP‐1 or SCAP by shRNAs markedly inhibited fatty acid synthetic enzyme expression in PLC cells (Appendix Fig S2H and I). More importantly, knockdown of SREBP‐1 or SCAP markedly attenuated Lin28A/B‐induced increase in the protein levels of fatty acid synthetic enzymes in PLC cells (Fig 2G and H). Consistently, knockdown of SREBP‐1 or SCAP significantly abolished the promoting effect of Lin28A/B on the accumulation of cellular neutral lipids as well as TG levels (Fig 2I and J, and Appendix Fig S2J). Collectively, these data demonstrate that both SREBP‐1 and SCAP are critical for Lin28A/B‐mediated fatty acid synthetic enzyme expression as well as the subsequent lipid metabolism.

Lin28A/B bind to SREBP‐1 and SCAP mRNA to enhance SREBP‐1 translation and maturation

To further elucidate how Lin28A and Lin28B increase SREBP‐1 expression, we first established a stable PLC cell line expressing shDGCR8 (Appendix Fig S3A), which lacks the expression of let‐7 mature microRNAs (Appendix Fig S3B) 32. Western blot analysis revealed that Lin28A/B still significantly enhance SREBP‐1, SCAP, and fatty acid synthetic enzyme expression in these stable cells, which was not affected remarkably by restoring the expression of let‐7 isoforms (Appendix Fig S3C), suggesting that Lin28A/B regulates SREBP‐1 and fatty acid synthesis through let‐7‐independent mechanisms. Since Lin28A and Lin28B have been reported to function as RNA binding proteins 13, we next performed an RNA immunoprecipitation experiment. Interestingly, our data showed that both endogenous Lin28A and Lin28B bind to SREBP‐1 as well as SCAP mRNA (Fig 3A). Bioinformatic analysis predicted two potential GGAGA binding motifs of Lin28A and Lin28B in the 3′UTR region of SREBP‐1 mRNA (Fig 3B). A luciferase assay using the reporter genes containing SREBP‐1 wild‐type 3′UTR or its 3′UTR with deletion of those two potential binding motifs demonstrated that those two GGAGA motifs in SREBP‐1 mRNA are important for its association with Lin28A/B (Fig 3C). A 35S pulse labeling experiment further proved that Lin28A and Lin28B enhance SREBP‐1 protein synthesis in PLC cells (Fig 3D and Appendix Fig S3D). Taken together, these data document that Lin28A/B bind to SREBP‐1 mRNA to facilitate SREBP‐1 protein translation.

Figure 3. Lin28A/B bind to SREBP‐1/SCAP mRNA to enhance SREBP‐1 translation and maturation.

Figure 3

Open in a new tab
  1. Binding of endogenous SREBP‐1 and SCAP mRNA by endogenous Lin28A and Lin28B was determined by RNA immunoprecipitation in PLC cell lines. ACSL6 served as negative control. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding IgG group; Student's t‐test.
  2. A diagram showing the predicted Lin28 GGAGA motif in the 3′UTR of SREBP‐1.
  3. Luciferase assays were performed to identify Lin28 binding sites in SREBP‐1 mRNA. Wild‐type or Δ3′UTR of SREBP‐1 sequence was inserted into dual‐luciferase vector. Lin28A and Lin28B were co‐transfected with pSI‐SREBP‐1‐3′UTR or pSI‐SREBP‐1‐3′UTR‐Δ1 or Δ2 into HEK293 cells followed by dual‐luciferase analysis. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding EV group; Student's t‐test.
  4. A 35S pulse labeling experiment was performed to test SREBP‐1 protein synthesis in PLC cells expressing NTC or shLin28A. PLC cells expressing NTC or shLin28A were incubated with medium containing 100 μCi/ml 35S methionine and cysteine for 1 h, followed by immunoprecipitation with SREBP‐1 antibody and autoradiography.
  5. CSD (M2) and CCHC (M1) RNA binding domain in Lin28A and Lin28B were mutated as indicated.
  6. Protein levels of SREBP‐1, SCAP, and metabolic enzymes in de novo fatty acid synthesis were determined by Western blot in PLC cells overexpressing Lin28A or Lin28B wild‐type (WT), CCHC mutant (M1), CSD mutant (M2), or double mutant (M1 + M2). β‐Actin served as loading control.
  7. Binding of endogenous SREBP‐1's mRNA by flag‐Lin28A or Lin28B wild‐type (WT), CCHC mutant (M1), CSD mutant (M2), and double mutant (M1 + M2) was determined by RNA immunoprecipitation in PLC cell lines. ACSL6 served as negative control. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding EV group; Student's t‐test.
  8. Luciferase assays were performed to identify which RNA binding domain of Lin28 bound to SREBP‐1 mRNA. Lin28A or Lin28B wild‐type (WT), CCHC mutant (M1), CSD mutant (M2), and double mutant (M1 + M2) were co‐transfected with pSI‐SREBP‐1‐3′UTR into HEK293 cells followed by dual‐luciferase analysis. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding EV group; Student's t‐test.
  9. Cellular TG was measured in PLC cells overexpressing Lin28A or Lin28B wild‐type (WT), CCHC mutant (M1), CSD mutant (M2), or double mutant (M1 + M2) by Biochemical Triglyceride Determination Kit. The values were normalized to cellular protein. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to EV group; Student's t‐test.
  10. Cellular neutral lipids were measured in PLC cells overexpressing Lin28A or Lin28B wild‐type (WT), CCHC mutant (M1), CSD mutant (M2), or double mutant (M1 + M2) by Nile red staining. Scale bars, 50 μm.

Source data are available online for this figure.

It has been reported that cys‐cys‐his‐cys zinc fingers (CCHC) and the cold‐shock domain (CSD) in Lin28A/B are important for their RNA binding activity 33. We generated different Lin28A/B expressing vectors, replacing three conserved surface aromatic residues in the CSD domain with alanines (M1) or two conserved histidines in CCHC domain with alanines (M2), or replacing all those residues in both domains (M1 + M2; Fig 3E). Western blot analysis showed that CCHC mutant (M1), CSD domain mutant (M2), and mutant of both domains (M1 + M2) in Lin28A/B lost the ability to enhance SREBP‐1 and SCAP protein expression (Fig 3F). Consistently, compared to wild‐type Lin28A/B, those mutants exhibited significantly reduced binding with SREBP‐1 or SCAP mRNA (Fig 3G) and markedly decreased luciferase activity using reporter gene containing the 3′UTR of SREBP‐1 (Fig 3H), suggesting that both CCHC and CSD domains in Lin28A/B are important for their binding to SREBP‐1 or SCAP mRNA to enhance their protein translation.

TG measurement and Nile red staining also revealed that mutation in CCHC domain (M1), CSD domain (M2), or both (M1 + M2) in Lin28A/B failed to increase the cellular TG levels and neutral lipid accumulation (Fig 3I and J), suggesting that both CCHC domain and CSD domain of Lin28A/B are necessary for their regulation on lipid metabolism. Collectively, these data documented that Lin28A/B bind to SREBP‐1 and SCAP mRNA to enhance their protein translation, SCAP‐mediated SREBP‐1 maturation, and subsequent fatty acid synthesis.

Lin28A/B protect cancer cells from saturated fatty acid‐induced ER stress

Although both unsaturated and saturated fatty acids are regulated by Lin28A/B (Figs 1G and 2C), cell growth data showed that replenishing monounsaturated fatty acid oleic acid (OA), but not saturated fatty acid stearic acid (SA), partially rescued shLin28A/B‐suppressed cell proliferation (Fig 4A). Through further analysis of the GC‐MS data as shown in Fig 1G, we found that knockdown of Lin28A/B significantly elevated the ratio of C16:0 to C16:1 and C18:0 to C18:1 in PLC cells (Fig 4B). More importantly, Raman spectrometry analysis of individual lipid droplets showed that knockdown of Lin28A/B markedly increased the ratio of saturated fatty acids (SFA) to monounsaturated fatty acids (MUFAs; Fig 4C and Appendix Fig S4A). Thus, our data showed that suppression of Lin28A/B led to disrupted long‐chain fatty acid homeostasis.

Figure 4. Lin28A/B protect cancer cells from saturated fatty acid‐induced ER stress.

Figure 4

Open in a new tab
  1. Cell growth curve was determined in PLC cells expressing NTC, shLin28A, or shLin28B in the presence or absence of OA or SA (100 μM). Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared between indicated groups. NS, not significant; Student's t‐test.
  2. Ratio between saturated fatty acids and monounsaturated fatty acids in PLC cells expressing NTC, shLin28A, or shLin28B was determined by GC‐MS. Average ion counts of metabolites were normalized to NTC. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding NTC group; Student's t‐test.
  3. Droplet‐level ratios of SFA to MUFA in PLC cells expressing NTC, shLin28A, or shLin28B were determined by Raman spectroscopy of individual lipid droplets. Each dot stood for a single lipid droplet, while the solid line represented the mean value among the population. Dark and light shaded regions corresponded to 1 and 2 standard deviations from the mean. Data are representative of 25 independent experiments. *P < 0.05 as compared to NTC group; Student's t‐test.
  4. mRNA splicing of XBP‐1 was determined by semi‐quantitative reverse transcription–PCR in PLC cells expressing NTC, shLin28A, or shLin28B. 18S rRNA served as loading control. Protein levels of ATF6 and CHOP were determined by Western blot in PLC cells expressing NTC, shLin28A, or shLin28B. β‐Actin served as loading control.
  5. PLC cells expressing NTC, shLin28A, or shLin28B were cultured in the presence or absence of OA (100 μM). mRNA splicing of XBP‐1 was determined by semi‐quantitative reverse transcription–PCR. 18S rRNA served as loading control. Protein levels of ATF6 and CHOP were determined by Western blot. β‐Actin served as loading control.
  6. PLC cells with Lin28A or Lin28B knockdown were further infected with viruses expressing SREBP‐1. mRNA splicing of XBP‐1 was determined by semi‐quantitative reverse transcription–PCR. 18S rRNA served as loading control. Protein levels of ATF6 and CHOP were determined by Western blot. β‐Actin served as loading control.
  7. PLC cells with Lin28A or Lin28B knockdown were further infected with viruses expressing SCD1. mRNA splicing of XBP‐1 was determined by semi‐quantitative reverse transcription–PCR. 18S rRNA served as loading control. Protein levels of ATF6 and CHOP were determined by Western blot. β‐Actin served as loading control.
  8. Cell growth curves of PLC cells expressing NTC, shLin28A, or shLin28B in the presence or absence of 0.5 mM 4‐PBA were determined by trypan blue counting. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding NTC group; Student's t‐test.

Source data are available online for this figure.

Deficiency in desaturation which alters the ratio of saturated fatty acids (SFA) to monounsaturated fatty acids (MUFAs) leads to cellular lipotoxicity and endoplasmic reticulum stress (ER stress) 34, 35, 36, 37, 38. Our data revealed that knockdown of Lin28A or Lin28B resulted in splicing of XBP‐1 mRNA, activated ATF6 and CHOP expression, and increased ROS in PLC cells (Fig 4D, and Appendix Fig S4B and C), indicating that Lin28A/B protect cells from ER stress. More interestingly, supplementation of exogenous oleic acid (OA), but not stearic acid (SA), attenuated shLin28A/B‐induced XBP‐1 mRNA splicing, ATF6 activation, CHOP expression, and ROS production (Fig 4E and Appendix Fig S4D–F), suggesting that disrupted long‐chain fatty acid homeostasis is involved in shLin28A/B‐induced ER stress. Since SCD1 is the key enzyme in catalyzing fatty acid desaturation, our data also confirmed that SCD1 is a downstream target of SREBP‐1. Next, we manipulated SREBP‐1 and SCD1 expression and found that forced expression of SREBP‐1 or SCD1 alleviated shLin28A/B‐induced ER stress (Fig 4F and G), suggesting that SREBP‐1‐activated SCD1 expression is involved in Lin28A/B‐regulated ER stress. Moreover, blocking ER stress by 4‐Phenylbutyric acid (4‐PBA) partially rescued the impaired cell growth by shLin28A or shLin28B (Fig 4H). Taken together, these data suggested that Lin28A and Lin28B protect cancer cells from saturated fatty acid‐induced ER stress by activating SREBP‐1 and its downstream SCD1 expression.

SREBP‐1 is critical for Lin28A/B‐mediated tumor growth

Next, we studied the role of SREBP‐1 regulated de novo fatty acid synthesis in tumor growth. Cell proliferation analysis revealed that knockdown of SREBP‐1 diminished the promoting effect of Lin28A and Lin28B on cell proliferation (Fig 5A). Meanwhile, treatment with fatostatin, the pharmacologic inhibitor of SREBP activity, also attenuated Lin28A/B‐enhanced cell proliferation in PLC cells (Appendix Fig S5A). Mouse xenograft experiments showed that overexpression of Lin28A/B enhanced tumor growth, which was abolished by knocking down of SREBP‐1 (Fig 5B and C, and Appendix Fig S5B). Oil red O staining demonstrated that the tumor tissues derived from Lin28A/B‐overexpressing cells exhibited increased lipid accumulation, which was blocked by shSREBP‐1 (Fig 5D). Consistently, knockdown of SREBP‐1 attenuated Lin28A/B‐induced increase in de novo fatty acid synthetic enzyme expression and strengthened Lin28A/B‐reduced ER stress in mouse xenograft tumor tissues (Fig 5E). Collectively, these data suggest that SREBP‐1 regulated fatty acid synthesis is critical for Lin28A/B‐enhanced tumor growth.

Figure 5. SREBP‐1 is critical for Lin28A/B‐mediated tumor growth.

Figure 5

Open in a new tab
  • A
    PLC cells overexpressing Lin28A or Lin28B were further infected with viruses expressing NTC or shSREBP‐1. Cell growth curves were determined by trypan blue counting. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to corresponding EV group; Student's t‐test.
  • B–E
    PLC cells overexpressing GFP‐Lin28A or GFP‐Lin28B were further infected with viruses expressing NTC or shSREBP‐1. Equal numbers of cells were injected subcutaneously into the flanks of BALB/c nude mice (n = 5 in each group). Tumor sizes were measured every 3 days using caliper (B). Photographs showed xenografts at the end of the experiment (C). Neutral lipids were measured in these tissues by oil red O staining (D). Protein levels of Lin28A, Lin28B, SREBP‐1, metabolic enzymes, ATF6, and CHOP were determined by Western blot (E). β‐Actin served as loading control. Data were presented as mean ± SE. *P < 0.05 as compared between indicated groups. NS, not significant; Student's t‐test. Scale bars, 100 μm.
  • F
    Cell growth curves of PLC cells overexpressing Lin28A or Lin28B wild‐type (WT), CCHC mutant (M1), CSD mutant (M2), or double mutant (M1 + M2) were determined by trypan blue counting. Data were presented as mean ± SD. Data are representative of three independent experiments. *P < 0.05 as compared to EV group; Student's t‐test.
  • G–J
    Equal numbers of PLC cells overexpressing GFP‐Lin28A, GFP‐Lin28B, or their double mutants (M1 + M2) were injected subcutaneously into the flanks of BALB/c nude mice (n = 5 in each group). Tumor sizes were measured every 3 days using caliper (G). Photographs showed xenografts at the end of the experiment (H). Neutral lipids were measured in these tissues by oil red O staining (I). Protein levels of Lin28A, Lin28B, SREBP‐1, SCAP, metabolic enzymes, ATF6, and CHOP were determined by Western blot (J). β‐Actin served as loading control. Data were presented as mean ± SE. *P < 0.05 as compared between indicated groups. NS, not significant; Student's t‐test. Scale bars, 100 μm.
  • K
    Protein level of Lin28A, Lin28B, SREBP‐1 and SCAP were determined by Western blot in HCC samples, normal (N) and tumor (T) tissues from the same patient were compared. Ponceau staining served as loading control in Appendix Fig S5D.

Source data are available online for this figure.

More importantly, cell proliferation analysis revealed that mutations in the CCHC domain (M1) or CSD domain (M2) or both (M1 + M2) abolished Lin28A/B‐mediated cell proliferation advantage in PLC cells (Fig 5F). Mouse xenograft experiments showed that, compared to Lin28A/B wild‐type, double mutations (M1 + M2) of Lin28A/B abrogated tumor growth in PLC cells (Fig 5G and H, and Appendix Fig S5C). Moreover, tumor tissues derived from PLC cells expressing Lin28A/B double mutations exhibited reduced lipid accumulation and consistently decreased de novo fatty acid synthetic enzyme expression but increased ER stress as compared to that of wild‐type Lin28A/B (Fig 5I and J). Collectively, these data documented that the RNA binding activity is important for Lin28A/B‐regulated lipid metabolism and tumor growth.

More interestingly, in YAP‐5SA‐induced mouse HCC model 39, oil red O staining demonstrated that the tumor tissues exhibited increased lipid accumulation (Appendix Fig S5D). Western blotting also demonstrated that the tumor tissues had higher expression of Lin28A, Lin28B, SREBP‐1, SCAP, and de novo fatty acid synthetic enzymes, and had lower expression of ER stress markers, ATF6 and CHOP (Appendix Fig S5E). Western blotting analysis of 32 pairs of human hepatocellular carcinoma (HCC) lesions and corresponding normal adjacent tissues demonstrated higher expression of Lin28A, Lin28B, SREBP‐1, SCAP, and de novo fatty acid synthetic enzymes in clinical HCC tumor tissues compared to corresponding adjacent normal tissues (Fig 5K, and Appendix Fig S5F and G). Results in these models suggested high correlation among these factors in human hepatocellular carcinoma patients.

Discussion

In this study, we demonstrate that Lin28A and Lin28B promote de novo fatty acid synthesis and cancer progression by enhancing SREBP‐1 expression. Mechanically, we find that Lin28A and Lin28B enhance SREBP‐1 expression at the post‐transcriptional level by binding with SREBP‐1/SCAP mRNAs. Furthermore, we show that suppression of Lin28A or Lin28B decreases monounsaturated fatty acids more significantly than saturated fatty acids and thus causes lipotoxicity and ER stress. Importantly, we also demonstrate that SREBP‐1 is critical for Lin28A‐ and Lin28B‐mediated tumor proliferation both in vitro and in vivo (Fig 6). Thus, we reveal a vital connection between Lin28 and lipid metabolism. Together with our previous results in cancer glucose metabolism 7, our data greatly extended the metabolic role of Lin28 in regulation of homeostasis and diseases.

Figure 6. Lin28A/B promote liver cancer by facilitating de novo fatty acid synthesis via SREBP‐1.

Figure 6

Open in a new tab

In this study, we demonstrate that Lin28A and Lin28B bind to SREBP‐1 and SCAP mRNAs to enhance SREBP‐1 translation and maturation, thereby promoting de novo fatty acid synthesis and conversion of saturated fatty acid to monounsaturated fatty acids. This regulatory mechanism leads to an inhibition of ER stress and cancer progression.

Previously, the Lin28/let‐7 axis was documented to regulate glucose metabolism in a diabetes model 14 and cancer cells 7. Lin28 was also reported to regulate one‐carbon and nucleotide metabolism in stem cells by let‐7‐dependent and let‐7‐independent regulatory mechanisms 16. Here, we discovered that Lin28A and Lin28B enhanced lipid accumulation by regulating SREBP‐1 at the post‐transcriptional level (Fig 2). As a key regulator of fatty acid synthesis, SREBP‐1 activates several enzymes in de novo fatty acid synthesis at transcriptional levels 26. SCAP is a key factor for SREBP‐1 maturation. Under normal conditions, SCAP binds to SREBP‐1 in the ER. When sterol is deficient, it takes SREBP‐1 to the Golgi where SREBP‐1 is processed to its mature form. The mature form of SREBP‐1 is then transported to the nucleus for its transcriptional function 27. In our study, interestingly, Lin28A and Lin28B were found to directly bind to both SREBP‐1 and SCAP mRNA to enhance the translation and maturation of SREBP‐1 (Fig 3). Lin28 regulates SREBP‐1 not only through directly binding to its mRNA at the translational level, but also via SCAP to enhance its protein maturation. This two‐step regulation strengthens the relationship between Lin28 and SREBP‐1, resulting in a robust regulation of fatty acid synthesis by Lin28. Although let‐7 is reported to target FASN and SCD1 40, the pathway we found here is independent of let‐7 (Appendix Fig S3A–C). By regulating SREBP‐1, Lin28 enhances lipid accumulation at multiple levels. Lin28 enhances de novo fatty acid synthesis, thus increases the levels of fatty acids, and ultimately enhances TG and phospholipids (Figs 1 and 2). While we appreciate that the LC‐MS we used in this study is suboptimal with some analytical limitations, the results clearly support our conclusions. In brief, we found Lin28 enhances lipid accumulation post‐transcriptionally via SREBP‐1, which is of great significance for cancer cell proliferation.

It is interesting to note that the decrease in monounsaturated fatty acid was more sensitive to the suppression of Lin28 (Fig 1G). We observed that, when we knocked down Lin28A and Lin28B, monounsaturated fatty acid had a great decrease while saturated fatty acid had a slight decrease, which is mainly caused by suppression of SCD1, a downstream target of SREBP‐1 that facilitates the conversion from saturated to unsaturated fatty acids. Although both saturated and unsaturated fatty acids are components of membrane and storage lipids, they are different functionally 34, 38. It is reported that accumulation of saturated fatty acids attenuated membrane mobility and caused lipotoxicity 35. Although Lin28A repressed ER‐associated protein translation to reduce cell surface receptors and secretory proteins in the secretory pathway in embryonic stem cells 41, interestingly, the relationship among Lin28, lipotoxicity, and ER stress has rarely been considered before. We thus detected markers of ER stress and found that suppression of Lin28A or Lin28B induced ER stress, which further impaired cancer cell proliferation (Fig 4). Importantly, by adding unsaturated fatty acid, the inhibition of cancer cell proliferation caused by shLin28A or shLin28B was significantly rescued (Fig 4A). We thus demonstrate that Lin28 plays a vital role for stress relief by balancing fatty acid saturation, thereby preventing cancer cells from lipotoxicity and ER stress.

Clinical sample analysis indicated that Lin28A, Lin28B, SREBP‐1, and SCAP protein levels were highly correlated in HCC samples (Fig 5K, and Appendix Fig S5F and G). Of note, this regulatory axis of Lin28/SREBP1/SCAP in HCC has never been reported previously. Furthermore, we used a xenograft model to confirm that SREBP‐1 was critical for Lin28‐mediated tumor development in vivo (Fig 5A–E and Appendix Fig S5B). Meanwhile, we confirmed the importance of the RNA binding function of Lin28 in tumor development (Fig 5F–J and Appendix Fig S5C). Since fatty acid blockade therapies are evaluated in clinical trials 42, we can envision a potential therapeutic strategy for HCC or other tumors with aberrant expression of Lin28, SREBP‐1, or SCAP. Based on our results, we propose that targeting de novo fatty acid synthesis and lipid balance could be of potential therapeutic significance for cancers with Lin28 aberrant expression.

Materials and Methods

Cell lines and cell culture

Human PLC, Hep3B, HepG2, HEK293T, and HEK293 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin. All cells were kept in a humidified incubator at 37°C and 5% CO2.

Establishment of stable cells

Lin28A, Lin28B, SREBP‐1, SCD1, and mutants of Lin28A and Lin28B were cloned into pSin‐EF2‐flag‐puro or into pSin‐EF2‐GFP‐puro vectors. shRNA was cloned into plko.1 vectors for cell transfection and lentivirus production. Cells were co‐transfected with plasmids encoding group antigen, polymerase, envelope protein, and vesicular stomatitis virus G protein into HEK293T cells using Lipofectamine 2000 (Invitrogen). Viral supernatant was collected 48 h post‐transfection and filtered (0.45 nm pore size), and PLC cells were infected in the presence of 8 μg/ml Polybrene (Sigma‐Aldrich). The transduced cells were selected by 1.0 μg/ml puromycin. Short hairpin RNAs (shRNAs) targeting Lin28A, Lin28B, SREBP‐1, SCAP, or DGCR8 were commercially purchased (Sigma‐Aldrich). The transduction was performed using the same procedure described above.

RNA isolation, quantitative real‐time PCR, and semi‐quantitative reverse transcription–PCR

Cellular total mRNA was extracted by TRIzol (Invitrogen) followed by DNase (Ambion) treatment and cDNA synthesis using the iScript cDNA Synthesis Kit (Bio‐Rad). cDNA samples were used for quantitative real‐time PCR (qRT–PCR) analysis using SYBR Green Master Mix (Vazyme) on a Bio‐Rad iCycler. For semi‐quantitative PCR assay, equal amounts of RNA for each treatment were subjected to semi‐quantitative reverse transcription–polymerase chain reaction, stained with SYBR Green and detected by Western ECL Substrate (Bio‐Rad). 18S rRNA served as the loading control. All primers were purchased from Invitrogen. Primer sequences used are shown in Appendix Table S4.

Western blot

Total protein was isolated from cells using RIPA buffer (50 mM Tris–HCl, pH 8.0, 150 mM NaCl, 5 mM EDTA, 0.1% SDS, 1% NP‐40) supplemented with cocktail. Protein concentration was measured using the Bradford assay kit. Equal amount of proteins was loaded and separated by SDS–PAGE. Antibodies for Lin28A, Lin28B, HK2, FASN, ACC1, ACLY, CPT1A, MCAD, SCAD, CD36, FABP1, FABP5, FATP2, FATP5, SREBP‐1, SCAP, ChREBP, LXRα, DGCR8, CHOP (Proteintech; 1:1,000); SCD1 (Abcam; 1:1,000); PDK1 (Enzo Life Sciences; 1:1,000); and ATF6 (Santa Cruz Biotechnology; 1:1,000) were used for blotting. β‐Actin (Abmart; 1:5,000) served as loading control. HRP‐conjugated anti‐rabbit and anti‐mouse (Bio‐Rad) secondary antibodies were used. Signal was detected using Western ECL Substrate (Bio‐Rad).

Nile red staining and triglyceride measurement

Lipid determination was performed following the protocol previously reported 43. To visualize lipid droplets, cultured cells were fixed in 4% paraformaldehyde solution on the 12‐well plates, stained with 0.05 μg/ml Nile red (Sigma) for 10 min, washed with PBS twice, then stained with DAPI. The images were visualized by immunofluorescence microscopy. For triglyceride measurement, collected cells were lysed in RIPA buffer with 1% NP‐40 for 30 min, and then, cell lysates were used to measure triglyceride by using a Biochemical Triglyceride Determination Kit (NJJC Bio). The values were normalized to cellular protein.

Total cellular fatty acid composition analysis

Extraction of intracellular metabolites was performed following the protocol reported previously 44. The total lipids were extracted by 1:1:1 methanol:chloroform:water (v:v). The bottom (chloroform) cellular layer was evaporated under nitrogen and heated by 2.5% H2SO4 in methanol for 1 h at 80°C to create fatty acid methyl esters (FAMEs). After two washes with hexane, FAMEs were determined by fractional 13C‐labeled enrichment and fatty acid composition with gas chromatography–mass spectrometry. For determination of total cellular fatty acid composition with stable isotopic 13C‐labeling, the quantification of isotopic 13C‐labeling and the correction of natural isotope abundance were performed following the protocol reported previously 23. PLC cell lines were incubated for 72 h in the presence of 25 mM 13C‐labeled glucose or 10 mM 13C‐labeled acetate before extraction of intracellular metabolites.

RNA immunoprecipitation

For endogenous immunoprecipitation, protein A/G beads were coated with antibody (2 μg) for 2 h at 4°C, mixed with prepared cell lysate, and rotated 4°C overnight. After six washes, RNA bound to Protein A/G beads was extracted with TRIzol (Invitrogen). IgG served as a control. For exogenous immunoprecipitation, cells were lysed, mixed with anti‐flag M2 agarose slurry, and rotated 4°C overnight. After six washes, RNA bound to anti‐flag M2 agarose slurry was competitively eluted with 3×flag peptide, and extracted with TRIzol (Invitrogen). Empty vector (EV) served as a control.

Luciferase assay

SREBP‐1‐3′UTR and SREBP‐1‐3′UTR‐Δ1 and Δ2 were inserted into the pSI‐CHECK‐2 dual‐luciferase reporter vector, designated as SREBP‐1‐3′UTR and pSI‐SREBP‐1‐3′UTR‐Δ1 and Δ2, respectively. HEK293 cells were co‐transfected with either SREBP‐1‐3′UTR or pSI‐SREBP‐1‐3′UTR‐Δ1 or Δ2 along with Lin28A or Lin28B wild‐type (WT), CCHC mutant (M1), CSD mutant (M2), and double mutant (M1 + M2) using Lipofectamine 2000 (Invitrogen) in a 48‐well plate. Luciferase activity was measured 48 h after transfection using the Dual‐Luciferase Reporter Assay System (Promega). Renilla luciferase was normalized to firefly luciferase activity.

Oil red O staining

Oil red O staining was performed following the protocol previously reported 45. Frozen sections from mice xenograft tumors were incubated with ORO working solution at room temperature, and counterstained the sections with Mayer's hematoxylin. The red lipid droplets were visualized by microscopy using 20× magnification. Data image was analyzed and quantified by Photoshop.

Liquid chromatography–mass spectrometry

Lipids were extracted from cell pellets by liquid–liquid extraction with 1.5 ml solvent of dichloromethane (OurChem)/methanol (Optima) mixture (2:1, v/v) following a previously reported protocol 46. Briefly, 1.5 ml of dichloromethane/methanol (2:1, v/v) with two internal standards, LPC (12:0) and PC (11:0/11:0; Birmingham, AL, USA), was added and vortexed for 1 min. Phase separation of aqueous and organic layers was performed by centrifugation at 1,000 g for 15 min at room temperature. After drying via freeze concentration in the centrifugal dryer, the lipid extract was reconstituted in 200 μl isopropanol (Optima)/methanol mixture (1:1, v/v) for lipidomic analyses.

Lipidomic analysis was conducted on an Ultimate‐3000 UPLC coupled to a Q Exactive hybrid quadrupole‐Orbitrap mass spectrometer (Thermo Scientific, Waltham, MA, USA) using a 100 × 2.1 mm hypersil GOLD 1.9‐μm C18 column (Thermo Scientific). The temperature of column oven was set at 45°C. The eluent A (60% acetonitrile and 40% water with 10 mmol/l ammonium formate) and eluent B (10% acetonitrile and 90% isopropanol with 10 mmol/l ammonium formate) were used. The flow rate was 0.35 ml/min, and the injection volume was 10 μl. The gradient was 30–100% B in 14.5 min, 100% B in 14.5–16.5 min, 100–30% B in 16.50–16.51 min, and 30% B in 16.51–20 min.

MS spectra were obtained in the positive and negative ion mode, respectively, with high mass accuracy MS analysis and data‐dependent MS analysis. The MS spray voltages were 3.0 kV and 2.8 kV in the ESI+ and ESI− mode, respectively. The MS was operated at a resolving power of 70,000 in full‐scan mode (automatic gain control target: 1e6) and of 17,500 in the Top10 data‐dependent MS mode [stepped normalized collision energy: 15, 25, and 35 both in positive and in negative ion mode; automatic gain control target: 1e5 (pos) and 2e5 (neg)] with dynamic exclusion setting of 6.0 s. All MS raw data were acquired using the software Xcalibur (version 3.0; Thermo Scientific). Lipid Search (version 4.0; Thermo Scientific) was used for identification and quantification. Lipids were identified according to the exact mass, retention time, and the pattern of precursor ions and MS/MS.

Raman spectrometry

Fixed cells adhered to a quartz coverslip were placed onto the stage of a home‐built Raman microscope (50 mW of 785‐nm laser light coupled to a 1 μm spot size via a Nikon 60X 1.27 NA objective, Raman spectra detected through a Andor Shamrock 500i spectrometer with DV416A‐LDC‐DD detector). Following collection of Raman spectra of individual lipid droplets, spectral processing was performed using scripts developed in‐house using MATLAB (The MathWorks, Natick, MA). The mean preprocessed spectra from treatment and control groups are shown as solid lines in Appendix Fig S4A. Then, the preprocessed spectra were modeled using asymmetric least squares with a penalty parameter of 0.1 and a model composed of albumin (as a protein stand‐in), 24‐hydroxycholesterol, and liquid‐phase oleic fatty acid and glyceryl stearate. The average fits are shown as dashed lines, showing good agreement between modeled and original spectra. Concentrations of monounsaturated fatty acids (MUFAs) and saturated fatty acids (SFA) from individual lipid droplets, determined by spectral modeling, are shown as individual points in Fig 4C. Bars and dark and light shaded regions correspond to mean concentrations, and ±1 and 2 standard deviations in the concentration, respectively.

35S Pulse labeling

Cells were cultured in 6‐cm dishes in medium lacking methionine and cysteine to which 100 μCi/ml EasyTag™ EXPRESS 35S Protein Labeling Mix (NEG772; PerkinElmer) was added for 35S labeling for 1 h. Then, cells were washed with PBS and immunoprecipitated by SREBP‐1 antibody. Proteins were then autoradiographed.

Animal studies

All animal studies were conducted with approval from the Animal Research Ethics Committee of the University of Science and Technology of China. Male BALB/c nude mice were purchased from SJA Laboratory Animal Company of China, which were randomly assigned to experimental groups. Equal numbers of the established PLC stable cells were injected subcutaneously into nude mice. Tumor volumes were measured every 3 days with a caliper and calculated using the equation, volume = width*depth*length*0.52.

Clinical human HCC samples

Snap‐frozen HCC tissues and the corresponding para‐cancerous tissues which were at least 2 cm away from the edge of the tumors were taken from 32 HCC patients by radical HCC resection in Anhui Provincial Hospital (Hefei, China). For using these clinical materials for research purposes, prior written informed consents from the patients and approval from the Institutional Research Ethics Committee of Anhui Provincial Hospital were obtained. Total proteins were extracted from paired tissues and detected by Western blotting.

Statistical analysis

The data were presented as the mean ± SD or mean ± SE of at least three independent experiments. Student's t‐test was used to calculate P values. Statistical significance is displayed as *P < 0.05.

Author contributions

HZ, WJ, and PG conceived and supervised this study. YZ, CL, HZ, and PG designed the experiments. YZ, CL, CH, QW, YC, SX, HL, LW, DH, LS, TL, XH, XZ, JW, and ZJS performed the experiments and analyzed the data. HZ, WJ, PG, YZ, and CL wrote the paper. All the authors read and approved the article.

Conflict of interest

The authors declare that they have no conflict of interest.

Supporting information

Appendix

Click here for additional data file. (1.8MB, pdf)

Source Data for Appendix

Click here for additional data file. (1.2MB, zip)

Review Process File

Click here for additional data file. (2.5MB, pdf)

Source Data for Figure 2

Click here for additional data file. (790.7KB, pdf)

Source Data for Figure 3

Click here for additional data file. (271.4KB, pdf)

Source Data for Figure 4

Click here for additional data file. (374KB, pdf)

Source Data for Figure 5

Click here for additional data file. (391.4KB, pdf)

Acknowledgements

This work is supported in part by National Basic Key Research Program of China (2014CB910600), National Key R&D Program of China (2018YFA0107103, 2017YFA0205600), National Natural Science Foundation of China (81525022, 31571472, 81530076, 81821001), The Chinese Academy of Sciences (XDPB10), and the Program of Development Foundation of Hefei Center for Physical Science and Technology (2017FXZY004). ZJS gratefully acknowledges support from the 1000 Young Talents Global Recruitment Program.

EMBO Reports (2019) 20: e48115

Contributor Information

Weidong Jia, Email: jwd1968@sina.com.

Huafeng Zhang, Email: hzhang22@ustc.edu.cn.

References

  • 1. Viswanathan SR, Powers JT, Einhorn W, Hoshida Y, Ng TL, Toffanin S, O'Sullivan M, Lu J, Phillips LA, Lockhart VL et al (2009) Lin28 promotes transformation and is associated with advanced human malignancies. Nat Genet 41: 843–848 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Powers JT, Tsanov KM, Pearson DS, Roels F, Spina CS, Ebright R, Seligson M, De Soysa Y, Cahan P, Theißen J (2016) Multiple mechanisms disrupt the let‐7 microRNA family in neuroblastoma. Nature 535: 246 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Molenaar JJ, Domingo‐Fernandez R, Ebus ME, Lindner S, Koster J, Drabek K, Mestdagh P, van Sluis P, Valentijn LJ, van Nes J et al (2012) LIN28B induces neuroblastoma and enhances MYCN levels via let‐7 suppression. Nat Genet 44: 1199–1206 [DOI] [PubMed] [Google Scholar]
  • 4. Nguyen LH, Robinton DA, Seligson MT, Wu L, Li L, Rakheja D, Comerford SA, Ramezani S, Sun X, Parikh MS et al (2014) Lin28b is sufficient to drive liver cancer and necessary for its maintenance in murine models. Cancer Cell 26: 248–261 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. King CE, Cuatrecasas M, Castells A, Sepulveda AR, Lee JS, Rustgi AK (2011) LIN28B promotes colon cancer progression and metastasis. Can Res 71: 4260–4268 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Urbach A, Yermalovich A, Zhang J, Spina CS, Zhu H, Perez‐Atayde AR, Shukrun R, Charlton J, Sebire N, Mifsud W et al (2014) Lin28 sustains early renal progenitors and induces Wilms tumor. Genes Dev 28: 971–982 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Ma X, Li C, Sun L, Huang D, Li T, He X, Wu G, Yang Z, Zhong X, Song L et al (2014) Lin28/let‐7 axis regulates aerobic glycolysis and cancer progression via PDK1. Nat Commun 5: 5212 [DOI] [PubMed] [Google Scholar]
  • 8. Tu H‐C, Schwitalla S, Qian Z, LaPier GS, Yermalovich A, Ku Y‐C, Chen S‐C, Viswanathan SR, Zhu H, Nishihara R (2015) LIN28 cooperates with WNT signaling to drive invasive intestinal and colorectal adenocarcinoma in mice and humans. Genes Dev 29: 1074–1086 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Madison BB, Liu Q, Zhong X, Hahn CM, Lin N, Emmett MJ, Stanger BZ, Lee J‐S, Rustgi AK (2013) LIN28B promotes growth and tumorigenesis of the intestinal epithelium via Let‐7. Genes Dev 27: 2233–2245 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Huang Y (2012) A mirror of two faces: Lin28 as a master regulator of both miRNA and mRNA. Wiley Interdiscip Rev RNA 3: 483–494 [DOI] [PubMed] [Google Scholar]
  • 11. Wilbert ML, Huelga SC, Kapeli K, Stark TJ, Liang TY, Chen SX, Yan BY, Nathanson JL, Hutt KR, Lovci MT et al (2012) LIN28 binds messenger RNAs at GGAGA motifs and regulates splicing factor abundance. Mol Cell 48: 195–206 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Li N, Zhong X, Lin X, Guo J, Zou L, Tanyi JL, Shao Z, Liang S, Wang LP, Hwang WT et al (2012) Lin‐28 homologue A (LIN28A) promotes cell cycle progression via regulation of cyclin‐dependent kinase 2 (CDK2), cyclin D1 (CCND1), and cell division cycle 25 homolog A (CDC25A) expression in cancer. J Biol Chem 287: 17386–17397 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Peng S, Chen LL, Lei XX, Yang L, Lin H, Carmichael GG, Huang Y (2011) Genome‐wide studies reveal that Lin28 enhances the translation of genes important for growth and survival of human embryonic stem cells. Stem Cells 29: 496–504 [DOI] [PubMed] [Google Scholar]
  • 14. Zhu H, Shyh‐Chang N, Segre AV, Shinoda G, Shah SP, Einhorn WS, Takeuchi A, Engreitz JM, Hagan JP, Kharas MG et al (2011) The Lin28/let‐7 axis regulates glucose metabolism. Cell 147: 81–94 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Shyh‐Chang N, Zhu H, Yvanka de Soysa T, Shinoda G, Seligson MT, Tsanov KM, Nguyen L, Asara JM, Cantley LC, Daley GQ (2013) Lin28 enhances tissue repair by reprogramming cellular metabolism. Cell 155: 778–792 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Zhang J, Ratanasirintrawoot S, Chandrasekaran S, Wu Z, Ficarro SB, Yu C, Ross CA, Cacchiarelli D, Xia Q, Seligson M et al (2016) LIN28 regulates stem cell metabolism and conversion to primed pluripotency. Cell Stem Cell 19: 66–80 [DOI] [PubMed] [Google Scholar]
  • 17. Ward PS, Thompson CB (2012) Metabolic reprogramming: a cancer hallmark even warburg did not anticipate. Cancer Cell 21: 297–308 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Sun L, Song L, Wan Q, Wu G, Li X, Wang Y, Wang J, Liu Z, Zhong X, He X et al (2015) cMyc‐mediated activation of serine biosynthesis pathway is critical for cancer progression under nutrient deprivation conditions. Cell Res 25: 429–444 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Ma X, Wang L, Huang D, Li Y, Yang D, Li T, Li F, Sun L, Wei H, He K et al (2017) Polo‐like kinase 1 coordinates biosynthesis during cell cycle progression by directly activating pentose phosphate pathway. Nat Commun 8: 1506 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Currie E, Schulze A, Zechner R, Walther TC, Farese RV Jr (2013) Cellular fatty acid metabolism and cancer. Cell Metab 18: 153–161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Baenke F, Peck B, Miess H, Schulze A (2013) Hooked on fat: the role of lipid synthesis in cancer metabolism and tumour development. Dis Models Mech 6: 1353–1363 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Mullen AR, Wheaton WW, Jin ES, Chen PH, Sullivan LB, Cheng T, Yang Y, Linehan WM, Chandel NS, DeBerardinis RJ (2011) Reductive carboxylation supports growth in tumour cells with defective mitochondria. Nature 481: 385–388 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Metallo CM, Gameiro PA, Bell EL, Mattaini KR, Yang J, Hiller K, Jewell CM, Johnson ZR, Irvine DJ, Guarente L et al (2011) Reductive glutamine metabolism by IDH1 mediates lipogenesis under hypoxia. Nature 481: 380–384 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Zaidi N, Swinnen JV, Smans K (2012) ATP‐citrate lyase: a key player in cancer metabolism. Can Res 72: 3709–3714 [DOI] [PubMed] [Google Scholar]
  • 25. Comerford SA, Huang Z, Du X, Wang Y, Cai L, Witkiewicz AK, Walters H, Tantawy MN, Fu A, Manning HC et al (2014) Acetate dependence of tumors. Cell 159: 1591–1602 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Horton JD, Goldstein JL, Brown MS (2002) SREBPs: activators of the complete program of cholesterol and fatty acid synthesis in the liver. J Clin Invest 109: 1125–1131 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Soyal SM, Nofziger C, Dossena S, Paulmichl M, Patsch W (2015) Targeting SREBPs for treatment of the metabolic syndrome. Trends Pharmacol Sci 36: 406–416 [DOI] [PubMed] [Google Scholar]
  • 28. Williams KJ, Argus JP, Zhu Y, Wilks MQ, Marbois BN, York AG, Kidani Y, Pourzia AL, Akhavan D, Lisiero DN et al (2013) An essential requirement for the SCAP/SREBP signaling axis to protect cancer cells from lipotoxicity. Can Res 73: 2850–2862 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Griffiths B, Lewis CA, Bensaad K, Ros S, Zhang Q, Ferber EC, Konisti S, Peck B, Miess H, East P et al (2013) Sterol regulatory element binding protein‐dependent regulation of lipid synthesis supports cell survival and tumor growth. Cancer Metab 1: 3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Uyeda K, Repa JJ (2006) Carbohydrate response element binding protein, ChREBP, a transcription factor coupling hepatic glucose utilization and lipid synthesis. Cell Metab 4: 107–110 [DOI] [PubMed] [Google Scholar]
  • 31. Lin C‐Y, Gustafsson J‐Å (2015) Targeting liver X receptors in cancer therapeutics. Nat Rev Cancer 15: 216–224 [DOI] [PubMed] [Google Scholar]
  • 32. Han J, Lee Y, Yeom KH, Kim YK, Jin H, Kim VN (2004) The Drosha‐DGCR32 complex in primary microRNA processing. Genes Dev 18: 3016–3027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Balzer E, Moss EG (2014) Localization of the developmental timing regulator Lin28 to mRNP complexes, P‐bodies and stress granules. RNA Biol 4: 16–25 [DOI] [PubMed] [Google Scholar]
  • 34. Alkhouri N, Dixon LJ, Feldstein AE (2009) Lipotoxicity in nonalcoholic fatty liver disease: not all lipids are created equal. Expert Rev Gastroenterol Hepatol 3: 445–451 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Borradaile NM, Han X, Harp JD, Gale SE, Ory DS, Schaffer JE (2006) Disruption of endoplasmic reticulum structure and integrity in lipotoxic cell death. J Lipid Res 47: 2726–2737 [DOI] [PubMed] [Google Scholar]
  • 36. Pineau L, Colas J, Dupont S, Beney L, Fleurat‐Lessard P, Berjeaud JM, Berges T, Ferreira T (2009) Lipid‐induced ER stress: synergistic effects of sterols and saturated fatty acids. Traffic 10: 673–690 [DOI] [PubMed] [Google Scholar]
  • 37. Listenberger LL, Han X, Lewis SE, Cases S, Farese RV Jr, Ory DS, Schaffer JE (2003) Triglyceride accumulation protects against fatty acid‐induced lipotoxicity. Proc Natl Acad Sci USA 100: 3077–3082 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Karaskov E, Scott C, Zhang L, Teodoro T, Ravazzola M, Volchuk A (2006) Chronic palmitate but not oleate exposure induces endoplasmic reticulum stress, which may contribute to INS‐1 pancreatic beta‐cell apoptosis. Endocrinology 147: 3398–3407 [DOI] [PubMed] [Google Scholar]
  • 39. Dai X, Liu H, Shen S, Guo X, Yan H, Ji X, Li L, Huang J, Feng X‐H, Zhao B (2015) YAP activates the Hippo pathway in a negative feedback loop. Cell Res 25: 1175 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Serguienko A, Grad I, Wennerstrøm AB, Meza‐Zepeda LA, Thiede B, Stratford EW, Myklebost O, Munthe E (2015) Metabolic reprogramming of metastatic breast cancer and melanoma by let‐7a microRNA. Oncotarget 6: 2451 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Cho J, Chang H, Kwon SC, Kim B, Kim Y, Choe J, Ha M, Kim YK, Kim VN (2012) LIN28A is a suppressor of ER‐associated translation in embryonic stem cells. Cell 151: 765–777 [DOI] [PubMed] [Google Scholar]
  • 42. Beloribi‐Djefaflia S, Vasseur S, Guillaumond F (2016) Lipid metabolic reprogramming in cancer cells. Oncogenesis 5: e189 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Huang LiT, Li X, Zhang L, Sun L, He X, Zhong X, Jia D, Song L, Semenza GL et al (2014) HIF‐1‐mediated suppression of acyl‐CoA dehydrogenases and fatty acid oxidation is critical for cancer progression. Cell Rep 8: 1930–1942 [DOI] [PubMed] [Google Scholar]
  • 44. Crown SB, Marze N, Antoniewicz MR (2015) Catabolism of branched chain amino acids contributes significantly to synthesis of odd‐chain and even‐chain fatty acids in 3T3‐L1 adipocytes. PLoS One 10: e0145850 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Mehlem A, Hagberg CE, Muhl L, Eriksson U, Falkevall A (2013) Imaging of neutral lipids by oil red O for analyzing the metabolic status in health and disease. Nat Protoc 8: 1149–1154 [DOI] [PubMed] [Google Scholar]
  • 46. Cequier‐Sánchez E, Rodriguez C, Ravelo AG, Zarate R (2008) Dichloromethane as a solvent for lipid extraction and assessment of lipid classes and fatty acids from samples of different natures. J Agric Food Chem 56: 4297–4303 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix

Click here for additional data file. (1.8MB, pdf)

Source Data for Appendix

Click here for additional data file. (1.2MB, zip)

Review Process File

Click here for additional data file. (2.5MB, pdf)

Source Data for Figure 2

Click here for additional data file. (790.7KB, pdf)

Source Data for Figure 3

Click here for additional data file. (271.4KB, pdf)

Source Data for Figure 4

Click here for additional data file. (374KB, pdf)

Source Data for Figure 5

Click here for additional data file. (391.4KB, pdf)

Articles from EMBO Reports are provided here courtesy of Nature Publishing Group

RESOURCES