Dietary cholesterol activates a Ral-dependent pathway driving LDLR turnover

28 min read Original article ↗

Data availability

All data supporting the findings of this study are available within the article and its Supplementary Information, the latter of which includes uncropped Western blot scans, genotyping primer sequences, qPCR primer sequences, shRNA oligo sequences and sgRNA oligo sequences. Publicly available rare-variant burden association results in humans (Table 1) were obtained from the GeneBass database (https://app.genebass.org). Mass spectrometry data from mouse liver tissues overexpressing RalA-WT, RalA(G23V) or RalA(S28N) have been deposited in FigShare under https://doi.org/10.6084/m9.figshare.32229999 (ref. 57). Source data are provided with this paper.

Code availability

No custom code was used in this study.

References

  1. Goldstein, J. L. & Brown, M. S. A century of cholesterol and coronaries: from plaques to genes to statins. Cell 161, 161–172 (2015).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  2. Luo, J., Yang, H. & Song, B. L. Mechanisms and regulation of cholesterol homeostasis. Nat. Rev. Mol. Cell Biol. 21, 225–245 (2020).

    Article  CAS  PubMed  Google Scholar 

  3. Applebaum-Bowden, D. et al. Down-regulation of the low-density lipoprotein receptor by dietary cholesterol. Am. J. Clin. Nutr. 39, 360–367 (1984).

    Article  CAS  PubMed  Google Scholar 

  4. Feingold, K. R. et al. in Endotext (eds K. R. Feingold et al.) https://www.ncbi.nlm.nih.gov/books/NBK570127/ (National Library of Medicine, 2000).

  5. Horton, J. D., Cuthbert, J. A. & Spady, D. K. Dietary fatty acids regulate hepatic low density lipoprotein (LDL) transport by altering LDL receptor protein and mRNA levels. J. Clin. Invest. 92, 743–749 (1993).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Lecerf, J. M. & de Lorgeril, M. Dietary cholesterol: from physiology to cardiovascular risk. Br. J. Nutr. 106, 6–14 (2011).

    Article  CAS  PubMed  Google Scholar 

  7. Silverman, M. G. et al. Association between lowering LDL-C and cardiovascular risk reduction among different therapeutic interventions: a systematic review and meta-analysis. J. Am. Med. Assoc. 316, 1289–1297 (2016).

    Article  CAS  Google Scholar 

  8. Horton, J. D., Cohen, J. C. & Hobbs, H. H. Molecular biology of PCSK9: its role in LDL metabolism. Trends Biochem. Sci. 32, 71–77 (2007).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Zelcer, N., Hong, C., Boyadjian, R. & Tontonoz, P. LXR regulates cholesterol uptake through Idol-dependent ubiquitination of the LDL receptor. Science 325, 100–104 (2009).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  10. Newman, C. B. et al. Statin safety and associated adverse events: a scientific statement from the American Heart Association. Arterioscler. Thromb. Vasc. Biol. 39, 38–81 (2019).

    Article  Google Scholar 

  11. Ah, Y. M., Jeong, M. & Choi, H. D. Comparative safety and efficacy of low- or moderate-intensity statin plus ezetimibe combination therapy and high-intensity statin monotherapy: a meta-analysis of randomized controlled studies. PLoS ONE 17, e0264437 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Calapai, F. et al. Adverse reactions to evolocumab: analysis of real-world data from EudraVigilance. Pharmaceuticals 17, 364 (2024).

  13. Tsouka, A. N., Tellis, C. C. & Tselepis, A. D. Pharmacology of PCSK9 inhibitors: current status and future perspectives. Curr. Pharm. Des. 24, 3622–3633 (2018).

    Article  CAS  PubMed  Google Scholar 

  14. van Dam, E. M. & Robinson, P. J. Ral: mediator of membrane trafficking. Int. J. Biochem. Cell Biol. 38, 1841–1847 (2006).

    Article  PubMed  Google Scholar 

  15. Shirakawa, R. & Horiuchi, H. Ral GTPases: crucial mediators of exocytosis and tumourigenesis. J. Biochem. 157, 285–299 (2015).

    Article  CAS  PubMed  Google Scholar 

  16. Bos, J. L., Rehmann, H. & Wittinghofer, A. GEFs and GAPs: critical elements in the control of small G proteins. Cell 129, 865–877 (2007).

    Article  ADS  CAS  PubMed  Google Scholar 

  17. Goldstein, J. L., DeBose-Boyd, R. A. & Brown, M. S. Protein sensors for membrane sterols. Cell 124, 35–46 (2006).

    Article  ADS  CAS  PubMed  Google Scholar 

  18. Emmer, B. T. et al. Genome-scale CRISPR screening for modifiers of cellular LDL uptake. PLoS Genet. 17, e1009285 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Chen, X. W. et al. A Ral GAP complex links PI 3-kinase/Akt signaling to RalA activation in insulin action. Mol. Biol Cell 22, 141–152 (2011).

    Article  CAS  PubMed  Google Scholar 

  20. Wolthuis, R. M., Zwartkruis, F., Moen, T. C. & Bos, J. L. Ras-dependent activation of the small GTPase Ral. Curr. Biol. 8, 471–474 (1998).

    Article  CAS  PubMed  Google Scholar 

  21. Welsch, M. E. et al. Multivalent small-molecule Pan-RAS inhibitors. Cell 168, 878–889.e829 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Graham, S. E. et al. The power of genetic diversity in genome-wide association studies of lipids. Nature 600, 675–679 (2021).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  23. Broadaway, K. A. et al. Liver eQTL meta-analysis illuminates potential molecular mechanisms of cardiometabolic traits. Am. J. Hum. Genet. 111, 1899–1913 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Karczewski, K. J. et al. Systematic single-variant and gene-based association testing of thousands of phenotypes in 394,841 UK Biobank exomes. Cell Genom. 2, 100168 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Tavori, H., Rashid, S. & Fazio, S. On the function and homeostasis of PCSK9: reciprocal interaction with LDLR and additional lipid effects. Atherosclerosis 238, 264–270 (2015).

    Article  CAS  PubMed  Google Scholar 

  26. Martin-Gonzalez, A. et al. Selective cargo and membrane recognition by SNX17 regulates its interaction with Retriever. EMBO Rep. 26, 470–493 (2025).

    Article  CAS  PubMed  Google Scholar 

  27. Bartuzi, P. et al. CCC- and WASH-mediated endosomal sorting of LDLR is required for normal clearance of circulating LDL. Nat. Commun. 7, 10961 (2016).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  28. Chen, X. W., Leto, D., Chiang, S. H., Wang, Q. & Saltiel, A. R. Activation of RalA is required for insulin-stimulated Glut4 trafficking to the plasma membrane via the exocyst and the motor protein Myo1c. Dev. Cell 13, 391–404 (2007).

    Article  CAS  PubMed  Google Scholar 

  29. Cantor, S. B., Urano, T. & Feig, L. A. Identification and characterization of Ral-binding protein 1, a potential downstream target of Ral GTPases. Mol. Cell. Biol. 15, 4578–4584 (1995).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Timur, Z. K., Akyildiz Demir, S. & Seyrantepe, V. Lysosomal cathepsin A plays a significant role in the processing of endogenous bioactive peptides. Front. Mol. Biosci. 3, 68 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  31. Galjart, N. J. et al. Human lysosomal protective protein has cathepsin A-like activity distinct from its protective function. J. Biol. Chem. 266, 14754–14762 (1991).

    Article  CAS  PubMed  Google Scholar 

  32. Kolli, N. & Garman, S. C. Proteolytic activation of human cathepsin A. J. Biol. Chem. 289, 11592–11600 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Seyrantepe, V. et al. Enzymatic activity of lysosomal carboxypeptidase (cathepsin) A is required for proper elastic fiber formation and inactivation of endothelin-1. Circulation 117, 1973–1981 (2008).

    Article  CAS  PubMed  Google Scholar 

  34. Yan, C. et al. Discovery and characterization of small molecules that target the GTPase Ral. Nature 515, 443–447 (2014).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  35. Cowling, R. T. Cathepsin A inhibitors to treat heart disease: much potential, many questions. JACC Basic Transl. Sci. 4, 345–347 (2019).

    Article  PubMed  PubMed Central  Google Scholar 

  36. Tillner, J. et al. Tolerability, safety, and pharmacokinetics of the novel cathepsin A inhibitor SAR164653 in healthy subjects. Clin. Pharmacol. Drug Dev. 5, 57–68 (2016).

    Article  CAS  PubMed  Google Scholar 

  37. Chen, X. W. & Saltiel, A. R. Ral’s engagement with the exocyst: breaking up is hard to do. Cell Cycle 10, 2299–2304 (2011).

    Article  CAS  PubMed  Google Scholar 

  38. Zhou, Y. & Hancock, J. F. Lipid profiles of RAS Nanoclusters regulate RAS function. Biomolecules 11, 1439 (2021).

  39. Zhu, Y. et al. Lipoprotein promotes caveolin-1 and Ras translocation to caveolae: role of cholesterol in endothelial signaling. Arterioscler. Thromb. Vasc. Biol. 20, 2465–2470 (2000).

    Article  CAS  PubMed  Google Scholar 

  40. Richardson, D. S., Spehar, J. M., Han, D. T., Chakravarthy, P. A. & Sizemore, S. T. The RAL enigma: distinct roles of RALA and RALB in cancer. Cells 11, 1645 (2022).

  41. Neyraud, V. et al. RalA and RalB proteins are ubiquitinated GTPases, and ubiquitinated RalA increases lipid raft exposure at the plasma membrane. J. Biol. Chem. 287, 29397–29405 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Martin, T. D., Mitin, N., Cox, A. D., Yeh, J. J. & Der, C. J. Phosphorylation by protein kinase Calpha regulates RalB small GTPase protein activation, subcellular localization, and effector utilization. J. Biol. Chem. 287, 14827–14836 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Goldstein, J. L. & Brown, M. S. The LDL receptor. Arterioscler. Thromb. Vasc. Biol. 29, 431–438 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Takeda, T. et al. Upregulation of IGF2R evades lysosomal dysfunction-induced apoptosis of cervical cancer cells via transport of cathepsins. Cell Death Dis. 10, 876 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Itoh, K. & Tsukimoto, J. Lysosomal sialidase NEU1, its intracellular properties, deficiency, and use as a therapeutic agent. Glycoconj. J. 40, 611–619 (2023).

    Article  CAS  PubMed  Google Scholar 

  46. Skorobogatko, Y. et al. RalA controls glucose homeostasis by regulating glucose uptake in brown fat. Proc. Natl Acad. Sci. USA 115, 7819–7824 (2018).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  47. Leto, D. & Saltiel, A. R. Regulation of glucose transport by insulin: traffic control of GLUT4. Nat. Rev. Mol. Cell Biol. 13, 383–396 (2012).

    Article  CAS  PubMed  Google Scholar 

  48. Xia, W. et al. Obesity causes mitochondrial fragmentation and dysfunction in white adipocytes due to RalA activation. Nat. Metab 6, 273–289 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Platt, R. J. et al. CRISPR-Cas9 knockin mice for genome editing and cancer modeling. Cell 159, 440–455 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. Feng, X. et al. Dual function of VGLL4 in muscle regeneration. EMBO J. 38, e101051 (2019).

    Article  PubMed  PubMed Central  Google Scholar 

  51. Abu-Remaileh, M. et al. Lysosomal metabolomics reveals V-ATPase- and mTOR-dependent regulation of amino acid efflux from lysosomes. Science 358, 807–813 (2017).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  52. Zhang, B. et al. Hepatic glycogen directly regulates gluconeogenesis through an AMPK/CRTC2 axis in mice. J. Clin. Invest. 135, e188363 (2025).

  53. Liu, X. et al. Isolation of primary human liver cells from normal and nonalcoholic steatohepatitis livers. STAR Protoc. 4, 102391 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Feng, X. et al. The tumor suppressor interferon regulatory factor 2 binding protein 2 regulates hippo pathway in liver cancer by a feedback loop in mice. Hepatology 71, 1988–2004 (2020).

    Article  CAS  PubMed  Google Scholar 

  55. de Leeuw, C. A., Mooij, J. M., Heskes, T. & Posthuma, D. MAGMA: generalized gene-set analysis of GWAS data. PLoS Comput. Biol. 11, e1004219 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  56. Giambartolomei, C. et al. Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet. 10, e1004383 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  57. Feng, X. Mouse liver proteomic data From RalA-WT, RalA-G23V or RalA-S28N overexpression. Figshare https://doi.org/10.6084/m9.figshare.32229999 (2026).

Download references

Acknowledgements

We thank all members of the Saltiel laboratory for their helpful suggestions. We thank J. Horton and E. Burstein for their helpful discussions. We thank Lifesharing OPO for providing deidentified donor liver tissues used for human primary hepatocyte isolation.

Funding

This work was supported by an American Diabetes Association postdoctoral fellowship to X.F. (1-25-PDF-76); National Institute of Health (NIH)–National Institute of Diabetes and Digestive and Kidney Diseases grants (grant nos. P30DK063491, P30DK120515, R01DK117551, R01DK128796 and R01DK135289 to A.R.S); NIH shared instrumentation grants S10OD016234 and S10OD021724 to UCSD Biomolecular and Proteomics Mass Spectrometry Facility, Cancer Institute Cancer Center Support Grant (CCSG grant no. P30CA23100) to UCSD Tissue Technology, and NS047101 to UCSD School of Medicine Microscopy Core. This work was also supported by NIH grants DK099205, AA028550, DK101737, AA029019 and DK091183 to T.K., and by the Sanford Stem Cell Fitness and Space Medicine Center at the Sanford Stem Cell Institute, UCSD.

Author information

Authors and Affiliations

  1. Department of Medicine, University of California, San Diego, San Diego, CA, USA

    Xue Feng, Shuo Zhang, Yuqi Wang, Twisha Kurlagunda, Allyssa Sit, Sadatsugu Sakane, Se Yong Park, Churaibhon Wisessaowapak, Kaylee Nguyen, Jamie Yan, Himani Pothulu, Catherine Dinh, Felicia Chu, Yuyao Ren, Bichen Zhang, Patrick Secrest, Chao-Wei Hung, Preethi Veeragandham, Yuliya Skorobogatko, Amit R. Majithia & Alan R. Saltiel

  2. Department of Pharmaceutical Chemistry, University of California, San Francisco, San Francisco, CA, USA

    Priyadarshini Jaishankar, Pusu Yang & Adam R. Renslo

  3. Department of Pharmacology, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA

    Linmeng Han

  4. Department of Surgery, University of California, San Diego, San Diego, CA, USA

    Tatiana Kisseleva

  5. Department of Pathology, University of Utah, Salt Lake City, UT, USA

    Philip L.S.M. Gordts

  6. Department of Biochemistry and Structural Biology, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA

    Peng Zhao

  7. Department of Pharmacology, University of California, San Diego, San Diego, CA, USA

    Alan R. Saltiel

Authors

  1. Xue Feng
  2. Shuo Zhang
  3. Yuqi Wang
  4. Twisha Kurlagunda
  5. Allyssa Sit
  6. Priyadarshini Jaishankar
  7. Pusu Yang
  8. Sadatsugu Sakane
  9. Se Yong Park
  10. Churaibhon Wisessaowapak
  11. Kaylee Nguyen
  12. Jamie Yan
  13. Himani Pothulu
  14. Catherine Dinh
  15. Felicia Chu
  16. Yuyao Ren
  17. Bichen Zhang
  18. Patrick Secrest
  19. Linmeng Han
  20. Chao-Wei Hung
  21. Preethi Veeragandham
  22. Yuliya Skorobogatko
  23. Tatiana Kisseleva
  24. Philip L.S.M. Gordts
  25. Adam R. Renslo
  26. Peng Zhao
  27. Amit R. Majithia
  28. Alan R. Saltiel

Contributions

X.F. performed experimental design, experiments, data acquisition and interpretation, and wrote the manuscript. A.R.S. conceptualized and supervised the study, performed data interpretation and wrote the manuscript. S.Z., T.K., A.S., C.W., K.N., J.Y., H.P., C.D., F.C., Y.R. and L.H. performed the experiments. X.F. and S.Z. drew the model figure. Y.W. and A.R.M. analysed the GWAS, eQTL and UK Biobank datasets, performed MAGMA analysis, interpreted the results and drafted the corresponding manuscript. P. J., P.Y. and A.R.R. synthesized the CTSA inhibitor. S.S., S.Y.P. and T.K. provided human primary hepatocytes. B.Z. provided technical support for primary hepatocyte isolation and contributed the Cas9 knock-in mouse model. P.S. performed the FPLC assay. C.-W.H. and P.V. performed RalA pulldown for proteomics analysis. Y.S. provided the RalGAPB flox mice and helpful suggestions. T.K., P.L.S.M.G., A.R., P.Z. and A.R.M. contributed valuable insights.

Corresponding authors

Correspondence to Xue Feng or Alan R. Saltiel.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature thanks Anja Zeigerer and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 Cholesterol exposure activates Ral GTPase to decrease LDLR protein expression in hepatocytes.

a-c. Relative mRNA levels of Ldlr and Srebp2 in liver tissues from C57BL/6J mice fed NCD or HCD for 4 weeks (a), 12 weeks (b), or 20 weeks (c). n = 5 mice. d. Representative immunoblots of LDLR protein levels in liver tissues from C57BL/6J mice fed NCD or HCD for 4 weeks (n = 5), 12 weeks (n = 5, 4), or 20 weeks (n = 4). e. Representative immunoblots in liver tissues from C57BL/6J mice fed NCD or HCD for 4 weeks or 20 weeks. n = 5 mice. f. Quantification of RalA and RalB GTPase activities, shown as the ratio of GTP-Ral to total Ral, in HepG2 cells treated with MβCD or Chol (20 μg ml−1, 24 h). n = 3. g, h. Plasma TC (g) or VLDL/LDL-C (h) levels from C57BL/6J mice fed NCD (n = 5 for TC, n = 3 for VLDL/LDL) or HCD (n = 3) for 4, 12, or 20 weeks. i. Quantification of RalA and RalB GTPase activities, shown as the ratio of GTP-Ral to total Ral, in RalGAPBf/f or RalGAPBHKO hepatocytes. n = 3. j. Relative mRNA levels of Ldlr in HepG2 cells overexpressing pLVX-FLAG, RalA-WT, -G23V, or -S28N. n = 3. k, l. Relative mRNA levels of Ralgapa1, Ralgapa2, Ralgapb, Rgl1, Rgl2, Rgl3, and Ralgds in liver tissues from C57BL/6J mice fed NCD (n = 5) or HCD (n = 4 for Ralgapa2 and Rgl3, n = 5 for others) for 20 weeks. m, n. Representative immunoblots in liver tissues from mice fed NCD or HCD for 20 weeks. n = 5. o. Fasting plasma insulin levels in C57BL/6J mice fed NCD or HCD for 20 weeks. n = 5. Data are mean ± s.e.m. Each dot (n) represents one biological replicate. ns, not significant.

Source data

Extended Data Fig. 2 Cholesterol-induced RAS activation suppresses LDLR in hepatocytes.

a. Quantification of RAS GTPase activity (active RAS, aRAS), shown as the ratio of GTP-RAS to total RAS, in HepG2 cells treated with MβCD or Chol (20 μg ml−1, 24 h). n = 3. b. Representative immunoblots in HepG2 cells treated under the indicated conditions for 24 h in LPDS medium. Chol: 20 μg ml−1. n = 3. c. Representative immunofluorescence staining of HA-KRAS and ATPA1 in HepG2 cells. Cells were infected with lentiviruses expressing HA-KRAS, and treated with either MβCD or cholesterol (20 μg ml−1) in LPDS medium. Scale bar: 5 μm. d. Quantification of Pearson’s correlation coefficient between HA-KRAS and ATPA1, as shown in (c). n = 5/6 independent fields of view. e. Representative immunoblots of liver samples from C57BL/6J mice fed NCD or HCD for 20 weeks. Livers were lysed and subjected to cellular fractionation to isolate the plasma membrane (PM) fraction. f, g. Representative immunoblots (f) and quantification of LDLR protein levels normalized to HSP90 (g) in AML12 cells expressing the pLEX vector or KRAS. n = 3. Data are mean ± s.e.m. Each dot (n) represents one biological replicate. Unpaired two-sided Student’s t-test (a, d, g). ns, not significant.

Source data

Extended Data Fig. 3 Activation of Ral in vivo impairs cholesterol homeostasis by reducing hepatic LDLR protein levels.

a-c. Body weight (a), plasma TG (b), and plasma PCSK9 (c) levels in RalGAPBf/f (n = 19) or RalGAPBHKO (n = 12) mice following 20 weeks of HCD. d-f. Body weight (d), plasma TC (e), and plasma TG (f) in RalGAPBf/f (n = 7) or RalGAPBHKO (n = 10/8/8) female mice after 20 weeks on HCD starting at 8 weeks of age. g. Representative immunoblots of livers from RalGAPBf/f (n = 4) or RalGAPBHKO (n = 4) female mice after 20 weeks on HCD. h-k. Body weight (h), plasma TC (i), liver TC (j), and plasma TG (k) in RalGAPBf/f (n = 8/8/8/7) or RalGAPBHKO (n = 10/12/12/12) mice after 20 weeks on NCD starting at 8 weeks of age. l. Representative immunoblots of livers from RalGAPBf/f (n = 4) or RalGAPBHKO (n = 4) mice after 20 weeks on NCD. m, n. Relative mRNA levels of Ldlr (m) and Srebp2 (n) in livers of RalGAPBf/f (n = 7) or RalGAPBHKO (n = 10) mice after 20 weeks on NCD. o. Plasma PCSK9 levels in RalGAPBf/f (n = 7) or RalGAPBHKO (n = 11) mice after 20 weeks on NCD. p, q. Body weight (p) and liver weight (q) in mice overexpressing AAV-GFP (n = 5), AAV-RalA-G23V (n = 5), or AAV-RalB-G23V (n = 5) after 8 weeks on HCD. r. Representative immunoblots showing total Ral (input) and GTP-bound Ral pulled-down using RalBP1 beads in liver lysates from mice overexpressing AAV-GFP, AAV-RalA-G23V, or AAV-RalB-G23V. n = 5. s. Plasma HDL-C from mice overexpressing AAV-GFP (n = 3), AAV-RalA-G23V (n = 5), or AAV-RalB-G23V (n = 5). Data are mean ± s.e.m. Each dot (n) represents one biological replicate. Unpaired two-sided Student’s t-test (a-f, h-k, m-q, s). ns, not significant.

Source data

Extended Data Fig. 4 Ral activation bypasses PCSK9 to promote lysosomal LDLR degradation.

a. Representative immunoblots in hepatocytes isolated from RalGAPBf/f or RalGAPBHKO mice. Cells were treated with 100 μg ml−1 cycloheximide (CHX) and harvested at the indicated time points. n = 2 per group. b. Representative live cell images of LysoTracker in primary hepatocytes from RalGAPBf/f or RalGAPBHKO mice. Scale bar, 5μm. c, d. Quantification of LysoTracker number (c) and fluorescence intensity (d) per cell as shown in (b). n = 29 cells from three independent views. e. Representative immunoblots of total Ral (input) and GTP-bound Ral pulled down with RalBP1 beads in WT and RalGAPB-KO AML12 cells. n = 2. f. Representative confocal images of DiI-LDL in WT and RalGAPB-KO AML12 cells. Cells were incubated overnight in serum-free medium, followed by a two-hour incubation with DiI-LDL (10 μg ml−1). Scale bar, 10 μm. g. Quantification of internalized LDL fluorescence intensity in WT (n = 89 cells) and RalGAPB-KO (n = 75 cells) AML12 cells as shown in (f). h. Representative immunofluorescence staining of LDLR and CD98 in WT and RalGAPB-KO AML12 cells. Cells were cultured overnight in serum-free medium. Scale bar, 10 μm. i. Representative immunofluorescence staining of LDLR and LAMP1 in WT and RalGAPB-KO AML12 cells. Pearson’s correlation coefficients between LDLR and LAMP1 are shown in the top right corner of each image. Scale bar, 5 μm. j, k. Representative immunoblots (j) and quantification of LDLR protein levels normalized to HSP90 (k) in WT or RalGAPB-KO AML12 cells. Cells were treated with DMSO or BafA1 (100 nM, 24 h) in LPDS medium. l. Representative live cell images of LysoTracker in WT or RalGAPB-KO AML12 cells. Scale bar, 5μm. m, n. Quantification of LysoTracker number (m) and fluorescence intensity (n) per cell as shown in (l). WT: n = 28 cells, RalGAPB-KO: n = 30 cells. o, p. PCSK9 levels in the culture medium collected from RalGAPBf/f or RalGAPBHKO primary hepatocytes cultured overnight in full medium (o) or LPDS medium (p). n = 3. q. HEK293T cells were transfected with GFP-LDLR and the indicated plasmids. GFP-LDLR protein levels were assessed by immunoblot two days post-transfection. n = 2 per group. r. Representative immunoblots in WT or RalGAPB-KO AML12 cells. Cells were treated with DMSO or 20 μM PCSK9 antagonist (SBC−115076) for 24 h in LPDS medium. Protein levels of LDLR were quantified and normalized to HSP90, with values shown below the blots. n = 2. s. Purified PCSK9 protein was added to WT or RalGAPB-KO AML12 cells at the indicated concentrations in serum-free medium. LDLR protein levels were assessed by Western blot. Data are mean ± s.e.m. Each dot (n) represents one biological replicate. Unpaired two-sided Student’s t-test (c, d, g, m-p); two-way ANOVA with Fisher’s test (k). ns, not significant.

Source data

Extended Data Fig. 5 SNX17-dependent recycling contributes to Ral-mediated regulation of LDLR.

a. Immunoprecipitation (IP) in HepG2 cells was performed using RalA as bait to pull down LDLR protein. The interaction between RalA and LDLR was detected by Western blot. b, c. Representative immunofluorescence staining (b) and quantification of the Pearson’s correlation coefficient between RalA and LDLR (c) in WT and RalGAPB-KO AML12 cells cultured overnight in serum-free medium. Scale bar, 10 μm. n = 5 fields. d. HEK293T cells were co-transfected with GFP-LDLR and FLAG-tagged RalB variants (WT, G23V, or S28N). Reciprocal immunoprecipitations were performed using either GFP-LDLR or FLAG-RalB as bait to pull down their binding partners. The interactions between GFP-LDLR and FLAG-RalB (including its mutants) were confirmed by Western blot. IB, immunoblot. e, f. Relative mRNA levels of Snx17 (e) and the indicated protein levels (f) in WT or RalGAPB-KO AML12 cells. Cells were infected with lentivirus carrying control shRNA (shCtrl) or SNX17 shRNA (shSNX17). n = 4 for (e), n = 2 for (f). Protein levels of LDLR were quantified and normalized to HSP90, with values shown below the blots in (f). g. Relative mRNA levels of Washc1, Washc2c, Ccdc22, and Commd1 in livers of C57BL/6J mice after 20 weeks on NCD or HCD. n = 5 mice. h, i. Representative immunoblots (h) and quantification of the indicated proteins normalized to HSP90 (i) in livers of C57BL/6J mice after 20 weeks on NCD or HCD. j. siRNA knockdown efficiency as determined by qPCR, showing mRNA levels of Ccdc22, Commd1, and Wash1. Cells were treated with the indicated siRNAs for three days before analysis. n = 3. k. Representative immunoblots in WT or RalGAPB-KO (KO) AML12 cells. Cells were treated with the indicated siRNAs for three days before analysis. Protein levels of LDLR were quantified and normalized to HSP90, with values shown below the blots. n = 2. Data are mean ± s.e.m. Each dot (n) represents one biological replicate. Unpaired two-sided Student’s t-test (c, e, g, i, j). ns, not significant.

Source data

Extended Data Fig. 6 The Ral-RalBP1-REPS1 axis promotes LDLR endocytosis and degradation.

a. WT or RalGAPB-KO (KO) AML12 cells were transfected with siRNAs targeting control (siNC), SEC5 (siSEC5), or SEC8 (siSEC8). Cells were harvested three days post-transfection, and protein levels of the indicated targets were assessed by Western blot. n = 2. b. WT or RalGAPB-KO (KO) AML12 cells were infected with lentivirus carrying either control shRNA (shCtrl) or RalBP1 shRNA (shRalBP1). Protein levels of the indicated targets were analysed by Western blot. n = 2. c. Co-IP of FLAG-RalBP1 and HA-REPS1. The interaction between the two proteins was detected by Western blots. IB, immunoblot. d, e. Representative immunoblots (d) and quantification of REPS1 protein levels normalized to HSP90 (e) in liver tissues from RalGAPBf/f or RalGAPBHKO mice fed NCD for 20 weeks. n = 5 mice. f, g. Representative immunoblots (f) and quantification of REPS1 protein levels normalized to HSP90 (g) in liver lysates from mice overexpressing AAV-GFP (n = 4), AAV-RalA-G23V (n = 5), or AAV-RalB-G23V (n = 5) after 8 weeks on HCD. h. siRNA knockdown efficiency of Reps1 as determined by qPCR in AML12 cells. n = 3 per group. i, j. LocusZoom plots of the SNPs for human total cholesterol (i) and human liver REPS1 expression (j). Lead SNP, rs62441842. PP.H4, posterior probability of a shared causal variant. Data are mean ± s.e.m. Each dot (n) represents one biological replicate. Unpaired two-sided Student’s t-test (e, g, h). ns, not significant.

Source data

Extended Data Fig. 7 CTSA mediates LDLR degradation downstream of Ral signalling.

a. Representative confocal images of HA-CTSA and LAMP1 in AML12 cells. Cells were transfected with HA-CTSA prior to staining. Co-localization was assessed using ImageJ, and Pearson’s correlation coefficient was calculated as r = 0.818. Scale bar, 10μm. b. Representative immunoblots of GTPγS/GDP loading assay showing the interaction between HA-CTSA and GTPγS- or GDP-loaded FLAG-RalA. Purified FLAG-RalA was preloaded with either GTPγS or GDP to mimic the active or inactive state of RalA, respectively, and used as bait to pull down HA-CTSA from HEK293T cell lysates. IB, immunoblot. c. Representative confocal images of RalA, CTSA, and HA-Tmem192 in AML12 cells. Cells were infected with lentiviruses expressing HA-Tmem192. Scale bar, 2μm. d, e. Representative immunoblots (d) and quantification of the CTSA-lyso/CTSA-pre ratio (e) in liver tissues from C57BL/6J mice after 4 weeks on NCD or HCD. CTSA-pre, CTSA precursor; CTSA-lyso, lysosomal CTSA. f. Representative confocal images of CTSA and HA-Tmem192 in WT or RalGAPB-KO AML12 cells. Cells were infected with lentiviruses expressing HA-Tmem192. Pearson’s correlation coefficients are indicated in each image. Scale bar, 5μm. g. Representative immunoblots of liver tissues from the indicated mice after 20 weeks on HCD. n = 4 mice per group. h. Plasma CTSA levels from the indicated mice after 20 weeks on HCD. n = 12 mice per group. i. Representative immunoblots of CTSA in culture supernatants and cell lysates from WT and RalGAPB-KO AML12 cells. The relative amount of CTSA in supernatants (sup) normalized to CTSA in corresponding lysates is shown as the ratio below each blot. n = 2 per group. j. CTSA levels in supernatants from WT and RalGAPB-KO AML12 cells. n = 3 per group. k. Representative immunoblots in AML12 cells. Cells were infected with lentiviruses expressing pLEX-HA or HA-CTSA prior to harvest. n = 3. l. Representative immunoblots in WT and RalGAPB-KO (KO) AML12 cells treated with DMSO or 100 μM AEBSF overnight in LPDS medium. n = 2. m, n. Representative confocal images (m) and quantification of plasma membrane LDLR fluorescence intensity (n) in the indicated non-permeabilized AML12 cells. Scale bar, 10 μm. n = 32/27/47/37 cells. o, p. Representative confocal images (o) and quantification of DiI-LDL fluorescence intensity (p) in the indicated AML12 cells. Cells were cultured overnight in LPDS medium, then incubated with 10 μg ml−1 DiI-LDL for two hours to assess LDL uptake. Scale bar, 10 μm. n = 100/58/92/91 cells. q. Relative mRNA levels of Ldlr in livers from control (sgLacZ, n = 9) or CTSA deletion (sgCTSA, n = 7) mice. r-u. Body weight (r), liver-to-body weight ratio (%) (s), plasma glucose (t), and plasma TG (u) from control (sgLacZ) or CTSA deletion (sgCTSA) mice. sgLacZ: n = 9/9/9/8; sgCTSA: n = 7. v, w. Quantification of LysoTracker number (v) and fluorescence intensity (w) per cell in control (sgLacZ) or CTSA deletion (sgCTSA) primary hepatocytes. n = 15 cells. Data are mean ± s.e.m. Each dot (n) represents one biological replicate. Unpaired two-sided Student’s t-test (e, h, j, q-w); one-way ANOVA with Tukey’s test (n, p). ns, not significant.

Source data

Extended Data Fig. 8 Ral and CTSA inhibition stabilize LDLR and improve plasma lipid profiles.

a. Relative mRNA levels of Ldlr in HepG2 cells treated with DMSO or RBC8 (10 μM, 24 h) in LPDS medium. n = 4. b. Representative immunoblots showing total Ral (input) and GTP-bound Ral pulled down by RalBP1-RBD agarose beads in HepG2 cells treated with DMSO or RBC8 (10 μM, 24 h) in LPDS medium. The ratio of GTP-Ral to total Ral is quantified below the blots. c, d. Representative confocal images (c) and quantification of plasma membrane LDLR fluorescence intensity (d) in non-permeabilized HepG2 cells. Cells were cultured overnight in LPDS medium. The next day, DiI-LDL was added on ice and incubated for 30 min. Cells were then shifted to 37 °C in PBS + /+ containing 25 μM Monensin to block recycling, along with either DMSO or 50 μM RBC8, for 30 min to allow LDL internalization. Afterward, cells were washed, fixed, and stained for LDLR without permeabilization to detect cell surface LDLR. Scale bar, 10 μm. n = 78 cells (DMSO), n = 86 cells (RBC8). e. Quantification of RalA and RalB GTPase activities, shown as the ratio of GTP-RalA to total RalA or GTP-RalB to total RalB, in livers from C57BL/6J mice injected with DMSO or RBC8 (50 mg kg−1). n = 4 mice per group. f-n. Body weight change ratio (f), liver-to-body weight ratio (%) (g), plasma alanine aminotransferase (ALT) (h), plasma aspartate aminotransferase (AST) (i), relative mRNA levels of Ldlr (j), liver TC (k), plasma PCSK9 (l), plasma HDL-C (m), and plasma TC (n) of C57BL/6J mice injected with DMSO or 50 mg kg−1 RBC8. DMSO: n = 9 (f, j, k, l), n = 8 (g-i, n), n = 6 (m); RBC8: n = 8 (f, g, j, l, n), n = 7 (h, i, k), n = 6 (m). o, p. Representative immunoblots (o) and quantification of LDLR protein levels normalized to HSP90 (p) in primary hepatocytes treated with DMSO or CTSA inhibitor (CTSAi, 10 μM, 24 h) in full medium. n = 3. q, r. Representative confocal images (q) and quantification of plasma membrane LDLR fluorescence intensity (r) in non-permeabilized human primary hepatocytes. Cells were treated with DMSO or CTSA inhibitor (CTSAi, 10 μM, 24 h) in LPDS medium. Scale bar, 5μm. n = 5 individual views. s. Human primary hepatocytes were pre-treated with DMSO or CTSA inhibitor (CTSAi, 10 μM, 24 h) in LPDS medium, followed by treatment with MβCD or cholesterol (Chol, 10 μg ml−1) together with lovastatin (0.5 μM) for 16 h. Representative immunoblots are shown. n = 3. t. 8-week-old C57BL/6J mice were fed HCD for 8 weeks, followed by oral gavage with either vehicle (0.6% hydroxyethyl cellulose plus 0.5% Tween) or a CTSA inhibitor (CTSAi, 100 mg kg−1) for an additional 4 weeks while maintained on HCD. Plasma HDL-C are shown (n = 4 mice per group). u, v. 8-week-old RalGAPBf/f and RalGAPBHKO mice were fed HCD for 4 weeks, followed by oral gavage with either vehicle (0.6% hydroxyethyl cellulose plus 0.5% Tween) or a CTSA inhibitor (CTSAi, 100 mg kg−1) for an additional 4 weeks while maintained on HCD. Plasma TC (u) and HDL-C (v) levels from RalGAPBf/f and RalGAPBHKO mice. n = 4 mice per group. Data are mean ± s.e.m. Each dot (n) represents one biological replicate. Unpaired two-sided Student’s t-test (a, d, e–n, p, r, t); two-way ANOVA with Fisher’s (u, v). ns, not significant.

Source data

Extended Data Fig. 9 Schematic model illustrating how high dietary cholesterol activates a Ral-dependent pathway to regulate LDLR trafficking and degradation in hepatocytes.

High dietary cholesterol activates RAS, which promotes Ral activation through RalGEFs. Activated Ral interacts with RalBP1, which serves as a scaffold to recruit REPS1 and other endocytic components, coordinating efficient internalization of LDLR from the plasma membrane. Active Ral promotes lysosomal targeting of LDLR and CTSA, enhancing CTSA maturation and lysosomal accumulation while limiting its secretion. CTSA then drives lysosomal LDLR degradation. Meanwhile, SNX17 normally facilitates LDLR recycling from endosomes back to the plasma membrane. However, when Ral is activated, it reduces SNX17 association with LDLR and suppresses LDLR recycling. RalGAPB inactivates Ral by stimulating GTP hydrolysis, thereby suppressing LDLR trafficking to lysosomes and CTSA-mediated degradation. Pharmacological inhibition of Ral (e.g., by RBC8) maintains Ral in its inactive GDP-bound state, attenuating LDLR degradation. Inhibiting CTSA reduces lysosomal LDLR degradation and turnover, supporting CTSA as a potential target to improve LDLR stability. Created in BioRender. Zhang, S. (2026) http://biorender.com/mrruex4.

Supplementary information

Source data

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Feng, X., Zhang, S., Wang, Y. et al. Dietary cholesterol activates a Ral-dependent pathway driving LDLR turnover. Nature (2026). https://doi.org/10.1038/s41586-026-10697-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Version of record:

  • DOI: https://doi.org/10.1038/s41586-026-10697-z