Inhibiting IP6K1 confers atheroprotection by elevating circulating apolipoprotein A-I

Abstract

Background and aims: Atherosclerotic cardiovascular diseases are the leading cause of death. Apolipoprotein A-I (apoA-I) mediates cholesterol efflux to lower the risks of atherosclerosis. Elevating circulating apoA-I is an effective strategy for atheroprotection. However, the regulatory mechanisms of apoA-I have been elusive.

Methods: Protein-protein interactions were examined by co-immunoprecipitations. Chemical biology tools were used to determine the binding of 5PP-InsP₅ to its target proteins and its roles in mediating protein-protein interactions. The mouse atherosclerotic model was generated by injecting AAV-PCSK9 and feeding a Western diet. Atherosclerotic plaques were determined by Oil Red O and H&E staining.

Results: We show that blocking IP6K1 activity increases apoA-I production in hepatocytes. IP6K1 binds to apoA-I and via its product 5PP-InsP₅ to induce apoA-I degradation, which requires ubiquitination factor E4A (UBE4A). Depleting 5PP-InsP₅ by deleting IP6K1 or blocking IP6K1 activity disrupts the interaction between UBE4A and apoA-I, preventing apoA-I degradation, leading to increased production of apoA-I. Hepatocyte-specific deletion of IP6K1 elevates circulating apoA-I levels, which augments cholesterol efflux and lowers the burden of atherosclerosis. Mice with both apoA-I KO and hepatocyte-specific IP6K1 KO were generated to validate that IP6K1 deletion-induced atheroprotection requires apoA-I.

Conclusions: Our findings reveal a mechanism by which blocking IP6K1 boosts apoA-I production. Blocking IP6K1 represents a potential treatment strategy to elevate circulating apoA-I for atheroprotection.

Keywords: Atherosclerosis; Cholesterol efflux; HDL; Hepatocyte; IP6K; Inositol pyrophosphate; UBE4A.

Graphical abstract

Fig. 1. Knockout of IP6K1 boosted the protein levels of apoA-I in the liver, resulting in an increase in HDL in the plasma.

A, Non-denatured plasma proteins of WT and IP6K1 KO mice were separated by protein gel electrophoresis in a native gel, which was stained by Coomassie blue. The protein band (arrow) was identified by mass spectrometry as apoA-I. n=4 mice per group. B, Protein levels of apoA-I in the plasma of WT and IP6K1 KO mice. Data represent mean±SEM, Student’s t-test, n=5 mice per group. C, ApoA-I concentrations in the plasma of WT and IP6K1 KO mice were determined by ELISA assay. Data were presented as mean ± SEM, Student’s t-test, n = 6 mice per group. D, Levels of HDL cholesterol in the plasma of WT and IP6K1 KO mice. Data represent mean±SEM, Student’s t-test, n=5 mice per group. E, Protein levels of apoA-I in the livers of WT and IP6K1 KO mice. Data represent mean±SEM, Student’s t-test, n=5 mice per group. F, Protein levels of apoA-I in the livers of control (IP6K1^f/f) and hepatocyte-specific IP6K1 KO mice (Alb-cre; IP6K1^f/f). Data represent mean±SEM, Student’s t-test, n=5 mice per group. G, Protein levels of apoA-I in the plasma of IP6K1^f/f and Alb-cre; IP6K1^f/f mice. Data represent mean±SEM, Student’s t-test, n=5 mice per group. H, ApoA-I concentrations in the plasma of IP6K1^f/f and Alb-cre; IP6K1^f/f mice were determined by ELISA assay. Data were presented as mean ± SEM, Student’s t-test, n=7 mice per group. I, Levels of HDL cholesterol in the plasma of IP6K1^f/f and Alb-cre; IP6K1^f/f mice. Data represent mean±SEM, Student’s t-test, n=5 mice per group. J, Protein levels of apoA-I in the in vitro cultured primary murine hepatocytes of WT and IP6K1 KO mice. Data represent mean±SEM, Student’s t-test, n=4 independent repeats. K, Protein levels of apoA-I in the in vitro cultured primary murine hepatocytes of the IP6K1^f/f and Alb-cre; IP6K1^f/f mice. Data represent mean±SEM, Student’s t-test, n=5 independent repeats. L, Protein levels of apoA-I in the cell culture media of primary murine hepatocytes of the IP6K1^f/f and Alb-cre; IP6K1^f/f mice. Data represent mean±SEM, Student’s t-test, n=5 independent repeats.

Fig. 2. Pharmacological inhibition of IP6K increased apoA-I protein levels.

A, Overexpressing WT but not mutant IP6K1 in the IP6K1 KO murine hepatocytes rescued apoA-I protein levels. Data represent mean±SEM, One-way ANOVA, n=3 independent repeats. B, Inhibiting IP6K activity by TNP (3 μM, 24 hours) treatment upregulated apoA-I in primary murine hepatocytes. Data represent mean±SEM, Student’s t-test, n=3 independent repeats. C, WT mice were treated with TNP (20 mg/kg) for 1 week. The apoA-I protein levels in the livers were upregulated. Data represent mean±SEM, Student’s t-test, n=3 mice per group. D, WT mice were treated with SC-919 (10 mg/kg) for 1 week. The apoA-I protein levels in the livers were upregulated. Data represent mean±SEM, Student’s t-test, n=3 mice per group. E, Administration of TNP (20 mg/kg, 1 week) increased apoA-I protein levels in the plasma. Data represent mean±SEM, Student’s t-test, n=3 mice per group. F, Administration of SC-919 (10 mg/kg, 1 week) increased apoA-I protein levels in the plasma. Data represent mean±SEM, Student’s t-test, n=3 mice per group. G, Myc-IP6K1 was overexpressed in primary murine hepatocytes. Myc-GFP was overexpressed as a control. The cells were treated with MG132 (10 μM, 4 hours) to block proteasomal activity. ApoA-I was immunoprecipitated and blotted for ubiquitin. H, WT and IP6K1 KO primary murine hepatocytes were treated with MG132 (10 μM, 4 hours) to block proteasomal activity. ApoA-I was immunoprecipitated and blotted for ubiquitin. I, Primary murine hepatocytes were treated with TNP (3 μM, 24 hours) or DMSO. The cells were then treated with MG132 (10 μM, 4 hours) to block proteasomal activity. ApoA-I was immunoprecipitated and blotted for ubiquitin. J, Primary murine hepatocytes were treated with SC-919 (1 μM, 24 hours) or DMSO. The cells were then treated with MG132 (10 μM, 4 hours) to block proteasomal activity. ApoA-I was immunoprecipitated and blotted for ubiquitin.

Fig. 3. IP6K1 formed a complex with apoA-I and UBE4A.

A, IP6K1 was immunoprecipitated in primary murine hepatocytes. Protein electrophoresis and silver staining were performed to look for its binding partners. The protein band (arrow) was identified by mass spectrometry as UBE4A. B, ApoA-I was immunoprecipitated in hepatocytes of WT and IP6K1 KOs. Protein electrophoresis and silver staining were performed to look for its binding partners. One protein band with molecular weight ~130 KDa (arrow) appeared darker in the WT preparations. C, Myc-IP6K1 and HA-UBE4A were overexpressed together, myc-GFP was overexpressed as a control. Immunoprecipitation of myc-IP6K1 co-pulled down HA-UBE4A. D, Immunoprecipitations of endogenous IP6K1 and UBE4A in primary murine hepatocytes co-pulled down each other. E and F, Flag-apoA-I and HA-UBE4A were overexpressed together, flag-GFP was overexpressed as a control. (E) Immunoprecipitation of flag-apoA-I co-pulled down HA-UBE4A. (F) Immunoprecipitation of HA-UBE4A co-pulled down flag-apoA-I. G and H, Myc-IP6K1 and flag-apoA-I were overexpressed together, myc-GFP was overexpressed as a control. (G) Immunoprecipitation of myc-IP6K1 co-pulled down flag-apoA-I. (H) Immunoprecipitation of flag-apoA-I co-pulled down myc-IP6K1 (arrow). I, Endogenous IP6K1 and UBE4A were co-pulled down by endogenous apoA-I in primary murine hepatocytes. J, IP6K1 (4 nm) and apoA-I (12 nm) were labeled in murine hepatocytes by immuno-electron microscopy. Scale bar 50 nm. K, UBE4A (4 nm) and apoA-I (12 nm) were labeled in murine hepatocytes by immuno-electron microscopy. Scale bar 50 nm.

Fig. 4. UBE4A was recruited by IP6K1 to ubiquitinate apoA-I.

A, Flag-apoA-I and HA-UBE4A were overexpressed together in primary murine hepatocytes. The cells were treated with M132 (10 μM, 4 hours) to block proteasomal activity. Flag-apoA-I was immunoprecipitated and blotted for ubiquitin. B, Flag-apoA-I was overexpressed in primary murine hepatocytes, and UBE4A was deleted by shRNA transduction (the shRNA2 as in panel D). The cells were treated with M132 (10 μM, 4 hours) to block proteasomal activity. Flag-apoA-I was immunoprecipitated and blotted for ubiquitin. C and D, UBE4A was deleted in AML12 murine hepatocytes. (C) Deletion of UBE4A upregulated apoA-I protein levels. Data represent mean±SEM, One-way ANOVA, n=3 independent repeats. (D) The protein levels of apoA-I in the cell culture media were higher in the UBE4A-deleted preparations. Data represent mean±SEM, One-way ANOVA, n=3 independent repeats. E and F, Flag-apoA-I was overexpressed in WT and IP6K1 KO primary murine hepatocytes. (E), Immunoprecipitation of flag-apoA-I co-pulled down less UBE4A in the IP6K1 KOs. Data represent mean±SEM, Student’s t-test, n=3 independent repeats. (F), Immunoprecipitation of UBE4A co-pulled down less flag-apoA-I in the IP6K1 KOs. Data represent mean±SEM, Student’s t-test, n=3 independent repeats. G and H, Flag-apoA-I was overexpressed in primary murine hepatocytes. The cells were treated with IP6K inhibitor SC-919 (1 μM, 24 hours). (G), Immunoprecipitation of flag-apoA-I co-pulled down less UBE4A in the SC-919-treated cells. Data represent mean±SEM, Student’s t-test, n=3 independent repeats. (H), Immunoprecipitation of UBE4A co-pulled down less flag-apoA-I in the SC-919-treated cells. Data represent mean±SEM, Student’s t-test, n=3 independent repeats.

Fig. 5. 5PP-InsP₅ enhanced the binding of UBE4A to apoA-I.

A, 5PCP resins pulled down endogenous UBE4A and apoA-I in whole cell lysates. B, 5PCP resins pulled down purified UBE4A and apoA-I, but not GST in an in vitro protein binding assay. C, 5PP-InsP₅ enhanced the binding of UBE4A to apoA-I. Data represent mean±SEM, Student’s t-test, n=3 independent repeats. D, Both 5-PCP-InsP₅ (5PCP) and 5-PCF₂Am-InsP₅ (CF2) promoted the binding of UBE4A to apoA-I. Data represent mean±SEM, One-way ANOVA, n=3 independent repeats. E, 5PP-InsP₅ but not InsP₃, InsP₄ or InsP₅ enhanced the binding of UBE4A to apoA-I. Data represent mean±SEM, One-way ANOVA, n=3 independent repeats. F, 5PP-InsP₅ but not 1PP-InsP₅ or 3PP-InsP₅ enhanced the binding of UBE4A to apoA-I. Data represent mean±SEM, One-way ANOVA, n=3 independent repeats.

Fig. 6. Knockout of IP6K1 augmented cholesterol efflux and attenuated atherosclerosis.

A-I, Hepatocyte-specific IP6K1 KO (Alb-cre; IP6K1^f/f) and control (IP6K1^f/f) mice were injected with AAV-PCSK9 and fed with a Western diet. (A), The protein levels of apoA-I were higher in the livers of Alb-cre; IP6K1^f/f mice than in controls (IP6K1^f/f). Data represent mean±SEM, Student’s t-test, n=3 mice per group. (B), The protein levels of apoA-I were higher in the plasma of Alb-cre; IP6K1^f/f mice than in controls (IP6K1^f/f). Data represent mean±SEM, Student’s t-test, n=3 mice per group. (C), HDL levels were higher in the plasma of Alb-cre; IP6K1^f/f mice than in controls (IP6K1^f/f). Data represent mean±SEM, Student’s t-test, n=6 mice per group. (D), HDL₂ and HDL₃ levels were higher in the plasma of Alb-cre; IP6K1^f/f mice than in controls (IP6K1^f/f). Data represent mean±SEM, Mann-Whitney U test, n=6 mice per group. (E), Pooled plasma samples were fractionated by FPLC for cholesterol analysis. VLDL, very-low-density lipoprotein. CM, Chylomicrons. n=6 mice per group. (F), The plasma of Alb-cre; IP6K1^f/f mice and IP6K1^f/f mice were collected and applied to the cellular cholesterol efflux assay. The Alb-cre; IP6K1^f/f preparations displayed higher cholesterol efflux activities. Data represent mean±SEM, Student’s t-test, n=6 mice per group. (G), Oil Red O staining of the whole aorta. Data represent mean±SEM, Student’s t-test, n=6 mice per group. Scale bar 5mm. (H), Oil Red O staining of the aorta root. Data represent mean±SEM, Student’s t-test, n=5 mice per group. Scale bar 100 μm. (I), H&E staining of the aorta root. Data represent mean±SEM, Student’s t-test for plaque area, Mann-Whitney U test for necrotic core, n=5 mice per group. Scale bar 100 μm. J-L, The IP6K1^f/f; apoA-I KO mice and Alb-cre; IP6K1^f/f; apoA-I KO mice were injected with AAV-PCSK9 and fed with a Western diet. (J), The plasma of the IP6K1^f/f; apoA-I KO mice and the Alb-cre; IP6K1^f/f; apoA-I KO mice were collected and applied to the cellular cholesterol efflux assay. There were no significant differences between the two groups. Data represent mean±SEM, Student’s t-test, n=6 mice per group. (K), Oil Red O staining of the whole aorta. Data represent mean±SEM, Mann-Whitney U test (lesion%), Student’s t-test (plaque area), n=6 mice per group. Scale bar 5 mm. (L), H&E and Oil Red O staining of the aorta root. Data represent mean±SEM, Student’s t-test, n=5 mice per group. Scale bar 100 μm.

Fig. 7. Model of 5PP-InsP₅ depletion increasing apoA-I expression, augmenting reverse cholesterol transport, and attenuating atherosclerosis.

(Left) IP6K1 physiologically binds to apoA-I and UBE4A. A local pool of 5PP-InsP₅ was produced by IP6K1 to enhance the interaction of apoA-I with UBE4A. This leads to apoA-I ubiquitination and degradation. (Right) Depleting 5PP-InsP₅ by genetic deletion or pharmacological inhibition of IP6K1 disrupts the binding of UBE4A to apoA-I. This allows apoA-I to escape from degradation. ABCA1 interacts with apoA-I to facilitate the formation of nascent HDL, which is then released into the plasma. The higher levels of apoA-I are linked to enhanced reverse cholesterol transport activity, which reduces atherosclerosis.