Abstract
Cancer cells increase lipogenesis for their proliferation and the activation of sterol regulatory element-binding proteins (SREBPs) has a central role in this process. SREBPs are inhibited by a complex composed of INSIG proteins, SREBP cleavage-activating protein (SCAP) and sterols in the endoplasmic reticulum. Regulation of the interaction between INSIG proteins and SCAP by sterol levels is critical for the dissociation of the SCAP–SREBP complex from the endoplasmic reticulum and the activation of SREBPs1,2. However, whether this protein interaction is regulated by a mechanism other than the abundance of sterol—and in particular, whether oncogenic signalling has a role—is unclear. Here we show that activated AKT in human hepatocellular carcinoma (HCC) cells phosphorylates cytosolic phosphoenolpyruvate carboxykinase 1 (PCK1), the rate-limiting enzyme in gluconeogenesis, at Ser90. Phosphorylated PCK1 translocates to the endoplasmic reticulum, where it uses GTP as a phosphate donor to phosphorylate INSIG1 at Ser207 and INSIG2 at Ser151. This phosphorylation reduces the binding of sterols to INSIG1 and INSIG2 and disrupts the interaction between INSIG proteins and SCAP, leading to the translocation of the SCAP–SREBP complex to the Golgi apparatus, the activation of SREBP proteins (SREBP1 or SREBP2) and the transcription of downstream lipogenesis-related genes, proliferation of tumour cells, and tumorigenesis in mice. In addition, phosphorylation of PCK1 at Ser90, INSIG1 at Ser207 and INSIG2 at Ser151 is not only positively correlated with the nuclear accumulation of SREBP1 in samples from patients with HCC, but also associated with poor HCC prognosis. Our findings highlight the importance of the protein kinase activity of PCK1 in the activation of SREBPs, lipogenesis and the development of HCC.
Similar content being viewed by others
Main
INSIG proteins are anchor proteins of the endoplasmic reticulum (ER) that have two isoforms, INSIG1 and INSIG23,4. The binding of cholesterol-derived oxysterols, including 22-, 24-, 25- and 27-hydroxycholesterol, is crucial for the binding of INSIG proteins to SREBP cleavage-activating protein (SCAP), and for the retention of the SREBP–SCAP complex in the ER5,6. Under sterol-limiting conditions, the SREBP–SCAP complex is captured by COPII-coated vesicles and transported to the Golgi apparatus, where S1P and S2P proteases cleave SREBPs (SREBP1 or SREBP2) to yield active amino-terminal fragments for nuclear translocation and gene transcription1. In cancer cells, whether the interaction between INSIG proteins and the SREBP–SCAP complex is regulated by a mechanism that is independent of sterol abundance is unclear.
AKT-phosphorylated PCK1 binds to INSIG1/2
To investigate the regulation of INSIG proteins by oncogenic signalling, we treated Huh7 human HCC cells with insulin-like growth factor 1 (IGF1) for 1 h to induce the signalling that is critical for HCC development7. Mass spectrometric analyses of immunoprecipitates of INSIG1 and INSIG2 (Extended Data Fig. 1a, Supplementary Tables 1, 2) showed that IGF1 induced an association between INSIG1 or INSIG2 (hereafter, INSIG1/2) and cytosolic PCK1 (also known as PEPCK1). PCK1 is the rate-limiting enzyme of gluconeogenesis in the liver and kidney, and converts oxaloacetate and GTP into phosphoenolpyruvate and CO28. In humans, cytosolic PCK1 shares 63.4% sequence identity with mitochondrial PCK29. Co-immunoprecipitation and immunofluorescence analyses showed that PCK1, but not PCK2, bound to INSIG1/2 in IGF1-stimulated Huh7 and Hep3B HCC cells (Fig. 1a, Extended Data Fig. 1b, c), and that IGF1 induced the colocalization of PCK1, but not PCK2, with INSIG1 (Extended Data Fig. 1d). In addition, cell fractionation analyses showed that a small amount of PCK1, but not PCK2, translocated to the ER (Extended Data Fig. 1e, f). Thus, PCK1 translocates to the ER and binds to INSIG1/2 upon IGF1 stimulation.
a–e, Immunoprecipitation (IP) and western blotting (WB) analyses were performed as indicated three times with similar results. a, Huh7 cells were treated with or without IGF1 for 1 h. p indicates a phosphorylated residue. b, Huh7 cells expressing Flag–PCK1 were pretreated with or without the indicated inhibitors for 30 min before treatment with or without IGF1 for 1 h. c, In vitro kinase assays were performed by mixing purified wild-type (WT) His–PCK1 or His–PCK1(S90A) with or without purified GST–AKT1 in the presence of [γ-32P]ATP. d, Huh7 cells pretreated with or without MK-2206 for 30 min were treated with IGF1 for the indicated time periods. e, Parental Huh7 cells and the indicated clones with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 for 1 h. C1, clone 1; C2, clone 2.
To determine the mechanism that regulates the interaction between PCK1 and INSIG1/2, we inhibited the signalling pathway downstream of IGF1 in Huh7 cells (Extended Data Fig. 1g). We found that treatment with the AKT inhibitor MK-2206 or expression of a dominant-negative AKT mutant (AKT-DN) blocked the IGF1-induced binding of PCK1 to INSIG1/2 (Fig. 1b, Extended Data Fig. 1h), the translocation of PCK1 to the ER (Extended Data Fig. 1i, j) and its colocalization with INSIG1 (Extended Data Fig. 1k). By contrast, expression of a constitutively active form of AKT (myr-AKT) induced the binding of PCK1, but not PCK2, to INSIG1/2 (Extended Data Fig. 1l). These results indicate that AKT induces the translocation of PCK1 to the ER, where it binds to INSIG1/2.
Co-immunoprecipitation analyses showed that stimulating HCC cells with IGF1 induced an interaction between AKT and PCK1 (Extended Data Fig. 1m). We also performed a pull-down assay that revealed that purified active glutathione S-transferase (GST)-tagged AKT1 bound directly to His-tagged PCK1 (Extended Data Fig. 2a) through its catalytic domain (as evidenced by the expression of different AKT1 truncation mutants; Extended Data Fig. 2b). An in vitro phosphorylation assay (Extended Data Fig. 2c) and liquid chromatography–tandem mass spectrometry (LC–MS/MS) analysis (Extended Data Fig. 2d) showed that active, but not inactive, AKT1 phosphorylated PCK1 at the evolutionally conserved residue Ser90 (Extended Data Fig. 2e) in an AKT phosphorylation motif RXXS/T10 (Extended Data Fig. 2f). Mutation of Ser90 in PCK1 to alanine (S90A) (Fig. 1c) or treatment with MK-2206 (Extended Data Fig. 2g) abolished this phosphorylation, which was also detected using an antibody that is specific to PCK1 phosphorylated at Ser90 (PCK1(pS90)) (Extended Data Fig. 2c, h). Treatment with IGF1 rapidly induced the phosphorylation of PCK1 in Huh7 cells, and this was inhibited by pretreating cells with the AKT inhibitor MK-2206 (Fig. 1d). An RNA-interference-resistant (r) PCK mutant (rPCK1(S90A)) expressed in endogenous PCK1-depleted Huh7 cells was resistant to IGF1-induced phosphorylation of PCK1 Ser90 (Extended Data Fig. 2i) and did not exhibit myr-AKT1- or IGF1-induced translocation to the ER (Extended Data Fig. 2j, k). By contrast, the phosphorylation-mimicking PCK1(S90E) mutant accumulated in the ER without IGF1 stimulation (Extended Data Fig. 2k). Knock-in expression of PCK1(S90A) in HCC cells using CRISPR–Cas9-mediated genome editing (Extended Data Fig. 2l–n) blocked the IGF1-induced translocation of PCK1 to the ER and its colocalization with INSIG1 (Extended Data Fig. 2o, p). These results suggest that AKT1-mediated phosphorylation of PCK1 Ser90 is necessary and sufficient for the translocation of PCK1 to the ER.
Of note, AKT-mediated phosphorylation of PCK1 (Extended Data Fig. 2q–s) or the introduction of an S90E mutation (Extended Data Fig. 2t–v) reduced the binding affinity of PCK1 to oxaloacetate and its enzymatic activity (that is, the production of phosphoenolpyruvate). Thus, AKT-mediated phosphorylation of PCK1 and its translocation to the ER inhibit the canonical function of PCK1 in gluconeogenesis.
To determine the role of the phosphorylation of PCK1 Ser90 in its binding to INSIG1/2, we mixed purified INSIG1/2 with wild-type GST–PCK1 or GST–PCK1(S90A). Only AKT-phosphorylated wild-type PCK1 interacted with INSIG1/2 (Extended Data Fig. 2w). This interaction was abolished by treatment with calf intestinal alkaline phosphatase (CIP), which dephosphorylated PCK1(pS90) residue. Consistent with this, treatment with IGF1 induced the binding of INSIG1/2 to S protein–Flag–streptavidin-binding-peptide (SFB)-tagged wild-type PCK1, but not SFB–PCK1(S90A) (Extended Data Fig. 2x). This binding was abrogated by treatment with CIP (Extended Data Fig. 2x) or by knock-in expression of PCK1(S90A) in HCC cells (Fig. 1e, Extended Data Fig. 2y). Thus, phosphorylation of PCK1 Ser90 is required for the binding of PCK1 to INSIG1/2.
To determine the role of INSIG1/2 in the translocation of PCK1 to the ER, we depleted INSIG1/2 in Huh7 cells, and found that this blocked the translocation of PCK1 to the ER that was induced by IGF1 or by myr-AKT1 (Extended Data Fig. 2z). Expression of INSIG1/2 truncation mutants revealed that loop 1 of INSIG1/2 bound to PCK1 (Extended Data Fig. 3a). This interaction was not blocked in a PCK1(pS90) peptide (Extended Data Fig. 3b), which suggests that it is a conformational change in PCK1 (mediated by phosphorylation at Ser90) that causes the binding of PCK1 to INSIG1/2, rather than the phosphorylation of Ser90 directly.
PCK1 phosphorylates INSIG1/2
As PCK1 is able to transfer a phosphate group from GTP to a metabolite, we next investigated whether PCK1 phosphorylates INSIG1/2. In the presence of radiolabelled [γ-32P]GTP and active AKT1, only purified wild-type PCK1 and not PCK1(S90A) or PCK1(C288S)—a kinase-dead mutant that could still be phosphorylated at Ser90 and interacted with INSIG1/2 (Extended Data Fig. 3c)—phosphorylated purified INSIG1 (Fig. 2a) and INSIG2 (Extended Data Fig. 3d). By contrast, this phosphorylation did not occur in the presence of [γ-32P]ATP (Extended Data Fig. 3e). LC–MS/MS analyses showed that INSIG1 was phosphorylated at Ser207 (Extended Data Fig. 3f), which corresponds to Ser151 in INSIG2. Both of these residues are evolutionally conserved and are located in the cytosolic loop 2 of INSIG1/2 (Extended Data Fig. 3g, h). INSIG1(S207A) (Fig. 2b) and INSIG2(S151A) (Extended Data Fig. 3i) mutants were not phosphorylated by PCK1 in vitro (as detected by an antibody that specifically recognizes both phosphorylated residues; Extended Data Fig. 3j, k). Compared to wild-type PCK1, PCK1(S90E) exhibited a much higher velocity of enzyme-catalysed reaction at infinite concentration of substrate (Vmax) and lower Michaelis constant (Km) in phosphorylating an INSIG1 peptide at Ser207 (Extended Data Fig. 3l). Notably, PCK1(S90E) or PCK1(S90D) mutants phosphorylated Ser207 of INSIG1 and Ser151 of INSIG2 in the absence of AKT (Extended Data Fig. 3m, n). In addition, INSIG1(S207A) and INSIG2(S151A), expressed in Huh7 cells, were resistant to phosphorylation that was induced by IGF1 stimulation or myr-AKT1 expression (Extended Data Fig. 3o, p). In line with this, IGF1-induced phosphorylation of INSIG1/2 was also abolished by knock-in expression of PCK1(S90A) (Extended Data Fig. 3q), or by that of INSIG1(S207A)/INSIG2(S151A) (Fig. 2c, Extended Data Fig. 4a–d) in HCC cells. Thus, AKT-phosphorylated PCK1 functions as a protein kinase and uses GTP as a phosphate donor to phosphorylate INSIG1/2.
a–e, Immunoblotting analyses were performed as indicated three times with similar results. a, b, Bacterially purified His–PCK1 proteins on Ni-NTA agarose beads were incubated with or without active GST–AKT1 in the presence of ATP for an in vitro kinase assay. The beads were then washed and incubated with or without the indicated SFB–INSIG1 proteins in the presence of [γ-32P]GTP. Autoradiography was performed. c, Parental Huh7 and Hep3B cells and the indicated clones with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants (2S/A) were stimulated with or without IGF1 for 1 h. d, e, Flag and His (Flag/His)-tagged wild-type INSIG1 (d) or INSIG1(S207A) (e) immunoprecipitated and purified from Huh7 cells were incubated with the indicated GST–PCK1 proteins with or without active GST–AKT1 in the presence of ATP and GTP for 1 h. The INSIG proteins on Ni-NTA agarose beads were washed and incubated with 400 nM [3H]25-hydroxycholesterol. Specifically bound [3H]25-hydroxycholesterol was measured (right) (n = 6). Data are mean ± s.d. NS, not significant (P = 0.859, 0.930, 0.795, 0.768 (left to right) (d); P = 0.720, 0.630 (left to right) (e)); **P < 0.001 (two-tailed t-test).
INSIG1/2 phosphorylation reduces sterol binding
We next examined whether phosphorylation of INSIG1/2 affects the binding of INSIG proteins to oxysterols. Phosphorylation of wild-type INSIG1 and INSIG2 by AKT-phosphorylated wild-type PCK1 (Fig. 2d, Extended Data Fig. 4e), PCK1(S90D) or PCK1(S90E) (Extended Data Fig. 4f, g)—but not PCK1(C288S) or PCK1(S90A)—reduced the binding affinity of INSIG1/2 to [3H]25-hydroxycholesterol. By contrast, the binding affinity was unchanged for INSIG1(S207A) (Fig. 2e) and INSIG2(S151A) (Extended Data Fig. 4h). Similarly, wild-type INSIG1/2—but not INSIG1(S207A) or INSIG2(S151A) (Extended Data Fig. 4i, j)—that was immunoprecipitated from IGF1-treated parental Huh7 cells showed a reduction in binding to [3H]25-hydroxycholesterol, whereas the binding affinity of wild-type INSIG1/2 that was immunoprecipitated from Huh7 cells with knock-in expression of PCK1(S90A) was not affected (Extended Data Fig. 4k, l). This reduction in binding also occurred for the phospho-mimicking mutants INSIG1(S207E) and INSIG2(S151E) (Extended Data Fig. 4m, n). These results indicate that phosphorylation of INSIG1/2 by PCK1 reduces the binding of INSIG1/2 to oxysterols.
The PCK1–INSIG1/2 axis activates SREBP
The release of oxysterols from INSIG1/2 results in the disruption of the interaction between INSIG1/2 and SCAP, the translocation of the SCAP–SREBP1 complex from the ER to the Golgi apparatus and nuclear accumulation of SREBP15. As expected, stimulation with IGF1 disrupted the association between INSIG1/2 and SCAP (Fig. 3a, Extended Data Fig. 5a). This disruption was inhibited in Huh7 and Hep3B cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) (Fig. 3a, Extended Data Fig. 5a) or PCK1(S90A) (Extended Data Fig. 5b, c), or with reconstituted expression of rPCK1(S90A) (Extended Data Fig. 5d, e) or rPCK1(C288S) (Extended Data Fig. 5f). Cell fractionation analyses showed that IGF1 treatment induced the translocation of SCAP from the ER to the Golgi apparatus, and that this was inhibited by knock-in expression of INSIG1(S207A)/INSIG2(S151A) (Extended Data Fig. 5g, h) or PCK1(S90A) (Extended Data Fig. 5i, j). Expression of these mutants also blocked the IGF-induced loss of colocalization of SCAP with the ER protein calnexin (Fig. 3b, Extended Data Fig. 5k, n) and the colocalization of SCAP with the Golgi apparatus protein golgin-97 (Extended Data Fig. 5l, m, o). Similar results were obtained when we expressed Myc-tagged SCAP (Extended Data Fig. 5p, q).
a–c, Parental Huh7 cells and INSIG1(S207A)/INSIG2(S151A) knock-in cells were transfected with the indicated plasmids and stimulated with or without IGF1 for 16 h. Immunoprecipitation and immunoblotting analyses (a) and immunofluorescence analyses (b, c) were performed as indicated. The arrows in b indicate Golgi-localized SCAP. The experiments were repeated three times independently with similar results. d, The mRNA expression levels of SREBP1 target genes were measured using quantitative PCR. *P = 0.009, 0.002 (left to right); **P < 0.001 (two-tailed t-test). e, The incorporation of 14C-glucose into triglycerides (TGs) (left) and fatty acids (FAs) (right) was measured. *P = 0.01, 0.03 (left to right); **P < 0.001 (two-tailed t-test). Data in d, e are mean ± s.d. (n = 6 biological replicates).
Stimulation with IGF1 increased the cleavage of SREBP1 (Extended Data Figs. 5r, 6a) and its nuclear accumulation (Fig. 3c, Extended Data Fig. 6b); SRE-promoter-driven luciferase activity (Extended Data Figs. 5s, 6c); and the expression of mRNA and proteins of SREBP1 target genes that are associated with lipogenesis, including fatty acid synthase (FASN), acetyl-CoA carboxylase alpha (ACACA, also known as ACC1), stearoyl-CoA desaturase (SCD, also known as SCD1) and glycerol-3-phosphate acyltransferase 1 (GPAM, also known as GPAT), as well as SREBF1 (which encodes SREBP1)1,11,12 (Fig. 3d, Extended Data Figs. 5t, 6d–f). Consequently, treatment with IGF1 resulted in increased incorporation of 14C-glucose into triglycerides and fatty acids (Fig. 3e, Extended Data Fig. 6g). Notably, all of these IGF1-induced effects were inhibited by knock-in expression of INSIG1(S207A)/INSIG2(S151A) (Fig. 3c–e, Extended Data Figs. 5r–t, 6f) or PCK1(S90A) (Extended Data Fig. 6a–g). Furthermore, the IGF1-induced incorporation of 14C-glucose into fatty acids was strongly blocked by expression of PCK1(S90A) in CHL-1 melanoma cells, U87 glioblastoma cells and H1993 non-small-cell lung cancer cells (Extended Data Fig. 6h), suggesting that the lipogenesis that is promoted by phosphorylation of PCK1 Ser90 occurs in several types of cancer. Consistently, knock-in expression of PCK1(S90D) (Extended Data Figs. 2l–n, 6i) or INSIG1(S207D)/INSIG2(S151D) (Extended Data Figs. 4a–d, 6j) in HCC cells was sufficient to induce SREBP1 cleavage.
Similar to our observations with SREBP1, expression of PCK1(S90A) or INSIG1(S207A)/INSIG2(S151A) also inhibited IGF1-induced cleavage of SREBP2 (Extended Data Fig. 6k, l) and SREBP2-mediated transcription of genes that are related to cholesterol biogenesis13, such as 3-hydroxy-3-methylglutaryl-CoA (HMGC) reductase (HMGCR), HMGC synthase 1 (HMGCS1, also known as HMGCS), low-density lipoprotein receptor (LDLR) and squalene synthase (FDFT1, also known as SS) (Extended Data Fig. 6m, n). Thus, PCK1-mediated phosphorylation of INSIG1/2 promotes the activation of both SREBP1 and SREBP2, and the expression of downstream genes that are involved in lipogenesis.
Because PCK1-mediated phosphorylation of INSIG1/2 promotes the release of sterols from their binding to INSIG1/2, we postulated that this phosphorylation does not affect the activation of SREBP1 that is induced by lipid depletion. As expected, knock-in expression of INSIG1(S207A)/INSIG2(S151A) or PCK1(S90A) did not affect the lipid-depletion-induced translocation of SCAP from the ER to the Golgi apparatus (Extended Data Fig. 6o–r), loss of colocalization of SCAP with calnexin (Extended Data Fig. 6s–u) and colocalization of SCAP with golgin-97 (Extended Data Fig. 6v–x), the cleavage and nuclear accumulation of SREBP1 (Extended Data Fig. 7a, b), and SRE-driven luciferase activity (Extended Data Fig. 7c). By contrast, incubation of HCC cells with 25-hydroxycholesterol—but not cholesterol—at a dosage (120 nM) much higher than physiological concentrations (40 nM)14 blocked the dissociation of INSIG1/2 from SCAP, the cleavage of SREBP and the increase of SRE luciferase activity that were induced by IGF1 (Extended Data Fig. 7d, f) or by expression of phosphorylation-mimicking PCK1 and INSIG1/2 mutants (Extended Data Fig. 7e, g). These results demonstrate that PCK1-mediated phosphorylation of INSIG1/2 promotes SREBP1 activation through the release of 25-hydroxycholesterol from INSIG1/2.
SREBP1 can be activated both in a manner that depends on AKT and mTORC1, and independently of AKT and mTORC115,16,17. It has been previously shown that insulin signalling and AKT activation suppress the expression of INSIG2 by promoting the decay of its mRNA17. In addition, INSIG1, but not INSIG2, is ubiquitinated and degraded upon sterol depletion18. A time-course experiment showed that INSIG2 was not obviously downregulated until prolonged treatment with IGF1 (24–48 h) (Extended Data Fig. 7h), and this downregulation was not affected in PCK1(S90A) or INSIG1(S207A)/INSIG2(S151A) mutants (Extended Data Fig. 7i). By contrast, treatment of HCC cells with IGF1 and cycloheximide (CHX)—which eliminates the translational regulation of protein expression—induced rapid degradation of wild-type INSIG1, but not INSIG1(S207A), wild-type INSIG2 or INSIG2(S151A) (Extended Data Fig. 7j, k). In addition, the half-life of INSIG1(S207D), but not INSIG2(S151D), was shortened (Extended Data Fig. 7l, m), indicating that phosphorylation of INSIG1 Ser207 promotes the degradation of the INSIG1 protein. However, the overall expression of INSIG1 in the presence of IGF1 and absence of CHX was not obviously altered (Extended Data Fig. 7h). This might be a result of an increase in INSIG1 transcription owing to SREBP activation19, which compensates for the degradation of INSIG1 in a feedback manner. In contrast to the delayed response of INSIG2 expression to IGF1 stimulation, immediate phosphorylation of PCK1 and INSIG1/2 (Extended Data Fig. 7n) and rapid cleavage of SREBP1 (Extended Data Fig. 7h) were detected after IGF1 treatment. Thus, PCK1-mediated rapid phosphorylation of INSIG1/2 and activation of SREBP1 is an immediate response to AKT activation.
De novo fatty acid synthesis occurs in the liver, adipose tissue and lactating breast20. Treatment of HL7702 and THLE-2 normal hepatocytes with IGF1 induced phosphorylation of PCK1 Ser90, subsequent phosphorylation of INSIG1 Ser207 and INSIG2 Ser151, and SREBP1 cleavage (Extended Data Fig. 8a). The latter effect was also induced by expression of PCK1(S90D) in these cells (Extended Data Fig. 8b), which suggests that PCK1-induced activation of SREBP1 also occurs in normal hepatocytes.
Under normal physiological conditions, high blood glucose levels after a meal increase the pancreatic secretion of insulin, which immediately activates the phosphoinositide 3-kinase (PI3K)–AKT signalling pathway and leads to an increase in glucose utilization and a reduction in gluconeogenesis in the liver21. Of note, phosphorylation of AKT, PCK1 Ser90, INSIG1 Ser207 and INSIG2 Ser151, and cleavage of SREBP1, were markedly enhanced in normal liver from mice that were refed with glucose after 24 h of fasting (Extended Data Fig. 8c)—suggesting that in vivo blood glucose levels regulate the PCK1-mediated phosphorylation of INSIG1/2 and the activation of SREBP1 in the liver. These results demonstrate the relevance of our findings in the context of the physiological functions of the liver.
We then examined a potential difference in the PCK1-mediated activation of SREBP1 between normal hepatocytes and HCC cells. We found that there was a substantial increase in the phosphorylation of AKT, PCK1 Ser90, INSIG1 Ser207 and INSIG2 Ser151—and in the cleavage of SREBP1—in HCC cells compared with normal human hepatocytes (Extended Data Fig. 8d). These increases were reduced by expression of PCK1(S90A) in HCC cells (Extended Data Fig. 8e), indicating that PCK1-mediated SREBP1 activation is increased in HCC cells with a high level of AKT activation.
Notably, these increases in the phosphorylation of AKT, PCK1 and INSIG1/2, and in the cleavage of SREBP1, were further enhanced or induced in Huh7, CHL-1, U87 and H1993 cancer cells that express the active KRAS(G12V) mutant (Extended Data Fig. 8f, g), a mutant version of the IGF1 receptor (IGF1R) (IGF1R(V922E)) (Extended Data Fig. 8h) or a mutant version of the epidermal growth factor receptor (EGFR) (EGFRvIII) (Extended Data Fig. 8i), and in cells treated with platelet-derived growth factor (PDGF) (Extended Data Fig. 8j). Expression of PCK1(S90A) (Extended Data Fig. 8f–j) or INSIG1(S207A)/INSIG2(S151A) (Extended Data Fig. 8f, h) inhibited INSIG1/2 phosphorylation and SREBP1 cleavage in these cells. These results suggest that PCK1-mediated activation of SREBP1 is induced by AKT activation that is elicited by different oncogenes or growth factors and occurs in several types of cancer.
The PCK1–INSIG1/2 axis promotes HCC growth
Notably, reconstituted expression of PCK1(S90A) in different HCC cell lines—which reduced the basal level of phosphorylation of INSIG1 Ser207 and INSIG2 Ser151 and the cleavage of SREBP1 (Extended Data Fig. 8k)—inhibited the proliferation of cells under normal culture conditions (Extended Data Fig. 8m), without altering the cleavage of SREBP1 and the decrease in cell survival induced by lipid depletion (Extended Data Fig. 8l, n). Similarly, knock-in expression of INSIG1(S207A)/INSIG2(S151A) or PCK1(S90A) inhibited the proliferation of Huh7 cells (Fig. 4a) and Hep3B cells (Extended Data Fig. 9a). By contrast, expression of PCK1(S90D) or INSIG1(S207D)/INSIG2(S151D) mutants enhanced cell proliferation (Extended Data Fig. 9b, c).
a, Parental Huh7 cells (2 × 105) and the indicated clones with knock-in expression of PCK1(S90A) (left) or INSIG1(S207A)/INSIG2(S151A) (right) were plated for 3 days. The cells were then collected and counted. **P < 0.001 (two-tailed t-test). b, Huh7 cells with or without knock-in expression of PCK1(S90A) (top) or INSIG1(S207A)/INSIG2(S151A) (bottom) were intrahepatically injected into athymic nude mice (n = 7 per group). Tumour growth was examined 28 days after injection. The arrows indicate tumours. Tumour volumes were calculated (right). *P = 0.03, 0.02 (left to right, top);*P = 0.02, 0.03 (left to right, bottom) (two-tailed t-test). Data in a, b are mean ± s.d. c–e, IHC analyses of xenograft tumours from nude mice (n = 7) (c, d) and 90 human HCC samples (e) were performed with the indicated antibodies. Representative staining images are shown. The regions in white boxes are shown at higher magnification below. f, Kaplan–Meier plots of the overall survival rates in 90 patients with HCC grouped according to high (staining score, 4–8) and low (staining score, 0–3) expression of PCK1(pS90) (top) and INSIG1(pS207)INSIG2(pS151) (bottom) (n = 38 (PCK1(pS90) low), n = 52 (PCK1(pS90) high); n = 39 (INSIG1(pS207)/INSIG2(pS151) low), n = 51 (INSIG1(pS207)/INSIG2(pS151) high). P values were calculated using a log-rank test (two-tailed).
Next, we intrahepatically (Fig. 4b) and subcutaneously (Extended Data Fig. 9d–f) injected Huh7 cells with or without knock-in expression of PCK1(S90A) or INSIG1(S207A)/INSIG2(S151A) into nude mice. Expression of these mutant proteins substantially inhibited tumour growth in the mice (Fig. 4b, Extended Data Fig. 9d–f), with a corresponding reduction in Ki67 expression (Extended Data Fig. 9g) and an increase in cell apoptosis (Extended Data Fig. 9h). In addition, expression of PCK1(S90A) reduced the phosphorylation of INSIG1 Ser207 and INSIG2 Ser151, and reduced the expression and nuclear distribution of SREBP1 (Fig. 4c, Extended Data Fig. 9i, j, left). In line with this, expression of INSIG1(S207A)/INSIG2(S151A) also decreased the nuclear levels of SREBP1 (Fig. 4d, Extended Data Fig. 9i, j, right). By contrast, expression of PCK1(S90D) or INSIG1(S207D)/INSIG2(S151D) promoted tumour growth (Extended Data Fig. 9k–m).
Consistent with the finding that AKT–PCK1 signalling activates SREBP1, expression of active IGF1R(V922E) (Extended Data Fig. 10a–c) or myr-AKT (Extended Data Fig. 10d–f) in Huh7 cells significantly increased the growth of liver tumours, with enhanced phosphorylation of PCK1 Ser90 and INSIG1 Ser207/INSIG2 Ser151 and enhanced nuclear expression of SREBP1. Knock-in expression of PCK1(S90A) or INSIG1(S207A)/INSIG2(S151A) inhibited both basal tumour growth and that induced by IGF1R(V922E) (Extended Data Fig. 10a–c) or myr-AKT (Extended Data Fig. 10d–f). By contrast, expression of a dominant-negative IGF1R(L1003R) mutant reduced both intrahepatic (Extended Data Fig. 10g, h) and subcutaneous (Extended Data Fig. 10i–n) tumour growth, with reduced phosphorylation of PCK1 and INSIG1/2 (Extended Data Fig. 10o) and reduced nuclear accumulation of SREBP1 (Extended Data Fig. 10p, q). The reduction in tumour growth (Extended Data Fig. 10g–n) and nuclear SREBP1 expression (Extended Data Fig. 10p, q) was partially reverted by expression of PCK1(S90D) or INSIG1(S207D)/INSIG2(S151D).
Furthermore, we used hydrodynamics-based transfection in mice to administer plasmids for the expression of active myr–AKT, c-Met and the sleeping beauty transposase (which induces rapid liver tumour growth22) together with wild-type PCK1 or PCK1(S90A). Expression of PCK1(S90A) reduced tumour growth, Ki67 expression, INSIG1/2 phosphorylation and nuclear SREBP1 expression (Extended Data Fig. 10r–t). Thus, PCK1-mediated phosphorylation of INSIG1/2 and subsequent activation of SREBP1 promote the development of HCC.
To determine the clinical relevance of PCK1-regulated SREBP1 activation, we performed immunohistochemistry (IHC) analyses of 30 paired samples of primary HCC and adjacent normal tissue. Phosphorylation of PCK1 Ser90 and INSIG1 Ser207/INSIG2 Ser151—and nuclear SREBP1 expression—were markedly increased in the HCC specimens compared with normal tissue (Extended Data Fig. 10u–w), and correlated with each other in 90 resected HCC tumours (Fig. 4e, Extended Data Fig. 10x, y). Notably, in HCC samples, high levels of phosphorylation of PCK1 Ser90 and INSIG1 Ser207/INSIG2 Ser151—and high levels of nuclear SREBP1 expression—were correlated with decreased overall durations of survival in patients with HCC (Fig. 4f). These results suggest that PCK1-mediated phosphorylation of INSIG1/2 has a critical role in the clinical aggressiveness of human HCC.
Metabolism and gene expression are two fundamental cellular processes that are essential for the proliferation of tumour cells and that can be mutually regulated23. PCK1 was originally characterized as a gluconeogenesis enzyme. Herein we have shown that PCK1 has protein kinase activity and translocates to the ER to regulate SREBP1 activation and SREBP1-mediated gene expression (Extended Data Fig. 10z). We also demonstrated that a metabolic enzyme uses GTP, rather than ATP, as a phosphate donor to phosphorylate a protein substrate. In addition, we have shown that oncogenic signalling can rapidly modulate the association between INSIG1/2 and sterol without reducing total cellular sterol levels. Our findings highlight the potential for inhibition of the protein kinase activity of PCK1 as a treatment strategy in human HCC.
Methods
Materials
Normal mouse IgG (sc-2025), normal rabbit IgG (sc-2027), GST (sc-138), tubulin (sc-8035), ERK1/2 (sc-514302) and INSIG1 (A-9) (sc-390504) (for immunoprecipitation or immunoblotting) antibodies were obtained from Santa Cruz Biotechnology. PCK1 (16754-1-AP) (for immunoprecipitation or immunoblotting), INSIG1 (22115-1-AP) (for immunofluorescence or immunoblotting) and INSIG2 (24766-1-AP) (for immunoprecipitation or immunoblotting) antibodies were purchased from Proteintech. Anti-AKT(pS473) (4060), rabbit AKT (9272), mouse AKT (2920), c-SRC (2108), haemagglutinin (HA) (3724), p44/42 MAPK (ERK1/2) (9102), c-Jun (9165), c-Jun(pS73) (3270), Myc-tag (9B11) (2276) and PCK1 (D12F5) (12940) antibodies (for immunoblotting) were purchased from Cell Signaling Technology. Rabbit antibodies that recognize PCK1(pS90), INSIG1(pS207) and INSIG2(pS151) were obtained from Signalway Biotechnology. IGF1 (85580C), PDGF (P8147), cyclohexamide (CHX) (01810), oxaloacetate (O7753), cholesterol (C8667), 25-hydroxycholesterol (H1015), mouse monoclonal anti-Flag (F1804), rabbit anti-Flag (F7425) and anti-His (SAB1305538) antibodies were purchased from Sigma-Aldrich. Anti-Flag M2 agarose beads were purchased from MP Biochemicals. [3H]25-hydroxycholesterol, [γ-32P]ATP and [γ-32P]GTP were obtained from PerkinElmer. Inactive AKT1 protein (14-279-D) and rabbit anti-Ki67 antibody (AB9260) were obtained from Millipore. Oxaloacetate assay kit (ab83428), anti-Myc tag rabbit (ab9106), calnexin (ab22595), SRC(pY418) (ab4816), SCAP (ab91323), golgin-97 (ab84340), SCD1 (CD.E10) (ab19862), FASN (ab22759) and anti-rabbit IgG heavy chain (HRP) (ab99702) antibodies were purchased from Abcam. U0126, SP600125, MK-2206 and SU6656 were purchased from EMD Biosciences. Active GST–AKT1 (A16-10G) was obtained from SignalChem. Active recombinant His–AKT1 protein (LS-G18427) was obtained from Lifespan Biosciences. SREBP1 (IgG 2A4) and SREBP2 antibodies (557037) for immunoblotting analyses were purchased from BD Biosciences. SREBP1 antibody (2A4) (NB100-2215) (for immunofluorescence and IHC analyses) and PCK1 mouse antibody (3E4) (H00005105-M1) (for immunofluorescence or immunoblotting) were purchased from Novus. INSIG2 (PA5-41707) and INSIG1 (PA5-97876) antibodies (for immunoblotting), calnexin (AF18) (MA3-027), golgin-97 (CDF4) (A-21270), glutathione agarose, 4′, 6-diamidino-2-phenylindole (DAPI), Alexa Fluor 488 goat anti-rabbit (A11008), Alexa Fluor 594 goat anti-rabbit (A11012), Alexa Fluor 488 goat anti-mouse (A11029) and Alexa Fluor 594 goat anti-mouse antibodies (A11005) were obtained from Thermo Fisher Scientific. The phosphoenolpyruvate carboxykinase activity assay kit (K359) was purchased from BioVision. PCK1(pS90) peptide (DVARIE-pS-KTVIVT), INSIG1/2(pS207/S151) peptide (WWTFDR-pS-RSGLGL) and PCK1(S90) covering peptide (WWTFDRSRSGLGL) were synthesized by SelleckChem. CIP was obtained from New England BioLabs. Ni-NTA agarose was obtained from Qiagen.
DNA construction and mutagenesis
PCR-amplified human wild-type PCK1, PCK2, INSIG1 and INSIG2 and IGF1R were cloned into pcDNA3.1/hygro(+)-Flag, -HA, -His, or -Myc, pCDH-CMV-MCSEF1-Puro-SFB or pET32a vectors. SCAP was cloned into pcDNA3.1/hygro(+)-Myc. pECE-Myr-HA-AKT1(delta4-129), MSCV-XZ066-EGFR vIII and pBabe-puro-KRAS(G12V) were purchased from Addgene. EGFR vIII and KRAS(G12V) were cloned into pcDNA3.1/hygro(+)-Flag. Flag/His-double-tagged INSIG1 and INSIG2 were constructed in a pFastBacHTa expression vector (Invitrogen) as described24. pT3-EF1a-c-Met was a gift from X. Chen (Addgene plasmid 31784). pCMV(CAT)T7-SB100 was a gift from Z. Izsvak (Addgene plasmid 34879). Flag–PCK1 (wild-type or S90A-mutant) and HA–myr-AKT were cloned into a pT3-EF1α vector. Myc-SCAP was a gift from Y. Chen at Shanghai Institute of Biological Sciences, Chinese Academy of Sciences.
pcDNA3.1 Flag–rPCK1, Flag–rPCK1(S90A), Flag–rPCK1(S90E), SFB–PCK1(S90A), pGEX-4T-1 PCK1(S90A), HA-AKT-DN (K179A, T308A, S473A), Flag–rPCK1(C288S), HA–rPCK1(C288S), PGEX-4T-1 PCK1(C228S), pET22b PCK1(S90A), pET22b PCK1(C288S), SFB–INSIG1(S207A), SFB–INSIG2(S151A), Flag–IGF1R(V922E), Flag–IGF1R(L1003R), Flag/His-double-tagged INSIG1(S207A) and INSIG1(S207E), Flag–INSIG1(S207D), Flag/His-double-tagged INSIG2(S151A) or INSIG2(S151E), Flag–INSIG2(S151D) and short hairpin RNA (shRNA)-resistant PCK1 constructs containing nonsense mutations of G948A, T951G, C954T and A957C were constructed using a QuikChange site-directed mutagenesis kit (Stratagene). pGIPZ shRNA was constructed via ligation of an oligonucleotide targeting human PCK1 into an XhoI/MluI-digested pGIPZ vector. The following pGIPZ shRNA target sequences were used: control shRNA oligonucleotide, 5′-GCTTCTAACACCGGAGGTCTT-3′; PCK1 shRNA oligonucleotide, 5′- TGTGCGTCAAACTTCATCC-3′. INSIG1 shRNA oligonucleotides, 5′- TAATGGTGTCTATCAGTATAC-3′ and 5′- GGAACATAGGACGACAGTTA-3′; INSIG2 shRNA oligonucleotides, 5′-CATCTAGGAGAACCTCATAAA-3′ and 5′- CTTCAGCTGTGATTGGGTT -3′; SCAP shRNA oligonucleotides, 5′- CTCTTCAGCTATTACAACA -3′ and 5′- AGGAAGAGGATGGTCTCCT -3′.
Cell lines and cell culture conditions
Hep3B, Huh7, H1993, CHL-1, SNU-398, SNU-475, HL7702, THLE-2 and 293T cells were from ATCC. The cells were maintained in complete medium containing Dulbecco’s modified Eagle’s medium (DMEM), 10% fetal bovine serum (FBS), 1,000 U ml−1 penicillin and 100 μg ml−1 streptomycin. Lipid-depleted medium contains DMEM supplemented with 5% LPDS and lovastatin (5 μM). Before IGF1 (100 ng ml−1) treatment, the cells were serum-starved for 16 h. U0126 (20 μM), SU6656 (4 μM), SP600125 (25 μM), or MK-2206 (10 μM) was added 30 min before treatment with or without IGF1 (100 ng ml−1) for 1 h. No cell lines used in this study were found in the database of commonly misidentified cell lines that is maintained by the International Cell Line Authentication Committee and NCBI Biosample. Cell lines were authenticated by short tandem repeat profiling and were routinely tested for mycoplasma contamination at The University of Texas MD Anderson Cancer Center. Cells were plated at a density of 4 × 105 per 60-mm dish or 1 × 105 per well of a 6-well plate 18 h before transfection. The transfection procedure was performed as previously described1,2.
Immunoprecipitation and immunoblotting analysis
The extraction of proteins using a modified buffer from cultured cells was followed by immunoprecipitation and immunoblotting using corresponding antibodies as described previously1,2.
GST pull-down assay
Equal amounts of His-tagged purified protein (200 ng per sample) were incubated with 100 ng of GST fusion proteins together with glutathione agarose beads in a modified binding buffer (50 mM Tris-HCl at pH 7.5, 1% Triton X-100, 150 mM NaCl, 1 mM DTT, 0.5 mM EDTA, 100 μM PMSF, 100 μM leupeptin, 1 μM aprotinin, 100 μM sodium orthovanadate, 100 μM sodium pyrophosphate, 1 mM sodium fluoride). The glutathione agarose beads were then washed four times with binding buffer and then subjected to immunoblotting analysis as previously described25.
Purification of recombinant proteins
Wild-type GST–PCK1, GST–PCK1(S90A), GST–PCK1(C288S), His–PCK1, His–PCK1(S90A), His–PCK1(C228S) were expressed in bacteria and purified as described previously26. Flag/His-double-tagged INSIG1 and INSIG2 were purified as previously described5.
[3H]25-hydroxycholesterol binding assay
[3H]25-hydroxycholesterol binding reactions were performed as previously described24. In brief, each reaction was performed in a final volume of 100 μl of buffer A (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 1 mM dithiothreitol, 0.1% Fos-choline 13, 0.005% sodium azide). It contained 1.2 μg (400 nM) of purified wild-type or mutants of Flag/His-double-tagged INSIG1 or INSIG2, 10–500 nM [3H]25-hydroxycholesterol and 25 mM phosphocholine chloride. After incubation for 4 h at room temperature, the mixture was passed through a column packed with 0.3 ml of Ni-NTA agarose beads (Qiagen). Each column was then washed 3 times with 10 ml of buffer B (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 1 mM dithiothreitol and 0.1% (w/v) Anapoe-C12E9). The protein-bound [3H]25-hydroxycholesterol was eluted with 250 mM imidazole and measured by scintillation counting.
Cell fractionation
The cell fraction assay was performed according to a previously reported method27. In brief, 40 10-cm dishes of parental Huh7 cells and the indicated clones of cells with INSIG1(S207A)/INSIG2(S151A) double knock-in expression or PCK1(S90A) knock-in expression were stimulated with or without IGF1 (100 ng ml−1) or incubated with or without lepid-depleted medium for 16 h. The cells were then placed on ice and washed twice with PBS and homogenization buffer (10 mM triethanolamine-acetic acid, pH 7.4, 0.25 M sucrose, 1 mM sodium EDTA, protease inhibitor cocktail (Roche)). The washed cells were collected in 0.8 ml homogenization buffer and homogenized by passing through a 25-gauge needle on a 1-ml syringe 13 times. After centrifugation at 2,000g for 15 min at 4 °C, the post-nuclear supernatant was collected and loaded on preformed iodixanol (Sigma-Aldrich) gradients. The discontinuous iodixanol gradients were prepared as 2.65 ml of 24%, 19.33%, 14.66% and 10%, which were made by diluting a 60% stock of iodixanol with cell suspension medium (0.85% (w/v) NaCl, 10 mM Tricine-NaOH, pH 7.4). After standing at room temperature for 2 h, the gradients were then centrifuged at 37,000 rpm in a SW40Ti rotor (Beckman Instruments) for 4 h. The post-nuclear supernatant was loaded on the top of the gradients and centrifuged at 37,000 rpm for another 2 h. Deceleration was performed without a brake. A total of 15 fractions (800 μl per fraction) were collected from the top to the bottom, and the bottom two fractions containing aggregated material were not analysed further. Aliquots of each fraction were used for further analysis by immunoblotting. ER fractions were isolated from the cells using an Endoplasmic Reticulum Isolation Kit (ER0100, Sigma-Aldrich). ER proteins were used in immunoblot analyses.
Conversion of radiolabelled glucose to triglycerides and fatty acids
The incorporation of the various radioactive substrates into triglycerides and fatty acids was measured as previously described28. In brief, cells were washed twice with PBS and incubated in 1 ml labelling medium (2.5% fatty-acid-free bovine serum albumin, 1% (v/v) penicillin/streptomycin, 0.5 mM d-glucose, 0.5 mM sodium acetate, 2 mM sodium pyruvate, 2 μCi ml−1 14C-U-glucose or 14C-acetate) at 37 °C in a humidified incubator (5% CO2) for 4.5 h before lipid extraction. All metabolic processes were stopped by washing cells twice with cold PBS and lysing them with the addition of modified Dole’s extraction mixture (80 ml isopropanol, 20 ml hexane, 2 ml 0.5 M H2SO4). Triglycerides were extracted with hexane and washed, and the solvent was evaporated. The incorporation of 14C-glucose into fatty acids and triglycerides was determined by evaporating the solvent from neutral lipids, adding 1 ml of KOH-ethanol (20 ml 95% ethanol, 1 ml water, 1 ml saturated KOH) and heating samples to 80 °C for 1 h. Sulfuric acid was added to the mixture to ensure complete saponification. Addition of hexane allowed for hydrophobic separation. The hydrophilic portion was evaporated and counted using liquid scintillation. Incorporation data were normalized according to cell number.
In vitro kinase assay
In vitro kinase assays were performed as previously reported29. In brief, for the AKT in vitro kinase assay, purified active GST–AKT (A16-10G, SignalChem) (500 ng) was incubated with bacterially purified His–PCK1 (200 ng) in 25 μl kinase buffer (50 mM Tris-HCl at pH 7.5, 100 mM KCl, 50 mM MgCl2, 1 mM Na3VO4, 1 mM DTT, 5% glycerol, 0.5 mM ATP and 10 μCi [γ-32P]ATP) at 25 °C for 1 h. The reaction was terminated by adding SDS–PAGE loading buffer and heated at 100 °C for 5 min. The reaction mixture was then subjected to an SDS–PAGE analysis. For PCK1 in vitro kinase assay, bacterially purified wild-type His–PCK1 or His–PCK1(S90A) on the Ni-NTA agarose beads was incubated with or without GST–AKT1 in the presence of ATP for 1 h. After the in vitro AKT kinase assay, the Ni-NTA agarose beads were washed in PBS five times and incubated with or without SFB-tagged wild-type or mutant INSIG1/2 using 50 μl kinase buffer in the presence of 0.5 mM GTP and 10 μCi [γ-32P]ATP or [γ-32P]GTP at 25 °C for 1 h. Autoradiography was performed.
Mass spectrometry analyses
For identification of interacting proteins, a protein band visualized via Coomassie blue staining was excised from an SDS–PAGE gel and digested in gel in 50 mM ammonium bicarbonate buffer containing RapiGest (Waters Corporation) overnight at 37 °C with 200 ng of modified sequencing-grade trypsin (Promega). The digested protein samples were analysed using high-sensitivity LC–MS/MS with an Orbitrap Elite mass spectrometer (Thermo Fisher Scientific). Proteins were identified by searching the fragment spectra against the UniProt protein database (EMBL-EBI) using the Mascot search engine (v.2.3; Matrix Science) with the Proteome Discoverer software program (v.1.4; Thermo Fisher Scientific). For detection of phosphorylation sites, in vitro phosphorylation of PCK1 by AKT and INSIG1/2 by PCK1 were performed according to the in vitro kinase assay protocol described above. Then the protein samples were digested using trypsin or chymotrypsin and analysed using LC–MS/MS with the Orbitrap Elite mass spectrometer as described previously30.
Determining the K m of PCK1
The Km of PCK1 was determined using a GTPase assay kit (MAK113, Sigma-Aldrich) according to the manufacturer’s instructions. In brief, purified recombinant wild-type PCK1 (10 ng) was incubated in 100 μl of reaction buffer (40 mM Tris-HCl at pH 7.5, 80 mM NaCl, 10 mM synthetic INSIG1 peptide substrate (LWWTFDRSRSGLGLG), 8 mM magnesium acetate, 1 mM DTT) with different concentrations of GTP at 37 °C in 96-well plates. The plates were read by multi-detection microplate readers (BMG Labtech) at 620 nm in kinetic mode for 5 min. Km and Vmax were calculated from a plot of 1/V versus 1/[substrate] according to the Lineweaver–Burke plot model.
Determination of oxaloacetate binding affinity and detection of PCK1 activity
Ni-NTA beads and purified recombinant wild-type His–PCK1, His–PCK1(S90E) or AKT-phosphorylated His–PCK1 were incubated in 200 μl binding buffer (50 mM Tris-HCl at pH 7.5, 100 mM KCl, 50 mM MgCl2, 50 mM MnCl2, 1 mM Na3VO4) containing 200 μM oxaloacetate (O7753, Sigma-Aldrich) at 30 °C for 30 min. The oxaloacetate associated with His–PCK1 on the beads was then centrifuged at 12,000 rpm for 10 min. The oxaloacetate remaining in the supernatant was quantified using an oxaloacetate assay kit (ab83428, Abcam). The oxaloacetate binding affinity was calculated according to the percentages of the remaining oxaloacetate in the supernatants. The activity of PCK1 was determined using a phosphoenolpyruvate carboxykinase activity assay kit (K359, BioVision).
CRISPR–Cas9-mediated genome editing
Genomic mutations were introduced into cells using the CRISPR–Cas9 system, as described previously30. Single-guide RNAs (sgRNAs) were designed to target the genomic area adjacent to mutation sites in PCK1(S90A), INSIG1(S207A) and INSIG2(S151A) using the CRISPR design tool (http://crispr.mit.edu/). The annealed guide RNA oligonucleotides were inserted into a PX458 vector (Addgene) digested with the BbsI restriction enzyme31. Cells were seeded at 60% confluence, followed by co-transfection of sgRNAs (0.5 μg) and single-stranded donor oligonucleotide (10 pmol) as a template to introduce mutations. Twenty-four hours after transfection, cells were trypsinized, diluted for single cells and seeded into 96-well plates. Genomic DNA was extracted from GFP-positive cells, followed by sequencing of the PCR products spanning the mutation sites. sgRNA targeting sequence for INSIG1(S207A): 5′-ACATTTGATCGTTCCAGAAG-3′; single-stranded donor oligonucleotide (ssODN) sequence for INSIG1(S207A): 5′-AAATTGGATTTTGCCAATAATGTCCAGCTGTCCTTGACTTTAGCAGCCCTATCTTTGGGCCTTTGGTGGACATTTGATCGcgCCAGgAGcGGCCTTGGGCTGGGGATCACCATAGCTTTTCTAGCTACGCTGATCACGCAGTTTCTCGTGTATAATGGTGTCTATCA-3′; ssODN sequence for INSIG1(S207D): 5′-AAATTGGATTTTGCCAATAATGTCCAGCTGTCCTTGACTTTAGCAGCCCTATCTTTGGGCCTTTGGTGGACATTTGATCGcgaCAGgAGcGGCCTTGGGCTGGGGATCACCATAGCTTTTCTAGCTACGCTGATCACGCAGTTTCTCGTGTATAATGGTGTCTATCA-3′; sgRNA targeting sequence for INSIG2(S151A): 5′-ACTTTTGATAGATCTAGAAG-3′; ssODN sequence for INSIG2(S151A): 5′-AAAGTGGATTTCGATAACAACATACAGTTGTCTCTCACACTGGCTGCACTATCCATTGGACTGTGGTGGACTTTTGATAGggCTAGgAGcGGTTTTGGCCTTGGAGTAGGAATTGCCTTCTTGGCAACTGTGGTCACTCAACTGCTAGTATATAATGGTGTTTACCA-3′. ssODN sequence for INSIG2(S151D): 5′-AAAGTGGATTTCGATAACAACATACAGTTGTCTCTCACACTGGCTGCACTATCCATTGGACTGTGGTGGACTTTTGATAGggaTAGgAGcGGTTTTGGCCTTGGAGTAGGAATTGCCTTCTTGGCAACTGTGGTCACTCAACTGCTAGTATATAATGGTGTTTACCA-3′. sgRNA targeting sequence for PCK1(S90A): 5′-GGCCAGGATCGAAAGCAAGA-3′; ssODN sequence for PCK1(S90A): 5′-CCGTGGTGCTTGGCTGAAAGGAAGCCTGTGA TTTTTGCAGCTGGTTGGCTCTCACTGACCCCAGGGATGTGGCCAGGATaGAggcCAAaACGGTTATCGTCACCCAAGAGCAAAGAGACACAGTGCCCATCCCCAAAACAGGCCTCAGCCAGCTCGGTCGCTGGATGTCAGAGG-3′. The lower-case letters in the ssODN sequences indicate the mutated nucleotides that will replace the endogenous nucleotides in the genomic DNA of parental cells using the CRISPR–Cas9 system. Genotyping was performed by sequencing PCR products amplified from the following primers: INSIG1 forward: 5′-AGAATGGGGCTATCGATGACTTC-3′; INSIG1 reverse: 5′-TGTAGTGGGGATATGCAGAACG-3′; INSIG2 forward: 5′- TCAAGTTCCTGTACGATTCTCAAGT-3′; INSIG2 reverse: 5′-AGCAAACAAGCACCAAAAATTG-3′; PCK1 forward: 5′-AAGGCCTTCGGGTAGTTTCAG-3′; PCK1 reverse: 5′-AGCCCCCTGGGTTAGAAGAG-3′.
SRE luciferase assay
SRE luciferase activity in cell lysates was measured with the luciferase assay system as previously described32. In brief, Huh7 cells in 24-well plates were transfected with 0.1 μg SRE-driven luciferase reporter and 0.075 μg β-galactosidase. At 24 h after the transfection, the cells were subjected to different culture media. Cell lysates were measured for the activity of luciferase and β-galactosidase on the basis of the manufacturer’s instruction (Promega).
Immunofluorescence analysis
Immunofluorescence analysis was performed as previously reported33. Cultured cells were fixed by 4% paraformaldehyde (PFA), treated with 0.1% Triton X-100 for 5 min and blocked in 3% BSA for 1 h. The cells were then incubated with primary antibodies at a dilution of 1:100. For tissue staining, tumour masses from mice were perfused with 0.1 M PBS (pH 7.4), embedded into optimal cutting temperature compound and frozen for cryostat section. Cryostat sections were fixed with 4% PFA for 15 min at room temperature. After PBS washing, cryostat sections were incubated in the blocking solution (PBS containing 3% donkey serum, 1% BSA, 0.3% Triton X-100 at pH 7.4) for 30 min at room temperature. In antibody reaction buffer (PBS plus 1% BSA, 0.3% Triton X-100 at pH 7.4), samples were stained with primary antibodies against SREBP1 antibody overnight at 4 °C. After incubation with fluorescent-dye-conjugated secondary antibodies and DAPI, immunofluorescent microscopic images of the cells were obtained and viewed using an IX81 confocal microscope (Olympus America). Colocalization of proteins was quantified by calculating Pearson’s correlation coefficient using the Coloc 2 plugin in Image J (National Institutes of Health).
Cell viability analysis
Cells (2 × 105) were plated in DMEM with 10% FBS (complete medium). After treatment with lipid-depleted medium for the indicated time, the viable cells were stained with trypan blue (0.5%) and counted using a Cell Viability Analyzer (Beckman Coulter).
Quantitative PCR
Total RNA was extracted from cells and tissue samples using TRIzol reagent according to the manufacturer’s instructions (Invitrogen). Equal amounts of RNA samples were used for cDNA synthesis with a TaqMan Reverse Transcription Reagents kit (Applied Biosystems). Quantitative PCR analysis was carried out using a 7500 Real-Time PCR system (Applied Biosystems) with a SYBR Premix Ex Taq kit (Takara Bio). The following primers were used for quantitative PCR: FASN, 5′-CACAGGGACAACCTGGAGTT-3′ and 5′-ACTCCACAGGTGGGAACAAG-3′; SCD, 5′-CGACGTGGCTTTTTCTTCTC-3′ and 5′-CCTTCTCTTTGACAGCTGGG-3′; ACACA, 5′-AGTGGGTCACCCCATTGTT-3′ and 5′- TTCTAACAGGAGCTGGAGCC-3′; GPAM, 5′-TTGTGGCTTGCCTGCTCCTCTA-3′ and 5′-AATCACGAGCCAGGACTTCCTC-3′; HMGCR, 5′-TCTGGCAGTCAGTGGGAACTATT-3′ and 5′-CCTCGTCCTTCGATCCAATTT-3′; HMGCS1, 5′-GATGTGGGAATTGTTGCCCTT-3′ and 5′-ATTGTCTCTGTTCCAACTTCCAG-3′; LDLR, 5′-AACGGTCATTCACCCAGGTC-3′ and 5′-GGCTGAAGAATAGGAGTTGCC-3′; FDFT1, 5′-CGATAGCTGTGTGCAAAGTAACT-3′ and 5′-CCATCTGCTGAGTGCTTTCTG-3′; GAPDH, 5′-AGCCACATCGCTCAGACAC-3′ and 5′-GCCCAATACGACCAATCC-3′.
TUNEL assay
Mouse tumour tissues were sectioned at 5-μm thickness. Apoptotic cells were counted using the DeadEnd Colorimetric TUNEL System (Promega) according to the manufacturer’s instructions.
IHC analysis and histological evaluation of human HCC specimens
Human HCC tissue collection and study approval were described previously34. Human HCC and adjacent matched non-tumour tissue samples (EHBH cohort) were obtained from Eastern Hepatobiliary Surgery Hospital in Shanghai, China. The use of human HCC samples and the relevant database was approved by the Eastern Hepatobiliary Surgery Hospital Research Ethics Committee and complied with all relevant ethical regulations. All tissue samples were collected in compliance with the informed consent policy. Sections of paraffin-embedded human HCC samples were stained with antibodies against AKT(pS473), PCK1(pS90), INSIG1(pS207)/INSIG2(pS151), SREBP1 or non-specific IgG as a negative control. The staining of the tissue sections was quantitatively scored according to the percentage of positive cells and the staining intensity as described previously30. The following proportion scores were assigned to the sections: 0 if 0% of the tumour cells exhibited positive staining, 1 for 0–1%, 2 for 2–10%, 3 for 11–30%, 4 for 31–70% and 5 for 71–100%. In addition, the staining intensity was rated on a scale of 0–3: 0, negative; 1, weak; 2, moderate; and 3, strong. The proportion and intensity scores were then combined to obtain a total score (range, 1–8) as described previously30. Scores were compared with overall survival duration, defined as the time from date of diagnosis to that of death or last known follow-up examination. All patients had received standard therapies after surgery.
Mouse studies
One million Huh7 cells with or without gene editing in PCK1 and INSIG1/2 were collected in 20 μl DMEM with 33% matrigel and intrahepatically or subcutaneously injected into 6-week-old male BALB/c athymic nude mice. The injections were performed as described previously35. Seven mice per group in each experiment were used. Mice were euthanized 28 days after injection. The liver of each mouse was dissected and then fixed in 4% formaldehyde and embedded in paraffin. The tumour volume was calculated using the formula: V = 1/2a2b (V, volume; a, shortest diameter; b, longest diameter). The mice were treated in accordance with relevant institutional and national guidelines and regulations.
The transgene HCC mouse model was established by overexpression of activated AKT combined with c-Met using hydrodynamic transfection22. Wild-type FVB/N mice were subjected to hydrodynamic injection as previously described36. In brief, the plasmids pT3-EF1α- Flag-PCK1 (wild-type or S90A-mutant) (20 μg), pT3-EF1α-HA-myr-AKT (20 μg) and pT3-EF1α-V5-c-Met (20 μg) together with pCMV/sleeping beauty transposase (SB) (2.4 μg), in a ratio of 12.5: 12.5: 12.5: 1.5, were diluted in 2 ml saline (0.9% NaCl), filtered through a 0.22-μm filter and injected into the lateral tail vein of the male mice in 6 s. After 14 weeks, the mice were euthanized and liver tumours were extracted.
The use of the mice was approved by the Institutional Review Board at MD Anderson Cancer Center and the Institutional Animal Care and Use Committee (IACUC) of Zhejiang University and Qingdao Cancer Institute, China. The maximum tumour diameter permitted by the committees is 1.5 cm. Mice arriving in the animal facility were randomly put into cages with five mice each. No statistical methods were used to predetermine sample size.
Statistics and reproducibility
All statistical data are presented as means ± s.d. All experiments were repeated at least twice independently with similar results. The mean values obtained in the control and experimental groups were analysed for significant differences. Pairwise comparisons were performed using a two-tailed t test. P values of less than 0.05 were considered significant. Unless stated otherwise, the experiments were not randomized and investigators were not blinded to allocation during experiments and outcome assessment.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this paper.
Data availability
All source data for immunoblotting are shown in Supplementary Fig. 1. All the other source data for Figs. 2–4 and Extended Data Figs. 1–10, containing the raw data for all experiments, are provided with the paper. All images and data were created and analysed by the authors and will be available from the lead corresponding author (Z.L.) on reasonable request.
References
Goldstein, J. L., DeBose-Boyd, R. A. & Brown, M. S. Protein sensors for membrane sterols. Cell 124, 35–46 (2006).
Röhrig, F. & Schulze, A. The multifaceted roles of fatty acid synthesis in cancer. Nat. Rev. Cancer 16, 732–749 (2016).
Feramisco, J. D., Goldstein, J. L. & Brown, M. S. Membrane topology of human Insig-1, a protein regulator of lipid synthesis. J. Biol. Chem. 279, 8487–8496 (2004).
Yabe, D., Brown, M. S. & Goldstein, J. L. Insig-2, a second endoplasmic reticulum protein that binds SCAP and blocks export of sterol regulatory element-binding proteins. Proc. Natl Acad. Sci. USA 99, 12753–12758 (2002).
Radhakrishnan, A., Ikeda, Y., Kwon, H. J., Brown, M. S. & Goldstein, J. L. Sterol-regulated transport of SREBPs from endoplasmic reticulum to Golgi: oxysterols block transport by binding to Insig. Proc. Natl Acad. Sci. USA 104, 6511–6518 (2007).
Shimano, H. & Sato, R. SREBP-regulated lipid metabolism: convergent physiology - divergent pathophysiology. Nat. Rev. Endocrinol. 13, 710–730 (2017).
Breuhahn, K., Longerich, T. & Schirmacher, P. Dysregulation of growth factor signaling in human hepatocellular carcinoma. Oncogene 25, 3787–3800 (2006).
Burgess, S. C. et al. Cytosolic phosphoenolpyruvate carboxykinase does not solely control the rate of hepatic gluconeogenesis in the intact mouse liver. Cell Metab. 5, 313–320 (2007).
Méndez-Lucas, A., Hyroššová, P., Novellasdemunt, L., Viñals, F. & Perales, J. C. Mitochondrial phosphoenolpyruvate carboxykinase (PEPCK-M) is a pro-survival, endoplasmic reticulum (ER) stress response gene involved in tumor cell adaptation to nutrient availability. J. Biol. Chem. 289, 22090–22102 (2014).
Vinayagam, A. et al. An integrative analysis of the InR/PI3K/Akt network identifies the dynamic response to insulin signaling. Cell Rep. 16, 3062–3074 (2016).
Espenshade, P. J. SREBPs: sterol-regulated transcription factors. J. Cell Sci. 119, 973–976 (2006).
Dif, N. et al. Insulin activates human sterol-regulatory-element-binding protein-1c (SREBP-1c) promoter through SRE motifs. Biochem. J. 400, 179–188 (2006).
Horton, J. D., Goldstein, J. L. & Brown, M. S. SREBPs: activators of the complete program of cholesterol and fatty acid synthesis in the liver. J. Clin. Invest. 109, 1125–1131 (2002).
Bakos, J. T., Johnson, B. H. & Thompson, E. B. Oxysterol-induced cell death in human leukemic T-cells correlates with oxysterol binding protein occupancy and is independent of glucocorticoid-induced apoptosis. J. Steroid Biochem. Mol. Biol. 46, 415–426 (1993).
Guo, D. et al. EGFR signaling through an Akt-SREBP-1-dependent, rapamycin-resistant pathway sensitizes glioblastomas to antilipogenic therapy. Sci. Signal. 2, ra82 (2009).
Düvel, K. et al. Activation of a metabolic gene regulatory network downstream of mTOR complex 1. Mol. Cell 39, 171–183 (2010).
Yecies, J. L. et al. Akt stimulates hepatic SREBP1c and lipogenesis through parallel mTORC1-dependent and independent pathways. Cell Metab. 14, 21–32 (2011).
Lee, J. N., Song, B., DeBose-Boyd, R. A. & Ye, J. Sterol-regulated degradation of Insig-1 mediated by the membrane-bound ubiquitin ligase gp78. J. Biol. Chem. 281, 39308–39315 (2006).
Shao, W. & Espenshade, P. J. Expanding roles for SREBP in metabolism. Cell Metab. 16, 414–419 (2012).
Menendez, J. A. & Lupu, R. Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis. Nat. Rev. Cancer 7, 763–777 (2007).
Huang, X., Liu, G., Guo, J. & Su, Z. The PI3K/AKT pathway in obesity and type 2 diabetes. Int. J. Biol. Sci. 14, 1483–1496 (2018).
Hu, J. et al. Co-activation of AKT and c-Met triggers rapid hepatocellular carcinoma development via the mTORC1/FASN pathway in mice. Sci. Rep. 6, 20484 (2016).
Li, X., Egervari, G., Wang, Y., Berger, S. L. & Lu, Z. Regulation of chromatin and gene expression by metabolic enzymes and metabolites. Nat. Rev. Mol. Cell Biol. 19, 563–578 (2018).
Radhakrishnan, A., Sun, L. P., Kwon, H. J., Brown, M. S. & Goldstein, J. L. Direct binding of cholesterol to the purified membrane region of SCAP: mechanism for a sterol-sensing domain. Mol. Cell 15, 259–268 (2004).
Xu, D. Q. et al. PAQR3 controls autophagy by integrating AMPK signaling to enhance ATG14L-associated PI3K activity. EMBO J. 35, 496–514 (2016).
Xia, Y. et al. c-Jun downregulation by HDAC3-dependent transcriptional repression promotes osmotic stress-induced cell apoptosis. Mol. Cell 25, 219–232 (2007).
Hammond, C. & Helenius, A. Quality control in the secretory pathway: retention of a misfolded viral membrane glycoprotein involves cycling between the ER, intermediate compartment, and Golgi apparatus. J. Cell Biol. 126, 41–52 (1994).
Danai, L. V. et al. Map4k4 suppresses Srebp-1 and adipocyte lipogenesis independent of JNK signaling. J. Lipid Res. 54, 2697–2707 (2013).
Lee, J. H. et al. EGFR-phosphorylated platelet isoform of phosphofructokinase 1 promotes PI3K activation. Mol. Cell 70, 197–210 (2018).
Li, X. et al. Mitochondria-translocated PGK1 functions as a protein kinase to coordinate glycolysis and the TCA cycle in tumorigenesis. Mol. Cell 61, 705–719 (2016).
Sander, J. D. & Joung, J. K. CRISPR–Cas systems for editing, regulating and targeting genomes. Nat. Biotechnol. 32, 347–355 (2014).
Xu, D. et al. PAQR3 modulates cholesterol homeostasis by anchoring Scap/SREBP complex to the Golgi apparatus. Nat. Commun. 6, 8100 (2015).
Jin, T. et al. Identification of the topology and functional domains of PAQR10. Biochem. J. 443, 643–653 (2012).
Xu, D. et al. The protein kinase activity of fructokinase A specifies the antioxidant responses of tumor cells by phosphorylating p62. Sci. Adv. 5, eaav4570 (2019).
Li, X. et al. A splicing switch from ketohexokinase-C to ketohexokinase-A drives hepatocellular carcinoma formation. Nat. Cell Biol. 18, 561–571 (2016).
Calvisi, D. F. et al. Increased lipogenesis, induced by AKT-mTORC1-RPS6 signaling, promotes development of human hepatocellular carcinoma. Gastroenterology 140, 1071–1083 (2011).
Acknowledgements
We thank L. Li for technical assistance. The mass spectrometry was supported in part by the Clinical and Translational Proteomics Service Center at the University of Texas Health Science Center. This work was supported by Zhejiang University Research Fund (188020*194221901/029) (Z.L.); the Leading Innovative and Entrepreneur Team Introduction Program of Zhejiang (2019R01001) (Z.L.); MOHW109-TDU-B-212-010001 (M.-C.H.); the Drug Development Center, China Medical University from the Ministry of Education in Taiwan (M.-C.H.); and The Odyssey Fellowship from The University of Texas MD Anderson Cancer Center (D.X.). Z.L. is Kuancheng Wang Distinguished Chair.
Author information
Authors and Affiliations
Contributions
Z.L. and D. Xu conceived and designed the study and wrote the manuscript; Z.L. and M.-C.H. acquired the funding support and supervised the study; D. Xu and Z.W. performed most experiments; Y.X., F.S., X.L. and X.Q. provided support for generating different knock-in mutation cell lines and protein purification; W.X., Y.W. and G.L. provided support for IHC staining; F.S., J.-H.L. and L.D. were involved in the studies with mice; M.-C.H., D. Xing and T.L. reviewed and edited the manuscript; H.W. provided the HCC samples and technical support; Y.Z., J.-s.L., D. Xing and T.L. performed statistical analysis and interpretation of the data.
Corresponding authors
Ethics declarations
Competing interests
Z.L. owns shares in Signalway Biotechnology (Pearland, TX), which supplied rabbit antibodies that recognize PCK1(pS90), INSIG1(pS207) and INSIG2(pS151). The interest of Z.L. in this company had no bearing on its being chosen to supply these reagents.
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 IGF1-induced AKT activation induces the translocation of PCK1 to the ER and the binding of PCK1 to INSIG1/2.
a, Huh7 cells with or without expression of Flag–INSIG1 (left) or Flag–INSIG2 (right) were treated with or without IGF1 (100 ng ml−1) for 1 h. An immunoprecipitation assay was performed using anti-Flag antibody, and immunoprecipitates of Flag–INSIG1 or Flag–INSIG2 were eluted with Flag peptide, separated using SDS–PAGE and stained with Coomassie Brilliant Blue. Selected peptide hits of proteins associated with Flag–INSIG1 or Flag–INSIG2, identified through mass spectrometry, are shown. b, Hep3B cells were treated with or without IGF1 (100 ng ml−1) for 1 h. c, Huh7 cells expressing Flag–PCK1 or Flag–PCK2 were stimulated with or without IGF1 (100 ng ml−1) for 1 h. d, Huh7 cells expressing Flag–PCK1 or Flag–PCK2 were stimulated with or without IGF1 (100 ng ml−1) for 1 h. Immunofluorescence analyses were performed with the indicated antibodies (top). The colocalization coefficients between the indicated proteins in the presence or absence of IGF1 are shown (bottom). At least n = 50 cells from each independent experiment were analysed and representative data are shown. Data are mean ± s.d. **P < 0.001 (two-tailed t-test). The regions in white boxes are shown at higher magnification on the right. e, Whole cell lysate (WCL), cytosolic and ER fractions were prepared from Huh7 and Hep3B cells stimulated with or without IGF1 (100 ng ml−1) for 1 h. Cellular fractions from equal numbers of cells were analysed using immunoblotting with the indicated antibodies. f, Huh7 cells expressing Flag–PCK1 or Flag–PCK2 were stimulated with or without IGF1 (100 ng ml−1) for 1 h. ER fractions and total lysates were prepared for immunoblotting analyses with the indicated antibodies. g, Total lysates were prepared from Huh7 cells pretreated with or without U0126 (20 μM), SP600125 (25 μM), SU6656 (4 μM) or MK-2206 (10 μM) for 30 min before treatment with or without IGF1 (100 ng ml−1) for 1 h. h, Huh7 cells expressing wild-type HA–AKT1 or HA–AKT1-DN (K179A, T308A, S473A were treated with or without IGF1 (100 ng ml−1) for 1 h. i, Huh7 cells were pretreated with or without U0126 (20 μM), SU6656 (4 μM), SP600125 (25 μM) or MK-2206 (10 μM) for 30 min before treatment with or without IGF1 (100 ng ml−1) for 1 h. ER fractions and total lysates were isolated for immunoblotting analyses with the indicated antibodies. j, Huh7 cells were stably transfected with a control vector or a vector expressing HA–AKT1-DN (K179A, T308A, S473A). The cells were treated with or without IGF1 (100 ng ml−1) for 1 h. ER fractions and total lysates were isolated for immunoblotting analyses with the indicated antibodies. k, Huh7 cells were pretreated with or without MK-2206 (10 μM) for 30 min and treated with or without IGF1 (100 ng ml−1) for 1 h. Immunofluorescence analyses were performed with the indicated antibodies (left). The colocalization coefficients between the indicated proteins are shown (right). At least n = 50 cells from each independent experiment were analysed and representative data are shown. Data are mean ± s.d. **P < 0.001 (two-tailed t-test). l, Huh7 cells expressing Flag–PCK1 or Flag–PCK2 were transfected with or without HA–myr-AKT1. m, Huh7 and Hep3B cells were treated with or without IGF1 (100 ng ml−1) for 1 h. In b, c, e–j, l, m, immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies, and the experiments were repeated three times independently with similar results.
Extended Data Fig. 2 AKT-mediated phosphorylation of PCK1 Ser90 is necessary and sufficient for the translocation of PCK1 to the ER and the binding of PCK1 to INSIG1/2, and inhibits the canonical role of PCK1 in gluconeogenesis.
a, A GST pull-down assay was performed by mixing purified His–PCK1 with purified GST or GST–AKT1. Immunoblotting analyses were performed as indicated. b, Left, schematic of AKT1 full-length and deletion mutants. Right, 293T cells were transfected with the indicated constructs, and GST pull-down assays and immunoblotting analyses with the indicated antibodies were performed. c, In vitro kinase assays were performed by mixing purified GST–PCK1 with purified active or inactive His–AKT1 in the presence of [γ-32P]ATP. Autoradiography and immunoblotting analyses were performed as indicated. d, In vitro kinase assays were performed by mixing purified His–PCK1 with or without purified GST–AKT1 in the presence of ATP. Mass spectrometric analysis of a tryptic fragment at m/z 805.91107 Da (−0.06 mmu/−0.08 ppm), which was matched with the +2 charged peptide 88-IESKTVIVTQEQR-100, suggested that PGK1 Ser90 was phosphorylated. The Mascot score was 31, and the expectation value was 0.25. e, Alignment of protein sequences spanning PCK1 Ser90 from different species. f, Alignment of PCK1 Ser90 to the non-canonical AKT-phosphorylated substrate motif (RXXS/T). The reported AKT substrates (TFEB, YAP1, CREB, CDK2, β-catenin, FAM129A, CDC25B and NPM) are shown. Red, phosphoacceptor residue; blue, basic residue (R). g, In vitro kinase assays were performed by mixing purified GST–PCK1 or GST–PCK1(S90A) and active His–AKT1 in the presence or absence of MK-2206 (10 μM) for 1 h. Immunoblotting analyses were performed as indicated. h, Left, IHC analyses of human HCC samples were performed with the indicated antibodies in the presence or absence of a blocking peptide for PCK1(pS90). Right, Huh7 cells expressing Flag–PCK1 were treated with or without IGF1 (100 ng ml−1) for 1 h. Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies in the presence or absence of a blocking peptide for PCK1(pS90). i, Huh7 cells expressing wild-type Flag–rPCK1 or Flag–rPCK1(S90A) were pretreated with or without MK-2206 (10 μM) for 30 min. The cells were treated with or without IGF1 (100 ng ml−1) for 1 h. Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. j, PCK1-depleted Huh7 cells were reconstituted with the indicated shRNA-resistant PCK1 proteins. After transfection with HA–myr-AKT1, ER fractions and total cell lysate were prepared for immunoblotting analyses as indicated. k, Huh7 and Hep3B cells expressing the indicated Flag–rPCK1 proteins were treated with or without IGF1 (100 ng ml−1) for 1 h. ER fractions and total cell lysate were prepared for immunoblotting analyses as indicated. l, m, Genomic DNA was extracted from two individual clones of parental Huh7 or Hep3B cells with knock-in expression of PCK1(S90A). PCR products were amplified from the indicated DNA fragment (l) and separated on an agarose gel (m). n, Sequencing of parental Huh7 and Hep3B cells and two individual clones of parental cells with knock-in expression of PCK1(S90A). The red line indicates the sgRNA-targeting sequence. The black line indicates the protospacer adjacent motif (PAM). Blue arrows indicate mutated nucleotides. A mutated amino acid and its wild-type counterpart are indicated by the solid red box. o, p, Parental Huh7 and Hep3B cells and the indicated clones of Huh7 and Hep3B cells with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 (100 ng ml−1) for 1 h. ER fractions and total cell lysates were prepared and immunoblotting analyses were performed as indicated (o). Immunofluorescence staining of Huh7 cells was performed with the indicated antibodies (p, left). The colocalization coefficients between the indicated proteins are shown (p, right). At least n = 50 cells from each independent experiment were analysed and representative data are shown. Data are mean ± s.d. **P < 0.001 (two-tailed t-test). q–s, Bacterially purified wild-type His–PCK1 or His–PCK1(S90A) on Ni-NTA agarose beads were incubated with or without purified active GST–AKT1 in the presence of ATP for an in vitro AKT kinase assay. Immunoblotting analyses were performed as indicated (q). After washing wild-type His–PCK1 or His–PCK1(S90A)-conjugated beads with PBS five times, the binding affinity of PCK1 to oxaloacetate (OAA) (r) and the relative PCK1 activity (s) were measured. Data are mean ± s.d. (n = 6 biological replicates). **P < 0.001 (two-tailed t-test); NS, not significant (P = 0.923 (r); P = 0.728 (s)). t–v, Wild-type His–PCK1 and His–PCK1(S90E) were purified from bacteria and Coomassie Brilliant Blue staining analyses were performed (t). The binding affinity of the His–PCK1 proteins to oxaloacetate (u) and the relative PCK1 activity (v) were measured. Data are mean ± s.d. (n = 6 biological replicates). **P < 0.001 (two-tailed t-test). w, Bacterially purified wild-type GST–PCK1 or GST–PCK1(S90A) on glutathione agarose beads were incubated with or without active His–AKT1 in the presence of ATP for an in vitro kinase assay. The GST-tagged proteins were then incubated with or without CIP (10 U) for 30 min at 37 °C, followed by incubation with Flag/His-tagged INSIG1 or INSIG2 purified from Huh7 cells for a pull-down assay. x, SFB-tagged wild-type PCK1 or PCK1(S90A) were pulled down from Huh7 cells treated with or without IGF1 (100 ng ml−1) for 1 h. These proteins were incubated with or without CIP (10 U) for 30 min at 37 °C. Immunoblotting analyses were performed as indicated. y, Parental Hep3B cells and the indicated clones of cells with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 for 1 h. Immunoprecipitation and immunoblotting analyses were performed as indicated. z, Huh7 cells expressing INSIG1 shRNA and INSIG2 shRNA were treated with or without IGF1 (100 ng ml−1) for 1 h (left) or transfected with HA–myr-AKT1 (right). ER fractions and total cell lysate were prepared for immunoblotting analyses as indicated. All experiments were repeated three times independently with similar results.
Extended Data Fig. 3 INSIG1/2 is required for the translocation of PCK1 to the ER and PCK1 functions as a protein kinase to phosphorylate INSIG1 Ser207 and INSIG2 Ser151.
a, Top, the topological structures of INSIG1 and INSIG2, which have 69% amino acid sequence identity and contain six transmembrane-spanning regions. Bottom, different INSIG1 (left) or INSIG2 (right) truncation mutants were expressed in 293T cells. These cells were treated with or without IGF1 (100 ng ml−1) for 1 h. Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. b, Bacterially purified GST–PCK1 was incubated with or without active His–AKT1 in the presence of ATP for 1 h. After the in vitro kinase assay, GST–PCK1-conjugated beads were washed with PBS five times and then incubated with purified Flag/His–INSIG1 or Flag/His–INSIG2 in the absence or presence of a PCK1(pS90) peptide for 2 h. Immunoblotting analyses were performed with the indicated antibodies. c, Endogenous PCK1-depleted Huh7 cells with reconstituted expression of shRNA-resistant wild-type rPCK1 or rPCK1(C228S) were transfected with Flag–INSIG1 (left) or Flag–INSIG2 (right), respectively. After being treated with or without IGF1 (100 ng ml−1) for 1 h, the cells were collected for immunoprecipitation and immunoblotting analyses as indicated. d, Bacterially purified wild-type His–PCK1, His–PCK1(S90A) or His–PCK1(C288S) on Ni-NTA agarose beads were incubated with or without active GST–AKT1 in the presence of ATP for an in vitro AKT kinase assay. The beads were washed with PBS five times and incubated with SFB–INSIG2 in the presence of [γ-32P]GTP. Autoradiography and immunoblotting analyses with the indicated antibodies were performed. e, In vitro kinase assays were performed by mixing purified His–PCK1 on Ni-NTA agarose beads with purified active GST–AKT1 in the presence of ATP for 1 h. His–PCK1-conjugated Ni-NTA agarose beads were washed with PBS five times and then incubated with SFB–INSIG1 or SFB–INSIG2 purified from Huh7 cells in the presence of [γ-32P]ATP or [γ-32P]GTP for 1 h. Autoradiography and immunoblotting analyses with the indicated antibodies were performed. f, An in vitro kinase assay was performed as in e, except that the His–PCK1-conjugated beads were incubated with SFB–INSIG1 purified from Huh7 cells in the presence of GTP for 1 h. Mass spectrometric analysis of a tryptic fragment at m/z 764.40387 Da (−1.96 mmu/−2.57 ppm), which was matched with the +2 charged peptide 205-RSRSGLGLGITIAF-218, suggested that INSIG1 Ser207 was phosphorylated. The Mascot score was 53, and the expectation value was 0.021. g, PCK1-mediated phosphorylation residues in the cytosolic loop 2 of both INSIG1 and INSIG2. h, Alignment of protein sequences spanning INSIG1 Ser207 and INSIG2 Ser151 from different species. i, Bacterially purified wild-type His–PCK1 on Ni-NTA agarose beads was incubated with or without purified active GST–AKT1 in the presence of ATP for 1 h for an in vitro AKT kinase assay. The beads were then washed with PBS five times and incubated with or without wild-type SFB–INSIG2 or SFB–INSIG2(S151A) in the presence of [γ-32P]GTP. Autoradiography and immunoblotting analyses with the indicated antibodies were performed. j, IHC analyses of human HCC samples were performed with the indicated antibodies in the presence or absence of a blocking peptide for INSIG1(pS207) and INSIG2(pS151). k, Huh7 cells expressing Flag–INSIG1 (top) or Flag–INSIG2 (bottom) were treated with or without IGF1 (100 ng ml−1) for 1 h. Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies in the presence or absence of a blocking peptide for INSIG1(pS207) and INSIG2(pS151). l, Enzyme kinetics plots of velocity relative to GTP concentration between purified wild-type His–PCK1 and His–PCK1(S90E). The Vmax and Km of PCK1 in phosphorylating an INSIG1 peptide at Ser207 were calculated (n = 6). Data are mean ±s.d. m, n, Bacterially purified wild-type His–PCK1, His–PCK1(S90E) or His–PCK1(S90D) on Ni-NTA agarose beads were incubated with wild-type SFB–INSIG1 or SFB–INSIG1(S207A) (m) or wild-type SFB–INSIG2 or SFB–INSIG2(S151A) (n) in the presence of [γ-32P]GTP. Autoradiography and immunoblotting assays with the indicated antibodies were performed. o, Wild-type Flag–INSIG1 or Flag–INSIG1(S207A) (left) or wild-type Flag–INSIG2 or Flag–INSIG2(S151A) (right) were transfected into Huh7 cells. Huh7 cells were stimulated with or without IGF1 (100 ng ml–1) for 1 h. Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. p, Wild-type Flag–INSIG1 or Flag–INSIG1(S207A) (left) or wild-type Flag–INSIG2 or Flag–INSIG2(S151A) (right) were co-transfected with or without HA–myr-AKT1 into Huh7 cells. Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. q, Parental Huh7 cells and the indicated clones of cells with knock-in expression of PCK1(S90A) were transfected with SFB–INSIG1 (left) or SFB–INSIG2 (right). These cells were then stimulated with or without IGF1 (100 ng ml–1) for 1 h. All experiments were repeated three times independently with similar results.
Extended Data Fig. 4 Generation of Huh7 and Hep3B cells with knock-in expression of both INSIG1(S207A) and INSIG2(S151A), and PCK1-mediated phosphorylation of INSIG1/2 reduces the binding of oxysterols to INSIG1/2.
a, b, Genomic DNA was extracted from two individual clones of parental Huh7 and Hep3B cells with knock-in expression of INSIG1(S207A) (a) and INSIG2(S151A) (b). PCR products amplified from the indicated DNA fragments were separated on an agarose gel. c, d, Sequencing of parental and two individual clones of parental Huh7 and Hep3B cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants. The red line indicates the sgRNA-targeting sequence. The black line indicates the PAM. Blue arrows indicate mutated nucleotides. A mutated amino acid and its wild-type counterpart are indicated by the solid red box. e, Top, Flag/His-tagged INSIG2 immunoprecipitated and purified from Huh7 cells was incubated with the indicated GST–PCK1 proteins with or without active GST–AKT1 in the presence of ATP and GTP for 1 h. Immunoblotting analyses were performed with the indicated antibodies. Bottom, INSIG2-conjugated Ni-NTA agarose beads were washed and incubated with 400 nM [3H]25-hydroxycholesterol. Specifically bound [3H]25-hydroxycholesterol was measured (n = 6). **P < 0.001 (two-tailed t-test); NS, not significant (P=0.875, 0.846, 0.969, 0.924 (left to right)). f, g, Top, Flag/His-tagged wild-type INSIG1 or INSIG1(S207A) (f) or Flag/His-tagged wild-type INSIG2 or INSIG2(S151A) (g) purified from Huh7 cells were incubated with purified wild-type GST–PCK1, GST–PCK1(S90D) or GST–PCK1(S90E) in the presence of GTP for 1 h. Immunoblotting analyses were performed with the indicated antibodies. Bottom, the INSIG1 or INSIG2 proteins on Ni-NTA agarose beads were washed with PBS five times and incubated with 400 nM [3H]25-hydroxycholesterol. Specifically bound [3H]25-hydroxycholesterol was measured (n = 6). **P < 0.001 (two-tailed t-test); NS, not significant (P=0.823, 0.445, 0.185 (left to right) (f); P=0.320, 0.196, 0.735 (left to right) (g)). h, Top, Flag/His-tagged wild-type INSIG2 or INSIG2(S151A) purified from Huh7 cells were incubated with purified wild-type GST–PCK1 with or without purified active GST–AKT1 in the presence of ATP and GTP for 1 h. Immunoblotting analyses were performed with the indicated antibodies. Bottom, the INSIG2 protein on Ni-NTA agarose beads was washed with PBS five times and incubated with 400 nM [3H]25-hydroxycholesterol. Specifically bound [3H]25-hydroxycholesterol was measured (n = 6). **P < 0.001 (two-tailed t-test); NS, not significant (P = 0.682, 0.947 (left to right)). i, j, Flag/His-tagged wild-type INSIG1 or INSIG1(S207A) (i) or wild-type INSIG2 or INSIG2(S151A) (j) were purified from Huh7 treated with or without IGF1 (100 ng ml−1) for 12 h. Immunoblotting analyses were performed with the indicated antibodies (left). The INSIG1 or INSIG2 proteins on Ni-NTA agarose beads were incubated with the indicated concentration of [3H]25-hydroxycholesterol. Specifically bound [3H]25-hydroxycholesterol was measured (n = 6) (right). k, l, Top, Flag/His-tagged INSIG1 (k) or INSIG2 (l) was expressed in parental Huh7 cells and the Huh7 cells with knock-in expression of PCK1(S90A). After these cells were treated with or without IGF1 (100 ng ml−1) for 12 h, immunoblotting analyses were performed with the indicated antibodies. Bottom, the immunoprecipitated and purified INSIG1 or INSIG2 was incubated with the indicated concentration of [3H]25-hydroxycholesterol. Specifically bound [3H]25-hydroxycholesterol was measured (n = 6). m, n, Top, the immunoprecipitated and purified Flag/His-tagged wild-type INSIG1 or INSIG1(S207E) (m) or Flag/His-tagged wild-type INSIG2 or INSIG2(S151E) (n) from Huh7 cells was incubated with the indicated concentration of [3H]25-hydroxycholesterol. Specifically bound [3H]25-hydroxycholesterol was measured (n = 6). **P < 0.001 (two-tailed t-test). Data in e–n are mean ± s.d. All experiments were repeated three times independently with similar results.
Extended Data Fig. 5 PCK1-mediated phosphorylation of INSIG1/2 promotes the translocation of SCAP from the ER to the Golgi apparatus.
a, Parental Hep3B cells and the indicated clones of Hep3B cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants were transfected with Myc–SCAP and stimulated with or without IGF1 (100 ng ml−1) for 16 h. Immunoprecipitation and immunoblotting analyses with the indicated antibodies were performed. b, Validation of the specificity of the SCAP antibody. SCAP shRNA was expressed in 293T cells. Immunoblotting analyses were performed with the indicated antibodies. c, Parental Huh7 or Hep3B cells and the indicated clones with knock-in expression of PCK1(S90A) were transfected with the indicated plasmids and stimulated with or without IGF1 (100 ng ml−1) for 16 h. Ni-NTA pull-down assays, immunoprecipitation and immunoblotting analyses with the indicated antibodies were performed. d, PCK1-depleted Huh7 (left) and Hep3B (right) cells were stably transfected with shRNA-resistant wild-type rPCK1 or rPCK1(S90A). Immunoblotting analyses were performed with the indicated antibodies. e, f, PCK1-depleted Huh7 (left) and Hep3B (right) cells with reconstituted expression of shRNA-resistant wild-type rPCK1 and rPCK1(S90A) (e) or rPCK1(C288S) (f) were transfected with vectors expressing Flag–INSIG1 and His–INSIG2. The cells were then stimulated with or without IGF1 (100 ng ml−1) for 16 h. Ni-NTA pull-down assays, immunoprecipitation and immunoblotting analyses with the indicated antibodies were performed. g, h, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. The cells were then subjected to homogenization and cell fractionation using gradient centrifugation. Immunoblotting analyses were performed with the indicated antibodies (g). The relative distribution of each protein in different fractions was quantified by densitometric analysis of the blots (n = 3) (h). i, j, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. The cells were then subjected to homogenization and cell fractionation using gradient centrifugation. Immunoblotting analyses were performed with the indicated antibodies (i). The relative distribution of each protein in different fractions was quantified by densitometric analysis of the blots (n = 3) (j). k, Colocalization coefficients between SCAP and calnexin in the indicated cells. At least n = 50 cells from each independent experiment were analysed and representative data are shown. **P < 0.001 (two-tailed t-test). l, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) were stimulated with or without IGF1 for 16 h. Immunofluorescence analyses were performed with the indicated antibodies. The white arrows indicate the Golgi-localized SCAP. Scale bar, 20 μm. m, Colocalization coefficients between SCAP and golgin-97. At least n = 50 cells from each independent experiment were analysed and representative data are shown. **P <0.001 (two-tailed t-test). n, o, Top, parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. Immunofluorescence analyses were performed with the indicated antibodies. The white arrows indicate the Golgi-localized SCAP. Bottom, colocalization coefficients between SCAP and calnexin (n) or golgin-97 (o). At least n = 50 cells from each independent experiment were analysed and representative data are shown. **P < 0.001 (two-tailed t-test). Scale bars, 20 μm. p, q, Top, parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants were stably transfected with Myc–SCAP and stimulated with or without IGF1 (100 ng ml−1) for 16 h. Immunofluorescence analyses were performed with the indicated antibodies. The white arrows indicate the Golgi-localized SCAP. Bottom, colocalization coefficients between Myc–SCAP and calnexin (p) or golgin-97 (q). At least n = 50 cells from each independent experiment were analysed and representative data are shown. **P < 0.001 (two-tailed t-test). Scale bars, 20 μm. r, Parental Huh7 and Hep3B cells and the indicated clones of these cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. N, N terminus of SREBP1; P, precursor of SREBP1. s, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) were transiently transfected with vectors expressing β-galactosidase and an SRE-driven luciferase reporter. Twenty-four hours later, the cells were stimulated with or without IGF1 (100 ng ml−1) for 16 h. The relative SRE luciferase activity after normalization to β-galactosidase activity is shown (n = 6). **P < 0.001 (two-tailed t-test). t, Parental Huh7 and Hep3B cells and the indicated clones of these cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. Immunoblotting analyses were performed with the indicated antibodies. Data in h, j, k, m–q, s are mean ± s.d. All experiments were repeated three times independently with similar results.
Extended Data Fig. 6 PCK1-mediated INSIG1/2 phosphorylation is required for IGF1- induced SREBP activation for lipogenesis and does not affect the lipid depletion-induced translocation of SCAP from the ER to the Golgi apparatus.
a, Parental Huh7 (left) and Hep3B (right) cells and the indicated clones with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. Immunoblotting analyses were performed as indicated. b, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. Immunofluorescence analyses were performed as indicated. Scale bar, 20 μm. c, Parental Huh7 cells and the indicated clones with knock-in expression of PCK1(S90A) were transiently transfected with vectors expressing β-galactosidase and an SRE-driven luciferase reporter and stimulated with or without IGF1 (100 ng ml−1) for 16 h. The relative SRE luciferase activity after normalization to β-galactosidase activity is shown (n = 6). Data are mean ± s.d. **P < 0.001 (two-tailed t-test). d, Parental Huh7 cells and the indicated clones with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. The mRNA expression levels for SREBP target genes were measured using quantitative PCR (n = 6). Data are mean ± s.d. **P < 0.001 (two-tailed t-test). e, Parental Huh7 (left) and Hep3B (right) cells and the indicated clones with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. Immunoblotting analyses were performed as indicated. f, Parental Huh7 cells and the indicated clones with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants (left) or PCK1(S90A) (right) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. The mRNA expression levels for SREBF1 were measured using quantitative PCR (n = 6). Data are mean ± s.d. *P = 0.002, **P < 0.001 (two-tailed t-test). g, Parental Huh7 cells and the indicated clones with knock-in expression of PCK1(S90A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. The incorporation of 14C-glucose into triglycerides (left) and fatty acids (right) was measured (n = 6). Data are mean ± s.d. and were compared between groups using a two-tailed t-test. *P = 0.014, 0.012 (left to right); **P = 0.004, 0.001 (left to right). h, Endogenous PCK1-depleted CHL-1 human melanoma cells, U87 human glioblastoma cells and H1993 human non-small-cell lung cancer cells with reconstituted expression of shRNA-resistant wild-type HA–rPCK1 or HA–rPCK1(S90A) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. The incorporation of 14C-glucose into fatty acids was measured (n = 6). Data are mean ± s.d. and were compared between groups using a two-tailed t-test. *P = 0.011, **P < 0.001. i, j, Parental Huh7 or Hep3B cells and the indicated clones with expression of PCK1(S90D) (i) or INSIG1(S207D)/INSIG2(S151D) double mutants (2S/D) (j) were collected for immunoblotting analyses as indicated. k, l, Parental Huh7 cells and the indicated clones with knock-in expression of PCK1(S90A) (k) or INSIG1(S207A)/INSIG2(S151A) (l) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. Immunoblotting analyses were performed as indicated. m, n, Parental Huh7 cells and the indicated clones with knock-in expression of PCK1(S90A) (m) or INSIG1(S207A)/INSIG2(S151A) (n) were stimulated with or without IGF1 (100 ng ml−1) for 16 h. The mRNA expression levels for SREBP2 target genes were measured using quantitative PCR (n = 6). Data are mean ± s.d. **P < 0.001 (two-tailed t-test). o, p, Parental Huh7 cells and the indicated clones with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants were incubated with a lipid-depleted medium and treated with lovastatin (an inhibitor of HMGCR) to inhibit cholesterol synthesis for 16 h. The cells were subjected to homogenization and cell fractionation using gradient centrifugation. Immunoblotting analyses were performed as indicated (o). The relative distribution of each protein in different fractions was quantified by densitometric analysis of the blots (n = 3) (p). Data are mean ± s.d. q, r, Parental Huh7 cells and the indicated clones with knock-in expression of PCK1(S90A) were incubated with or without lipid-depleted medium for 16 h. The cells were subjected to homogenization and cell fractionation using gradient centrifugation. Immunoblotting analyses were performed as indicated (q). The relative distribution of each protein in different fractions was quantified by densitometric analysis of the blots (n = 3) (r). Data are mean ± s.d. s–x, Parental Huh7 cells and the indicated clones with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants (s, v) or PCK1(S90A) (t, w) were incubated with or without lipid-depleted medium for 16 h. Immunofluorescence analyses were performed as indicated. Scale bars, 20 μm. The white arrows indicate the Golgi-localized SCAP. The colocalization coefficients between SCAP and the ER marker calnexin (u) and the Golgi apparatus marker golgin-97 (x) are shown. At least n = 50 cells from each independent experiment were analysed and representative data are shown. Data are mean ± s.d. and were compared between groups using a two-tailed t-test. NS, not significant (P = 0.788, 0.514, 0.689, 0.693 (left to right) (u); P = 0.976, 0.606, 0.750, 0.940 (left to right) (x)). All experiments were repeated three times independently with similar results.
Extended Data Fig. 7 PCK1-mediated phosphorylation of INSIG1/2, which does not affect lipid-depletion-induced activity of SREBP1, promotes rapid activation of SREBP1 and degradation of INSIG1.
a, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants (top) or knock-in expression of PCK1(S90A) (bottom) were incubated with or without lipid-depleted medium for 16 h. Immunoblotting analyses were performed with the indicated antibodies. b, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of NSIG1(S207A)/INSIG2(S151A) double mutants (left) or knock-in expression of PCK1(S90A) (right) were incubated with or without lipid-depleted medium for 16 h. Immunofluorescence analyses were performed with the indicated antibodies. c, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants (left) or knock-in expression of PCK1(S90A) (right) were transiently transfected with vectors expressing β-galactosidase and an SRE-driven luciferase reporter. Luciferase activity was determined after the cells were incubated with or without lipid-depleted medium for 16 h. The relative SRE luciferase activity after normalization to β-galactosidase activity is shown (n = 6). NS, not significant (P = 0.965, 0.699, 0.967, 0.767 (left to right)). d, Huh7 cells expressing Flag–INSIG1 and His–INSIG2 were stimulated with or without IGF1 (100 ng ml−1) for 16 h in the presence of 40 nM or 120 nM cholesterol or 25-hydroxycholesterol. Ni-NTA pull-down assays, immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. 25-HC, 25-hydroxycholesterol. e, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of both INSIG1(S207D) and INSIG2(S151D) (left) or knock-in expression of PCK1(S90D) (right) were incubated with 40 nM or 120 nM cholesterol or 25-hydroxycholesterol for 16 h. Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. f, Huh7 cells were transiently transfected with vectors expressing β-galactosidase and an SRE-driven luciferase reporter. Twenty-four hours later, the cells were stimulated with or without IGF1 (100 ng ml−1) for 16 h in the presence of 40 nM or 120 nM cholesterol or 25-hydroxycholesterol. The relative SRE luciferase activity after normalization to β-galactosidase activity is shown (n = 6). *P = 0.001, **P < 0.001 (two-tailed t-test); NS, not significant (P = 0.921, 0.579 (left to right)). g, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of INSIG1(S207D)/INSIG2(S151D) double mutants (left) or knock-in expression of PCK1(S90D) (right) were transiently transfected with vectors expressing β-galactosidase and an SRE-driven luciferase reporter. Luciferase activity was determined after the cells were incubated with 40 nM or 120 nM cholesterol or 25-hydroxycholesterol for 16 h. The relative SRE luciferase activity after normalization to β-galactosidase activity is shown (n = 6). *P = 0.002, **P < 0.001 (two-tailed t-test); NS, not significant (P = 0.642, 0.957, 0.842, 0.372 (left to right)). h, Huh7 cells were treated with IGF1 (100 ng ml−1) for the indicated time periods. Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. i, Parental Huh7 cells and the indicated clones of Huh7 cells with knock-in expression of PCK1(S90A) (left) or INSIG1(S207A)/INSIG2(S151A) double mutants (right) were treated with or without IGF1 (100 ng ml−1) for 48 h. Immunoblotting analyses were performed with the indicated antibodies. j, k, Serum-starved parental Huh7 (j) and Hep3B (k) cells or the indicated clones with knock-in expression of INSIG1(S207A)/INSIG2(S151A) double mutants were stimulated with or without IGF1 (100 ng ml−1) in the presence or absence of CHX (100 μg ml−1) for the indicated time periods. Immunoblotting analyses were performed with the indicated antibodies (left). The relative abundance of remaining INSIG1 or INSIG2 protein was quantified (right) (n = 6). **P < 0.001 (two-tailed t test); NS, not significant (P = 0.721, 0.637 (left to right) (j); P = 0.310, 0.853 (left to right) (k)). l, Huh7 cells expressing Flag-tagged wild-type INSIG1 (top) or INSIG1(S207D) (bottom) were treated with CHX (100 μg ml−1) for the indicated time periods. Immunoblotting analyses were performed with the indicated antibodies (left). The relative abundance of the remaining INSIG1 protein was quantified (right) (n = 6). **P < 0.001 (two-tailed t test). m, Huh7 cells expressing Flag-tagged wild-type INSIG2 (top) or INSIG2(S151D) (bottom) were treated with CHX (100 μg ml−1) for the indicated time periods. Immunoblotting analyses were performed with the indicated antibodies (left). The relative abundance of the remaining INSIG2 protein was quantified (right) (n = 6). **P < 0.001 (two-tailed t-test); NS, not significant (P = 0.395, 0.973, 0.636, 0.882 (left to right)). n, Endogenous PCK1-depleted Huh7 cells with reconstituted expression of shRNA-resistant wild-type rPCK1 or rPCK1(S90A) were treated with IGF1 (100 ng ml−1) for the indicated time periods. Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. Data in c, f, g, j–m are mean ± s.d. All experiments were repeated three times independently with similar results.
Extended Data Fig. 8 PCK1-mediated phosphorylation of INSIG1 Ser207 and INSIG2 Ser151 and SREBP1 activation are induced by oncogene- or growth-factor-mediated activation of AKT in several cancer types.
a, b, e–l, Cells were transfected with the indicated plasmids and immunoprecipitation or immunoblotting analyses were performed with the indicated antibodies. a, Endogenous PCK1-depleted HL7702 (left) and THLE-2 (right) cells with reconstituted expression of shRNA resistant wild-type HA–rPCK1 or HA–rPCK1(S90A) were treated with or without IGF1 (100 ng ml−1) for 16 h. b, Endogenous PCK1-depleted HL7702 (left) and THLE-2 (right) cells with reconstituted expression of shRNA resistant wild-type HA–rPCK1 or HA–rPCK1(S90D) were analysed. c, C57BL/6J male mice (eight-week-old) fasted for 24 h were intraperitoneally injected with or without glucose (1 g per kg body weight). After 6 h, the mouse livers were dissected for immunoprecipitation and immunoblotting analyses. d, HL7702, THLE-2, Huh7 and Hep3B cells were analysed by Ni-NTA pull-down, immunoprecipitation and immunoblotting assays. e, Endogenous PCK1-depleted HL7702, THLE-2, Huh7 and Hep3B cells with reconstituted expression of shRNA resistant wild-type HA–rPCK1 or HA–rPCK1(S90A) were analysed. f, Parental Huh7 cells and the indicated clones with knock-in expression of PCK1(S90A) (left) or INSIG1(S207A)/INSIG2(S151A) double mutants (right) were transfected with or without Flag–KRAS(G12V). g, Endogenous PCK1-depleted CHL-1, U87 and H1993 cells with reconstituted expression of shRNA resistant wild-type HA–rPCK1 or HA–rPCK1(S90A) were transfected with the indicated plasmids. h, Parental Huh7 cells and the indicated clones with knock-in expression of PCK1(S90A) (left) or INSIG1(S207A)/INSIG2(S151A) double mutants (right) were transfected with or without constitutively active Flag–IGF1R(V922E). i, Endogenous PCK1-depleted CHL-1, U87 and H1993 cells with reconstituted expression of shRNA resistant wild-type HA–rPCK1 or HA–rPCK1(S90A) were transfected with the indicated plasmids. j, Endogenous PCK1-depleted CHL-1, U87 and H1993 cells with reconstituted expression of shRNA resistant wild-type HA–rPCK1 or HA–rPCK1(S90A) were transfected with the indicated plasmids and treated with or without PDGF (30 ng ml−1) for 16 h. k, m, Endogenous PCK1-depleted Huh7, Hep3B, SNU-398 and SNU-475 cells with reconstituted expression of shRNA resistant wild-type HA–rPCK1 or HA–rPCK1(S90A) were analysed by immunoprecipitation and immunoblotting analyses with the indicated antibodies (k). The cells were plated and then collected and counted for 3 days (n = 6) (m). Data are mean ± s.d. **P < 0.001 (two-tailed t-test). l, n, Endogenous PCK1-depleted Huh7, Hep3B, SNU-398 and SNU-475 cells with reconstituted expression of shRNA resistant wild-type HA–rPCK1 or HA–rPCK1(S90A) were transfected with Flag–INSIG1 and His–INSIG2. After incubation with or without lipid-depleted medium for 16 h, the cells were collected for Ni-NTA pull-down, immunoprecipitation and immunoblotting analyses as indicated (l). Viable cells were counted for 3 days after lipid depletion (n=6) (n). Data are mean ± s.d. NS, not significant (P = 0.708, 0.619, 0.888, 0.901, 0.633, 0.788, 0.902, 0.764 (left to right)). All experiments were repeated three times independently with similar results.
Extended Data Fig. 9 PCK1-mediated phosphorylation of INSIG1/2 promotes the proliferation of liver cancer cells and the growth of tumours in mice.
a, Parental Hep3B cells (2 × 105) and the indicated clones of Hep3B cells (2 × 105) with knock-in expression of PCK1(S90A) (left) or INSIG1(S207A)/INSIG2(S151A) double mutants (right) were plated for 3 days. The cells were then collected and counted. Data are mean ± s.d. (n = 6). **P < 0.001 (two-tailed t-test). b, c, Parental Huh7 and Hep3B cells (2 × 105) and the indicated clones of Huh7 and Hep3B cells with knock-in expression of PCK1(S90D) (b) or INSIG1(S207D)/INSIG2(S151D) double mutants (c) were plated for 3 days. The cells were then collected and counted. Data are mean ± s.d. (n = 6). **P < 0.001 (two-tailed t-test). d–h, Parental Huh7 cells (1 × 106) or the clones with knock-in expression of PCK1(S90A) or INSIG1(S207A)/INSIG2(S151A) double mutants were subcutaneously injected into the left or right flanks of athymic nude mice, respectively (n= 7 per group) (d, left). The resulting tumours were resected 28 days after injection (d, right). The growth of xenografted tumours in the mice was measured (e) and the tumours were weighed (f). Data are mean ± s.d. (n = 7). *P = 0.002, **P < 0.001 compared with the wild-type group (two-tailed t-test). g, IHC analyses of tumour samples were performed with an anti-Ki67 antibody. Ki67-positive cells were quantified (right). h, TUNEL analyses of the indicated tumour samples were performed. Apoptotic cells were stained brown and quantified in n = 10 microscopic fields (right). Data are mean ± s.d. **P < 0.001 compared with the wild-type group (two-tailed t-test). i, j, Huh7 cells (1 × 106) with or without knock-in expression of PCK1(S90A) or INSIG1(S207A)/INSIG2(S151A) double mutants were intrahepatically injected into athymic nude mice (n = 7 per group). At 28 days after injection, the mice were euthanized and the liver tumours were dissected for immunoprecipitation, immunoblotting (i) and immunofluorescence (j) analyses with the indicated antibodies. k, Huh7 cells (1 × 106) with or without knock-in expression of PCK1(S90D) (top) or INSIG1(S207D)/INSIG2(S151D) double mutants (bottom) were intrahepatically injected into athymic nude mice (n = 7 per group). The mice were euthanized and examined for tumour growth 22 days after injection. The arrows indicate tumours. Tumour volumes were calculated (right). Data are mean ± s.d. (n = 7). **P < 0.001 compared with the wild-type group (two-tailed t-test). l, m, Huh7 cells (1 × 106) and Huh7 cells with knock-in expression of PCK1(S90D) (l) or INSIG1(S207D)/INSIG2(S151D) double mutants (m) were subcutaneously injected into the left or right flanks of athymic nude mice, respectively (n = 7 per group). The resulting tumors were resected 22 days after injection (left). The growth of xenografted tumours in the mice was measured (middle) and the tumours were weighed (right). Data are mean ± s.d. (n = 7). **P < 0.001 compared with the wild-type group (two-tailed t-test). All experiments were repeated three times independently with similar results.
Extended Data Fig. 10 Activation of the IGF1R–AKT–PCK1–INSIG1/2 signalling cascade is required for the growth of liver tumours and correlates with poor prognosis for HCC.
a, b, Huh7 cells (1 × 106) with or without knock-in expression of PCK1(S90A) (a) or INSIG1(S207A)/INSIG2(S151A) double mutants (b) were stably transfected with or without Flag–IGF1R(V922E) and intrahepatically injected into athymic nude mice (n = 7 per group). The mice were euthanized and examined for tumour growth 28 days after injection. The arrows indicate tumours. Tumour volumes were calculated (right). Data are mean ± s.d. (n = 7). *P =0.011 (a), 0.013 (b), **P <0.001 (two-tailed t-test). c, IHC analyses of xenografted tumours from nude mice were performed with the indicated antibodies. The regions in white boxes are shown at higher magnification below. d, e, Huh7 cells (1 × 106) with or without knock-in expression of PCK1(S90A) (d) or INSIG1(S207A)/INSIG2(S151A) double mutants (e) were stably transfected with or without Flag–myr-AKT1 and intrahepatically injected into athymic nude mice (n = 7 per group). The mice were euthanized and examined for tumour growth 28 days after injection. The arrows indicate tumours. Tumour volumes were calculated (right). Data are mean ± s.d. (n = 7). *P = 0.015 (d), 0.010 (e), **P < 0.001 (two-tailed t-test). f, IHC analyses of xenografted tumours from nude mice were performed with the indicated antibodies. The regions in white boxes are shown at higher magnification below. g, h, Huh7 cells (1 × 106) with or without knock-in expression of PCK1(S90D) (g) or INSIG1(S207D)/INSIG2(S151D) double mutants (h) were stably transfected with Flag–IGF1R(L1003R) and intrahepatically injected into athymic nude mice (n = 7 per group). The mice were euthanized and examined for tumour growth 28 days after injection. The arrows indicate tumours. Tumour volumes were calculated (right). Data are mean ± s.d. (n = 7). *P = 0.003 (g), 0.005 (h), **P < 0.001 (two-tailed t-test). i–n, Huh7 cells (1 × 106) with or without knock-in expression of PCK1(S90D) (i–k) or INSIG1(S207D)/INSIG2(S151D) double mutants (l–n) were stably transfected with or without Flag–IGF1R(L1003R) and subcutaneously injected into the left flanks of athymic nude mice (n = 7 per group). The resulting tumours were resected 28 days after injection (i, l). The growth of xenografted tumours in the mice was measured (j, m). The tumours were weighed (k, n). Data are mean ± s.d. (n = 7). *P= 0.002 (j), 0.011 (k), 0.009 (m), 0.016 (n), **P < 0.001 compared with the PCK1 wild-type + IGF1R(L1003R) group (j, k) or compared with the INSIG1/2 wild-type + IGF1R(L1003R) group (m, n) (two-tailed t-test). o–q, Huh7 cells (1 × 106) with or without knock-in expression of PCK1(S90D) or INSIG1(S207D)/INSIG2(S151D) double mutants were stably transfected with or without Flag–IGF1R(L1003R) and intrahepatically injected into athymic nude mice (n = 7 per group). The mice were euthanized 28 days after injection. IHC analyses of xenografted tumours from nude mice were performed with the indicated antibodies. The regions in white boxes are shown at higher magnification below. r–t, Wild-type pT3-EF1α-Flag-PCK1 (or pT3-EF1α-Flag-PCK1(S90A)), pT3-EF1α-HA-myr-AKT1 and pT3-EF1α-V5-c-Met, along with the sleeping beauty transposase (SB), were stably expressed in the mouse liver using hydrodynamic transfection into FVB/N mice (n = 7 per group). After 14 weeks, the mice were euthanized and representative liver tumours are shown (r, left). The average tumour volumes were measured (r, right) (n = 7 per group). **P < 0.001 (two-tailed t-test). s, IHC analyses of tumour samples were performed with an anti-Ki67 antibody (left). Ki67-positive cells were quantified in 10 microscopic fields (right). **P < 0.001 (two-tailed t-test). t, IHC analyses of liver tumours from nude mice were performed with the indicated antibodies. The regions in white boxes are shown at higher magnification below. u, IHC staining of 30 human HCC and matched non-tumour tissue samples was performed with the indicated antibodies. Representative images of two cases are shown. The regions in white boxes are shown at higher magnification below. v, Representative immunoblotting analyses of two cases of human HCC and matched non-tumour tissue samples was performed with the indicated antibodies. w, The indicated staining scores for PCK1(pS90), INSIG1(pS207)/INSIG2(pS151) and SREBP1 expression levels in HCC and matched non-tumour liver samples (n = 30) were compared using a paired t-test (two-tailed). Data are mean ± s.d. **P < 0.001 compared with the non-tumour adjacent tissue. x, Representative immunoblotting analyses of four different human HCC samples were performed with the indicated antibodies. y, IHC staining of human HCC samples with the indicated antibodies was scored, and correlation analyses were performed. A Pearson correlation test was used (two-tailed) (n = 40). Note that the scores for some samples overlap. z, Mechanism for PCK1-promoted activation of SREBP1 and lipogenesis. PEP, phosphoenolpyruvate; OAA, oxaloacetate; HC, hydroxycholesterol; bHLH, basic helix-loop-helix protein; Reg, regulatory domain of SREBP. All experiments were repeated three times independently with similar results.
Supplementary information
Supplementary Information
This file contains a Supplementary Discussion and associated references, legends to two Supplementary Tables, and Supplementary Figure 1 (raw gel data from Figures 1-3 and Extended Data Figures 1-10).
Source data
Rights and permissions
About this article
Cite this article
Xu, D., Wang, Z., Xia, Y. et al. The gluconeogenic enzyme PCK1 phosphorylates INSIG1/2 for lipogenesis. Nature 580, 530–535 (2020). https://doi.org/10.1038/s41586-020-2183-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41586-020-2183-2
This article is cited by
-
Metabolic reprogramming in the tumor microenvironment of liver cancer
Journal of Hematology & Oncology (2024)
-
Characterization of diseased primary human hepatocytes in an all-human cell-based triculture system
Scientific Reports (2024)
-
The integrate profiling of single-cell and spatial transcriptome RNA-seq reveals tumor heterogeneity, therapeutic targets, and prognostic subtypes in ccRCC
Cancer Gene Therapy (2024)
-
Upregulation of p300 in paclitaxel-resistant TNBC: implications for cell proliferation via the PCK1/AMPK axis
The Pharmacogenomics Journal (2024)
-
VHL suppresses autophagy and tumor growth through PHD1-dependent Beclin1 hydroxylation
The EMBO Journal (2024)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
