Tumor protein D52 (TPD52) affects cancer cell metabolism by negatively regulating AMPK

Abstract

Background: The AMP-activated protein kinase (AMPK) is a central regulator of energy homeostasis, with deregulation leading to cancer and other diseases. However, how this pathway is dysregulated in cancer has not been well clarified.

Methods: Using a tandem affinity purification/mass-spec technique and biochemical analyses, we identified tumor protein D52 (TPD52) as an AMPKα-interacting molecule. To explore the biological effects of TPD52 in cancers, we conducted biochemical and metabolic assays in vitro and in vivo with cancer cells and TPD52 transgenic mice. Finally, we assessed the clinical significance of TPD52 expression in breast cancer patients using bioinformatics techniques.

Results: TPD52, initially identified to be overexpressed in many human cancers, was found to form a stable complex with AMPK in cancer cells. TPD52 directly interacts with AMPKα and inhibits AMPKα kinase activity in vitro and in vivo. In TPD52 transgenic mice, overexpression of TPD52 leads to AMPK inhibition and multiple metabolic defects. Clinically, high TPD52 expression predicts poor survival of breast cancer patients.

Conclusion: The findings revealed that TPD52 is a novel regulator of energy stress-induced AMPK activation and cell metabolism. These results shed new light on AMPK regulation and understanding of the etiology of cancers with TPD52 overexpression.

Keywords: AMP-activated protein kinase (AMPK); cell metabolism; tumor protein D52 (TPD52).

FIGURE 1

TPD52 interacts with AMPKα. (A) Tandem affinity purification was performed using HEK293T cells stably expressing N‐FLAG‐S‐tagged TPD52. The major hits from mass spectrometry analysis were shown in the table. (B) His‐tagged AMPKα1, α2, β1, β2, and γ1 were overexpressed and purified from Escherichia coli, and a GST pull‐down assay using GST‐TPD52 was carried out to determine the in vitro interactions. (C) TPD52 interacts with AMPK endogenously in 293T cells. Irrelevant IgG was used as the immunoprecipitation control. The whole‐cell lysate was used as an input. (D) His‐tagged AMPKα1 kinase domain (AMPKαE) was overexpressed and purified from E. coli, and a GST pull‐down assay of TPD52 using the purified proteins was carried out to determine the interactions. (E) Mapping the regions of TPD52 required for AMPK binding. Upper panel, schematic representation of TPD52 constructs and the minimum interaction region. Lower panel, GST‐tagged TPD52 full‐length (FL) or deletion mutants were purified from E. coli, and a GST pull‐down assay was performed.

FIGURE 2

TPD52 regulates AMPK activity and cellular metabolism. (A) SK‐BR‐3 cells were transfected with indicated siRNAs. The phosphorylations of AMPKα, ACC1, and TSC2 in cell lysates were detected by Western Blot. (B) MDA‐MB‐231 cells were transfected with the indicated constructs, and the phosphorylations of AMPKα, ACC1, and TSC2 in cell lysates were detected by Western Blot. (C) MDA‐MB‐231 cells were transfected with the indicated constructs, and the intracellular lipid droplet level was assessed by Oil Red O staining. All error bars represent the SD from the mean value of three independent experiments. *p < 0.05. (D) Overexpression of TPD52 increases lactate production. The lactate production was measured in a medium collected at 48 h after transfection of control or TPD52 expressing constructs in MDA‐MB‐231 cells. The results represent the mean ± SEM of three independent experiments. *p < 0.05. (E) Overexpression of TPD52 increases glucose intake. The glucose intake was measured in a medium collected at 48 h after transfection of control and TPD52 overexpressing constructs in MDA‐MB‐231 cells. The results represent the mean ± SEM of three independent experiments. *p < 0.05. (F) Overexpression of TPD52 increases lipogenesis and glycolytic gene expression. Relative expression levels of fas, ldha, and pdk1 mRNA in control or TPD52 overexpressed MDA‐MB‐231 cells were determined by qPCR. Transcript levels were determined relative to actin mRNA levels and normalized to control cells. The results represent the mean ± SEM of three independent experiments. *p < 0.05.

FIGURE 3

TPD52 directly regulates AMPK kinase activity. (A) SK‐BR‐3 cells were treated as indicated for 24 h, and the interaction between TPD52 and AMPKα was examined by co‐IP. (B) MDA‐MB‐231 cells were stably transfected with TPD52, and LKB1–AMPKα interaction was examined by co‐IP. (C) TPD52 directly inhibits ACC1 Ser79 phosphorylation by AMPKα in vitro. GST‐TPD52, AMPKα, and ACC1N were purified from E. coli as indicated in the Methods, and an in vitro kinase assay was performed. (D) MDA‐MB‐231 cells were transfected with the indicated constructs, and the intracellular lipid droplets were assessed by Oil Red O staining. All error bars represent the SD from the mean value of three independent biological replicates. (E) The lactate production was measured in a medium collected at 48 h after transfection of control or TPD52 WT or D4 mutant expression constructs in MDA‐MB‐231 cells. The results represent the mean ± SEM of three independent experiments. *, p < 0.05. (F) The glucose intake was measured under the same condition as the previous ones. The results represent the mean ± SEM of three independent experiments. *p < 0.05.

FIGURE 4

TPD52 regulates AMPK activity and cell metabolism in vivo. (A, B) The Oil Red O staining and H&E staining of the liver tissues from the NFD‐ and HFD‐fed WT and HA‐TPD52 transgenic mice. (C) Blood glucose levels were measured in tail vein blood samples using a glucometer. Values are expressed as mean ± SD (n = 6). *p < 0.05. (D) Serial changes in blood glucose levels after intraperitoneal injection of glucose in the indicated mice (n = 6). Values are expressed as mean ± SD. *p < 0.05. (E) AMPK phosphorylation, precursor (−p), and nuclear‐processed (−n) SREBP1c and HA‐TPD52 levels in liver tissues from the indicated mice were examined by Western Blot. (F) Expression of TPD52 and phosphorylated ACC1 and AMPK in livers from the NFD‐ and HFD‐fed WT and HA‐TPD52 transgenic mice was determined by IHC staining. (G, H) mRNA levels of lipogenesis gene (fas and scd1) and glycolytic gene (pdk1 and ldha) in indicated liver cells from the WT and HA‐TPD52 transgenic mice were determined by qRT‐PCR. Relative mRNA levels were corrected to actin mRNA levels and normalized relative to control cells. The results represent the mean ± SEM of three independent experiments. ***p < 0.001, **p < 0.01, *p < 0.05.

FIGURE 5

TPD52 expression is associated with clinical outcome and metabolic gene expression in breast cancer patients. (A) Kaplan–Meier (KM) survival analysis of the breast invasive carcinoma patients in TCGA cohort (https://portal.gdc.cancer.gov, TCGA‐BRCA, V13.0, 2018). The patients were divided into high‐ and low‐expression groups based on the medium FPKM cutoff value of the TPD52 level. **p < 0.01. (B) TPD52 and the lipogenesis genes (fas, scd1, and acaca/acc1) are coexpressed in breast invasive carcinomas. The correlation was calculated using Pearson's correlation coefficients (R). (C) TPD52 and the glycolysis genes (aldoa and ldha) are coexpressed in breast invasive carcinomas. The correlation was calculated using Pearson's correlation coefficients (R).