An autophagy-independent function of Beclin 1 in cancer

Abstract

The involvement of Beclin 1 in cancer has been attributed primarily to its role in autophagy initiation. Our recent findings identify a novel alternative mechanism by which Beclin 1 can affect tumor progression through the regulation of growth factor receptor signaling.

The involvement of Beclin 1 in cancer was first indicated over a decade ago in studies demonstrating that its overexpression inhibits MCF-7 breast carcinoma xenograft tumor growth.¹ The subsequent generation of Beclin 1 knockout mice (Becn1^−/−) revealed that heterozygous disruption of Becn1 expression promotes spontaneous tumorigenesis in mice, identifying Beclin 1 as a haploinsufficient tumor suppressor.^2,3 In human tumors, Beclin 1 mRNA and protein levels are frequently decreased and these lower levels are often associated with poor prognosis and decreased overall patient survival.^4,5 Taken together, the mouse model and human tumor data support the hypothesis that reduced Beclin 1 expression in human cancer may contribute to disease progression.

Beclin 1 is the mammalian homolog of yeast vacuolar protein sorting-associated protein 30 (Vps30; also known as autophagy-related protein 6 [Atg6]) and functions within a core complex that includes class III phosphatidylinositol (PI)-3 kinase (PI3KC3; also known as Vps34) and PI3-kinase p150 subunit (p150; also known as Vps15).⁶ Beclin 1 stimulates the activity of PI3KC3 to generate phosphatidylinositol-3 phosphate (PI3P), which is necessary for the recruitment of effector proteins containing a FYVE (Fab1p, YOTB, Vac1p, EEA1) or PX (Phox homology) domain that facilitate membrane fusion and trafficking events. Beclin 1 has been implicated in multiple membrane trafficking pathways that require PI3P including autophagy, vacuolar protein sorting, cytokinesis, phagocytosis, and endocytosis.⁶ This promiscuous activity raises an important question regarding which of these Beclin 1-regulated pathways are important for its tumor suppressor function. To date, the contribution of Beclin 1 to cancer has been ascribed primarily to its role in the regulation of the degradative process of autophagy. In contrast, alternative functions of Beclin 1 have been relatively understudied in the context of cancer. This is important given that disruption of other essential autophagy pathway genes has not recapitulated the results from the Becn1^+/− mice with regard to tumor development. It is this discrepancy that supports a potential role for autophagy-independent Beclin 1 functions in cancer.

Our recent studies have identified a role for Beclin 1 in the control of growth factor receptor signaling.⁷ Regulation of epidermal growth factor receptor (EGFR) degradation by Beclin 1 had been reported previously, but the functional consequences of this regulation were not known.⁸ Furthermore, the involvement of Beclin 1 in regulating other growth factor receptors had not been investigated. Our studies revealed that Beclin 1 regulates both EGF and insulin-like growth factor-1 (IGF-1) stimulated activation of AKT and extracellular signal-regulated kinase (ERK) in breast carcinoma cells. Beclin 1 controls this signaling by regulating the maturation of early, signaling competent endosomes.⁷ Beclin 1 is required for the growth factor-stimulated production of PI3P by PI3KC3, and in the absence of Beclin 1 the transition of PI3P-negative endosomes to PI3P-positive endosomes is delayed.⁷ The extended residency time of growth factor receptors in the PI3P-negative compartment sustains the longevity of downstream signals.

Our data indicate an autophagy-independent mechanism for the regulation of growth factor receptor signaling by Beclin 1. Reduced expression of ATG5, an essential autophagy pathway gene, does not recapitulate the early endosome maturation defects that we observed in Beclin 1-deficient cells.⁷ Beclin 1 recruits distinct binding partners to regulate autophagy initiation (ATG14L; also known as Beclin 1-associated autophagy-related key regulator (Barkor)/Atg14; Complex I) or endocytic receptor trafficking (UV radiation resistance-associated gene [UVRAG]; also known as Vps38; Complex II).⁶ Suppression of the expression of UVRAG, but not of ATG14L, delays EGFR degradation.⁸ Moreover, impaired PI3P-positive endosome formation in Becn1^−/− mouse embryo fibroblasts is rescued by overexpression of UVRAG but not of ATG14L or a UVRAG-binding deficient Beclin 1 mutant.⁹ These latter studies implicate the involvement of Complex II in the regulation of endosome maturation that leads to enhanced signaling, and provide additional support for an autophagy-independent mechanism of action in this regulation.

An important question that arises from cell-based studies of Beclin 1 loss is whether there is evidence to support autophagy-independent functions of Beclin 1 in vivo. Beclin 1 associates with endosomes in neurons and early endosome antigen 1 (EEA1), a FYVE-domain protein, fails to localize to early endosomes in Becn1^−/− neurons, indicating deficient PI3P levels.⁹ A decrease in late endosome formation is also observed in Beclin 1-deficient neurons. In human breast tumors we identified a negative correlation between Beclin 1 expression and AKT and ERK activation, supporting a potential role for Beclin 1 loss in enhancing these signaling pathways in tumors.⁷ However, these studies do not allow conclusions to be drawn regarding which Beclin 1-regulated pathways are responsible for the functional outcomes of Beclin 1 loss, since both autophagy and autophagy-independent functions are lost when total Beclin 1 expression is decreased. Future studies that can distinguish between each pathway are needed to further discriminate the contributions of Beclin 1 action in cancer, as well as other pathologic conditions such as neurodegenerative disease.

There is growing interest in targeting autophagy for cancer therapy. In this regard, several groups have recently reported the development of selective PI3KC3 inhibitors.¹⁰ Although this approach was developed with the goal of inhibiting autophagy in cancer cells, inhibiting PI3KC3 activity would mimic Beclin 1 suppression by decreasing PI3P production, increasing growth factor-dependent signaling pathways concomitant with inhibition of the autophagic pathway (Fig. 1). The dual impact of this inhibition raises concern, especially in light of our data showing that increased signaling enhances the invasive potential of breast carcinoma cells.⁷ Understanding how the simultaneous loss of autophagy and gain of signaling affects tumor growth and progression is essential in order to move forward with approaches to target this pathway for the treatment of cancer.

Figure 1.

Dual impact of Beclin 1 loss on cancer. Beclin 1 regulates both early endosome maturation and autophagy initiation through its interactions with and activation of Class III phosphatidylinositol (PI)-3 kinase (PI3KC3/VPS34). Decreased Beclin 1 expression inhibits autophagy, which can promote tumor initiation by increasing oxidative stress that promotes DNA damage and genomic instability. Diminished autophagy would also contribute to increases in necrosis and associated inflammation. Beclin 1 deficiency would also delay early endosome maturation (PI3P- to PI3P+ endosomes) and in doing so, enhance the intensity and longevity of growth factor receptor signaling. This increased signaling stimulates tumor invasion that would promote tumor progression. Sustained growth factor signaling may also further suppress autophagy initiation through mTOR (mechanistic target of rapamycin) activation. BECN1, Beclin 1; p150, PI3-kinase p150 subunit; ULK 1/2, Unc-51-like kinase 1/2; PI3P, phosphatidylinositol-3 phosphate; ROS, reactive oxygen species.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.