Introduction

The cellular prion protein (PrPc) is a highly conserved cell surface glycoprotein encoded by the PRNP gene, which is located on the short arm of chromosome 20 [1]. In humans, PrPc is expressed in various peripheral tissues, and to a higher extent in the nervous system [2]. Although the physiological role of PrPc remains to be fully established, its misfolded isoform scrapie PrP (PrPSc) is known to be key in the pathogenesis and transmission of prion diseases [3,4]. Prion diseases can be sporadic, inherited or infectious, and they include Creutzfeldt–Jakob disease (CJD), Gerstmann–Sträussler–Sheinker syndrome (GSS), fatal familial insomnia (FFI), kuru, bovine spongiform encephalopathy (BSE), and chronic wasting disease (CWD) [5].

PrPc misfolding occurs due to modifications in its secondary structure consisting of a decreased length of coiling α-helixes that are replaced by a long strip of β-sheets. The latter contributes to forming insoluble and protease-resistant PrPSc [6]. Within prion-infected brains, PrPSc forms pathological protein aggregates, which act as seeds for normal PrPc [7]. The accumulation of misfolded PrP may also derive from slowed PrPc clearance, which may be due, at least in part, to alterations in cell clearing pathways, mostly autophagy (ATG) (Figure 1). As proof of concept, ATG inducers foster PrPsc removal [8,9]. This is not surprising, as a wide class of “prion-like”, prone-to-misfold proteins (such as alpha-synuclein, SOD1, TDP-43, and FUS) may accumulate when a failure in cell clearing systems occurs [10,11].

Figure 1. Structure and turnover of the cellular prion protein (PrPc). The synthesis of PrPc requires the entry of the nascent protein into the lumen of the endoplasmic reticulum (ER), where the N-terminal signal peptide is removed, while a glycosyl-phosphatidyl-inositol (GPI) anchor remains attached to the C-terminal domain. Then, the protein moves to the Golgi apparatus to undergo post-translational modifications. Once completely folded, PrPc moves along the secretory pathway towards the outer leaflet of the plasma membrane, where it anchors via the GPI lipid moiety. Here, GPI-anchored PrPc is strategically associated with lipid rafts, suggesting an involvement in signal transduction and cell-to-cell communication. The clearance of PrPc depends on autophagy (ATG) and P26S proteasome systems. The accumulation of misfolded PrP leads to the formation of insoluble scrapie PrP (PrPSc), which may also derive from slowed PrPc clearance due to a failure of ATG. When ATG is impaired, endocytosed PrPc and PrPSc are rapidly recycled back to the plasma membrane or released extracellularly through exosomes. Black solid arrows indicate molecular steps (PrPc endocytosis, PrPc conversion into PrPSc, PrPs ubiquitination and recognition by the proteasome); black dotted arrows indicate ATG progression (fusion of PrPc-containing endosomes with autophagosomes and formation of autophagolysosomes), and PrPc/PrPSc degradation; red solid lines indicate the effects of ATG impairment; red dotted arrows indicate the exosomal release of undigested PrPc/PrPSc in the extracellular space.

Besides supporting the role of PrPsc as an infectious agent of prion disease, PRNP knockout (KO) experimental models have provided some insights into the physiological function of PrPc [12,13]. In the nervous system, PrPc is involved in neurite extension, neuronal differentiation, and neuroprotection [14,15]. More in general, PrPc is involved in copper metabolism, signal transduction, cell proliferation, adhesion, and migration [16]. Thus, albeit promoting differentiation of resident stem cells, PrPc may also promote stemness and cell proliferation, depending on specific conditions [17,18,19].

The discovery of PrPc expression in different types of stem cells joined with evidence on PrPc overexpression in a variety of tumors has recently prompted its investigation in cancer stem cell (CSC) research [20,21,22,23]. CSCs are endowed with enhanced self-renewal, sustained proliferation, and tumor-initiating potential. Thus, they are pivotal in fueling tumor growth and conferring therapeutic resistance, while sustaining tumor infiltration and relapse. This applies to both hematopoietic and solid tumors, where PrPc is markedly overexpressed, including pancreatic ductal adenocarcinoma (PDAC), breast cancer, gastric and colorectal cancer, and gliomas [21,22,23,24,25,26,27,28,29,30,31,32]. High levels of PrPc are associated with an enhanced CSCs’ tumorigenic potential, proliferation, and invasion, along with greater metastatic capacity, drug resistance, and angiogenesis. On the other hand, PrPc downregulation/inhibition suppresses tumor stemness, growth, proliferation, invasiveness, and angiogenesis [21,22,24,25,26,27,28,29,30,31,32].

Targeting PrPc in Cancer Stem Cells

Although targeting of PrPc has not yet been tested therapeutically in cancer, several experimental studies point to PrP downregulation/inhibition as a potential anti-cancer strategy in a variety of tumors. For instance, in colon cancer cells, treatment with anti-PrP antibodies reduces cell proliferation and invasiveness, even though an increased efficacy is observed in combination with the chemotherapeutic drugs irinotecan, 5-fluorouracil, cisplatin, and doxorubicin [27]. Similarly, combining PRNP silencing with fucoidan provides an enhanced efficacy against colorectal CSCs’ proliferation and migration in vitro, while reducing tumor volume and angiogenesis in vivo [28].

Either RNAi-mediated downregulation of PrPc or administration of anti-PrPc monoclonal antibodies inhibits both in vitro and in vivo tumorigenicity and invasiveness of colorectal CSCs by abrogating epithelial to mesenchymal transition (EMT) related to the ERK2 (MAPK1) pathway [21]. Similarly, PRNP silencing abrogates colorectal cancer cell stem-like and mesenchymal-like phenotype through inhibiting the recruitment of the Hippo pathway effectors YAP and TAZ, and the TGFβ pathway [32]. Knockdown of PrPc expression decreases lung adenocarcinoma cells’ lamellipodium formation, in vitro migration, and invasion, as well as in vivo experimental lung metastasis, which is associated with reduced JNK phosphorylation and reduced protein levels of a transcriptional activator of the PRNP promoter, namely, the nuclear factor interleukin 3 (NFIL3) [64].

In human PDAC and melanoma cell lines, PrP occurs as a pro-PrP isoform, being neither glycosylated nor GPI-anchored, albeit retaining the GPI anchor peptide signal sequence (GPI-PSS) [26,89]. This latter interacts with filamin A (FLNA) and integrin β1 to disrupt the cytoskeletal organization and promote cancer cell invasiveness and migration. Inhibiting PrP expression by shRNA or via GPI-PSS-targeting peptides reduces PDAC and melanoma cell proliferation and invasiveness in vitro as well as tumor growth in vivo [26,89]. In addition to filamin A, PrPc interacts with Notch1, forming a PrPc/FLNA/Notch1 complex, which is associated with enhanced PDAC proliferation, invasiveness, and xenograft tumor growth [29]. These effects are reverted by PrPc silencing through Notch1 downregulation, and combining PrPc and Notch1 inhibition is more effective than targeting single pathways alone.

Targeting the GPI anchor of PrPc was investigated as a device for targeting cancer cell proliferation and metastasis in renal carcinoma through tissue inhibitor of metalloproteinase (TIMP) engineering [90]. Fusing the TIMP-1 protein to the GPI anchor of PrP creates a membrane-tethered complex, which co-localizes on the cell surface with membrane type 1-matrix metalloproteinase (MT1-MMP). This prevents MMP-mediated proteolysis of ECM components while reducing cancer cell growth and proliferation in vitro as well as in mouse xenografts [90].

Downregulating PrPc abrogates CSCs’ resistance to chemotherapy. Secreted PrPc can directly sequester chemotherapeutic drugs, blocking their cytotoxic activity, as shown in breast cancer cells. Genetic depletion of PrPc prevents such an interaction while sensitizing breast CSCs to chemotherapy [91]. Down-regulation of PrPc by siRNA sensitizes breast cancer cells to adriamycin and tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) [20]. Moreover, administration of chlorpromazine, which exerts anti-prion effects, inhibits CSCs’ proliferation while preventing resistance to ionizing radiation and chemotherapeutic drugs in melanoma, breast cancer, and glioma [92,93,94].

Many of the beneficial effects of PrPc inhibition, including sensitization to chemotherapy, rely on the downregulation of the PI3K/Akt pathway, which, in turn, is bound to ATG stimulation. For instance, PrPc silencing counteracts the increased colorectal CSCs’ survival, proliferation, and 5-fluorouracil (5-FU) resistance through downregulation of PI3K/Akt [30]. In gastric cancer cells, in vitro administration of the Akt inhibitor LY294002 or Akt siRNAs leads to inhibition of PrPc-induced and CyclinD1-related CSC proliferation, G1/S phase transition, and multidrug drug resistance [25]. PrPc-induced multi-drug-resistance in gastric cancer cells is also attenuated by inhibition of PI3K/Akt following the knockdown of the PrPc-interacting protein LRP37 [95].

Although these studies did not directly address the role of ATG, evidence was provided in GBM indicating that PrPc silencing by DNA-antisense oligonucleotides promotes mTOR-dependent ATG activation to halt GSCs’ proliferation and growth [85]. These consist of the induction of LC3-II, Beclin-1, and a simultaneous decrease in p62, Bcl-2, along with inhibition of mTOR. Even PrPc degradation by the proteasome system was shown to decrease tumor progression, which is remarkable because the very same Akt/mTOR pathway synergistically modulates both p26S proteasome and ATG [96,97]. In detail, the tumorigenicity of colorectal cancer cells is associated with PrPc accumulation and upregulation of heat-shock 70 kDa protein-1-like (HSPA1L), which, in turn, stabilizes the hypoxia-inducible factor-1α (HIF-1α) protein. This latter interacts with the ubiquitin-protein E3 ligase glycoprotein 78 (GP78) to inhibit proteasome-dependent degradation of PrPc. Thus, targeting the HSPA1L/HIF-1α/GP78 axis may counteract PrPc accumulation and tumor progression by stimulating cell-clearing systems [98].

Noteworthy, similar to what reported for PI3K/Akt/mTOR inhibition and ATG activation [76], PrPc downregulation counteracts GBM growth and self-renewal by promoting CSCs’ differentiation. For instance, the blockade of PrPc-HOP/STI-1 interaction by means of a HOP peptide mimicking the PrPc-binding site (HOP230–245) counteracts GSCs’ proliferation and self-renewal through inhibition of PI3K/Akt [66]. Similarly, administration of HOP230–245 peptide to mice bearing GBM xenografts decreases tumor volume while extending animals’ survival [66]. In turn, PrPc/HOP silencing reduces GSCs’ proliferation by downregulating the stemness markers CD133, CD15, Oct4, and Sox2, while promoting GSCs’ differentiation [22]. These effects are replicated in vivo where the tumorigenic potential of GSCs is markedly inhibited in GBM xenografts lacking PrPc and/or HOP. PrPc depletion also impairs GSCs’ migration and invasiveness by downregulating cell adhesion-related proteins [22].

Again, the downregulation of Notch1, which occurs as a downstream effect of PrPc silencing [29], inhibits the PI3K/Akt/mTOR pathway to abolish GSCs’ stemness, self-renewal, invasiveness and in vivo tumor growth [99]. It is remarkable that these effects are reproduced by the ATG inducers AZD8055 and rapamycin, which suppress GSCs’ self-renewal and abolish GSCs tumorigenicity through degradation and inhibition of Notch1 [84].

Indirect evidence for ATG involvement is provided in colorectal CSCs as well, where the combination of 5-fluorouracil (5-FU) with the ATG inducer melatonin inhibits PrPc, along with the expression of the stem cell markers Oct4, Nanog, Sox2, and ALDH1A1 while suppressing tumor growth, proliferation, and angiogenesis [31].

In summary, strategies aimed at downregulating PrPc expression, including stimulation of ATG-dependent PrPc clearance may produce beneficial effects in cancer in general, and in GBM in particular, by inhibiting CSCs’ stemness, self-renewal, proliferation, invasiveness, and resistance to radio-/chemo-therapy (Figure 2).

Figure 2. Targeting the cellular prion protein (PrPc) in cancer stem cells (CSCs). Strategies aimed at downregulating/inhibiting PrPc in CSCs include anti-PrP antibodies, PrP silencing (siRNA), administration of chlorpromazine, anti-PrP/stress-inducible protein 1 (STI-1)/Hsp70/Hsp90 organizing protein (HOP) peptides, PI3K/Akt/mTOR inhibitors, and combined strategies such as anti-PrP antibodies or PrP silencing (siRNA) with chemotherapy (CT), and PrP silencing with Notch inhibitors. These strategies counteract CSCs’ stemness, self-renewal, growth, proliferation, epithelial to mesenchymal transition (EMT), invasiveness, and resistance to radio/chemotherapy. PI3K/Akt/mTOR inhibitors also counteract PrPc accumulation by fostering its degradation through the cell-clearing pathways’ autophagy (ATG) and proteasome (red cross PrP degradation). ATG activation also downregulates Notch signaling, which is enhanced upon association with PrPc (red cross Notch1). In this way, PrP-induced pathways sustaining CSCs’ phenotype and PrP exosome release are toned down. This is key, as secreted PrP acts via autocrine and paracrine mechanisms to foster CSCs’ phenotype and can limit therapeutic drugs’ efficacy via direct sequestration of CT. Black solid lines indicate molecular steps (arrows: Stimulation; “T”-shaped lines: Inhibition); black dotted arrows indicate PrPc-related molecular steps leading to the activation of intracellular signalling pathways.