Cancer stem cells: The potential role of autophagy,proteolysis,and cathepsins in glioblastoma stem cells

Abstract

One major obstacle in cancer therapy is chemoresistance leading to tumor recurrence and metastasis. Cancer stem cells, in particular glioblastoma stem cells, are highly resistant to chemotherapy, radiation, and immune recognition. In case of immune recognition, several survival mechanisms including, regulation of autophagy, proteases, and cell surface major histocompatibility complex class I molecules, are found in glioblastoma stem cells. In different pathways, cathepsins play a crucial role in processing functional proteins that are necessary for several processes and proper cell function. Consequently, strategies targeting these pathways in glioblastoma stem cells are promising approaches to interfere with tumor cell survival and will be discussed in this review.

Keywords

Major histocompatibility complex class I cathepsin glioblastoma autophagy

Introduction of cancer stem cell research

The concept of cancer stem cells (CSCs), which is considered to be one of the milestones in our understanding of cancer,¹ was first proposed 80 years ago (1937) for leukemia.² While key experiments were performed in the 1960s and 1970s,^3–5 it was not until the 1990s that the first CSCs were isolated, initially from leukemia.^6,7 This was followed by the identification of putative stem cells in a range of solid tumors, first in breast cancer,⁸ then in brain,⁹ prostate,¹⁰ head and neck,¹¹ pancreas,¹² colon,¹³ mesenchymal,¹⁴ and lung cancer.¹⁵

CSCs are the only tumor cells showing the capability to self-renewal and multi-potent differentiation usually found in stem cells. These abilities of CSCs are often condensed to “stemness” which gives rise to the assumption that CSCs derive from mutated stem cells, indicating that the CSCs model resolves many previously debated issues.^16,17 CSCs developed different mechanisms to evade and resist therapies. DNA repair is more efficient in CSCs in contrast to differentiated tumor cells.¹⁸ This might explain, at least in part, why radiation or chemotherapy does effect CSCs only poorly. In general, CSCs feature a high basal level of genomic instability finally leading to higher resistance to therapy, initiation of regression and metastasis, and is linked to a poor prognosis.¹⁹ Furthermore, high expression of ATP-binding cassette (ABC) transporters, which transport chemotherapeutics to the cell membrane for exocytosis²⁰; an increase of aldehyde dehydrogenase (ALDH) accompanied with high levels of detoxification; or the accumulation of anti-apoptotic proteins might account for a restricted success of chemotherapy.²¹

However, the concept of CSCs is not without controversy,²² and, importantly, CSCs should not be confused with the cell of origin,²³ but with its hierarchical order, it emphasizes the view of tumors as complex ecological systems which act along strict Darwinian principles found in ecology and population genetics. While it has been proposed that the cells of origin are tissue stem cells or progenitor cells,^24,25 this view is not universally accepted²⁶ and recent modeling data can be seen to support the notion that any tissue cell should be drawn as putative cell of origin giving rise to CSCs.²⁷ An alternative model, shown by us²⁸ and others^29–31 to be potentially applicable for glioblastoma stem cells, postulates that differentiated (tumor) cells can de-differentiate to stem cells, in essence completely separating the concepts of CSCs and cell of origin. Interestingly, while de-differentiation has also been observed to occur in normal tissue cells,³² it takes place at a much higher frequency within populations of differentiated cancer cells.³³

Taken together, CSCs are heterogenic cell populations from different origin, which have not been fully elucidated yet, and present many mechanisms explaining resistance to cancer therapy and recurrence. In the next section, we will discuss autophagy and their potential targets to interfere with the survival of CSCs.

Autophagy and its dual role in cancer development

Cells need specific mechanisms to respond to extra- and intracellular stresses such as nutrient starvation, changes in insulin concentrations, growth factor levels, endoplasmic reticulum (ER) stress, or hypoxia. Autophagy is an intracellular recycling process that transports damaged proteins and organelles to the lysosome for degradation. The degraded proteins are then transported back into the cytosol and can be metabolized to energy carriers. Autophagy is not only a response to cell stress but also maintains homeostasis in healthy cells by eliminating damaged organelles, such as peroxisomes and mitochondria.³⁴

Autophagy is a (patho-)physiological process which occurs in both healthy and malignant cells and can act as a tumor-suppressing or tumor-promoting (oncogenic) factor.³⁵ In case of tumor suppression, autophagy protects from accumulation of reactive oxygen species (ROS), damaged proteins, or organelles by degradation, thereby maintaining cellular homeostasis. Accumulation of ROS can lead to DNA damage and genomic instability and by this means supports tumorigenesis. Thus, autophagy prevents healthy cells from cancer development.³⁶ Furthermore, autophagy is important by the fact that monoallelic loss of the autophagy gene encoding Beclin1 increases the risk of human ovarian, breast, and prostate tumors.³⁷ However, more recent studies questioned a distinct role of Beclin1 in tumor growth.³⁸ Nevertheless, defects in other autophagy-associated genes (e.g. ATG5) are linked to cancer.³⁵

Once tumor cells are established, autophagy can be tumor-promoting by protecting malignant cells from stressors and facilitates their survival. If the tumor reaches a diameter larger than 1–2 mm, angiogenesis is essential to supply tumor cells with nutrition and oxygen. However, tumor-induced neoangiogenesis is normally incomplete, de facto tumor cells suffer from hypoxia and subsequent metabolic stress. CSCs are often located in a hypoxic core region of a tumor and have to fight against consequences of oxygen shortage.^39,40 Interestingly, hypoxia plays a crucial role in regulating autophagy since the absence of oxygen leads to inhibition of mammalian target of rapamycin complex 1 (mTORC1) and subsequent decreased inhibition of the unc-51 like autophagy activating kinase 1 (Ulk1) complex finally resulting in activation of autophagy. Consequently, application of anti-angiogenic drugs, which lead to hypoxia, can be cytotoxic for tumor cells but might also generate adapted and even more potent CSCs as demonstrated by Conley et al.⁴¹ for breast CSCs.

Another important role of autophagy in CSCs is linked to the Warburg effect. The Warburg effect demonstrates a hallmark of cancer development since all tumor cells upregulate their levels of aerobic glycolysis and reduce mitochondrial energy supply. Consequently, the metabolic process rises and autophagy provides CSCs with necessary nutrients and supports metabolic reconfiguration.^42,43 In addition, autophagy supports metastasis since the performance of autophagy protects detached cells from anoikis which is a form of programmed cell death.⁴⁴ However, there is evidence that high autophagic activity can diminish CSCs by excessive degradation of organelles and proteins.³⁵ This underlines the assumption that autophagy plays a dual role in tumorigenesis.

Different types of autophagy pathways

There are several types of autophagy pathways: microautophagy, macroautophagy, chaperone-mediated autophagy, pexophagy, mitophagy, and xenophagy.^45–47 Of these, macroautophagy is the best-known autophagy mechanism being involved in antigen presentation and is therefore a potential target for cancer therapy.⁴⁸ During the process of macroautophagy, cytoplasmic constituents are separated by a double membrane forming the autophagosome by fusing with the lysosome to build the autolysosome. The enclosed content can be degraded by lysosomal proteases mostly cathepsins.^49,50 Microtubule-associated proteins (MAPs) 1A/1B light chain 3 (LC3) and ubiquitin-binding protein p62 (sequestosome 1, p62) play a critical role in autophagy and serve as common indicators for the autophagic flux. LC3-I is processed to LC3-II by the cytosolic cysteine protease autophagy-related 4 (ATG4) and is covalently conjugated to phosphatidylethanolamine which is located at the outer autophagosome membrane. Thereby, LC3-II accumulates and represents an early marker for autophagy.^51,52 p62 directs polyubiquitinated protein aggregates to the autophagosome via binding to LC3-I. As a result, the protein aggregates and p62 and LC3-II are also degraded by resident cathepsins. A decrease of p62 indicates autophagic degradation and is a marker for the late stage of autophagy.^51,53,54 Inhibition of autophagy leads to an accumulation of p62 which subsequently interferes with tumor proliferation by deregulation of NF-kB signaling or by an increase in ROS.⁵⁵ In this section, we will focus on the regulation of autophagy.

Regulation of autophagy

Mutations in the core autophagy signaling cascade, that is, in the Atg genes, have been found to be associated with several groups of disorders (metabolic, pulmonary, infectious, vascular, neurodegenerative diseases as well as cancer⁵⁶). While dysregulation of proteins involved in autophagy has been demonstrated for several cancers, such as hepatocellular carcinoma, endometrial adenocarcinomas, and colorectal cancer,⁵⁷ no study has yet found similar aberrations in glioblastoma. However, the PI3K/Akt/mTOR signaling cascade is activated in almost 90% of all glioblastoma, predominantly via inactivation (frequently mutated in cancer cells) of its negative regulator PTEN.⁵⁸

Macroautophagy is regulated by the presence or absence of intra- and extracellular stressors like intracellular metabolic stress, nutrient starvation, hypoxia, growth factor deprivation, and stress related to the ER.^59–61 Inhibition of autophagy under favorable conditions (presence of amino acids, growth factors, oxygen, and glucose) is usually mediated via two major components, namely, cyclic adenosine monophosphate (cAMP)-dependent protein kinase A (PKA) and mTORC1.⁶² Active PKA and mTORC1 inhibit autophagy processes by phosphorylating Ulk1 and Ulk2, thereby preventing Ulk1/Ulk2-complex formation and subsequent autophagosome formation.^63,64

There is a crosstalk between the stemness-related pathways (Wnt/β-catenin, Hedgehog, and Notch) and the regulation of autophagy on different levels. A recent report showed that autophagy leads to degradation of Notch1, thereby influencing stem cell differentiation. Knockdown of autophagy-regulating components like ATG7 or ATG16L1 resulted in increased levels of Notch, the transcription factor fragment notch intracellular domain (NICD), and its target gene Hes1. Consequently, in neuronal cultures prepared from mice with partial loss of function of the Atg16L1 gene, more stem cells could be found in comparison to wild-type mice.⁶⁵ Conversely, Natsumeda et al.⁶⁶ observed that pharmacological inhibition of Notch leads to protective induction of autophagy in glioma neurospheres finally resulting in chemoresistance.

Targeting the autophagy pathway

Autophagy can have tumor-promoting capacity in tumor cells; an optimal drug combines the ability to target autophagy and is selective for CSCs. However, one obstacle of standard drugs is specificity since CSCs and stem cells share the same surface markers, survival signals, DNA repair, self-renewal, and differentiation mechanisms. Examples of autophagy inhibitors are chloroquine (CQ) and hydroxychloroquine (HCQ), known as anti-malarial drugs exhibiting anti-tumor capacity by preventing autophagosomal maturation and inhibition of lysosomal degradation^67,68 (Table 1). Administration of CQ and HCQ is promising in autophagy-targeting therapy, and some studies are currently under way to investigate the effects of supplementing conventional anti-cancer therapy with CQ/HCQ.⁹⁰ Furthermore, the epidermal growth factor (EGF) receptor is a target in CSC therapy, thereby showing promising clinical results in epithelial as well as head and neck cancer when combined with conventional therapies.⁹¹

Table 1.

Summary of different autophagy inhibitors.

Compound	Mechanism	Reference
Bafilomycin A1	Inhibits autophagosome–lysosome fusion by inhibiting vacuolar ATPase and lysosomal acidification	Yamamoto et al.⁶⁹
(Hydroxy-)chloroquine	Prevents lysosomal acidification and prevents fusion of endo-/lysosomes, inhibits also CatB	Steinman et al.,⁷⁰ Seglen et al.,⁷¹ and Wibo and Poole⁷²
DBeQ	Inhibits autophagosome maturation by inhibition of p97 ATPase	Chou et al.⁷³
E-64d	Interferes with autolysosomal digestion. E-64d is a cell permeable inhibitor of cysteine proteases	Wilcox and Mason⁷⁴
l-asparagine	Selectively inhibits autophagic–lysosomal delivery	Hoyvik et al.⁷⁵
Leupeptin	Inhibitor of serine and cysteine proteases, inhibits plasmin, trypsin, papain, and calpain as well as several cathepsins	Seglen et al.⁷¹
LY294002	Inhibits autophagic sequestration, inhibitor of PI3K	Blommaart et al.⁷⁶
3-Methyladenine	Inhibits autophagic sequestration, inhibitor of PI3K	Blommaart et al.⁷⁶ and Seglen and Gordon⁷⁷
Methylamine	Prevents lysosomal acidification, thereby inhibiting lysosomal enzymes	Seglen et al.⁷¹
MHY1485	Activates mTOR and suppresses fusion of autophagosomes and lysosomes	Choi et al.⁷⁸
Monensin	Inhibits lysosomal protein degradation by disturbing lysosomal pH	Grinde⁷⁹
N-acetyl-l-cysteine	Decreases phosphorylation of c-Jun N-terminal kinase (JNK) and its substrate Bcl-2 which dissociates from Beclin1	Underwood et al.⁸⁰
Nocodazole	Depolymerizes microtubules, thereby preventing transport of endo-/lysosomes and fusion of autophagosomes with the lysosomal compartment	Webb et al.⁸¹ and Reunane et al.⁸²
Pepstatin A	Interferes with autolysosomal digestion. Irreversibly inhibits aspartic proteases (CatD, CatE)	Deng et al.⁸³
Pyrvinium	Inhibits autophagosome formation and reduces the transcription of autophagy-related genes	Deng et al.⁸³
SBI-0206965	Inhibits autophagy-initiating kinases Ulk1 and Ulk2 and suppresses autophagy which is induced by mTOR inhibition	Egan et al.⁸⁴
Spautin-1	Inhibits ubiquitin-specific peptidases and thereby promotes Beclin1 ubiquitylation and degradation	Liu et al.⁸⁵
Thapsigargin	Blocks autophagosome fusion with the endocytic system by blocking recruitment of Rab7 to autophagosomes	Ganley et al.⁸⁶
Vinblastine	Depolymerizes microtubules and thereby prevents transport of endo-/lysosomes and fusion of the autophagosome with the lysosomal compartment	Hoyvik et al.,⁸⁷ Punnonen and Reunanen,⁸⁸ and Munafo and Colombo⁸⁹
Vitamin E	Leads to increased phosphorylation of mTOR substrates resulting in autophagy inhibition	Underwood et al.⁸⁰
Wortmannin	Inhibits autophagic sequestration, inhibitor of PI3K	Blommaart et al.⁷⁶

mTOR: mammalian target of rapamycin.

For a long time, plant-derived natural compounds are used as alternative medicine and described as potential anti-cancer drugs. Curcumin, an extract of turmeric (Curcumin longa), was reported to regulate autophagy via elevated levels of LC3-II and Beclin1 finally leading to autophagosome formation.^92,93 Resveratrol, a phytoalexin found among grapes, induces Beclin1-independent autophagy in breast cancer cells by inhibition of Akt/protein kinase B (PKB)–mediated phosphorylation and mTOR signaling.⁹⁴ Similarly, the flavonoid quercetin, occurring in many fruits and vegetables, is known to trigger the accumulation of hypoxia-inducible factor-1α (HIF-1α) by blocking mTOR signaling.^95,96 Thus, plant-derived natural extracts are promising components in targeting autophagy and do not require time- and money-consuming approval.

In the next part, we will have a closer look into glioblastoma stem cell development, their resistance to treatment, and tumor recurrence which is the most serious challenge in glioblastoma research. Furthermore, we will elucidate the proteolytic machinery of glioblastoma stem cells and how these cells can escape immune detection.

A novel cathepsin-based therapeutic approach that reduces MHC I–mediated immune evasion in glioblastoma stem cells

Glioblastoma

Glioblastoma, also known as astrocytoma grade IV, is the most common and aggressive tumor of the central nervous system characterized by rapid cell proliferation, high vascularity, and diffuse invasion of the surrounding healthy tissue. The current therapeutic approach after diagnosis comprises neurologically safe maximal surgical removal and courses of both radio- and chemotherapy.^97,98 Although many novel treatment strategies have been investigated in the last decade following the addition of the chemotherapeutic drug temozolomide⁹⁹ to the standard treatment protocol, most of them have led to no significant therapeutic improvement or—in the case of bevacizumab—have yielded rather controversial results.¹⁰⁰ This is, at least partially, due to the presence of the blood–brain barrier limiting the application of many drugs as well as genetic alterations causing resistance to multimodal therapy approaches, leaving a limited range of treatment options that only prolong survival and quality of life for a few months.¹⁰¹ Therefore, despite intense efforts, glioblastoma remains one of the most lethal tumors with a mean patient survival of 14.6 months.¹⁰² Besides increased apoptotic resistance, there are several other cellular aspects that make glioblastoma a highly lethal and difficult-to-treat disease: Glioblastoma grows diffusely and highly invasive, thereby infiltrating the surrounding brain tissue and making localized treatment, for example surgery, particularly ineffective. For example, after surgical removal tumor recurrence manifests within 2–3 cm of the resection cavity wall in more than 95% of cases.¹⁰³ Furthermore, systemic chemotherapy and whole brain irradiation have so far proven insufficient to eradicate invasive cells and micrometastasis.¹⁰⁴ This might be due, at least in part, to invading cells associated with the anatomic structures and increased apoptotic resistance.^105–107 Crucially, the presence of these invasive glioblastoma stem cells¹⁰⁸ is sufficient to cause progressive neurologic dysfunction.

The central role of proteolysis in tumor progression has long been established and extensively described. Several classes of proteases (metalloproteases, and cysteine-, aspartic-, serine proteases) have been identified and shown to destroy the tissue structure surrounding the invasive and—by extension—metastatic process.¹⁰⁹ Even though glioblastoma lacks the ability of most malignant tumors to metastasize to distant sites, motility and invasion of glioblastoma stem cells are a defining feature of malignancy. Upon clinical presentation, glioblastoma stem cells have invariably spread and infiltrated large areas of the brain.¹¹⁰ We recently described a mechanism whereby glioblastoma stem cells re-model their microenvironment to be more permissive toward invasion¹¹¹: Stress-induced NF-κB activates matrix metalloproteinases (MMPs) via the conversion of plasminogen to plasmin by the proteolytic performance of urokinase-type plasminogen activator (uPA). This, in turn, allows for the effective processing of fibronectin which is then incorporated into a tumor-specific matrix. In essence, our work postulates that the role of proteolysis in glioblastoma stem cell invasion is not limited to the destruction of obstacles in the paths of invading cells and the truncation of tethering points which might otherwise retain tumor cells in the bulk structure but leads to the formation of novel fibronectin “roads.” These protect glioblastoma stem cells from detrimental influences, such as anoikis, and provide a route along where glioblastoma stem cells can transverse the brain.¹¹¹

Recent data clearly indicate that established cell lines used as experimental surrogates for glioblastoma poorly reflect the in vivo situation in terms of gene and protein expression.^112,113 Furthermore, these cell lines also lack some key features of glioblastoma when used in orthotopic animal models; no micrometastasis can be observed when using U87MG cells, one of the few cell lines that are actually tumorigenic when used in xenotransplants.²⁸ Culturing primary patient material as spheroids stably preserves tumor expression patterns allowing short-term expansion under adherent conditions.¹¹⁴ Importantly, these cells are also able to recapitulate the patients’ tumor in a murine environment.¹¹⁵ Letting spheroid cultures differentiate into adherent cells allows us to examine the different types of cancer cells present in a tumor. In addition to the various lineages of differentiated cells obtained, one can also assess the behavior of CSCs or tumor-initiating cells.^116,117 Glioblastoma stem cells are suggested to emerge from neural stem cells or partially from differentiated progenitor cells and are characterized by stem cell markers, such as Sox-2, Nestin, Musashi, and the disputed CD133. However, the latter certainly does not identify the most stem cell–like population in glioblastoma,^118,119 and recent work of Schneider et al.²⁸ shows that a high-level state of plasticity exists between the stem cell and differentiated phenotypes. Nevertheless, glioblastoma stem cells are highly resistant to treatment, partially due to the high expression of the multidrug resistance ABC transporter protein breakpoint cluster region pseudogene 1 (BCRP1), DNA repair proteins, and gene products that inhibit apoptosis. Furthermore, these cells are frequently not within the cell cycle, are not eradicated by conventional treatment, and most likely contribute to recurrence.¹²⁰

The importance of cathepsins in tumorigenesis and glioblastoma cells

Lysosomal cathepsins play an essential role in proteome homeostasis by catalyzing the degradation of proteins for proper recycling of amino acids. In addition, cathepsins are responsible for proteolytic maturation of antigens to antigenic peptides for an efficient antigen presentation to the major histocompatibility complex (MHC). Most cathepsins are intracellular proteases, active in an acidic environment, and their expression level varies depending on the cell type. However, a dysfunction of the proteolytic activity of cathepsins is associated with various pathophysiological processes.¹²¹

Cysteine cathepsins of the C1A subfamily are monomeric enzymes with a molecular weight between 20 and 35 kDa. Cathepsin C is the only papain-like protease which significantly exceeds the characteristic size of the subfamily due to its tetrameric composition (200 kDa).¹²² C1A cysteine cathepsins contain cathepsin B (CatB), -C, -F, -H, -K, -L, -O, -S, -V, -W, and -X. These enzymes are mainly active within the endosomes and lysosomes and are involved in protein turnover together with the aspartic proteases CatD and CatE.¹²³ CatG belongs to the class of serine proteases (S1) and is found in azurophilic granules of neutrophils^124,125 and primary antigen-presenting cells (APCs).^126,127 The G refers to its capability to degrade globulins of the α-, β-, and γ types.¹²⁸ The 28-kDa human CatG has both chymotrypsin- and trypsin-like activity and is active in a broad pH range (pH 5–8; optimum at neutral pH).¹²⁹ During activation, among other proteases, CatG is released by neutrophils at the site of inflammation to initiate and support an immune response.^129,130

According to their common role in proteolytic degradation, an uncontrolled cathepsin activity, particularly of cysteine cathepsins, can lead to a majority of clinical disorders, including tumors. Consequently, the regulation of high cathepsin activity to a steady-state level by specific protease inhibitors is an encouraging tool to interfere with tumorigenesis.^131,132 The acidic extracellular environment of tumor cells is ideal for the proteolytic activity of cathepsins since most of the cathepsins have their highest activity in a slightly acidic condition and are often found outside of tumor cells. Subsequently, they activate growth factors and proteases or degrade components of the extracellular matrix.¹³³ For example, CatB is involved in the function of MMPs which leads to the detachment of cells and thus to the initiation of the cell migration process.¹³⁴ In addition, CatB is responsible for survival and growth of tumor cells, activation of signaling pathways of angiogenesis;¹³⁵ converts pro-uPA to uPA;¹³⁶ and inactivates MMP inhibitors, such as tissue inhibitors of metalloproteinase (TIMP1 and TIMP2) which are important for migration of tumor cells.¹³⁷ The protein level of CatB occurs to be ten times higher in glioblastoma in contrast to low-grade glioma or normal brain tissue. Furthermore, high levels of CatB correlate with invasiveness of glioblastoma cells.^138,139 Conversely, levels of natural cysteine protease inhibitors are downregulated in glioblastoma.¹⁴⁰ Indeed, inhibition of CatB by the selective CatB inhibitor CA074 in glioma cells reduced invasiveness of these cells.¹⁴¹ We used three pairs of previously described, distinct patient-derived cell populations, each pair consisting of stem cells and genetically identically, but epigenetically diverse differentiated cells.¹⁴² These cells, termed SC35, SC38, and SC40, were obtained from two male patients and one female patient, between the ages of 44 and 75 years at the time of diagnosis, and are shown to grow invasive in our murine xenotransplant model.²⁸ The general observation that high CatB levels are characteristic for glioblastoma cells¹³⁸ was confirmed in glioblastoma stem cells from several patients (SC35, 38, and 40), differentiated glioblastoma cells (PC35, 38, and 40), and glioblastoma cell lines (A172, U87).¹⁴³

Similar to CatB, levels of CatL are higher in glioblastoma cells and responsible for glioblastoma cell invasion.¹⁴⁴ CatS is also known to be highly expressed in tumor cells; for instance, CatS is involved in microtubule formation and enhances the collagenolytic activity which accounts for tumor invasion.^145,146 In fact, CatS is upregulated in astrocytoma cells suggesting a crucial role of CatS in invasion of these cells.¹⁴⁷ Until now, only little information has been acquired regarding CatH. Lower amounts have been found in lung cancer while there are higher levels existing in sera of lung cancer patients¹⁴⁸ and in glioblastoma cells,¹⁴⁹ and similar findings were described for CatK which is highly expressed in glioblastoma cells.¹⁵⁰

The activity of cysteine proteases is regulated by many factors including endogenous inhibitors of the cystatin family. Lower cystatin expression levels like for cystatin C (CysC) are associated with tumor progression. It has been reported that cystatins decrease with malignant invasiveness of glioblastoma cells.¹⁵¹ To this end, it is plausible to target cathepsin activity in glioblastoma stem cells to avoid recurrence of glioblastoma. Modulating of their proteolytic activity by specific protease inhibitors is a logical consequence and might interfere with viability, proliferation, and migration. On time, a huge variety of selective protease inhibitors are available and summarized in¹²¹ and Sienczyk and Oleksyszyn.¹⁵² Notably, it is important that such protease inhibitors are applicable to restore the balance of cathepsin activity in glioblastoma cells since the disruption of the functional properties of cathepsins supports tumorigenesis.

In order to gain a better insight why glioblastoma stem cells are not recognized and eliminated by immune cells, which is also called immune evasion, the immune system will be specified in more details in the following section.

MHC I–mediated immune evasion mechanisms of glioblastoma stem cells

Natural killer (NK)¹⁵³ cells are bone marrow–derived large granular lymphocytes and involved in the defense of viral infections as well as the destruction of tumor cells.^154–156 Due to the lack of clonal antigen receptors, NK cells are traditionally counted to the innate immune system. However, recent evidence suggests that they have a long-term memory to certain antigens and can adapt to the environment by releasing cytokines.¹⁵⁷ NK cells are characterized by their ability to initiate apoptosis in target cells without the necessity of MHC-mediated antigen presentation. An absence of MHC I on the target cell surface is recognized by NK cells resulting in an NK-driven cytolysis (“missing-self” recognition).¹⁵⁶ In turn, NK cells are not activated by the inspected cell when cell surface MHC I- or non-classical MHC molecules are present on the cell surface.¹⁵⁶ More precisely, NK cells express the killer inhibitory receptor (KIR) on the cell surface which can bind to MHC I molecules and inhibit NK cell activation, whereas the natural killer cell receptor protein 1 (NKR-P1) probably recognizes a modified glycocalyx and retains NK cells activated. Activated NK cells secrete perforins from granules which penetrate the plasma membrane of the target cell and subsequently polymerize a cylindrical trans-membrane pore leading to osmotic lysis. Thereby, the target cell will be eliminated.^158,159

A T cell–mediated immune response is qualified by recognition of pathogen-derived antigens presented by APCs and non-immune cells. While these antigens (peptides) are recognized by CD8⁺ cytotoxic T lymphocytes (CTLs) when bound to MHC I, CD4⁺ T helper cells inspect MHC II when presented by professional APCs. Activated CTLs can eliminate target cells by different mechanisms, on one hand, by releasing perforins and granzymes, which leads to the activation of the caspase cascade resulting in apoptosis, and, on the other hand, by secreting cytokines mainly interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α). Both cytokines play an important role in inhibition of viral replication, promotion of MHC-mediated antigen presentation, and activation of the immune system.^160,161

The T cell receptor (TCR) of CD8⁺ and CD4⁺ T cells recognizes antigens bound to the binding groove of MHC molecules. Endogenous antigens such as viral proteins or defective ribosomal translation products are proteolytically degraded by the proteasome and transported to the lumen of the ER by the transporter associated with antigen presentation (TAP).¹⁶² In the ER lumen, antigens are further processed to a length of 8–9 amino acids by amino peptidases. The ER lumen is also the place of MHC I folding and assembly. After binding of peptides to nascent MHC I molecules, the MHC I peptide complex is transported to the cell surface via the Golgi network. MHC I molecules are heterodimers composed of a polymorphic α-chain, which is associated with a β2-microglobulin chain. They are encoded by the three classical loci human leukocyte antigen-A (HLA-A), HLA-B, and HLA-C.^160,163 MHC class II glycoproteins are responsible for the cell surface presentation of exogenous antigens which are previously processed in the endocytic compartment by the concerted action of cathepsins. MHC II molecules (HLA-DQ, HLA-DP, and HLA-DR) are mainly expressed in professional APCs, such as B cells, dendritic cells (DCs), and macrophages.¹⁶⁰

APCs are also able to present extracellular antigens to CD8⁺ T cells by MHC I molecules, and intracellular antigens can be presented to CD4⁺ T cells by MHC II molecules. This phenomenon is named cross-presentation and is important to recruit CD8⁺ T cells in response to endocytosed viral- and tumor-derived antigens. The latter plays a crucial role in tumor immune surveillance.¹⁶⁴ Since it is not completely clear how cross-presentation is accomplished, several mechanisms are currently discussed. Cell surface MHC I molecules can be endocytosed (recycling pathway), MHC I molecules in the ER traffic to the trans-Golgi network (TGN) and then to the cell surface or to the endocytic compartment (vacuolar pathway), or reach the phagosome where MHC I are loaded with a new round of exogenous or endocytic antigens and reach the cell surface for inspection by CTLs or are degraded by a yet to be identified protease.^165–168 We found that in these pathways, MHC I molecules can be degraded by CatG (Figure 1).¹⁴³ The autophagosome arranges the delivery of cytosolic proteins to the MHC II antigen-presenting pathway by merging with the MHC II loading compartment.¹⁶⁹

Figure 1.

MHC I is degraded by CatG. (a) MHC I molecules are mainly delivered to the cell surface or partly to the endocytic compartment via the trans-Golgi network (TGN) for degradation. Cell surface MHC I molecules can be endocytosed and either loaded with a new set of antigenic peptides in order to reach the cell surface (recycling pathway) or degraded by the proteolytic activity of CatG. (b) If immune cells are treated with a CatG inhibitor, the surface expression of MHC I is increased. This is probably due by the inhibition of the enzymatic activity of CatG in the endocytic compartment.

The pathological cell surface reduction of MHC I, which causes an impaired immune response, is an immune evasion mechanism by several viruses, for instance, immune deficiency virus (HIV-1) and tumor cells.¹⁵⁶ In HIV-1-infected cells, HIV-1-derived Nef protein prevents the cell surface expression of MHC molecules which accumulate in the intracellular space.^170,171 Similarly, invading glioblastoma stem cells reduce newly synthesized MHC I cell surface expression which diminishes tumor-associated antigen presentation and elimination by CTLs.^172,173 However, glioblastoma stem cells still maintain a limited set of MHC I presumably to escape from recognition by NK cells.¹⁵⁶ Recently, we demonstrated that MHC I molecules are catalytically degraded by CatG in vitro and in peripheral blood mononuclear cells (PBMCs) in contrast to glioblastoma stem cells. In order to determine whether the absence of CatG in glioblastoma stem cells might be the reason of complete downregulation of MHC I, levels of MHC I were analyzed in CatG-overexpressing glioblastoma stem cells (Figure 2). Indeed, MHC I was effectively downregulated when CatG was introduced in glioblastoma stem cells. In addition, the amount of intracellular MHC I was also reduced confirming that MHC I is proteolytically degraded by CatG and is responsible for reduced cell surface expression of MHC I in CatG-expressing cells.¹⁴³ Even though CatG is an essential protease for post-transcriptional regulation of MHC I molecules, glioblastoma stem cells do not have the ability to proteolytically degrade MHC I by CatG which prevents the complete reduction of cell surface MHC I to avoid recognition by NK cells. Restoration of CatG in glioblastoma might be a supportive therapeutic opportunity by using oncolytic viruses¹⁷⁴ loaded with a CatG encoding gene which possibly sensitizes glioblastoma stem cells to be targeted by NK cells.

Figure 2.

Immune evasion mechanism of glioblastoma stem cells. (a) Invading glioblastoma stem cells reduce surface expression of MHC I to diminish tumor-associated antigen presentation and evade recognition by CTLs. A limited set of cell surface MHC I molecules is maintained to escape recognition by NK cells. (b) Overexpression of CatG in glioblastoma stem cells effectively downregulates MHC I molecules. However, it remains unclear whether NK cells are activated by CatG-mediated reduction of cell surface MHC I.

Conclusion and further perspectives

Autophagy is a process that ensures nutrient supply and homeostasis in the presence of stressors and supports cell survival. The main problem in targeting autophagy is the similarity between glioblastoma stem cells and normal stem cells. In healthy cells, autophagy acts tumor-suppressive by maintaining homeostasis. In cancer cells; however, autophagy can act in either a tumor-promoting (oncogenic) or tumor-suppressing manner, and the imbalance between these dictates the outcome of cell survival.³⁵

Lysosomal cathepsins control the proteome steady state by catalyzing protein degradation; however, a dysfunctional or uncontrolled cathepsin activity can cause harm. Distinct cathepsins are upregulated in tumor cells; of these, increased CatB activity correlates with glioblastoma stem cell mobility, and the obvious consequence to inhibit CatB activity impaired the invasiveness of glioblastoma stem cells. Furthermore, levels of MHC I are regulated by the proteolytic performance of CatG. On the other hand, glioblastoma stem cells do not harbor CatG allowing a limited set of MHC I molecules on the cell surface to escape NK cell recognition. Cathepsin inhibitors or activators, for instance, lactoferrin, which is a natural CatG activator as published by us just recently,¹⁷⁵ or bioactive tripeptides, which increase CatG activity,¹⁷⁶ might reduce high cathepsin activity found in glioblastoma stem cells or reinstall such cathepsins necessary for supportive immune recognition.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

T.B. was supported by the Alexander von Humboldt Polish Honorary Research Scholarship (DPK-422-1658/2013).

References

Nature milestones cancer, http://www.Nature.Com/milestones/milecancer/timeline.Html

Furth

Kahn

MC.

The transmission of leukaemia of mice with a single cell. Am J Cancer 1937; 31: 276–282.

Bruce

Van Der Gaag

A quantitative assay for the number of murine lymphoma cells capable of proliferation in vivo. Nature 1963; 199: 79–80.

Becker

Till

JE.

Cytological demonstration of the clonal nature of spleen colonies derived from transplanted mouse marrow cells. Nature 1963; 197: 452–454.

Buick

Till

McCulloch

EA.

Colony assay for proliferative blast cells circulating in myeloblastic leukaemia. Lancet 1977; 1: 862–863.

Lapidot

Sirard

Vormoor

. A cell initiating human acute myeloid leukaemia after transplantation into scid mice. Nature 1994; 367: 645–648.

Bonnet

Dick

JE.

Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 1997; 3: 730–737.

Al-Hajj

Wicha

Benito-Hernandez

. Prospective identification of tumorigenic breast cancer cells. P Natl Acad Sci USA 2003; 100: 3983–3988.

Singh

Hawkins

Clarke

. Identification of human brain tumour initiating cells. Nature 2004; 432: 396–401.

10.

Collins

Berry

Hyde

. Prospective identification of tumorigenic prostate cancer stem cells. Cancer Res 2005; 65: 10946–10951.

11.

Prince

Sivanandan

Kaczorowski

. Identification of a subpopulation of cells with cancer stem cell properties in head and neck squamous cell carcinoma. P Natl Acad Sci USA 2007; 104: 973–978.

12.

Hermann

Huber

Herrler

. Distinct populations of cancer stem cells determine tumor growth and metastatic activity in human pancreatic cancer. Cell Stem Cell 2007; 1: 313–323.

13.

O’Brien

Pollett

Gallinger

. A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature 2007; 445: 106–110.

14.

Wei

Utomo

. Side population cells isolated from mesenchymal neoplasms have tumor initiating potential. Cancer Res 2007; 67: 8216–8222.

15.

Eramo

Lotti

Sette

. Identification and expansion of the tumorigenic lung cancer stem cell population. Cell Death Differ 2008; 15: 504–514.

16.

Clarke

Dick

Dirks

. Cancer stem cells—perspectives on current status and future directions: AACR workshop on cancer stem cells. Cancer Res 2006; 66: 9339–9344.

17.

Kreso

Dick

JE.

Evolution of the cancer stem cell model. Cell Stem Cell 2014; 14: 275–291.

18.

Blanpain

Mohrin

Sotiropoulou

. DNA-damage response in tissue-specific and cancer stem cells. Cell Stem Cell 2011; 8: 16–29.

19.

Odoux

Fohrer

Hoppo

. A stochastic model for cancer stem cell origin in metastatic colon cancer. Cancer Res 2008; 68: 6932–6941.

20.

Hirschmann-Jax

Foster

Wulf

. A distinct “Side population” of cells with high drug efflux capacity in human tumor cells. P Natl Acad Sci USA 2004; 101: 14228–14233.

21.

Allan

AL.

The role of human aldehyde dehydrogenase in normal and cancer stem cells. Stem Cell Rev 2011; 7: 292–306.

22.

Magee

Piskounova

Morrison

SJ.

Cancer stem cells: impact, heterogeneity, and uncertainty. Cancer cell 2012; 21: 283–296.

23.

Rycaj

Tang

DG.

Cell-of-origin of cancer versus cancer stem cells: assays and interpretations. Cancer Res 2015; 75: 4003–4011.

24.

Reya

Morrison

Clarke

. Stem cells, cancer, and cancer stem cells. Nature 2001; 414: 105–111.

25.

Cairns

Mutation selection and the natural history of cancer. Nature 1975; 255: 197–200.

26.

Visvader

JE.

Cells of origin in cancer. Nature 2011; 469: 314–322.

27.

Powers

Zhu

. Substantial contribution of extrinsic risk factors to cancer development. Nature 2016; 529: 43–47.

28.

Schneider

Strobele

Nonnenmacher

. A paired comparison between glioblastoma “stem cells” and differentiated cells. Int J Cancer 2016; 138(7): 1709–1718.

29.

Dahan

Martinez Gala

Delmas

. Ionizing radiations sustain glioblastoma cell dedifferentiation to a stem-like phenotype through survivin: possible involvement in radioresistance. Cell Death Dis 2014; 5: e1543.

30.

Leder

Pitter

Laplant

. Mathematical modeling of PDGF-driven glioblastoma reveals optimized radiation dosing schedules. Cell 2014; 156: 603–616.

31.

Olmez

Shen

McDonald

. Dedifferentiation of patient-derived glioblastoma multiforme cell lines results in a cancer stem cell-like state with mitogen-independent growth. J Cell Mol Med 2015; 19: 1262–1272.

32.

Beck

Blanpain

Unravelling cancer stem cell potential. Nat Rev 2013; 13: 727–738.

33.

Gupta

Fillmore

Jiang

. Stochastic state transitions give rise to phenotypic equilibrium in populations of cancer cells. Cell 2011; 146: 633–644.

34.

Klionsky

Emr

SD.

Autophagy as a regulated pathway of cellular degradation. Science 2000; 290: 1717–1721.

35.

Liu

Wen

Cheng

Survival or death: disequilibrating the oncogenic and tumor suppressive autophagy in cancer. Cell Death Dis 2013; 4: e892.

36.

Lorin

Hamai

Mehrpour

. Autophagy regulation and its role in cancer. Semin Cancer Biol 2013; 23: 361–379.

37.

Liang

Jackson

Seaman

. Induction of autophagy and inhibition of tumorigenesis by beclin 1. Nature 1999; 402: 672–676.

38.

Laddha

Ganesan

Chan

. Mutational landscape of the essential autophagy gene BECN1 in human cancers. Mol Cancer Res 2014; 12: 485–490.

39.

Bao

. Hypoxia-inducible factors regulate tumorigenic capacity of glioma stem cells. Cancer Cell 2009; 15: 501–513.

40.

Clarke

van der Kooy

Low oxygen enhances primitive and definitive neural stem cell colony formation by inhibiting distinct cell death pathways. Stem Cells 2009; 27: 1879–1886.

41.

Conley

Gheordunescu

Kakarala

. Antiangiogenic agents increase breast cancer stem cells via the generation of tumor hypoxia. P Natl Acad Sci USA 2012; 109: 2784–2789.

42.

Warburg

Posener

Negelein

Ueber den stoffwechsel der tumoren; biochemische zeitschrift. Biochem Z 1924; 152: 319–344.

43.

Kim

Lee

. Involvement of mitophagy in oncogenic k-ras-induced transformation: overcoming a cellular energy deficit from glucose deficiency. Autophagy 2011; 7: 1187–1198.

44.

Fung

Lock

Gao

. Induction of autophagy during extracellular matrix detachment promotes cell survival. Mol Biol Cell 2008; 19: 797–806.

45.

Strandberg

Lenardo

MJ.

The selectivity of autophagy and its role in cell death and survival. Autophagy 2008; 4: 567–573.

46.

Cuervo

Dice

JF.

A receptor for the selective uptake and degradation of proteins by lysosomes. Science 1996; 273: 501–503.

47.

Dunn

Jr.

Studies on the mechanisms of autophagy: formation of the autophagic vacuole. J Cell Biol 1990; 110: 1923–1933.

48.

Crotzer

Blum

JS.

Autophagy and its role in MHC-mediated antigen presentation. J Immunol 2009; 182: 3335–3341.

49.

Baba

Takeshige

Baba

. Ultrastructural analysis of the autophagic process in yeast: detection of autophagosomes and their characterization. J Cell Biol 1994; 124: 903–913.

50.

Munz

Enhancing immunity through autophagy. Annu Rev Immunol 2009; 27: 423–449.

51.

Pankiv

Clausen

Lamark

. p62/SQSTM1 binds directly to Atg8/LC3 to facilitate degradation of ubiquitinated protein aggregates by autophagy. J Biol Chem 2007; 282: 24131–24145.

52.

Mizushima

Yoshimori

How to interpret LC3 immunoblotting. Autophagy 2007; 3: 542–545.

53.

Racoma

Meisen

Wang

. Thymoquinone inhibits autophagy and induces cathepsin-mediated, caspase-independent cell death in glioblastoma cells. PLoS ONE 2013; 8: e72882.

54.

Bjorkoy

Lamark

Brech

. p62/SQSTM1 forms protein aggregates degraded by autophagy and has a protective effect on huntingtin-induced cell death. J Cell Biol 2005; 171: 603–614.

55.

Mathew

Karp

Beaudoin

. Autophagy suppresses tumorigenesis through elimination of p62. Cell 2009; 137: 1062–1075.

56.

Choi

Ryter

Levine

Autophagy in human health and disease. New Engl J Med 2013; 368: 1845–1846.

57.

Bonora

Wieckowsk

Chinopoulos

. Molecular mechanisms of cell death: central implication of ATP synthase in mitochondrial permeability transition. Oncogene 2015; 34: 1608.

58.

Westhoff

Karpel-Massler

Bruhl

. A critical evaluation of PI3K inhibition in glioblastoma and neuroblastoma therapy. Mol Cell Ther 2014; 2: 32.

59.

Inoki

Zhu

Guan

KL.

TSC2 mediates cellular energy response to control cell growth and survival. Cell 2003; 115: 577–590.

60.

Takeshige

Baba

Tsuboi

. Autophagy in yeast demonstrated with proteinase-deficient mutants and conditions for its induction. J Cell Biol 1992; 119: 301–311.

61.

Lum

Bauer

Kong

. Growth factor regulation of autophagy and cell survival in the absence of apoptosis. Cell 2005; 120: 237–248.

62.

Stephan

Yeh

Ramachandran

. The Tor and cAMP-dependent protein kinase signaling pathways coordinately control autophagy in Saccharomyces cerevisiae. Autophagy 2010; 6: 294–295.

63.

Jung

Jun

. ULK-Atg13-FIP200 complexes mediate mTOR signaling to the autophagy machinery. Mol Biol Cell 2009; 20: 1992–2003.

64.

Hosokawa

Hara

Kaizuka

. Nutrient-dependent mTORC1 association with the ULK1-Atg13-FIP200 complex required for autophagy. Mol Biol Cell 2009; 20: 1981–1991.

65.

Fleming

Ricketts

. Autophagy regulates notch degradation and modulates stem cell development and neurogenesis. Nat Commun 2016; 7: 10533.

66.

Natsumeda

Maitani

Liu

. Targeting notch signaling and autophagy increases cytotoxicity in glioblastoma neurospheres. Brain Pathol 2016; 26: 713–723.

67.

Glaumann

Ahlberg

Comparison of different autophagic vacuoles with regard to ultrastructure, enzymatic composition, and degradation capacity—formation of crinosomes. Exp Mol Pathol 1987; 47: 346–362.

68.

Zhao

Cai

Santi

. Chloroquine-mediated radiosensitization is due to the destabilization of the lysosomal membrane and subsequent induction of cell death by necrosis. Radiat Res 2005; 164: 250–257.

69.

Yamamoto

Tagawa

Yoshimori

. Bafilomycin A1 prevents maturation of autophagic vacuoles by inhibiting fusion between autophagosomes and lysosomes in rat hepatoma cell line, H-4-II-E cells. Cell Struct Funct 1998; 23: 33–42.

70.

Steinman

Mellman

Muller

. Endocytosis and the recycling of plasma membrane. J Cell Biol 1983; 96: 1–27.

71.

Seglen

Grinde

Solheim

AE.

Inhibition of the lysosomal pathway of protein degradation in isolated rat hepatocytes by ammonia, methylamine, chloroquine and leupeptin. Eur J Biochem 1979; 95: 215–225.

72.

Wibo

Poole

Protein degradation in cultured cells. II. The uptake of chloroquine by rat fibroblasts and the inhibition of cellular protein degradation and cathepsin B1. J Cell Biol 1974; 63: 430–440.

73.

Chou

Brown

Minond

. Reversible inhibitor of p97, DBeQ, impairs both ubiquitin-dependent and autophagic protein clearance pathways. P Natl Acad Sci USA 2011; 108: 4834–4839.

74.

Wilcox

Mason

RW.

Inhibition of cysteine proteinases in lysosomes and whole cells. Biochem J 1992; 285(Pt 2): 495–502.

75.

Hoyvik

Gordon

Berg

. Inhibition of autophagic-lysosomal delivery and autophagic lactolysis by asparagine. J Cell Biol 1991; 113: 1305–1312.

76.

Blommaart

Krause

Schellens

. The phosphatidylinositol 3-kinase inhibitors wortmannin and LY294002 inhibit autophagy in isolated rat hepatocytes. Eur J Biochem 1997; 243: 240–246.

77.

Seglen

Gordon

PB.

3-methyladenine: specific inhibitor of autophagic/lysosomal protein degradation in isolated rat hepatocytes. P Natl Acad Sci USA 1982; 79: 1889–1892.

78.

Choi

Park

. Inhibitory effect of mTOR activator MHY1485 on autophagy: suppression of lysosomal fusion. PLoS ONE 2012; 7: e43418.

79.

Grinde

Effect of carboxylic ionophores on lysosomal protein degradation in rat hepatocytes. Exp Cell Res 1983; 149: 27–35.

80.

Underwood

Imarisio

Fleming

. Antioxidants can inhibit basal autophagy and enhance neurodegeneration in models of polyglutamine disease. Hum Mol Genet 2010; 19: 3413–3429.

81.

Webb

Ravikumar

Rubinsztein

DC.

Microtubule disruption inhibits autophagosome-lysosome fusion: implications for studying the roles of aggresomes in polyglutamine diseases. Int J Biochem Cell B 2004; 36: 2541–2550.

82.

Reunanen

Marttinen

Hirsimaki

Effects of griseofulvin and nocodazole on the accumulation of autophagic vacuoles in Ehrlich ascites tumor cells. Exp Mol Pathol 1988; 48: 97–102.

83.

Deng

Lei

Liu

. Pyrvinium targets autophagy addiction to promote cancer cell death. Cell Death Dis 2013; 4: e614.

84.

Egan

Chun

Vamos

. Small molecule inhibition of the autophagy kinase ULK1 and identification of ULK1 substrates. Mol Cell 2015; 59: 285–297.

85.

Liu

Xia

Kim

. Beclin1 controls the levels of p53 by regulating the deubiquitination activity of USP10 and USP13. Cell 2011; 147: 223–234.

86.

Ganley

Wong

Gammoh

. Distinct autophagosomal-lysosomal fusion mechanism revealed by thapsigargin-induced autophagy arrest. Mol Cell 2011; 42: 731–743.

87.

Hoyvik

Gordon

Seglen

PO.

Use of a hydrolysable probe, [14C]lactose, to distinguish between pre-lysosomal and lysosomal steps in the autophagic pathway. Exp Cell Res 1986; 166: 1–14.

88.

Punnonen

Reunanen

Effects of vinblastine, leucine, and histidine, and 3-methyladenine on autophagy in Ehrlich ascites cells. Exp Mol Pathol 1990; 52: 87–97.

89.

Munafo

Colombo

MI.

A novel assay to study autophagy: regulation of autophagosome vacuole size by amino acid deprivation. J Cell Sci 2001; 114: 3619–3629.

90.

Carew

Kelly

Nawrocki

ST.

Autophagy as a target for cancer therapy: new developments. Cancer Manag Res 2012; 4: 357–365.

91.

Crombet

Osorio

Cruz

. Use of the humanized anti-epidermal growth factor receptor monoclonal antibody h-R3 in combination with radiotherapy in the treatment of locally advanced head and neck cancer patients. J Clin Oncol 2004; 22: 1646–1654.

92.

Zhuang

Long

Zheng

. Curcumin promotes differentiation of glioma-initiating cells by inducing autophagy. Cancer Sci 2012; 103: 684–690.

93.

Aoki

Takada

Kondo

. Evidence that curcumin suppresses the growth of malignant gliomas in vitro and in vivo through induction of autophagy: role of Akt and extracellular signal-regulated kinase signaling pathways. Mol Pharmacol 2007; 72: 29–39.

94.

Scarlatti

Maffei

Beau

. Role of non-canonical beclin 1-independent autophagy in cell death induced by resveratrol in human breast cancer cells. Cell Death Differ 2008; 15: 1318–1329.

95.

Wang

Liu

. Quercetin induces protective autophagy in gastric cancer cells: involvement of akt-mtor- and hypoxia-induced factor 1alpha-mediated signaling. Autophagy 2011; 7: 966–978.

96.

Kim

Moon

Ahn

. Quercetin induces mitochondrial mediated apoptosis and protective autophagy in human glioblastoma U373MG cells. Oxid Med Cell Longev 2013; 2013: 596496.

97.

Behin

Hoang-Xuan

Carpentier

. Primary brain tumours in adults. Lancet 2003; 361: 323–331.

98.

Furnari

Fenton

Bachoo

. Malignant astrocytic glioma: genetics, biology, and paths to treatment. Gene Dev 2007; 21: 2683–2710.

99.

Stupp

van den Bent

Hegi

ME.

Optimal role of temozolomide in the treatment of malignant gliomas. Curr Neurol Neurosci 2005; 5: 198–206.

100.

de Groot

Fuller

Kumar

. Tumor invasion after treatment of glioblastoma with bevacizumab: radiographic and pathologic correlation in humans and mice. Neuro-oncology 2010; 12: 233–242.

101.

Delgado-Lopez

Corrales-Garcia

EM.

Survival in glioblastoma: a review on the impact of treatment modalities. Clin Transl Oncol 2016; 18: 1062–1071.

102.

Kesari

Schiff

Henson

. Phase ii study of temozolomide, thalidomide, and celecoxib for newly diagnosed glioblastoma in adults. Neuro-oncology 2008; 10: 300–308.

103.

Jue

McDonald

KL.

The challenges associated with molecular targeted therapies for glioblastoma. J Neuro-oncol 2016; 127: 427–434.

104.

Giese

Bjerkvig

Berens

. Cost of migration: invasion of malignant gliomas and implications for treatment. J Clin Oncol 2003; 21: 1624–1636.

105.

Scherer

HJ.

A critical review: the pathology of cerebral gliomas. J Neurol Psychiatry 1940; 3: 147–177.

106.

Pullan

Wilson

Metcalfe

. Requirement of basement membrane for the suppression of programmed cell death in mammary epithelium. J Cell Sci 1996; 109(Pt 3): 631–642.

107.

Box

Rogers

Mendiola

. Tumour-microenvironmental interactions: paths to progression and targets for treatment. Semin Cancer Biol 2010; 20: 128–138.

108.

Singh

Clarke

Terasaki

. Identification of a cancer stem cell in human brain tumors. Cancer Res 2003; 63: 5821–5828.

109.

Duffy

MJ.

The role of proteolytic enzymes in cancer invasion and metastasis. Clin Exp Metastas 1992; 10: 145–155.

110.

Hambardzumyan

Bergers

Glioblastoma: defining tumor niches. Trends Cancer 2015; 1: 252–265.

111.

Westhoff

Zhou

Nonnenmacher

. Inhibition of NF-kappaB signaling ablates the invasive phenotype of glioblastoma. Mol Cancer Res 2013; 11: 1611–1623.

112.

Walling

Kotliarov

. Genomic changes and gene expression profiles reveal that established glioma cell lines are poorly representative of primary human gliomas. Mol Cancer Res 2008; 6: 21–30.

113.

Ortensi

Setti

Osti

. Cancer stem cell contribution to glioblastoma invasiveness. Stem Cell Res 2013; 4: 18.

114.

De Witt Hamer

Van Tilborg

Eijk

. The genomic profile of human malignant glioma is altered early in primary cell culture and preserved in spheroids. Oncogene 2008; 27: 2091–2096.

115.

Pollard

Yoshikawa

Clarke

. Glioma stem cell lines expanded in adherent culture have tumor-specific phenotypes and are suitable for chemical and genetic screens. Cell Stem Cell 2009; 4: 568–580.

116.

Fan

Salford

Widegren

Glioma stem cells: evidence and limitation. Semin Cancer Biol 2007; 17: 214–218.

117.

Lathia

Gallagher

Myers

. Direct in vivo evidence for tumor propagation by glioblastoma cancer stem cells. PLoS ONE 2011; 6: e24807.

118.

Beier

CD133 negative cancer stem cells in glioblastoma. Front Biosci 2011; 3: 701–710.

119.

Kim

Lee

Kim

. The presence of stem cell marker-expressing cells is not prognostically significant in glioblastomas. Neuropathology 2011; 31: 494–502.

120.

Liu

Yuan

Zeng

. Analysis of gene expression and chemoresistance of cd133+ cancer stem cells in glioblastoma. Mol Cancer 2006; 5: 67.

121.

Burster

Processing and regulation mechanisms within antigen presenting cells: a possibility for therapeutic modulation. Curr Pharm Design 2013; 19: 13.

122.

Dolenc

Turk

Pungercic

. Oligomeric structure and substrate induced inhibition of human cathepsin C. J Biol Chem 1995; 270: 21626–21631.

123.

Colbert

Matthews

Miller

. Diverse regulatory roles for lysosomal proteases in the immune response. Eur J Immunol 2009; 39: 2955–2965.

124.

Starkey

Barrett

AJ.

Human cathepsin G. Catalytic and immunological properties. Biochem J 1976; 155: 273–278.

125.

Mounter

Atiyeh

Proteases of human leukocytes. Blood 1960; 15: 52–59.

126.

Burster

Beck

Tolosa

. Cathepsin G, and not the asparagine-specific endoprotease, controls the processing of myelin basic protein in lysosomes from human B lymphocytes. J Immunol 2004; 172: 5495–5503.

127.

Burster

Beck

Tolosa

. Differential processing of autoantigens in lysosomes from human monocyte-derived and peripheral blood dendritic cells. J Immunol 2005; 175: 5940–5949.

128.

Lebez

Kopitar

Leucocyte proteinases. I. Low molecular weight cathepsins of F and G type. Enzymologia 1970; 39: 271–283.

129.

Korkmaz

Horwitz

Jenne

. Neutrophil elastase, proteinase 3, and cathepsin G as therapeutic targets in human diseases. Pharmacol Rev 2010; 62: 726–759.

130.

Baggiolini

Schnyder

Bretz

. Cellular mechanisms of proteinase release from inflammatory cells and the degradation of extracellular proteins. CIBA F Symp 1979; 75: 105–121.

131.

Affara

Andreu

Coussens

LM.

Delineating protease functions during cancer development. Methods Mol Biol 2009; 539: 1–32.

132.

Vandooren

Opdenakker

Loadman

. Proteases in cancer drug delivery. Adv Drug Deliver Rev 2016; 97: 144–155.

133.

Gocheva

Joyce

JA.

Cysteine cathepsins and the cutting edge of cancer invasion. Cell Cycle 2007; 6: 60–64.

134.

Eeckhout

Vaes

Further studies on the activation of procollagenase, the latent precursor of bone collagenase. Effects of lysosomal cathepsin B, plasmin and kallikrein, and spontaneous activation. Biochem J 1977; 166: 21–31.

135.

Lakka

Gondi

Rao

JS.

Proteases and glioma angiogenesis. Brain Pathol 2005; 15: 327–341.

136.

Kobayashi

Ohi

Sugimura

. Inhibition of in vitro ovarian cancer cell invasion by modulation of urokinase-type plasminogen activator and cathepsin B. Cancer Res 1992; 52: 3610–3614.

137.

Kostoulas

Lang

Nagase

. Stimulation of angiogenesis through cathepsin B inactivation of the tissue inhibitors of matrix metalloproteinases. FEBS Lett 1999; 455: 286–290.

138.

Rempel

Rosenblum

Mikkelsen

. Cathepsin B expression and localization in glioma progression and invasion. Cancer Res 1994; 54: 6027–6031.

139.

Sivaparvathi

Sawaya

Wang

. Overexpression and localization of cathepsin B during the progression of human gliomas. Clin Exp Metastas 1995; 13: 49–56.

140.

Sivaparvathi

McCutcheon

Sawaya

. Expression of cysteine protease inhibitors in human gliomas and meningiomas. Clin Exp Metastas 1996; 14: 344–350.

141.

Demchik

Sameni

Nelson

. Cathepsin B and glioma invasion. Int J Dev Neurosci 1999; 17: 483–494.

142.

Nonnenmacher

Westhoff

Fulda

. Rist: a potent new combination therapy for glioblastoma. Int J Cancer 2015; 136: E173–E187.

143.

Palesch

Wagner

Meid

. Cathepsin G-mediated proteolytic degradation of MHC class I molecules to facilitate immune detection of human glioblastoma cells. Cancer Immunol Immun 2016; 65: 283–291.

144.

Sivaparvathi

Yamamoto

Nicolson

. Expression and immunohistochemical localization of cathepsin l during the progression of human gliomas. Clin Exp Metastas 1996; 14: 27–34.

145.

Flannery

McQuaid

McGoohan

. Cathepsin S expression: an independent prognostic factor in glioblastoma tumours—a pilot study. Int J Cancer 2006; 119: 854–860.

146.

Burden

Gormley

Jaquin

. Antibody-mediated inhibition of cathepsin S blocks colorectal tumor invasion and angiogenesis. Clin Cancer Res 2009; 15: 6042–6051.

147.

Flannery

Gibson

Mirakhur

. The clinical significance of cathepsin S expression in human astrocytomas. Am J Pathol 2003; 163: 175–182.

148.

Schweiger

Staib

Werle

. Cysteine proteinase cathepsin H in tumours and sera of lung cancer patients: relation to prognosis and cigarette smoking. Brit J Cancer 2000; 82: 782–788.

149.

Sivaparvathi

Sawaya

Gokaslan

. Expression and the role of cathepsin H in human glioma progression and invasion. Cancer Lett 1996; 104: 121–126.

150.

Verbovsek

Motaln

Rotter

. Expression analysis of all protease genes reveals cathepsin K to be overexpressed in glioblastoma. PLoS ONE 2014; 9: e111819.

151.

Konduri

Yanamandra

Siddique

. Modulation of cystatin c expression impairs the invasive and tumorigenic potential of human glioblastoma cells. Oncogene 2002; 21: 8705–8712.

152.

Sienczyk

Oleksyszyn

Irreversible inhibition of serine proteases—design and in vivo activity of diaryl alpha-aminophosphonate derivatives. Curr Med Chem 2009; 16: 1673–1687.

153.

Miller

Brunner

Sprent

. Thymus-derived cells as killer cells in cell-mediated immunity. Transplant Proc 1971; 3: 915–917.

154.

Kiessling

Klein

Wigzell

. “Natural” killer cells in the mouse. I. Cytotoxic cells with specificity for mouse moloney leukemia cells. Specificity and distribution according to genotype. Eur J Immunol 1975; 5: 112–117.

155.

Herberman

Nunn

Lavrin

. Natural cytotoxic reactivity of mouse lymphoid cells against syngeneic acid allogeneic tumors. I. Distribution of reactivity and specificity. Int J Cancer 1975; 16: 216–229.

156.

Karre

Natural killer cell recognition of missing self. Nat Immunol 2008; 9: 477–480.

157.

Paust

von Andrian

UH.

Natural killer cell memory. Nat Immunol 2011; 12: 500–508.

158.

Young

Damiano

DiNome

. Dissociation of membrane binding and lytic activities of the lymphocyte pore-forming protein (perforin). J Exp Med 1987; 165: 1371–1382.

159.

Pahl

Cerwenka

Tricking the balance: NK cells in anti-cancer immunity. Immunobiology 2017; 222: 11–20.

160.

Blum

Wearsch

Cresswell

Pathways of antigen processing. Annu Rev Immunol 2013; 31: 443–473.

161.

Edwards

Davis

Browne

. Anti-viral strategies of cytotoxic T lymphocytes are manifested through a variety of granule-bound pathways of apoptosis induction. Immunol Cell Biol 1999; 77: 76–89.

162.

Neefjes

Jongsma

Paul

. Towards a systems understanding of MHC class I and MHC class II antigen presentation. Nat Rev Immunol 2011; 11: 823–836.

163.

Rammensee

HG.

Peptides made to order. Immunity 2006; 25: 693–695.

164.

Adiko

Babdor

Gutierrez-Martinez

. Intracellular transport routes for MHC I and their relevance for antigen cross-presentation. Front Immunol 2015; 6: 335.

165.

Gromme

Uytdehaag

Janssen

. Recycling MHC class I molecules and endosomal peptide loading. P Natl Acad Sci USA 1999; 96: 10326–10331.

166.

Reid

Watts

Cycling of cell-surface MHC glycoproteins through primaquine-sensitive intracellular compartments. Nature 1990; 346: 655–657.

167.

Hochman

Jiang

Matyus

. Endocytosis and dissociation of class I MHC molecules labeled with fluorescent beta-2 microglobulin. J Immunol 1991; 146: 1862–1867.

168.

Schirmbeck

Melber

Reimann

Hepatitis B virus small surface antigen particles are processed in a novel endosomal pathway for major histocompatibility complex class I-restricted epitope presentation. Eur J Immunol 1995; 25: 1063–1070.

169.

Munz

Antigen processing for MHC class II presentation via autophagy. Front Immunol 2012; 3: 9.

170.

Hewitt

: The MHC class I antigen presentation pathway: strategies for viral immune evasion. Immunology 2003; 110: 163–169.

171.

Schwartz

Marechal

Le Gall

. Endocytosis of major histocompatibility complex class I molecules is induced by the HIV-1 Nef protein. Nature Med 1996; 2: 338–342.

172.

Seidel

Oliveira

Lampen

. A novel category of antigens enabling CTL immunity to tumor escape variants: Cinderella antigens. Cancer Immunol Immunother 2012; 61: 119–125.

173.

Zagzag

Salnikow

Chiriboga

. Downregulation of major histocompatibility complex antigens in invading glioma cells: stealth invasion of the brain. Lab Invest 2005; 85: 328–341.

174.

Pisklakova

McKenzie

Zemp

. M011l-deficient oncolytic myxoma virus induces apoptosis in brain tumor-initiating cells and enhances survival in a novel immunocompetent mouse model of glioblastoma. Neuro-Oncology. Epub ahead of print 8 March 2016. DOI: 10.1093/neuonc/now006.

175.

Eipper

Steiner

Lesner

. Lactoferrin is an allosteric enhancer of the proteolytic activity of cathepsin G. PLoS ONE 2016; 11: e0151509.

176.

Siltari

Kivimaki

Ehlers

. Effects of milk casein derived tripeptides on endothelial enzymes in vitro; a study with synthetic tripeptides. Arznei-Forschung 2012; 62: 477–481.