Upregulation of Cathepsin X in Glioblastoma: Interplay with γ-Enolase and the Effects of Selective Cathepsin X Inhibitors

Abstract

Glioblastoma (GBM) is the most common and deadly primary brain tumor in adults. Understanding GBM pathobiology and discovering novel therapeutic targets are critical to finding efficient treatments. Upregulation of the lysosomal cysteine carboxypeptidase cathepsin X has been linked to immune dysfunction and neurodegenerative diseases, but its role in cancer and particularly in GBM progression in patients is unknown. In this study, cathepsin X expression and activity were found to be upregulated in human GBM tissues compared to low-grade gliomas and nontumor brain tissues. Cathepsin X was localized in GBM cells as well as in tumor-associated macrophages and microglia. Subsequently, potent irreversible (AMS36) and reversible (Z7) selective cathepsin X inhibitors were tested in vitro. Selective cathepsin X inhibitors decreased the viability of patient-derived GBM cells as well as macrophages and microglia that were cultured in conditioned media of GBM cells. We next examined the expression pattern of neuron-specific enzyme γ-enolase, which is the target of cathepsin X. We found that there was a correlation between high proteolytic activity of cathepsin X and C-terminal cleavage of γ-enolase and that cathepsin X and γ-enolase were colocalized in GBM tissues, preferentially in GBM-associated macrophages and microglia. Taken together, our results on patient-derived material suggest that cathepsin X is involved in GBM progression and is a potential target for therapeutic approaches against GBM.

Keywords: cathepsin X; cathepsin X inhibitors; glioblastoma; glioblastoma stem cells; tumor microenvironment; γ-enolase.

Figure 1

Cathepsin X mRNA levels, protein expression, and enzymatic activity in GBM. (A) Relative mRNA level of cathepsin X was increased in de novo GBM tissues (n = 43) and recurrent GBM (GBM rec, n = 5) compared to LGG (n = 14) and nontumor brain tissues (N, n = 16). (B) Cathepsin X at the mRNA level was expressed in primary GBM cells (n = 17) and normal astrocytes (NA, n = 1) and at a low level in GSCs (n = 6). (C) Cathepsin X mRNA was expressed in classical (CL, n = 18), proneural (PN, n = 3), mesenchymal (MES, n = 17), and mixed (MIX, n = 25) GBM subtypes. The boxplots show relative cathepsin X mRNA expression in different sample groups. Different Y-axis scales are presented. (D) The cathepsin X protein level was not different in GBM tissues (n = 14) compared to the nontumor tissue control (N, n = 7). (E) The cathepsin X enzyme activity was significantly increased in GBM tissues (n = 14) compared to the nontumor tissue control (N, n = 7). The data are presented as the mean values $\pm$SEM. * p < 0.05, ** p < 0.01.

Figure 2

Localization of cathepsin X in CD68-positive and Iba-1-positive cells. Representative images of triple immunofluorescence staining of cathepsin X (green) and markers of (A) GSCs (SOX2—red, CD133—grey), (B) macrophages and microglia (CD68—red) and microglia (Iba1—purple), and (C) GBM cells and astrocytes (GFAP—red) and GSCs (SOX2—grey) are shown. The nuclei were counterstained with Hoechst 33258 (Hoechst, blue). Scale bar = 50 μm.

Figure 3

Effects of cathepsin X inhibitors AMS36 and Z7 on viability of GBM cells and GBM-associated cells. GBM cells NIB140 (A), differentiated macrophage THP-1 cells (B), and BV-2 microglial cells (C) exposed to the NIB140- and NCH421k-conditioned media were treated with cathepsin X inhibitors AMS36 and Z7 at various concentrations. Cell viability was then assessed using the MTS assay. The control means a DMSO solvent (0.25%). The control medium was a blank GBM cell/GSC medium without soluble molecules secreted from GBM cells/GSCs. The data are presented as the mean values $\pm$SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001.

Figure 4

Colocalization of cathepsin X and γ-enolase in GBM tissues. (A) A structure of the γ-enolase dimer and cleavage of C-terminal amino acids by cathepsin X (A-1). Three different primary antibodies against γ-enolase were used: A γ-enolase (D-7) antibody specific for the epitope between amino acids 41–73 near the N-terminus (purple amino acid sequence), a γ-enolase (NSE-P2) antibody against the internal region (amino acids 271–285, blue amino acid sequence), and a γ-enolase (NSE-P1) antibody against amino acids 416–433 in the C-terminal region (red amino acid sequence) (A-2). Total γ-enolase was detected using the NSE-P2 antibody, whereas the intact active form was detected using the NSE-P1 antibody. (B) Double immunofluorescence staining for cathepsin X and γ-enolase. In Figure B-1 and Figure B-3, no or partial colocalization of cathepsin X and γ-enolase was observed. In Figure B-2, colocalization of both proteins was observed, suggesting that cathepsin X colocalizes mainly with cleaved γ-enolase. The cell nuclei were counterstained with Hoechst 33258 (Hoechst, blue). Scale bar = 50 μm. (C) Significantly lower levels of the intact active form of γ-enolase were observed in GBM tissues compared to the nontumor brain tissue control (N). Protein levels of both forms of γ-enolase in GBM and nontumor brain tissues were obtained by means of ELISA. The bars represent the means with individual values (n = 14 of GBM and n = 7 of nontumor tissue samples) ± SEM. **** p < 0.0001. Image A-1 was created with Mol*Viewer [51] based on RCSB PDB (rcsb.org) ID 1TE6 [52] and image A was created with BioRender.com.

Figure 5

Colocalization of cathepsin X and γ-enolase in CD68-positive cells in GBM tissues. Representative images of triple immunofluorescence staining for cathepsin X (green), γ-enolase NSE-P2 (red), and marker of macrophages and microglia (CD68—purple) show overlapping expression of cathepsin X and γ-enolase NSE-P2 in the CD68-positive cells. The cell nuclei were counterstained with Hoechst 33258 (Hoechst, blue). Scale bar = 50 μm.

Figure 6

Effects of the γ-Eno peptide on proliferation of GBM cells, GSCs, and GBM-associated cells. GBM cells NIB140, GSCs NCH421k (A), as well as differentiated macrophage THP-1 cells (B) and BV-2 microglial cells (C) exposed to the NIB140- and NCH421k-conditioned media were treated with increasing concentrations of the γ-enolase peptide, C-terminal 30-amino-acid sequence of human brain γ-enolase (γ-Eno). Cell proliferation was evaluated using CFSE staining and flow cytometry. The control means a blank culture GBM cell/GSC medium without addition of γ-Eno (A). The control medium is a blank GBM cell/GSC medium without soluble molecules secreted from GBM cells/GSCs and without addition of γ-Eno (B,C). The data are presented as the means $\pm$SEM. * p < 0.05, ** p < 0.01.