Chaperone-Mediated Autophagy in Pericytes: A Key Target for the Development of New Treatments against Glioblastoma Progression

Abstract

Glioblastoma (GB) cells physically interact with peritumoral pericytes (PCs) present in the brain microvasculature. These interactions facilitate tumor cells to aberrantly increase and benefit from chaperone-mediated autophagy (CMA) in the PC. GB-induced CMA leads to major changes in PC immunomodulatory phenotypes, which, in turn, support cancer progression. In this review, we focus on the consequences of the GB-induced up-regulation of CMA activity in PCs and evaluate how manipulation of this process could offer new strategies to fight glioblastoma, increasing the availability of treatments for this cancer that escapes conventional therapies. We finally discuss the use of modified PCs unable to increase CMA in response to GB as a cell therapy alternative to minimize undesired off-target effects associated with a generalized CMA inhibition.

Keywords: autophagy; cell therapy; chaperone-mediated autophagy; glioblastoma; glioblastoma therapy; pericytes; perivascular niche.

Figure 1

Schematic representation of cell type differences in peritumoral and intratumoral areas of GB cancer. In the peritumoral area, peripheral GB cells interact with peritumoral PCs from pre-existent blood vessels and promote several changes in their phenotype. These changes facilitate processes such as co-optation or peritumoral immune evasion, which help with tumor survival, tumor cell infiltration, and tumor spreading in the brain. In the intratumoral area, most studies identify PCs as cells derived from the tumor, which includes not only regular PCs, but also PCs derived from GSC. These PCs are defined to help in angiogenesis or immune evasion of the tumor. In both areas, the cellular changes are intended to promote tumor survival and growth.

Figure 2

Model of the effects of GB-induced CMA on PCs. GB interaction with PCs promotes a ROS burst that triggers the increased upregulation of CMA in PCs through the expression of LAMP-2A (big garnet arrow). KFERQ motifs in CMA substrate proteins are recognized by the Hsc70 chaperone in a complex with other co-chaperones. This complex binds to substrates and carries them to the lysosomal membrane where they interact with LAMP-2A, and the lysosomal Hsc70 to translocate them into the lysosomal lumen to be degraded. Aberrant induction of CMA by GB leads to changes in the PC proteome, promoting an immunosuppressive function in PCs that exerts a negative regulation on T cells and antigen-presenting cells as a result of changes in gene expression programs that up-regulate the anti-inflammatory cytokines TGF-β and IL-10 (red dashed arrows). The expression of PD-L1 (negative regulator) and the lack of CD80 and CD86 (co-stimulatory molecules) promote regulatory T cell generation (red arrows). GB-induced CMA in PCs also facilitates tumor growth by stabilizing GB–PC interaction and preventing the secretion of anti-tumor proteins and phagocytic ability. In addition, it induces pro-tumor immune molecules and modulates the MSC-like properties such as the increase in the microvesicles secretion with pro-regenerative factors and cytokines that assist tumor proliferation (black dashed arrows).

Figure 3

Possible immune responses after CMA inhibition therapies to prevent GB progression. In immune and brain-resident cells, CMA may be altered in different ways in response to GB, and its inhibition might favor tumor progression (green arrows) or promote anti-tumor activity (red arrows). Thus, (A) future gene and pharmacological therapies against CMA should take into account that cells of the adaptive immune response, such as CD4⁺ T cells and B cells, require CMA to develop their anti-tumor responses. CD4⁺ T cells need CMA for proper activation, including cytokine release and proliferation through degradation of negative regulators of T cell activation. B cells also need CMA for maintaining antigen presentation. However, inhibition of CMA in TME cells could have an anti-tumor effect. For instance, CMA decreases the inflammasome-mediated responses of TAMs and TNF-α secretion in microglial cells. Astrocytes also require CMA to acquire an anti-inflammatory phenotype dependent on physical contact with tumor cells. Thus, CMA inhibition would maintain an inflammatory phenotype that would aid tumor elimination. In addition, aberrant GB-induced CMA upregulation in PCs promotes cancer cell survival and progression through cell–cell stable interactions, contributing to the immunosuppressive peritumoral niche with the secretion of pro-tumoral proteins, angiogenesis promotion, and pro-tumoral macrophages recruitment. Therefore, CMA ablation in PCs would prevent GB–PC interaction, maintaining a pro-inflammatory phenotype, among other anti-tumor changes, that reduce tumor progression and promote tumor clearance. On the other hand, (B) CMA-deficient PCs can be used as an alternative therapy that can avoid possible side-effects of CMA inhibition in healthy and GB-conditioned cells. Modified PCs are able to inhibit GB tumor growth and induce tumor clearance preventing PC–GB interactions. They could no longer be modulated by GB cells and present a pro-inflammatory function that prevents tumor growth and helps tumor elimination by stimulation of the anti-tumor immune responses. This effect is partly due to the secretion of pro-tumor molecules (lumican, vitamin D, among others), increased phagocytosis ability, and antigen presentation. Hence, CMA-deficient PCs can stimulate TME cells, increasing phagocytic and inflammatory populations (macrophages, PCs, and microglia) and promoting T cell inflammatory responses.