Page 275 - Read Online

Dello Russo et al. Neuroimmunol Neuroinflammation 2018;5:36 I http://dx.doi.org/10.20517/2347-8659.2018.42 Page 7 of 13

GAMs are directly involved in the degradation of the extracellular matrix. Thus, these cells are key regula-
tors of a central process involved in the expansion of tumors as well as in the invasion of normal brain pa-
[69]
renchyma . In fact, microglial cells significantly increase the invasive phenotype of GL261 glioma cells in
[70]
vivo . Consistently, the invasiveness of glioma cells is diminished in microglial-depleted organotypic brain
[71]
slices inoculated with GL261 glioma cells . Matrix metalloproteases (MMPs), i.e., the enzymes involved in
[72]
the remodeling of the extracellular environment , are largely produced by tumor cells, infiltrating microg-
lia/macrophages, or other infiltrating leukocytes, particularly at the invasive tumor edge facilitating tumor
[4]
growth and invasion [71,73,74] . As detailed in our recent review , a complex crosstalk exists between glioma
cells and infiltrating GAMs which increases the activity of MMP enzymes, including MMP-2 and MMP-9.
[75]
Notably, the latter is over-expressed in GAM cells sorted from human glioblastoma tissues .
Consistently, several pharmacological treatments displayed beneficial effects in glioblastoma by limiting
the release of MMPs. For example, minocycline, a highly lipophilic tetracycline antibiotic with a good BBB
penetration property, reduced the expression of MT1-MPP in invading microglia/macrophages by suppress-
[75]
[76]
ing p38 MAPK activation . The drug also reduced secretion of MMP-9 and other pro-inflammatory
[76]
cytokines from microglia and tumor cells resulting in an overall decrease of glioblastoma cell migration .
Notably, minocycline is also able to reduce MCP-1 secretion by glioblastoma cells, thus potentially limit-
ing GAMs’ recruitment at tumor site (as discussed above). The same inhibitory effects on MT1-MMP were
[77]
displayed in vitro by the lipid lowering agent, atorvastatin . In addition, propentofylline, an atypical meth-
ylxanthine with central nervous system (CNS) glial modulating and antinflammatory actions, significantly
reduced tumor growth by targeting microglial production of MMP-9. The drug restricted also the migratory
[78]
capacity of both glioma CNS-1 cells and microglia in vitro . Invasion and infiltration of the normal brain
parenchyma interfere with radical surgical resections of glioblastoma, that often recur after the first aggres-
sive treatment. Pharmacological reduction of glioma cell motility and invasiveness thus hold the potential to
improve the outcome of current therapeutic approaches, by limiting the infiltration extent of normal brain
[69]
parenchyma .
Other features of GAMs
In vitro, microglia co-cultured in the presence of glioma cells appear to be morphologically activated al-
[10]
though phagocytosis is largely impaired . Nevertheless, another promising therapeutic approach for the
treatment of glioblastoma consists in the use of nanoparticles which are internalized by GAMs increasing
their antitumor immune activation [79,80] . Moreover, GAMs produce a vast array of growth and angiogenic
[81]
factors which further sustain proliferation of tumor cells [8,48,52] as well as tumor vessel formation . Interest-
ingly, genetic and pharmacological ablation in GAMs of neuropilin 1, a co-receptor that amplifies signaling
through the VEGF-A and TGFβ pathways, is associated with reduced glioma growth and blood vessel for-
[82]
mation and increased survival time of glioma bearing mice .

CONCLUSION
GAMs represent the most relevant population of tumor infiltrating cells that significantly contribute to the
pathogenesis of glioblastoma by favoring tumor growth and invasion of the normal brain parenchyma. Pre-
clinical evidence supports the notion that GAMs are a viable pharmacological target whose function can be
modulated in order to prevent their pathological activation. Current available data, summarized in Table 1,
suggest that the immune activation of GAMs can be genetically or pharmacologically modulated so that
these cells can be efficiently instructed to perform anti-tumor activities. In addition, it is possible to control
their recruitment at the tumor site, and the production of extracellular matrix remodeling enzymes, thus
limiting tumor growth and the ability to infiltrate normal brain parenchyma. One of the main limitations to
systemic chemotherapy for glioblastoma is represented by the inability of most drugs to effectively penetrate
the BBB and achieve cytotoxic concentrations in the cerebrospinal fluid and brain parenchyma. In fact, sev-