S100A4 is a potential target for immunotherapy for human glioma

Nourhan Abdelfattah, Department of Neurology, Houston Methodist Institute, Texas, USA, analyzed multiregional and multidimensional analysis of human gliomas and mapped GBM cell atypia and spatial, molecular, and functional heterogeneity
of gliomas and associated stromal cells, including immune cells.

The results were published online in the February 2022 issue of Nature Communications
.

- Excerpted from the article chapter

【Ref: Abedelfattah N, et al.
Nat Commun.
2022 Feb 9; 13(1):767.
doi: 10.
1038/s41467-022-28372-y.
】

Research background

For the treatment of glioblastoma (GBM), GBM should theoretically be an ideal candidate for immunotherapy, because immune cells can cross the blood-brain barrier, track infiltrating glioma cells, and selectively kill the index, tumor cells, and preserve normal brain tissue
.

But in fact, most GBM immunotherapy trials, including vaccines, CAR-T cells, and immune checkpoint inhibitors, have shown only limited survival benefit
in GBM patients.

An important barrier to the effectiveness of immunotherapy in GBM is due to the lack of tumor infiltrating lymphocytes (TIL; <5%), and a large number of immunosuppressive myelocytes are present, making it an "immunocold tumor"<b14>.

Tumor-associated myelocytes are key regulators of tumor progression, metastasis, and immune evasion, and are highly promising therapeutic targets
.

Nourhan Abdelfattah, Department of Neurology, Houston Methodist Institute, Texas, USA, analyzed multiregional and multidimensional analysis of human gliomas and mapped GBM cell atypia and spatial, molecular, and functional heterogeneity
of gliomas and associated stromal cells, including immune cells.

The results were published online in the February 2022 issue of Nature Communications
.

Research methods

The authors first performed multi-omics analysis
on multi-point samples from 18 patients with glioma.

After unsupervised clustering analysis of single-cell data, 12 subpopulations
with different gene expression patterns were identified.

According to the copy number and marker gene expression, they are divided into myelocytes, glioma cells, T cells, B cells, or other stromal cells
.

Among them, glioma cells accounted for 40.
5%, myeloid cells 45.
0%, and T cells accounted for 9.
7%.


Through pathway analysis of characteristic genes, these subpopulations were found to be significantly enriched in
epithelial-mesenchymal transition (EMT), hypoxia, Myc-target-v1, interferon-γ (IFNG)-response, TNFa-transduction-NFkB signaling, and G2M checkpoint (cell cycle) markers.

Study results

T cells and natural killer cells (NK) accounted for 6.
4% ±2.
5% of primary GBM and 14.
3% ±8.
9% of recurrent GBM, indicating increased
T cell infiltration during glioma progression.

PDCD1/PD1 expression was low in all samples, which may explain the low efficacy of anti-PD1/PDL1 inhibitors on GBM
.

Myelocytes, including microglia and bone marrow-derived macrophages (BMDM), form the largest stromal region
in gliomas.

The authors identified nine bone marrow subsets (MC 1-9)
with unique gene expression patterns.

MC1, MC2, MC6, and MC7 express previously detected microglial markers P2RY12 and TMEM119 and high levels of markers BHLHE41, SORL1, SPRY1, and SRGAP28
.

MC1 (i-Mic) expresses high levels of activated microglial markers CCL3 or macrophage inflammatory protein-1 (MIP-a), CCL4/MIP-β, CCL3L3, CCL4L2 and CD83 as well as TNF, IL1B and NFKBIZ
.

In BMDM cells, MC8(DC) stands for antigen presenting cells (APCs) and expresses the traditional dendritic cell markers CD1C, BATF3, and MHC-II genes
.

Compared to the rest of the MC subsets, MC4 (MDSC) expressed high levels of MIF and lower levels of the mature macrophage markers CD68, CD163, CD204/MSR1, CD206/MRC1, and CD49d/ITGA4
.

MC3 (s-Mac1) expresses high levels of the monocytes marker CD14 and the polarized M2-like macrophage markers CD163 and CD204/MSR1
.

MC5 (s-Mac2) expresses high levels of CD163, S100A4, LYZ, and immunosuppressive markers such as VEGFA, TGFB1, and IL10
.

After classifying various glioma and immune cell subtypes in humans, the authors sought to identify immunomodulatory targets
.

S100A4 stands out
as a highly expressed characteristic gene in T cell subsets as well as proneoplastic myeloid cells.

Dual immunofluorescence analysis of human GBM and mouse glioma samples confirmed that S100A4 was associated
with immunosuppressive macrophages (CD206 and CD163) and T cells (FOXP3).

Elevated expression of S100A4 is significantly associated with poor prognosis in patients with
glioma and GBM.

Multivariate analysis of S100A4 expression and tumor subtype, sex, recurrence, IDH status, and MGMT status showed that S100A4 was an independent prognostic factor
.

Conclusion of the study

In summary, the authors' comprehensive analysis of 201,986 human gliomas, immune cells, and other stromal cells at the single-cell level revealed extensive spatial and molecular heterogeneity
in immune infiltration.

The authors also used S100A4 as a modulator of immunosuppressive T cells and myeloid cells of GBM and demonstrated that deletion of S100A4 in tumor cells reprogrammed immune processes and significantly improved survival
.