News

Pathways of myeloid cell differentiation in cancer

29 Jan 2013
Pathways of myeloid cell differentiation in cancer

by ecancer reporter Clare Sansom

 

The development and differentiation of myeloid cells from their precursor cells, a process that is known as myelopoiesis, is altered in cancer leading to the numerical expansion of relatively immature and activated cells.

 

These cells, together known as myeloid-derived suppressor cells (MDSCs), regulate immune responses in diverse disease states including infectious disease, sepsis and trauma, and they can also promote tumour metastasis and angiogenesis.

 

Two groups of MDSCs with slightly different morphologies and functions have been identified, one derived from monocytic cells (M-MDSCs) and the other from polymorphonuclear cells (PMN-MDSCs).

 

It has been assumed that the two MDSC types develop along different pathways, involving slightly different sets of growth factors, cytokines and transcription factors.

 

A group of researchers led by Dmitry Gabrilovich of H. Lee Moffitt Cancer, Tampa, Florida, USA have now investigated the pathways through which myeloid cells differentiate.

 

These results suggest that this pathway is altered in cancer so cells differentiate preferentially into PMN-MDSCs rather than into macrophages and dendritic cells.

 

Gabrilovich and his co-workers measured the proliferation and expansion of both types of MDSC in mice with transplanted and spontaneous tumours, finding that the PMN-MDSCs expanded more than M-MDSCs; this PMN-MDSC expansion was substantial in the blood and the spleen and statistically significant in the bone marrow.

 

They then injected a marker of DNA replication, 5-bromodeoxyuridine, into tumour-bearing mice and measured its incorporation into MDSCs.

 

Interestingly, the M-MDSCs were found to have a higher proliferation rate than the PMN-MDSCs even though fewer M-MDSCs were observed, suggesting that M-MDSCs might be able to differentate into PMN-MDSCs.

 

To test this, Gabrilovich and his co-workers sorted M-MDSCs and PMN-MDSCs from the spleen of tumour-bearing mice and cultured them in vitro for 3-5 days.

 

They found that more than 40% of the M-MDSCs differentiated into cells with the typical morphology of PMN-MDSCs, whereas few if any PMN-MDSCs differentiated into M-MDSCs.

 

These PMN-MDSCs had identical phenotypes to PMN-MDSCs originally isolated from normal and tumour-bearing mice, and all types of PMN-MDSC showed immuno-suppressive activity.

 

M-MDSCs were also shown to be able to differentiate into PMN-MDSCs when they were transplanted into tumour-bearing mice, but not when transplanted into normal mice.

 

The researchers compared the gene expression profiles of monoctyes, M-MDSCs and PMN-MDSCs in mouse spleen, looking most closely at patterns of expression of the retinoblastoma gene RB1 which had previously been observed to be lowered in PMN-MDSCs.

 

Expression of RNA and protein from RB1 was found to be fairly high and similar in M-MDSCs and in mature myeloid and lymphoid cells, but much lower in PMN-MDSCs; expression was lowest in cells derived from tumour-bearing mice.

 

Mature polymorphonuclear myeloid cells derived from mouse bone marrow also had low expression of the Rb1 protein; when the cells were removed from the bone marrow Rb1 was up-regulated in these cells but not in PMN-MDSCs.

 

The researchers next used immunofluorescence to show that down-regulation of RB1 expression is required in order for M-MDSCs to differentiate into PMN-MDSCs.

 

Mononuclear cells isolated from human patients diagnosed with several types of solid cancer have been found to include a significant proportion – generally from 10-20% – of cells characterised as MDSCs, with the majority of these having the PMN-MDSC phenotype.

 

The researchers isolated monocytes from the bone marrow of patients with multiple myeloma and volunteers without cancer, and cultured these in the presence or absence of medium that had been conditioned with tumour cells.

 

Some of the M-MDSCs isolated from the bone marrow of myeloma patients were seen to differentate into PMN-MDSCs, and this process was again shown to be associated with low expression of Rb1.

 

Furthermore, myeloid cells were found to acquire the characteristics of PMN-MDSCs and to accumulate in the spleens of transgenic mice in which the retinoblastoma gene had been deleted, indicating that this myeloid differentiation is directly regulated by the Rb1 protein.

 

Finally, the researchers showed that inhibition of the histone deacetylase HDAC-2 upregulated retinoblastoma expression and consequently prevented this aberrant differentiation of immature myeloid cells.

 

Taken together, these results implicate epigenetic changes modulated through HDAC-2 in the down-regulation of retinoblastoma expression and thus in the differentiation of immature myeloid cells into PMN-MDSCs, which are the primary type of myeloid-derived suppressor cell to accumulate in cancer.

 

This represents a new mechanism for regulation of myeloid cells in cancer, which might possibly be targeted by selective and specific anti-cancer drugs.

 

 

Reference 

Youn, J.-I., Kumar, V., Collazo, M. and 11 others (2013). Epigenetic silencing of retinoblastoma gene regulates pathologic differentiation of myeloid cells in cancer. Nature Immunology, published ahead of print 27 January 2013. doi:10.1038/ni.2526