5. T´ ECNICAS DE RESOLUCI ´ ON DEL PROBLEMA
6.1.1. Caracter´ısticas, par´ ametros y t´ ecnicas m´ as utilizadas
Gielen P.R., Schulte B.M., Kers-Rebel E.D., Verrijp K., Petersen-
Baltussen H.M.J.M., ter Laan M., Wesseling P., Adema G.J.
Chapter 4
4
Abstract
M
yeloid-derived suppressor cells (MDSCs), defined as CD33+MHC-II- cells, are in-creased in a variety of human tumours and are associated with immunosuppression.
MDSCs can be further subdivided in CD14+ monocytic (M-MDSCs) and CD15+ granulocytic
(PMN-MDSCs) subpopulations. Here we set out to determine the presence of MDSC sub-
sets in the blood and tumour tissue of patients with a glioma, including the most malig-
nant variant i.e. glioblastoma (GBM). CD33+MHC-II- MDSCs in blood from 21 patients with glioma and 12 healthy individuals were phenotyped and quantified by flowcytometry.
Myeloid populations of M-MDSC and PMN-MDSC phenotype were both significantly in-
creased in the blood of GBM patients versus to healthy controls. The myeloid activation markers CD80 and PD-L1 could not be detected on either of these MDSC subsets, while CD124, CD86 and CD40 were detected at similar levels on MDSCs in patients with glioma
and healthy donors. By contrast, in tumour cell suspensions, the MDSC population con-
sisted almost exclusively of CD15+ cells. Immunohistochemistry confirmed infiltration of CD15+MHC-II- cells in glioma tissue samples. These data support a role for cells with an MDSC phenotype in blood and tumour microenvironment of patients with GBM.
Introduction
G
liomas are the most common primary brain tumours, originate from glial cells or precur- sors thereof. Based on their histology, gliomas are classified and graded for malignancy according to guidelines of the World Health Organization (WHO) (Louis, Ohgaki et al. 2007). Glioblastoma (GBM) is the most malignant and, unfortunately, the most frequent glioma. Despite multimodal treatment (i.e. surgical resection followed by chemoradiation), the me- dian survival for patients with a GBM is only 14.2 months (Gousias, Markou et al. 2010). Thus, there is an urgent need for the development of novel, more effective therapies. Im- munotherapy is emerging as an effective therapy for the treatment of various cancers (Car- son, Doose et al. 2006; Couzin-Frankel 2013; Gielen, van der Valk et al. 2014), but the brain is unique in that it is an immune-specialized site, capable of tightly regulating the immune response.Multiple immunotherapeutic strategies have been exploited to treat patients with a glioma. Data from phase I and phase II trials have suggested that immunotherapy can produce a meaningful and sometimes durable beneficial response in patients with GBMs. The report- ed studies were generally performed in only a limited number of patients, and immunologi- cal responses could only be measured in a portion of these patients (Marsh, Goldfarb et al. 2013). The limited efficacy of immunotherapy in patients so far might be the result of an immunosuppressive network implemented by the tumour (Grauer, Wesseling et al. 2009). Such a network encompasses soluble mediators, inhibitory receptors and suppressive cells that may prevent the initiation and execution of proper immune responses (Finocchiaro and Pellegatta 2011; Hanahan and Weinberg 2011; Lindau, Gielen et al. 2013). One of these im- munosuppressive cell types is the regulatory T cell (Treg). We have previously reported that Tregs accumulate within high-grade gliomas but not in the blood of patients with glioma. Furthermore, Tregs isolated from glioma were shown to be able to suppress tumour-infil- trating lymphocyte (TIL) function (Fecci, Mitchell et al. 2006; Heimberger, Abou-Ghazal et al. 2008; Jacobs, Idema et al. 2009; Jacobs, Idema et al. 2010).
4
Recently, another immune suppressive cell type, myeloid-derived suppressor cell (MDSC), was found to be increased in blood and in tumour tissue of patients with various tumours (Diaz-Montero, Salem et al. 2009; Gabitass, Annels et al. 2011). MDSCs are a heterogeneous group of immature myeloid-derived cells that are capable of suppressing the immune sys- tem. In mice, these cells are divided in a polymorphonuclear subpopulation (PMN-MDSCs), which is defined by expression of CD11b+Gr-1highLy-6G+Ly-6ClowCD49d-, and a monocytic
Donor # Tumor type WHO grade age sex
PT 1 glioblastoma IV 64 f
PT 2 low grade oligodendroglioma II 52 m
PT 3 glioblastoma IV 62 m
PT 4 anaplastic oligoastrocytoma III 42 f
PT 5 glioblastoma IV 61 m
PT 6 low grade ependymoma II 29 f
PT7 anaplastic astrocytoma III 40 m
PT8 glioblastoma IV 62 f
PT9 anaplastic oligodendroglioma III 53 m
PT10 low grade glioma n.o.s. II 21 f
PT11 glioblastoma IV 67 m
PT12 glioblastoma IV 58 f
PT13 anaplastic oligodendroglioma III 45 m
PT14 glioblastoma IV 67 f
PT15 glioblastoma/gliosarcoma IV 52 m
PT16 glioblastoma IV 64 f
PT17 anaplastic oligodendroglioma III 44 f
PT18 glioblastoma IV 60 m
PT19 low grade oligoastrocytoma II 45 m
PT20 anaplastic oligodendroglioma III 49 f
PT21 glioblastoma IV 52 f HC1 N/A 57 f HC2 N/A 53 m HC3 N/A 63 f HC4 N/A 49 m HC5 N/A 66 f HC6 N/A 51 m HC7 N/A 30 m HC8 N/A 64 m HC9 N/A 39 m HC10 N/A 62 m HC11 N/A 60 f HC12 N/A 56 m
Table 1.Sample details.HC: Healthy control; N/A: Not Applicable; n.o.s.: not otherwise specified; PT: patient; m: male; f: female
4
population (M-MDSCs), which is characterized by a CD11b+Gr-1highLy-6G-Ly-6ChighCD49d+ phenotype (Movahedi, Guilliams et al. 2008; Youn, Nagaraj et al. 2008). In humans, various studies have used different criteria for recognizing different phenotypes of MDSCs (Solito, Marigo et al. 2014). Most commonly, MDSCs are defined by the expression of the myeloid markers CD33 and/or CD11b and the absence or low expression of HLA-DR. Similar to mu- rine MDSCs, there are monocytic and granulocytic subpopulations of human MDSCs that ex- press CD14 or CD15, respectively. Both populations are present in low numbers in the blood of healthy donors (Lindau, Gielen et al. 2013). Interestingly, in patients with tumours, either CD14+ M-MDSCs or CD15+ PMN-MDSCs were most abundantly increased in the blood de- pending on the tumour type (Lindau, Gielen et al. 2013). To date, human MDSCs have often been assessed in peripheral blood mononuclear cells (PBMCs) samples after depletion of mature neutrophils by separation using a ficoll gradient. In addition, most studies have in- vestigated the presence of either PMN-MDSCs or M-MDSCs in peripheral blood, whereas both populations might be important to sustaining an immunosuppressive network in both the tumour as well as in peripheral blood of tumour patients.
MDSCs can suppress T-cell function by several mechanisms, including production and se- cretion of arginase, nitric oxide and reactive oxygen species (Lindau, Gielen et al. 2013). In addition, several myeloid markers, including CD80, PD-L1, CD124 and CD40, have been shown to be expressed on murine (Yang, Cai et al. 2006; Youn, Nagaraj et al. 2008; Pan, Ma et al. 2010) as well as human (Mandruzzato, Solito et al. 2009; Poschke, Mougiakakos et al. 2010; Zhang, Wang et al. 2013) MDSCs. For murine MDSCs, CD80 and CD40 were found to be important for their suppressive capacity and induction of Tregs, respectively (Yang, Cai et al. 2006; Pan, Ma et al. 2010). For human MDSCs, however, the expression of these markers has not been determined in detail and may differ for different tumour types. Because these cells may interfere with the proper induction of an immune response after immunotherapy, it is important to map their presence and phenotype, to determine their role in the tumour environment and to link their presence to therapeutic efficacy.
We therefore studied MDSC subsets in blood and in tumour tissue of patients with glioma. MDSCs were defined by a lack of major histocompatibility complex class II (MHC-II) expres- sion and the presence of the myeloid markers CD33 (in blood) or CD11b (in tumour tissue). We found an increase in both PMN- and M-MDSCs in the blood of patients with GBM com- pared with the blood of healthy controls. In contrast, only PMN-MDSCs were highly preva- lent in tumour tissue, suggesting that these cells are the main MDSC population contribut- ing to the immunosuppressive tumour microenvironment within glial neoplasms.