Human Myeloid-derived Suppressor Cells Derived From the Bone Marrow Are Decreased With Age but Maintain Their Functional Ability

Myeloid-derived suppressor cells (MDSCs) are a heterogeneous population of immunosuppressive cells developing from myeloid progenitors, which are enriched in pathological conditions such as cancer, and are known toinhibit the functions of effector T cells. During aging, several changes occur both at the adaptive and innate immune system level, in a process dened as immunoscenescence. In particular, the low-grade inammation state observed in the elderly appears to affect hematopoiesis. We previously demonstrated that the combination of GM-CSF and G-CSF drives the in vitro generation of bone marrow-derived MDSCs (BM-MDSCs) from precursors present in human bone marrow aspirates of healthy donors, and that these cells are endowed with a strong immune suppressive ability, resembling that of cancer-associated MDSCs. In the present work we investigated BM-MDSCs induction and functional ability in a cohort of pediatric versus elderly donors. To this aim, we analyzed the differences in maturation stages and ability to suppress T cell proliferation. We found that the ex vivo distribution of myeloid progenitors is similar between pediatric and elderly individuals, whereas after cytokine treatment a signicant reduction in the more immature compartment is observed in the elderly. Despite the decreased frequency, BM-MDSCs maintain their suppressive capacity in aged donors. Taken together, these results indicate that in vitro induction of MDSCs from the BM is reduced with aging and opens new hypotheses onthe role of age-related processes in myelopoiesis.

cancer-associated MDSCs. In the present work we investigated BM-MDSCs induction and functional ability in a cohort of pediatric versus elderly donors. To this aim, we analyzed the differences in maturation stages and ability to suppress T cell proliferation. We found that the ex vivo distribution of myeloid progenitors is similar between pediatric and elderly individuals, whereas after cytokine treatment a signi cant reduction in the more immature compartment is observed in the elderly. Despite the decreased frequency, BM-MDSCs maintain their suppressive capacity in aged donors. Taken together, these results indicate that in vitro induction of MDSCs from the BM is reduced with aging and opens new hypotheses onthe role of age-related processes in myelopoiesis. These cells are released into the bloodstream and recruited to the affected tissues, where they proliferate and are activated by in ammatory factors, and suppress acute in ammatory reactions by inhibiting the functions of distinct components of innate and adaptive immunity 5 .In tumors and in ammatory disordersT cells represent the main target of MDSC-induced immune tolerance 1,6 .Indeed, MDSCs are involved in tumor angiogenesis, drug resistance and tumor progression and could represent a potential therapeutic target both in cancer and in chronic in ammatory diseases 3,7 .
During aging several changes take place such as the dysregulation of the immune, central and peripheral nervous, endocrine and metabolic system. In particular, aging is associated witha decline of functional capacity of both the adaptive and innate immune systems, in a process de ned as immunoscenescence;however these changes have a more signi cant impactonthe cells of the adaptive immunity than those of innate immunity 6 .There is an age-related loss of CD4 + and CD8 + T cells and an alteration of their functional capacities. The reduction of these cells seems to be linked to the decrease of the expansion of T cell clones and also to the proliferation ability after maturation 6 . Moreover, there is an increase of regulatory T cells and a reshaping in the number of T helper populations associated with aging 6 .
Aging is also associated with a chronic, low-grade in ammation state called in ammaging 8 .It appears that in the microenvironment of the bone marrow (BM)in ammaging affects haematopoietic stem cells, with a possible rebound on the myelopoiesis and lymphopoiesis process 9 . In particular, one of the hallmarks of the alterations in the BM during aging is the increased myelopoiesis, associatedwith a concomitantdecrease in lymphopoiesis.
Our group demonstrated that GM-CSF, G-CSF, and IL-6 allow thein vitrogeneration of MDSCs from precursors present in human bone marrow aspirates of healthy donors, and named such cells BM-derived MDSCs (BM-MDSCs).Of note, these cells share the phenotype and the suppressive function of MDSCs isolated from cancer patients 10  Most of the works that advance that MDSC levels increase during aging have been obtained in mouse models in which these suppressive populations have been evaluated in the BM, spleen and circulation 11,12 , while only a few works have documented it also in humans 13,14 . It is not clear if the increase of myelopoiesis, which occurs in aging, is associated with an increase of MDSCs generation in the BM of humans, and consequently with an increase of these cells in the blood circulation. In addition, there are no data showing whether aging impacts on the immunosuppressive ability of MDSCs.
Aim of this study is to comparethe induction of MDSCs from the BM of young and old individuals by using an optimized method to generate in four days MDSCs from precursor cells through cytokines treatment. Such cells are equivalent to MDSCs present in the blood of cancer patients and gives us the possibility to evaluate not only the expansion ability of the precursors, but also the immunosuppressive ability of the induced myeloid suppressor cells.

Results And Discussion
In this study a total of 20 pediatric and 24 elderly donors were enrolled. Demographic data are reported in Table 1, showing a median age of 6 years for pediatric donors, and of 79 years for elderly individuals. BM aspirates from both pediatric and elderly donors with normal cytologic characteristicswere freshly characterized for the myeloid differentiation by ow cytometry using by combining the expression of the markers CD11b and CD16.To induce the expansion of BM-MDSCs from myeloid cell precursors, we cultured BM cells for four days with the combination of the cytokines G-CSF and GM-CSF, and assessed the maturation of the myelomonocytic precursors by ow cytometry, as previously reported 10,15 . A representative example of the immunophenotype of a BM before and after cell culture of a pediatric and an adult individual is shown in gure 1A, left panels, while the cumulative result of several independent experiments is shown in Figure 1, right panels. Three different myeloid subsetswere analyzed:the most immatureand immune suppressive fraction corresponding to CD11b low/-/CD16 -(immature-BM-MDSCs, i-BM-MDSCs), and the two more differentiated but immature and non-suppressive CD11b + /CD16and CD11 + /CD16 + subsets 10 . Only in the freshly isolated BM cells a more differentiated cell population corresponding to CD11b + /CD16 high polymorphonuclear (PMN)cells is present, while the cell culture with the cytokines gives rise to a heterogeneous cell population in which the most differentiated myeloid cells, corresponding to PMN, is lacking (Fig. 1A).
When we compared theex vivo distribution of the three myeloid cell subsetsbetween pediatric and elderly donors, we found a similar distribution of the myeloid differentiating cells (Fig. 1A, left panels; Fig. 1B suggesting that there is no signi cant change in their distribution with aging, despite a reported shift toward myelopoiesis 16 . Instead, after 4 days of cell culture in the presence of myelomonocytic cytokines we observed a signi cant reduction in the percentage of suppressivei-BM-MDSCs in the group of elderly, (p=0.0001 by pairwise comparison), but not in the two other groups of more differentiated cells (Fig. 1C).
To assess whetheri-BM-MDSCs isolated from the two groups of donors displayed immune suppression on T cells, after 4 days of cell culture two fractions of BM-MDSCs were isolated, corresponding to the more immature CD11band the more differentiated CD11b + cell fraction. A functional assay was set up withallogenic peripheral blood mononuclear cells (PBMCs) activated with anti-CD3 and anti-CD28 for 4 days in the presence of the two myeloid cell fractions.Previous work from our laboratory already demonstrated that the CD11bfraction corresponds to the suppressive i-BM-MDSC, while CD11b + cells are devoid of signi cant immune suppressive activity 10 .Evaluationof the proliferation of T cells was assessed by a qualitative (Fig. 2A) or a quantitative (Fig. 2b) fashion. In line with previous results, only the most immature CD11bcell fraction showed the ability to suppress the proliferation of T cells in both groups of donors (Fig. 2 A-B) 10 and,importantly, no signi cant differences were found comparing the suppression ability of CD11b -BM-MDSCs isolated from pediatric and elderly donors, thus suggesting that the suppressive ability of MDSCs is maintained with aging.Of note, the immune suppressive capacity is acquired only after cytokines induction, as demonstrated by functional tests performed on ex vivo sorted immature CD11bcells from the BM of both the pediatric 10 and elderly (data not shown) donors,that failed to suppress T cell proliferation.
We previously demonstrated that MDSCs can be induced in vitro by using a combination of cytokines and that CD11b -/CD16cells phenotypically and functionally resembled to myeloid suppressor cells present in the blood of cancer patients 10,15 . In the present work, we extend these data and observe that a reduced percentage of immune suppressive myeloid cells can be generated with G-CSF and GM-CSFcombinationfrom the BM of elderly donors, suggesting a less effective ability to induce this mechanism.Of note, the reduced frequency of BM-MDSCs in elderly donors does not translate in a worsened MDSCs function, since this population maintains the same immunosuppressive capacity of pediatric donors, as demonstrated by the elevated suppression observed in a functional assay.Literature data report that the aging process is associated with a signi cant increase in MDSCs 14 , andaccompanied by an augmented myeloid-to-lymphoid ratio 16 , although it was never clari ed the mechanism of this age-related process. Actually, con icting results exist concerning the cellular complexity of the bone marrow nicheduring aging 17 , and our understanding of MDSCgeneration during aging remains limited 13 . However, our ndings suggest thatin vitroinduction of MDSCs from the BM is reduced with aging and, thus, may be related to an impaired response to cytokine stimuli 18 . Therefore,the increased peripheral MDSC levels observed in cancer patients might be dependenton cancer-associated mechanisms, rather than on the aging process itself.

Patients information
Two series of consecutive patients were recruited in this study.One of these corresponded to elderly who underwent total hip arthroplasty (THA) or hip endoprosthesis at the Orthopedic Clinic, and the second included pediatric patients enrolled in the protocol AIEOP-BFM-ALL 2000,with suspected leukemia or lymphomas, with lymphatic leukemiaafter 78 days without recurrences, and with lymphatic leukemiaafter BM transplantation as a part of the diagnostic follow-up. For adult patients one of the inclusion criteria was suspension of steroid therapy for at least 3 months. Exclusion criteria were cancer, infections, sickle cell anemia and autoimmune diseases. Informed consent was obtained from all participants, in compliance with all the relevant national regulations, institutional policies and in accordance the tenets of the Helsinki Declaration, and has been approved by the local Ethic Committee(CESC Code: n. 95).

BM-MDSC induction and cell subsets separation
Fresh BM aspirates were treated with K 2 EDTA to prevent coagulation, and lysed to remove red blood cells with a hypotonic solution of ammonium chloride. Myeloid populations were isolated through magnetic sorting by the depletion of CD3 + /CD19 + /CD56 + lymphocytes, with a cocktail of immunomagnetic beads obtained by combining anti-human CD3, CD19, and CD56 beads (MiltenyiBiotec). Cell purity was checked by FACS analysis on forward/side scatter parameters. Subsequently, the CD3 − /CD19 − /CD56 − fraction was cultured with 40 ng/ml G-CSF and GM-CSF (MiltenyiBiotec) for 4 days at 37°C, 8% CO 2 in order to obtain BM-MDSCs, as previously described 10,15 . On the fourth day, cells were collected andseparated into CD11band CD11b + fractions with immunomagnetic anti-human CD11b beads (MiltenyiBiotec). The purity of sorted cells was checked by staining both fractions with anti-CD16 (BD Pharmingen) and anti-CD11b (Beckman Coulter) antibodies and analyzing cells by FACS Calibur cytometer (BD Biosciences). All the fractions were obtained with a purity of ≥ 90%.

Immunophenotyping analysis by ow cytometry
Cell surface staining for ow cytometry was performed as previously described 10 . Brie y, cellswere incubated with FcReceptor (FcR) Blocking Reagent (MiltenyiBiotec) and then labeled for 20 min at 4°C with monoclonal antibodies (Abs) anti-CD11b (BeckmanCoulter) and anti-CD16 (BD Pharmingen). Data acquisition was performedwith FACSCalibur (BD Biosciences) and samples were analyzed by FlowJo software (Tree Star Inc).

Proliferation assay and immune suppression evaluation
Peripheral blood mononuclear cells (PBMCs) were isolated from the peripheral blood of healthy donors by density gradient centrifugation on FicollPaque PLUS (GE Healthcare-Amersham, NJ, USA), as previously described 10 . Brie y, PBMCs were stained with 0.5 µM CellTrace™ Violet Cell Proliferation Kit (Invitrogen, Molecular Probes, MA, USA), according to manufacturer's instructions. CellTrace-labelled PBMCs were activated with coated 1 µg/ml anti-CD3 and 5 µg/ml soluble anti-CD28 (BioLegend) for 4 days and co-cultured in at bottom 96 well plates at the 1:1 ratio withCD11b -, CD11b + and unsorted fractions. Cell cultures were incubated at 37°C and 5% CO 2 in argininefree-RPMI (Biological Industries), supplemented with 150 µM arginine, 10% FBS (Sigma-Aldrich),10 U/ml penicillin and streptomycin, and 10mM HEPES. After 4 days, cells were harvested, stained with anti-CD3 (Beckman Coulter) and analyzed by ow cytometry. Proliferation of T cells was evaluated by assessing the signal of CellTrace on CD3 + cells, and considering as proliferating the cells present from generation 2onwards, or calculating the absolute number of CD3 + cells in each sample by BD TruCount™ tubes (BD Biosciences). In both cases data were normalized assuming the proliferation of T cells cultured alone as 100%.

Statistical analysis
The Shapiro-Wilk test was used to determine whether data were distributed normally. Chi-square (χ2) test was performed to compare dichotomous data. Mann-Whitney test or unpaired t-test were used to compare continuous variables. Continuous variables were reported as median ± interquartile range (IQR). A p<0.05 was considered as statistically signi cant. All analyses were performed withGraphpad Prism All participants gave a written informed consent for participation in the study. The study was approved bythe local Ethic Committee(CESC Code: n. 95).

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests. Authors' contributions SM: Study concept; drafting the manuscript, data analysis and interpretation. EM,SM and SS performed laboratory studies. SF performed cytological evaluation of the BM aspirates and interpreted data. AP and EB: sample collection, drafting the manuscript, data analysis and interpretation. AB and PR performed surgery and collected data. All authors provided critical comments, read and approved the nal version of the manuscript.  Suppression of PBMCsproliferation by myeloid subsets sorted from BM or BM-MDSCs.CellTrace-labelled PBMCs activated with anti-CD3 and anti-CD28 were cultured in the presence of 1:1 ratio of the different myeloid populations sorted from BM or BM-MDSCs. Immunosuppression of these populations was evaluated on activated T cells, (gated as CellTrace+/CD3+) normalized on the control without myeloid cells.A Qualitative suppression was calculated, analyzing the number of proliferating cellsfrom generation 2 to generation 10, assumed to be 100% without myeloid cells.BQuantitative suppression was calculated by analyzing the absolute number of proliferating CD3+cells byTruCount™ tubes (n=20 independent experiments for pediatric patients, n=20 for elderly patients).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. lettereditor.pdf