Influence of murine mesenchymal stem cells on proliferation, phenotype, vitality, and cytotoxicity of murine cytokine-induced killer cells in coculture

Abstract

Stimulating lymphocytes with Ifn-γ, anti-CD3, and interleukin-2 promotes the proliferation of a cell population coexpressing T-lymphocyte surface antigens such as CD3, CD8a, and CD25 as well as natural killer cell markers such as NK1.1, CD49, and CD69. These cells, referred to as cytokine-induced killer cells (CIKs), display cytotoxic activity against tumour cells, even without prior antigen presentation, and offer a new cell-based approach to the treatment of malignant diseases. Because CIKs are limited in vivo, strategies to optimize in vitro culture yield are required. In the last 10 years, mesenchymal stem cells (MSCs) have gathered considerable attention. Aside from their uses in tissue engineering and as support in haematopoietic stem cell transplantations, MSCs show notable immunomodulatory characteristics, providing further possibilities for therapeutic applications. In this study, we investigated the influence of murine MSCs on proliferation, phenotype, vitality, and cytotoxicity of murine CIKs in a coculture system. We found that CIKs in coculture proliferated within 7 days, with an average growth factor of 18.84, whereas controls grew with an average factor of 3.7 in the same period. Furthermore, higher vitality was noted in cocultured CIKs than in controls. Cell phenotype was unaffected by coculture with MSCs and, notably, coculture did not impact cytotoxicity against the tumour cells analysed. The findings suggest that cell-cell contact is primarily responsible for these effects. Humoral interactions play only a minor role. Furthermore, no phenotypical MSCs were detected after coculture for 4 h, suggesting the occurrence of immune reactions between CIKs and MSCs. Further investigations with DiD-labelled MSCs revealed that the observed disappearance of MSCs appears not to be due to differentiation processes.

Figure 1. Comparison of cytokine-induced killer cell (CIK) proliferation.

CIKs were cultivated according to protocol (controls). On day 7, some of the CIKs were cultivated either with mesenchymal stem cells (cCIKs) or in MSC-conditioned medium (mediaCIKs), and proliferation was compared with that of controls. After starting the coculture, cCIKs proliferated faster than controls (p<0.05). The mediaCIKs displayed increased proliferation (p>0.1), too, but proliferation rates did not reach the level of cCIK proliferation.

Figure 2. Comparison of CIK phenotypes.

Fluorescence-activated cell sorter (FACS) analyses of control CIKs and cCIKs were performed on day 14. Controls expressed typical CIK markers, including T-cell markers CD3, CD8a, and CD25 and natural killer (NK) cell markers such as CD49, CD69, and NK1.1. No differences were observed between control CIKs and cCIKs.

Figure 3. Comparison of CIK vitality.

Vitalities of control CIKs and cCIKs were assessed on day 14 using two different methods. Part A displays vitalities determined using trypan blue staining. No significant differences in CIK viability were found at the 0.05 level (p = 0.1). Part B shows comparisons of CIK vitality, as assessed using 7-AAD staining. No significant differences were found (p = 0.1).

Figure 4. Comparison of CIK lysis potential against tumour cells.

Lysis rates of control CIKs compared with those of cCIKs against fibrosarcoma cells (Wehi 164 S) on day 14 for different effector-cell/target-cell ratios. Error indicator represents standard error of the mean. Initially, lysis rates increased with increasing effector-cell/target-cell ratio up to a lysis rate of 27% at a ratio of approximately 30∶1. Further increases in the ratios did not increase lysis rates further. No significant differences were observed between control CIKs and cCIKs (p>0.1).

Figure 5. MSC characterization.

A: MSCs were evaluated under a light microscope and representative photographs were captured on days 4, 7, 12, and 42 of incubation. Cells show typical fibroblast morphology and undergo typical time-dependent transitions among three morphologically distinct cell types: thin spindle-shaped cells, wider spindle-shaped cells, and still wider spindle-shaped cells. B: Fibroblastic colony-forming unit assays (CFU-F) were performed for 14 days. Colonies were then stained for ALP and methylene blue. When cultivated in normal medium, only a few colonies expressed ALP (line 1), whereas nearly all of the colonies stained with methylene blue also showed ALP expression when cultivated in osteogenic medium (line 2).

Figure 6. MSC phenotyping.

MSCs were harvested and analysed with flow cytometry for a panel of MSC-characterizing markers on day 7. Thus, the presented phenotype resembles that of MSCs used in the cocultures. MSCs were negative for CD3, CD31, CD34, and CD90. Positive signals were measured for CD44, CD73, and CD166. Positive as well as negative signals were detected for CD45.

Figure 7. Comparison phenotypes: CIKs – adherent cells after coculture - MSCs.

After 7/MSC coculture (day 14), adherent cells were harvested separately from suspended cells and analysed using flow cytometry. In addition, 14-day-old MSCs and CIKs were harvested and analysed for the same markers. Adherent cells after coculture (column in the middle) display the phenotype of CIKs (left column) and differ considerably from the phenotype expected for MSCs (right column).

Figure 8. DiD-labelled MSCs.

To investigate whether cocultured MSCs were differentiated, we carried out coculture experiments using DiD-labelled MSCs. Suspended cells (cCIKs) and adherent cells were harvested separately from each other after 4 and 24 h of coculture, stained with anti-CD3 and anti-CD25, and analysed using flow cytometry. The first line shows the results of flow cytometry analysis just before the addition of CIKs to the labelled MSCs. Clearly positive DiD signals were seen at the MSCs (positive control). However, CD3 and CD25 were negative, which was expected for MSCs. The CIKs were not labelled and were therefore DiD negative. As for CIKs characteristic, CD3 and CD25 were positive. The second line shows the results of FACS analysis of adherent cells and suspended cells after 4 h of coculture. Adherent cells have switched phenotype from CD3−/CD25− to CD3+/CD25+. DiD signals of adherent cells have disappeared. Suspended cells were still CD3+/CD25+ and DiD−. The third line (after 24 h) confirms the results seen after 4 h. Adherent cells showed the phenotype of the suspended cells and were DiD−.