Efficient Expansion of Human Umbilical Cord Blood-Derived NK Cells Ex Vivo without Requiring Feeder Layers

Introduction: Natural killer (NK) cells are central components of the innate immunity. They have ability to kill a wide range of cancer cells and are a promising tool for both autologous and allogeneic immune enhancement therapy in cancer treatment. Actually, NK cells can be derived from multiple sources such as: peripheral blood, cord blood... Among these, cord blood (CB) is known as an ideal for NK cells expansion because it occupies a higher percentage of NK cells compare to the peripheral blood and is a rich source of hematopoietic stem cell as well as progenitor immune cells. The objectives of this study were to invent a reasonable approach for expanding a relevant number of NK cells from human cord blood to clinical application. Methods: At the initial step, cord blood was collected from the patients and then the mononuclear cells were obtained by density centrifugation with Ficoll. After that, these cells were cultured in 2 stages: stimulation and expansion with inactivated autologous plasma and BINKIT expansion kit contains several kinds of growth factors which are specified for NK cells. The immunophenotype of NK cells was analyzed every 2 days through %CD56+CD3by Flow cytometry technique. The ex-vivo activation and expansion of NK cells was performed in GMP grade -clean room for about 3 weeks. Results: After culturing periods, the average NK cell count post-expansion was 2040.6 ± 1463.5 x10^6 and the average cell fold expansion was 814.4 ± 560.9, the expanded cells presented 90.6% ± 8.9% purify of CD56+CD3-, the percentage of CD56 bright CD3cells was 84.2% ± 14.4%. Conclusion: Human cord blood has a significant potential to expand NK cells to relevant number for clinical use and our method are suitable for getting a relevant number and quality of NK cells to the clinical use.

The function of NK cells is regulated through the interaction between activating receptors and inhibitory receptors [4]. The lack of combination between inhibitory KIR receptor on NK cells surface and MHC-I on target cell surface lead to the stimulation of activating receptors, and then helps NK cells to kill target cells.
Therefore, the KIR mismatch assists NK cells in expressing more killing activity [8]. This is also the reason why adoptive cell transfer (ACT) approaches using allogeneic NK cells have been more effective for cancer immunotherapy [9]. In order to conduct the clinical application about NK cell-based immunotherapy, it is needs to obtain a sufficient number of NK cells with high cytotoxicity.
The sources of NK cells include umbilical cord blood (UCB), bone marrow, peripheral blood (PB) and embryonic stem cells. Actually, UCB not only has higher proportion of NK cells but is also an easy collecting source.
Furthermore, the establishment of UCB bank in the world now assists UCB being preserved for a long time. This leads to UCB can be used as a ready source for NK cell-based therapy. Nevertheless, although UCB contains greater frequencies of NK cells as opposed to PB, the numbers of obtained NK cells are still really small because of the limited UCB volume. This is a major hindrance in providing adequate numbers of NK cells for clinical trial [10]. The main objectives of this study were to optimize the procedure of expansion of NK cells from human cord blood in order to obtain a larger number of NK cells and higher purify of CD56+CD3-in population. The next stage is activation of these NK cells to increase the capacity of killing cancer cells effectively.

Collection of Cord Blood Samples
Human cord blood which is collected directly from the umbilical cord of the new born baby at Vinmec International Hospital. Before collection, the mother was diagnosed healthy and does not carry any of the following viruses: HIV, HBV, HPV, HCV. All the pregnant woman singed the written consent which is approved by the Ethics Committee of Vinmec International Hospital.

Isolation and Inactivation of Autologous Plasma
The whole cord blood was firstly centrifuged at 1700×g for 10 minutes. After centrifugation, whole blood is separated into 2 layers: plasma layer above and blood cell layer below. Then, the plasma layer was collected by pipetting. This plasma was then heat inactivated at 58 o C for 60 minutes. After inactivating, plasma continued to be centrifuged at 1700×g for 5 minutes to collect the supernatant plasma and discard the pellet at the bottom of the tube.

Isolating Mononuclear Cells
Cord blood after plasma collection is diluted with PBS, followed by dropping down slightly to Ficoll-Paque/Lymphoprep (Stem Cell Technologies, CA) with volume proportion 2:1, and then the mixture was density centrifuged at 840×g for 20 minutes. The Ficoll solution plays a role as a tool to separate the whole blood into many layers follow the density gradient. After centrifugation, the mixture was divided into 4 layers, from top to bottom: plasma, a layer of mononuclear cells (MNC) called buffy coat, Ficoll, and erythrocytes. The MNC were collected and then washed with PBS 2 times.

Stimulation Culture
The stimulation culture process was implemented within 3 days (day 0 to day 3). UCB-MNCs were cultured in initial NK medium and initial NK cocktail (Biotherapy Institute Japan) containing IL-2, OK432, zoledronic acid and 10% heat-inactivated autologous plasma in an initial NK flask (Biotherapy Institute Japan) for initial activation. This flask was immobilized with anti-CD3 monoclonal antibody and anti-CD16 monoclonal antibody. The cell density at seeding was 1 × 106 cells/ml.

Expansion Culture
After 3 days of stimulating cultivation, the cells need to be cultured in new condition which is free from the anti-CD3 antibody, anti-CD16 antibody, OK432, and zoledronic acid or the like. These supplemented with IL-2 and 10% heat-inactivated human plasma.
Subculture medium was added depending to the cell number every 2 days. The cells were incubated at 37 °C with 5% CO 2 .
The viability of cells at the day of ending culture was 98.6% ± 0.6%.

Immunophenotyping Analysis by Flow Cytometry
In average, the expanded NK cells (CD56+CD3-) presented 90.6% ± 8.9% purify of immune cell population (ranging from 80.7% to 96.8%). After 21 days of expansion culture, the percentage of NK cell increased by 17.8 ± 9.3 fold. There was a significant change in markers expression during culturing process ( Figure   2A). The percentages of CD56dim CD3-cells at day 0 was 7.4 ± 3 %.
At the initial day, the CD56-CD3+ population was dominant, until day 3 the CD56+CD3-population started rising sharply and reached >43% at day 5. From day 3 to day 10, the rate of NK cells increased dramatically and rapidly. From day 10, the rate of NK cell continued to slowly increase and peaked at 88.5 ± 11.5 %.
There were a gradually rise in the percentage of CD56bright

Discussion
The NK expansion kit BINKIT which we used for culturing includes initial flask, initial medium and initial cocktail. The initial flask is coated with anti-CD3 antibody and anti-CD16 antibody.
These antibodies may bind to the surface receptors of cells other than NK cells (CD3, CD16) and then stimulating these cells to release many liquid growth factors which are needed for NK expansion. IL-2 is known as the most optimal for NK proliferation. Actually, IL-2 enhanced at least 10-fold more NK cell expansion compared to IL-4, IL-7, or IL-12 [11]. The initial cocktail contains IL-2 (Interleukin -2) which joins in signal pathways to proliferate NK cells and form the NK function. Concretely, these factors enhance the cytotoxicity of NK cells by stimulating the secretion of IFN-γ which is needed for killing other cells (cancer cells or infected cells) [12]. The initial medium contains OK-432, this serve as an immune adjuvant capable of activating, for example, monocytes, through the binding to the surface TLR of the monocytes so that immune response is activated [13]. were cultured in 2 ways. In the first way, they used the culture bag for CD34+ expansion and combine with medium containing SCF, IL-7, IL-15 and IL-2 within 6 weeks. As the result, the mean total cell expansion was 1300 fold, the NK cells product was 900-1900 x 10^6 and %CD56+CD3-was 71%±9% [16]. Their process prolonged 6 weeks as opposed to our 3 week-process, which can lead to the cell exhausted as we observed at the end of the culture (3 weeks) the dominant of crooked cells. In the second way, they used a bioreactor system for cell culturing for 6 weeks and obtained much better results: the mean total cell expansion was 2100 fold, high pure NK product with 92%±2% and total NK cells were 1600 -3700 x 10^6 [16].

As in Shah and his colleagues' report which was published in
October 2013, the proportion of CD56+CD3-obtained was > 95%

Conclusion
Human cord blood has a significant potential to expand NK cells to relevant number for clinical use. We were success to expand NK cells from Vietnamese human cord-blood with significantly pure (more than 90%), high viability (more than 98%) and the number of NK cell were relevant to clinical application (2040.6 × 10^6, ranging from 871.7 to 3681.9 × 10^6). These NK cell products are suitable for clinical trials in order to apply in cancer treatment.

Competing Interests
The author(s) declare that they have no competing interests