NK feeder cells of mIL-21 drive strong expansion and metabolic activation

News

The NK feeder cells of mIL-21 drive the strong expansion and metabolic activation of NK cells

NK feeder cells of mIL-21 drive the strong expansion and metabolic activation. NK cell therapy is a promising anti-cancer strategy, but the widespread clinical success of NK cell therapy is limited to a certain extent by the challenge of producing large doses of NK cells, which may be necessary for clinical efficacy [12].

NK cells account for only 10-15% of peripheral blood lymphocytes. However, clinical trials have shown that high-dose NK cells (>109/kg) are both safe and may be necessary for efficacy [13,14,15,16].

Another major advantage of producing large numbers of NK cells is their potential use as a universal donor “off the shelf” therapy. This expansion method supports the use of expanded NK cells from a single donor to treat multiple patients.

Several NK expansion platforms have been reported before, but few clinical-grade expansion platforms can support the large-scale expansion of highly cytotoxic NK cells. For example, NK cells have been expanded with IL-2 and various other cytokine combinations such as IL-12, IL-15, IL-18, and IL-21. These cytokine-based expansion methods result in highly cytotoxic NK cells with similar memory characteristics, but it is reported that due to NK cell senescence, the multiple expansion is limited (approximately 4 times at the 10th day of expansion) [17,18,19 ].

The use of irradiated helper cells as an expansion method of antigen presenting “feeder” cells can increase yield [20,21,22]. For example, using irradiated PBMC and OKT3 to expand NK cells can expand NK cells 2300 times in 17 days [23]. Another system involves Epstein-Barr virus-transformed lymphoblastic feeder cells, which leads to a strong expansion of NK cells 2-4 weeks before senescence [24].

In order to solve the problem of aging, K562 feeder cells are designed to express membrane-bound IL-21 (mbIL-21) with 4-1BB ligand, allowing longer NK cell culture [21,22,25,26,27] .

Other methods to expand NK cells for ACT include the use of immortalized NK cell lines, such as NK-92 cells. One of the main challenges of this approach is that the cells must be irradiated before the patient is administered, which limits the efficacy of this treatment strategy because the cells cannot expand in the patient and maintain anti-tumor activity [28,29].

In this paper, we report the creation of a new NK cell feeder cell line based on mbIL-21, which can support the production of high doses of highly activated NK cells. We recently used this platform to manufacture “universal donor” NK cells for the recently launched Phase 1 clinical trial. In addition, we characterize the mechanism by which mbIL-21 drives the growth and activation of NK cells by activating IL-21-dependent signal transduction leading to metabolic changes that make cells proliferate and kill cancer cells.

The test process and results are as follows:

1. NK cell separation/purification

Peripheral blood mononuclear cells (PBMC) were separated from the peripheral blood of healthy donors by ficoll (GE Healthcare) gradient centrifugation. NK cells are separated from PBMC by magnetic beads CD3 depletion and CD56 separation (Miltenyi biotec). NK cells are cultured with IL-2 (IL-2-NK) for 24 hours, or with designated irradiated NKF cells and IL-2 (NKF-NK). Unless otherwise stated, all studies using NKF to expand NK cells were performed 2 weeks after expansion.
NK feeder cells of mIL-21 drive the strong expansion and metabolic activation

In order to develop a new NK cell expansion platform, our goal is to select suspension cell lines as feeder cells, which are effectively lysed by NK cells and exhibit low levels of HLA I expression. Low levels of HLA class I proteins are beneficial because certain epitopes are recognized by the NK cell inhibitory kill inhibitor receptor (KIR), which impairs NK cell activation.

Initially, the leukemia cell lines HL-60 and OCI-AML3 were chosen because they meet these required characteristics. These two cell lines were irradiated and cultured with freshly isolated NK cells for 1 week to evaluate their expansion potential. OCI-AML3 is a myeloid leukemia cell line, which is 4.6 times larger than the 2.9 times that of HL-60 cells.

It has been reported that the presence of membrane-bound IL-21 (mbIL-21) can prevent NK cells from senescence and significantly improve their ability to expand in vitro [25]. A new type of NK feeder cell was developed using OCI-AML3 cells transduced with mbIL-21 (NKF) (Supplementary Figure 1C).

NK feeder cells of mIL-21 drive the strong expansion and metabolic activation

In order to expand the NK cells isolated from peripheral blood, NK cells (9% of total PBMC (CI 95%: 6.178-15.14) were isolated from PBMC and co-cultured with irradiated NKF cells added weekly (Figure 1A). Flow cytometry evaluated the purity of the expanded NK cells from 15 donors.

After 2 weeks of expansion, CD56+/CD3− cells accounted for about 94% of the expanded cells (CI 95%: 92.32–96.53), and CD3+ cells accounted for about 94% of the expanded cells. <1% (CI95%: 0.069–1.03), B cells are almost undetectable (Figure 1B, C). About 87% of expanded NK cells are CD56+/CD16+, indicating that a large part of these cells can mediate ADCC.

Low T cell contamination after expansion is important to avoid potential graft-versus-host disease of universal donor NK cells.

The NKF expansion platform is optimized based on the ratio of NKF to NK and the concentration of IL-2. The goal of adding feeder cells is to limit the number of feeder cells while producing robust expansion to avoid unnecessary dead feeder cells in the final product.

After 3 weeks, expansion at a ratio of 5-1 (NKF to NK) resulted in NK cell yield 8.3 times higher than that of 1-1 (p = 0.017), and NK cell yield was 1-1 times higher than that of 2-1 ( p = 0.057) (Figure 1D). The amplification yield at the IL-2 concentration range of 10-1000 U/mL did not result in a statistically significant difference (Figure 1E).

This is consistent with the previous report, that is, high IL-2 concentration does not affect the proliferation ability of NK cell feeder cell expansion 31,32,33,34. The yield of 200 U/mL IL-2 is the highest, so the subsequent expansion is 5 -1 ratio of feeder cells to NK cells and 200 U/mL IL-2.

2. Cytotoxicity test

Compared with traditional NK cell-derived IL-2 overnight activated NK (IL-2-NK) cells used for adoptive cell therapy, the cytotoxic activity of NKF-expanded NK cells (NKF-NK) expanded in 2 weeks was evaluated using flow-based Cytotoxicity determination by cytometry. details as follows:

The cytotoxic function of NK cells is evaluated by measuring the number of living cells identified by the calcein-AM (CAM) label. Target cells and NK cells were labeled with CAM (BD Pharmingen) and Calcein (CV) (eBioscience), respectively. NK cells and target cells were co-cultured at the specified ratio for 4 hours in triplicate, and the samples were analyzed in 96-well plates by flow cytometry (Attune NXT, Invitrogen). Gating CV-positive NK cells for analysis. The percentage of cell lysis is calculated as follows:

NK feeder cells of mIL-21 drive the strong expansion and metabolic activation

Compared with IL-2-NK cells, NKF-NK cells showed significantly increased cytotoxic activity against a variety of cancer cell lines (Figure 2A, B and Supplementary Figure 3). The killing rate of NKF-NK cells on Jurkat and TC106 cells was 31% (p = 0.003) and 37% (p = 0.009) higher than IL-2-NK cells, respectively, and the NK target ratio was 1-1. In order to further evaluate the cytotoxic activity of NKF-NK, these cells were co-cultured with leukemia, lymphoma and colon cancer cell lines at various NK target cell ratios. NKF-NK cells showed a dose-dependent killing of all cell types tested (Figure 2C). We further compared the cytotoxic activity of NKF-expanded NK cells and mbIL21-K562-expanded NK cells.

NK feeder cells of mIL-21 drive the strong expansion and metabolic activation

3. Cell phenotype

Use the following antibodies: Biolegend (NKG2D-APC/Cy7, NKp46-FITC, NKp30-PE, CD158-FITC, CXCR6-PE, CD54-FITC), Novus Biologics (c-myc-BB421), BD Biosciences (NKP44-BB515, CD57-BV421, DNAM-1-PE, 2B4-BV421, Ki67-BB786, p-STAT3-Percp-cy5.5) and R&D system (NKG2C-PE, NKG2A-alexa-488). Use “transcription factor phosphate buffer group” (BD Biosciences) for intracellular staining for phenotypic analysis.

Using flow cytometry, baseline expression levels of key NK cell surface molecules were measured in IL-2-NK and 2-week amplified NKF-NK. For direct comparison, the same donor was used for the two NK cell populations (n=4). Compared with IL-2NK cells, the expansion of NKF cells resulted in increased expression of the activated receptors NKG2D, NKp30 and NKp44 (Figure 3A and Supplementary Figure 4). Activated receptors such as NKG2D and natural cytotoxic receptors are essential for the activation and function of NK cells [35].

NK feeder cells of mIL-21 drive the strong expansion and metabolic activation

The expression of the killing immunoglobulin receptor (KIR) of NKF-NK cells was also significantly reduced (Figure 3B). The expansion of NKF cells also leads to an increase in adhesion receptors (LFA-1 and CD54), which are important for the binding of NK cells to tumor targets, and enhance cell lysis (Figure 3C and Supplementary Figure 4) CD57 expression, which indicates the expression of NK Terminally differentiated cells were also measured and found to be reduced in NKF-NK cells (Figure 3D).

The ability of NK cells that express high levels of CD57 to continue to proliferate in the body is limited. Finally, the expression of key chemokine receptors that regulate the transport of NK cells in vivo is analyzed. The expansion of NKF leads to a decrease in CXCR4, a receptor that is reported to isolate NK cells in the bone marrow (Figure 3E). In addition, the increase in CXCR6 indicates that NKF-NK cells can metastasize to the liver, which is a common site of cancer metastasis. CXCR6 is also indicated in the development of memory-like NK cells that persist after hapten or virus exposure.

4. mbIL-21 signal promotes continuous NK cell expansion and increased metabolic activity

Although short-term co-culture with NKF cells can lead to NK cell proliferation and activation, continuous proliferation is necessary to support clinical research on “universal donor” NK cells. Unlike the rapidly aging IL-2 treated NK cells, NKF cells can achieve long-term NK cell proliferation. For example, after 5 weeks of expansion, the average expansion is 10,973 times (Supplementary Figure 3A).

In order to support the clinical transformation of NKF-expanded NK cells, a high-capacity culture device is also needed to adapt to the high cell number. We tested the ability of freshly isolated NK cells to expand in the G-REX device. NK cells were expanded with NKF cells for 2 weeks, and the average NK cell yield was 89 times (Supplementary Figure 3B). This shows that the semi-automated G-REX system can use NKF feeder cells to expand NK cells.

The proliferation of NK cells expanded with NKF cells was compared with the proliferation of parental OCI-AML3 cells (OCI-NK). Compared with 200 times the use of OCI-AML3 cells, NKF cells resulted in an average 843-fold expansion of NK cells after 3 weeks (p = 0.027) (Figure 4A). Consistent with the enhanced proliferation of NK cells, Ki67, a proliferation marker on NKF-NK cells, was up-regulated compared with OCI-NK cells (p = 0.038). As expected, Ki67 levels were higher when using either feeder cell line compared to IL-2-NK cells (p = 0.0021) (Figure 4A).

Next, we will evaluate the activating molecules downstream of IL-21 receptor (IL-21R) signaling to clarify the mechanism of IL-21-dependent continuous NK cell proliferation. It is known that IL-21R signal can activate STAT3, and STAT3 can directly induce c-myc expression. NKF-NK cells showed higher expression of p-Stat3 than OCI-NK cells (p=0.041) and IL-2-NK cells (p=0.00038) (Figure 4B). In addition, compared with OCI-NK cells (p = 0.043) and IL-2-NK cells (p = 0.00085), NKF-NK cells showed higher expression of c-myc (Figure 4B).

Since the activity and proliferation status of immune cells depend to a large extent on their metabolic capacity, the effect of mbIL-21 signal transduction on NK cell metabolism was studied. After activation, the metabolism of immune cells usually shifts from producing energy mainly through oxidative phosphorylation to aerobic glycolysis. The oxygen consumption rate (OCR) to measure the rate of oxidative phosphorylation (oxphos) and the extracellular acidification rate (ECAR) to measure glycolysis were measured in NKF-NK, OCI-NK and IL-2-NK cells (Figure 4C) .

At baseline, NKF-NK cells and OCI-NK cells had the same oxphos and glycolysis rate (Figure 4D, E). Under pressure, NKF-NK cells have higher OCR (p = 0.043) and ECAR (p = 0.0015) values than OCI-NK cells (Figure 4D, E). Both expanded NK cell populations have higher OCR and ECAR values than IL-2-NK (Figure 4D, E). The OCR/ECAR ratio of NKF-NK cells is close to 1, indicating a balance between glycolysis and oxphos pathways (Figure 4F). The energy state of NKF-NK is higher than that of OCI-NK and IL-2-NK, as observed in the graphs of OCR and ECAR (Figure 4G).

Since mbIL-21 leads to enhanced NK cell proliferation and metabolism during feeder cell expansion, we also evaluated its effect on the cytotoxic activity of NK cells. Although NKF-NK cells showed a significant increase in cytotoxic activity compared with non-feeder expanded cells (Figure 2A, B), there was no significant difference in the cytotoxic function of NKF-NK and OCI-NK cells (Figure 2A, B). 4H). The results indicate that, at least in the presence of feeder cells, mbIL-21 may not significantly affect the cytotoxicity of NK cells.

5. NKF-NK cells reduce tumor burden in mouse tumor xenografts and improve survival rate

Nod-SCID-IL-2Rgamma-null mice (NSG, Jackson Laboratory) were injected bilaterally with 1 × 105 TC106 (sarcoma) cells. Ten days after TC106 injection, when the tumor was palpable in time, mice (n = 5 per group) received weekly intravenous (IV) injection of 1 × 106 NK cells or vectors and IL-2 (75,000 U/ml). The tumor volume was measured twice a week. The mice were sacrificed after losing 15% of their body weight.

A xenotransplantation model of leukemia was established by IV injecting 1×106 Jurkat cells into NSG mice. Seven days after the injection, the mice (n=5 per group) received 5×106 NK cells or vectors and IL-2 (75,000 U/ml) per week. According to institutional guidelines, the mice were sacrificed after they were dying or after losing 15% of their initial body weight. The tissues of the mice were harvested for immunostaining.

To evaluate the therapeutic potential of NKF-NK cells for cancer treatment, mouse models of sarcoma and lymphoid leukemia were used. For the sarcoma model, Ewing’s sarcoma cell line TC106 was injected subcutaneously into immunodeficient NSG mice. The sarcoma model is used because it can cause lung metastasis, which is the most common metastatic site of human sarcoma and the known site of NK cell transport in the body [45,46]. In this model, we not only observed a reduction in the growth of primary sarcoma after administration of NKF-NK cells, but also a significant reduction in tumor metastasis to the lung (Figure 5A-C). Ki67 staining of lung specimens from vehicle-treated mice and NKF-NK-treated mice showed that the proliferation of tumor cells in NKF-NK-treated mice was reduced compared with vehicle-treated mice (p = 0.023) ( Figure 5D, E).

In addition to evaluating the ability of NKF-expanded NK cells to reduce tumor growth, the ability of these cells to prolong survival in a highly aggressive circulating lymphocytic leukemia model was also evaluated. NSG mice were injected intravenously with Jurkat cells to establish circulatory diseases. Compared with control-treated mice, the median survival rate of mice injected with NKF-NK cells increased by approximately 13 days (p = 0.0016) (Figure 5F).

Discussion:

Adoptive NK cell therapy has shown promise in the field of cancer treatment, but the development of other robust methods to expand a large number of highly activated NK cells is important to continue advancing the field. In addition, a better understanding of the mechanisms for achieving this extension is important for formulating the best strategy.

Here, we report the development of a new feeder line, NKF, which can support clinical trials of “universal donor” NK cell therapy and clarify how mbIL-21 affects NK cell expansion and activation. Due to the limited availability of powerful in vitro amplification platforms that can be used in the clinic, it is important to develop new feeders. Compared with K562-mbIL21 feeder cells, NKF cells behave similarly in terms of NK cell expansion and cytotoxicity to cancer cells. Since K562-mbIL21 feeder cells are no longer available for a wide range of clinical applications, the NKF platform may provide a valuable alternative.

In order to use NKF cells to make clinical-grade NK cells, a master cell bank was created to support recently initiated clinical trials to test the universal donor NKF in cancer patients to expand NK cells (NCT02890758). Based on an average expansion of more than 10,000 times at 5 weeks, it should be feasible to produce more than 4 × 1012 NK cells from a single donor sample. Methods to characterize the feasibility of feeder cell-based expansion in large-capacity bioreactors such as Xuri are in progress.

Since NK cells are different from T cells, they are not considered to cause graft-versus-host disease. Therefore, there are many advantages to developing “universal donor” ready-made NK cells. Logically, this strategy will significantly reduce costs and increase the global accessibility of this treatment strategy. Although the method of in vitro expansion is promising, the donor needs of each prospective patient and the associated costs (logistics and finances) of cell processing for each patient still limit the feasibility of adoptive NK cell therapy. A powerful expansion system, such as NKF feeder cells, should be able to produce NK cell doses from a single donor for 100 or more recipients. Being able to harvest NK cells from a donor will greatly reduce the cost of this treatment. Utilizing donors and recipients that do not match HLA also increases the potential of NK cell allogeneic reactions to promote the “graft anti-leukemia (GVL)” effect (without GVHD). This has been found to improve disease-free survival in certain cancer patient groups 47,48,49.

In order to improve current NK cell therapy, preclinical studies have shown that the combination of NK cells and immunomodulators (such as TGFβ inhibitors) offers hope. For example, the combination of expanded NK cells and Galunisertib resulted in a significant increase in the anti-tumor efficacy of a mouse model of colon cancer metastasis [51]. In addition, the immunomodulator lenalidomide can enhance the cytotoxicity of NK cells to multiple myeloma cells [52].

Using unexpanded NK cells and NKF-expanded NK cells or parental feeder cells lacking mbIL-21, we further evaluated the mechanism by which mbIL-21 maintains NK cell expansion. Our research shows that mbIL-21 activates the well-characterized IL-21-dependent pathway composed of STAT3 and cMyc. STAT3 activation is necessary for the downstream effects of IL-21 signaling and is a known inducer of c-Myc. It is known that the activation of cMyc can regulate various cellular processes, which are important for the proliferation and activity of NK cells, including induction of glycolysis, mitochondrial biogenesis, and cell cycle [53-55]. Interestingly, NK expansion of parent OCI-AML3 feeder cells lacking mbIL-21 resulted in partial activation of the STAT3/cMyc pathway, which may be because activation of many other receptor signaling pathways can also lead to minimal STAT3 activation.

In general, NK adoptive cell therapy is a promising treatment method that relies on the development of a powerful in vitro expansion platform, such as NKF cells that can support the production of highly active clinical-grade NK cells.