- Top of page
- PATIENTS AND METHODS
Despite recent advances in the early diagnosis of prostate cancer and treatment of localized disease, there is no effective treatment for advanced prostate cancer, especially if it is hormone-refractory . Therefore, several new treatment methods are being investigated, amongst which cell therapy using immune cells with antitumour activity is a promising candidate.
Generally, both cytotoxic T lymphocytes (CTLs) and natural killer (NK) cells are considered to be important in tumour immunity. CTLs are known to produce specific cytotoxic activity against tumour cells by recognizing tumour-associated antigens in a MHC-restricted manner . However, frequent loss of MHC class I and II expression has been reported in prostate cancer cells and tissues [3–6]. The loss of MHC molecules allows prostate cancer to evade the cellular immune system. In contrast, NK cells can kill tumour cells without restrictions from loss of MHC . Moreover, NK cells have strong killing activity against MHC-negative target cells. Thus, if the expansion of NK cells with potent antitumour activity is possible, NK cells could be promising effector cells in cell therapy for prostate cancer.
Lymphokine-activated killer (LAK) cells also have non-MHC restricted cytotoxicity. Generally LAK cells represent a composite of CD3– NK cells and CD3+ T cells, and have the capacity to kill a variety of tumour cells and MHC class I-negative target cells. However, most importantly, activated NK cells but not T cells have a major role in LAK cell activity [8,9].
Based on this background, we investigated a selective and efficient culture method to induce NK cells. To establish a method for selective expansion of NK cells in vitro, we screened 11 cell lines that scarcely expressed surface MHC class I molecules as candidate stimulators of NK cells. Among them, we found that HFWT , a cell line originating from a Wilms’ tumour, can efficiently stimulate the selective expansion of human NK cells from peripheral blood mononuclear cells (PBMCs) of patients with brain tumour and in healthy volunteers . However, both cellular and humoral immunity are severely suppressed in patients with prostate cancer, especially those with bone metastases . In the present study we examined whether activated antitumour NK cells can be induced from PBMCs from these patients.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
The characteristics of the 10 patients with histologically confirmed prostate cancer (aged 52–80 years) in this study are summarized in Table 1. PBMCs were obtained from these patients; four had stage D2 disease with bone metastases and hormone-refractory status, and the other six had no metastatic disease (one stage B and five stage C). At the time of NK cell induction, the four patients with metastatic disease were treated with dexamethasone or a combination of estramustine phosphate and etoposide. The six patients with stage B or C disease were treated with a combination of LHRH analogue and antiandrogen. Written informed consent for blood sampling and cell culture was obtained from all patients.
Table 1. The patients’ characteristics
| 1||68||D2||poor|| 97||dexamethasone|
| 2||52||D2||poor|| 68||EMP/VP-16*|
| 4||79||D2||poor|| 965||dexamethasone|
| 5||75||B||poor|| 1.4||MAB†|
| 6||79||C||well|| 0.1||MAB|
| 7||73||C||moderate|| 0.1||MAB|
| 8||72||C||poor|| 0.3||MAB|
| 9||74||C||poor|| 0.5||MAB|
All the cell lines used in the present study were taken from routine cultures at the RIKEN Cell Bank. Cell lines were maintained in a basal medium, RPMI 1640 or Ham's F12, containing 10% or 15% fetal bovine serum. HFWT cells, an anchorage-dependent Wilms’ tumour cell line, were used as feeder cells for the selective expansion of NK cells. Two established prostate cancer cell lines, PC-3 and LNCaP, were used as targets for cytotoxic assay.
For the expansion of NK cells, PBMCs were prepared from 20 mL of heparinized peripheral blood with a conventional preparation kit (Lymphoprep, Nycomed Pharma AS, Oslo, Norway). The cells were washed twice with calcium- and magnesium-free Dulbecco's PBS. One day before the start of NK cell induction, 1 × 105 HFWT cells were seeded into each well of 24-well tissue culture plates. After overnight incubation these cells were irradiated with 50 Gy. The PBMCs (1 × 106 cells/mL, 1 mL/well) were then seeded on HFWT cells at a responder : stimulator ratio of 10 : 1. RHAM α medium  supplemented with 5% autologous plasma and interleukin-2 (200 U/mL) was used for culture of the lymphocytes. The NK cell culture was continued with appropriate changes of the medium including interleukin-2 (at least half of the medium was changed every 2 days) until HFWT cells completely disappeared. After this, the cell suspension was diluted to 5 × 105/mL and the culture continued. After 2 weeks in culture the number of lymphocytes was counted and the phenotypes of the lymphocytes analysed by flow cytometry. As a control, PBMCs were cultured with interleukin-2 but with no HFWT cells at the same time. Expansion of NK cells from PBMCs of four normal healthy subjects was also investigated.
For flow cytometry, lymphocytes were stained with monoclonal antibodies (mAbs), i.e. fluorescein isothiocyanate (FITC)-labelled anti-CD3 (UCHT1, IgG1) and PE-labelled anti-CD56 (MOC-1, IgG1, DAKO Japan, Kyoto), and FITC-labelled anti-CD16 (3G8, IgG1) and PE-labelled anti-CD56 (V NK75, IgG1, BD PharMingen, San Diego, CA). Cell-surface MHC-class I expression of PC-3, LNCaP and HFWT cells was analysed using PE-labelled anti-MHC-class I (W6/32, IgG2a, DAKO Japan). Isotype-matched control mAbs were used as negative controls. Cells were stained with these mAbs for 30 min at 4 °C. After washing, the cells were immediately analysed by a FACScan flow cytometer (Becton Dickinson, San Jose, CA).
For the cytotoxicity assay, the non-radioisotopic crystal violet staining assay was used to measure cytotoxic activity of lymphocytes against anchorage-dependent target tumour cells, as described previously [14,15]. This assay is compatible with the standard 51Cr-release assay at an effector : target (E : T) ratio of ≤ 10 . Briefly, 1 × 104 target cells in 100 µL culture medium were seeded in each well of 96-well plates and pre-cultured overnight. After washing culture plates with PBS, the cultured lymphocytes suspended in 100 µL of medium were added as effector cells to each well at the indicated E : T ratio. The cells were co-cultured for 24 h and then washed once gently with appropriate amounts of PBS. Adherent target cells were fixed for 30 min with 10% (v/v) formalin (100 µL/well) and then stained with crystal violet solution (0.4% in water, 80 µL/well) for 30 min at room temperature. The plate was washed with tap water and dried at room temperature. To each well, 200 µL of 70% methanol was added and the optical density at 570 nm of each well determined. The percentage of surviving target cells was expressed as (B − C)/A × 100, where A is the absorbance of control target cells pre-cultured on a separate plate just before adding the effector cells, B is the absorbance of target cells remaining after effector cells were added, and C is that of effector cells only. Each value represents the mean of triplicates. Cytotoxicity was also quantified in triplicate by the standard 4-h 51Cr-release assay described elsewhere .
Data are expressed as the mean (sd); a paired Student's t-test was used to compare two culture conditions, analysing the mean cell growth, phenotypes and cytotoxicity of lymphocytes cultured with or with no HFWT cells.
- Top of page
- PATIENTS AND METHODS
The expression of MHC class I antigens by the three cell lines is shown in Fig. 1. As reported by Bander et al. PC-3 cells strongly expressed MHC class I antigens, whereas LNCaP cells expressed relatively low levels. In contrast, HFWT cells scarcely expressed MHC class I antigens.
Figure 1. A FACS histogram of cell-surface MHC-class I expression by a, PC-3, b, LNCaP and c, HFWT cells (filled peaks). Open peaks represent control staining with isotype-matched control antibody.
Download figure to PowerPoint
When co-cultured with irradiated HFWT cells the lymphocytes from unfractionated PBMCs grew 20- to 130- fold (mean 53). In contrast, lymphocytes cultured with no stimulation by HFWT cells showed 4- to 70-fold (mean 34) expansion, as shown in Table 2. The proportion of CD16+CD56+ cells was significantly higher in the expanded lymphocyte population co-cultured with HFWT cells than in the control culture (Table 2). Lymphocytes other than CD16+CD56+ cells consisted mainly of CD3+CD56– cells, which correspond to the phenotype of LAK cells or nonspecific T cells. Representative results of phenotypes from the patient with stage D2 disease (patient 3) are shown in Fig. 2. There was no significant difference in growth and phenotypes of lymphocytes between patients with stage D2 and B or C. When compared to normal healthy subjects, there was no significant difference in the growth of lymphocytes. However, the proportion of CD16+CD56+ cells of the expanded lymphocyte population co-cultured with HFWT cells was higher in the patients (Table 2).
Table 2. Growth and phenotypes of lymphocytes, comparing the expansion of lymphocytes co-cultured with irradiated HFWT cells supplemented with interleukin-2 for 2 weeks, and those cultured with interleukin-2 but with no HFWT cells, cytotoxicity against the prostate cancer cell lines and HFWT cells (E : T ratio of 4)
|Cell type||Patient no.||Mean (sd) |
|Lymphocyte growth, -fold|| || || || || || || || || || || ||44.0 (15.0)|
|HFWT + interleukin-2||30||29|| 25||70|| 19||89|| 60||30||127|| 53||53.2 (34.5)|| |
|Interleukin-2||ND||ND|| 4||31|| 44||20|| 33||72|| 58|| 8||33.8 (23.6)|| |
|CD16+ CD56+, %|| || || || || || || || || || || ||70.3 (9.1)*|
|HFWT + interleukin-2||85.7||89.8|| 90.1||86.4|| 70.2||91.6|| 96.1||82.2|| 93.8|| 77.1||86.3 (7.9)|| |
|Interleukin-2||ND||ND|| 22.0||41.7|| 44.9||50.8|| 82.2||14.9|| 91.5|| 32.6||47.6 (27.1)†|| |
|CD3+ CD56–, %|| || || || || || || || || || || ||14.4 (10.5)*|
|HFWT + interleukin-2||10.4|| 0.9|| 2.6|| 1.6|| 3.1|| 4.8|| 1.2||10.3|| 0.4|| 14.7|| 5.0 (5.0)|| |
|Interleukin-2||ND||ND|| 58.0||44.0|| 14.5||46.6|| 12.08||79.89|| 1.8|| 51.3||38.5 (26.6)†|| |
|Cytotoxic activity against prostate cancer cell lines and HFWT cells|
|E : T ratio 4|
|PC-3|| || || || || || || || || || || ||16.7 (16.7)*|
|HFWT + interleukin-2||75.5||39.3|| 21.1||17.6||101.8||65.3|| 61.7||28.5|| 63.0|| 60.0||53.4 (26.5)|| |
|Interleukin-2||ND||ND||117.6||97.0|| 90.5||83.0||100.7||95.0|| 49.5||110.4||93.0 (20.7)*|| |
|LNCaP|| || || || || || || || || || || || 7.0 (12.1)|
|HFWT + interleukin-2|| 0|| 0|| 3.5|| 1.5|| 0|| 0|| 20.9|| 3.0|| 33.4|| 27.8|| 9.0 (13.1)|| |
|Interleukin-2||ND||ND|| 11.0||44.1|| 0|| 0|| 52.1||48.9|| 22.7|| 66.5||30.7 (25.6)*|| |
|HFWT|| || || || || || || || || || || || 0.3 (0.5)|
|HFWT + interleukin-2|| 0|| 0|| 6.4|| 3.6|| 0|| 0|| 15.3|| 0|| 15.2|| 9.8|| 5.0 (6.3)|| |
|Interleukin-2||ND||ND|| 17.44||64.1|| 0|| 0|| 28.4||52|| 10.8||6 0.8||29.2 (26.5)*|| |
|E : T ratio 8|
|PC-3|| || || || || || || || || || || || 4.0 (3.4)|
|HFWT + interleukin-2||45.8|| 2.8|| 1.7|| 3.7|| 69.8||37.8|| 20.2|| 5.9|| 4.7|| 31.9||22.4 (23.3)|| |
|Interleukin-2||ND||ND|| 98.5||83.3|| 73.2||60.6|| 80.9||85.1|| 34.3||105.4||77.7 (22.3)†|| |
|LNCaP|| || || || || || || || || || || || 9.7 (16.7)|
|HFWT + interleukin-2|| 0|| 0|| 8.6|| 2.6|| 0|| 0|| 15.3|| 5.6|| 0|| 12.4|| 4.5 (5.8)|| |
|Interleukin-2||ND||ND|| 0||10.0|| 0|| 0|| 12.1||29.0|| 0|| 43.5|| 11.8 (16.3)|| |
|HFWT|| || || || || || || || || || || || 5.2 (10.4)|
|HFWT + interleukin-2|| 0|| 0|| 9.3|| 3.7|| 0|| 0|| 4.0|| 3.7|| 0|| 0|| 2.1 (3.1)|| |
|Interleukin-2||ND||ND|| 0||21.3|| 0|| 0|| 17.6||21.4|| 0|| 37.1||12.2 (14.2)|| |
Figure 2. Representative results of phenotypes (a,b HFWT + interleukin-2; c,d interleukin-2 only) from the patient with stage D2 disease (no. 3) determined by flow cytometry.
Download figure to PowerPoint
The activity of the induced lymphocytes was tested initially against K562 cells, which is the standard target of NK cells, by the 4-h 51Cr-release assay. Lymphocytes co-cultured with HFWT cells had stronger NK activity than those in the control culture (Fig. 3A). Figure 3B shows cytotoxicity of the induced lymphocytes against HFWT cells by the 51Cr-release assay, which was compatible with NK activity against K562 cells. Cytotoxicity against HFWT cells by the crystal violet staining assay was well correlated with NK activity against K562 cells (Fig. 3C). These NK cells also showed strong cytotoxic activity against PC-3 and LNCaP, and HFWT cells, at E : T ratios of 4 and 8 (Table 2). Representative results of cytotoxicity from the patient with stage D2 disease (no. 4) and stage C disease (no. 8) are shown in Fig. 4. NK cells expanded on the irradiated HFWT cells almost completely killed LNCaP and HFWT cells at an E : T ratio of 4 in 24 h. Also, the NK cells killed 80% of PC-3 cells at an E : T ratio of 4 (Fig. 4A). Similarly, lymphocytes cultured with no HFWT cells had killing activity against LNCaP and HFWT cells, but did not efficiently kill PC-3 cells (Fig. 4B). The cytotoxicity of lymphocytes was similar in patients with no metastasis (Fig. 4C,D). The mean of these NK activities against PC-3, LNCaP and HFWT cells was significantly greater than the mean cytotoxicity of the lymphocytes in the control culture at an E : T ratio of 4. The cytotoxicity was also significantly high against PC-3 at an E : T ratio of 8. There was no difference in the induced NK activity between stage D2 and B or C disease. When compared with normal healthy subjects, the activities of NK cells from the patients was not significantly different, except for the cytotoxicity against PC-3 at an E : T ratio of 4 (P = 0.048; Table 2).
Figure 3. Comparison of NK activities of the lymphocytes (HFWT + interleukin-2, green circles; interleukin-2 only, red squares) from a patient with stage C prostate cancer, assayed by two methods. Cytotoxic activities against: a, K562 cells and b, HFWT cells by the standard 4-h 51Cr-release assay; c, against HFWT cells by the 24-h crystal violet staining assay. Each value represents the mean of triplicates.
Download figure to PowerPoint
Figure 4. Cytotoxic activities of the lymphocytes from patients with stage D2 (no. 4, a and b) and c (no. 8, c and d) disease against PC-3 (green open circles), LNCaP (red open squares) and HFWT (light red closed squares) cells assayed by the 24-h crystal violet staining assay. a,c, lymphocytes co-cultured with and b,d with no HFWT cells.
Download figure to PowerPoint
- Top of page
- PATIENTS AND METHODS
Recent progress in human tumour immunology, especially based on the molecular identification of tumour antigens, has developed effective and active cell-transfer immunotherapies . Also in prostate cancer, new treatment approaches aim to eradicate cancer cells by inducing systemic immunity to tumour-associated antigens [17–20]. However, several tumour factors that allow cancer cells to escape from the immune system remain to be overcome . Although dendritic cells pulsed with these antigens are capable of stimulating potent cytotoxic T lymphocytes, most of these antigens are also known to be expressed in normal prostate tissues. Thus overcoming immune tolerance to proteins expressed in normal prostate is a major obstacle to developing rational strategies in prostate cancer immunotherapy. Loss of MHC expression can permit cancer cells to evade T cell-mediated immune recognition. In contrast, NK cells have strong killing activity against MHC class I-negative target cells. Based on this background, therapy using NK cells, not based on T cells, was investigated as a promising alternative therapy for prostate cancer .
In patients with prostate cancer the activity of NK cells from PBMCs is reduced more in patients with metastatic disease than in those without or in healthy controls [23,24]. Kastelan et al. reported that this reduced NK activity level might correlate with the presence of tumour cells in the circulation. Therefore, an efficient method to promote the selective expansion of potent NK cells is needed to overcome the immunosuppressive state.
To establish a method for the selective expansion of NK cells in vitro, we screened 11 cell lines that scarcely expressed surface MHC class I molecules as candidate stimulators of NK cells; the HFWT cells can efficiently stimulate the expansion of NK cells or NK precursors in PBMCs . In the present study in patients with prostate cancer, not only with localized disease but also with advanced hormone-refractory metastatic disease, activated NK cells can be selectively induced by co-culture with HFWT cells. We also compared the inducibility of activated NK cells between patients and healthy volunteers. There were no significant differences between them in cell growth and cytotoxicity against LNCaP and HFWT (Table 2). Although cytotoxicity against PC-3 was higher in healthy volunteers, the differences were not significant when the same assay was at an E : T ratio of 8 (Table 2). Most importantly, this method allows efficient NK cell expansion even from PBMCs of patients treated with steroidal agents and anticancer drugs. These expanded NK cells showed strong cytotoxicity against MHC class I-positive PC-3 cells and LNCaP or HFWT cells, which weakly or scarcely express it. We are also planning to evaluate the NK activity against DU145, which is another MHC class I-positive prostate cancer cell line , and other prostate cancer cell lines. Preliminary data were obtained from two patients with stage C disease; the killing activities of NK cells induced from PBMCs of patients 9 and 10 was 37% and 75% at an E : T ratio of 8, respectively. In contrast, LAK cells expanded by interleukin-2 did not kill PC-3 cells. Generally LAK cells represent a mixture of NK cells and non-MHC-restricted CTLs that have the capacity to kill a variety of tumour cells and MHC class I-negative target cells. Falk et al. reported that the cytotoxicity of LAK cells was negatively regulated through interactions with MHC class I molecules in a manner similar to that seen with NK cells, and was independent of the individual human leukocyte antigen background of the tumour cells in an allogeneic system. The present data support the greater susceptibility of LNCaP and HFWT cells than PC-3 cells to LAK cells. Moreover, NK cells expanded by stimulation with HFWT cells can also efficiently kill LAK-resistant PC-3 cells. Although the molecular mechanisms of NK cell-mediated cytotoxicity have not been clarified yet in allogeneic systems, we speculate that the expanded NK cells are strongly activated, sufficient to overcome the inhibitory signal transduced through interactions with MHC class I molecules and killer inhibitory receptors. Further studies are required to clarify the mechanisms of HFWT cell-dependent NK cell activation and NK cell-mediated cytotoxicity in both autologous and allogeneic systems.
In conclusion, strongly activated NK cells were induced in all 10 patients with prostate cancer, with or without metastatic disease. These data suggest that the efficient and selective expansion of autologous NK cells is feasible even in such patients. Further investigation is needed to determine whether this approach could be a candidate for cell therapy for advanced prostate cancer.