- Email: [email protected]
Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes
Biochemical and Biophysical Research Communications xxx (2017) 1e6
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes Akira Takagi, Yutaka Horiuchi, Masanori Matsui* Department of Microbiology, Faculty of Medicine, Saitama Medical University, Moroyama-cho, Iruma-gun, Saitama 350-0495, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 August 2017 Accepted 12 August 2017 Available online xxx
Several non-radioactive methods have widely been utilized to detect antigen-specific cytotoxic T lymphocyte (CTL) responses instead of the classical 51Cr-release assay. These methods include intracellular cytokine staining, major histocompatibility complex-class I tetramers, and the CD107a mobilization assay. However, they do not directly measure target-cell death. In contrast, several attempts have been made to develop the flow cytometric CTL (FC-CTL) assay for evaluation of cytotoxicity. However, further improvement is necessary for it to become standardized. Here, we evaluated the characteristics of the FCCTL assay based on the uptake of propidium iodide (PI) using target cell lines expressing the green fluorescent protein (GFP). The FC-CTL assay was found to be sensitive enough to detect primary CTL responses. The usage of a pre-established GFP-expressing target cell line facilitated the procedure of the assay, and enabled a clear discrimination between target and effector cells. Time-course analyses demonstrated that PI-stained target cells were detected as early as surface CD107a expression after antigenic stimulation. Thus, the PI/GFP-based FC-CTL assay is sufficiently sensitive to practically detect the early stages of target-cell death, and may have a great potential for becoming a standard tool to measure CTL activity. © 2017 Elsevier Inc. All rights reserved.
Keywords: Flow cytometry CTL assay Target-cell death Propidium iodide Green fluorescent protein Time-course analyses
1. Introduction After the unproductive history of the cancer immunotherapy, a novel approach of the immune checkpoint blockade has brought therapeutic benefits to some patients suffering from malignancies [1]. This strategy involves blockade with antibodies targeting immunosuppressive molecules to reactivate antitumor immune cells. Successful results in the clinical trials have confirmed that tumor-specific CD8þ cytotoxic T lymphocytes (CTLs) play a pivotal role for the clearance of tumor cells, and therefore, efficient and accurate evaluation of CTL function becomes more and more important for the development of the cancer immunotherapy. In the past, the 51Cr-release assay [2] was most often used to measure killing activities of antigen-specific CTLs because this was relatively simple and reproducible. However, this assay has several
Abbreviations: FC-CTL assay, flow cytometric CTL assay; ICS, intracellular cytokine staining; PI, propidium iodide; GFP, green fluorescent protein; FMP, influenza A virus matrix protein; Ad, adenovirus; E/T, effector to target; % CD, % cell death; % SCD, % specific cell death; % E, % expression; % SE, % specific expression. * Corresponding author. E-mail address: [email protected] (M. Matsui).
disadvantages including low sensitivity, radioisotope usage, and lack of information about effector T cells. Additionally, data provided by this assay show only the information regarding the end point of target-cell death. More recently, several non-radioactive methods have widely been utilized to assess antigen-specific CTL responses. These alternative methods including the interferon (IFN)-g enzyme-linked immunospot assay [3], intracellular cytokine staining (ICS) [4], major histocompatibility complex-class I tetramers [5], and the CD107a mobilization assay [6] have improved sensitivity and convenience to enumerate antigenspecific CTLs. However, they do not measure target-cell death that directly demonstrates the cytolytic function of CTLs. On the other hand, several attempts have been made to develop the flow cytometric CTL (FC-CTL) assay for the evaluation of target-cell death. Among them, the most popular strategies depend on the uptake of DNA intercalating fluorescent agents such as propidium iodide (PI) [7], 7-amino-actinomycin D [8] and TO-PRO-3 iodide [9] in target cells that have been labelled with fluorescent dyes [7,9], or transfected with a gene encoding a fluorescent protein [10,11]. These data demonstrated that the FC-CTL assay had various advantages over the 51Cr-release assay, since it was more sensitive, more informative and safer. However, further improvement seems
http://dx.doi.org/10.1016/j.bbrc.2017.08.045 0006-291X/© 2017 Elsevier Inc. All rights reserved.
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
2
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
to be necessary for this assay to become standardized in the basic and clinical studies. One of the most critical issues may be the cumbersome preparation of labelled target cells. In the current study, we performed the PI-based FC-CTL assay using tumor target cell lines that stably express the green fluorescent protein (GFP). The stable expression of GFP in target cells facilitated the procedure of the assay, and enabled a clear discrimination between target and effector cells. We evaluated the FC-CTL assay for sensitivity in comparison with the 51Cr-release assay. We also performed the time-course analyses of the FC-CTL assay along with other assays to examine how the frequency of apoptotic target cells changed after antigen stimulation. 2. Materials and methods 2.1. Mice HHD II mice express a transgenic HLA-A*0201 monochain, designated as HHD, in which human b2-microglobulin is covalently linked to a chimeric heavy chain composed of HLA-A*0201 and H2Db [12]. C57BL/6J mice were purchased from the Charles River Laboratories Japan, Inc. (Yokohama, Japan). Eight- to twelve-weekold mice were used for all experiments. Mice were housed in appropriate animal care facilities at Saitama Medical University, Saitama, Japan, and handled according to international guidelines for experiments with animals. 2.2. Synthetic peptides Influenza A virus matrix protein 1 (FMP)-derived peptides, FMP58-66 (GILFGVFTL) [13] and FMP128-135 (MGLIYNRM) [14], which were restricted by HLA-A*0201 and H-2Kb, respectively, were synthesized by Operon Biotechnologies (Tokyo, Japan). 2.3. Construction of recombinant adenovirus Recombinant adenovirus expressing the full-length FMP protein of influenza A virus (Ad-FMP) was previously generated using the AdEasy adenoviral vector system (Agilent Technologies, Santa Clara, CA) [15]. Virus titers were determined by calculating the 50% tissue culture infectious dose (TCID50) on AD-293 cells. Wild-type adenovirus (Ad-WT) was used as a control. 2.4. Cell lines A mouse lymphoma cell line, RMA (H-2b), was cultured in RPMI1640 medium (Nacalai Tesque, Kyoto, Japan) with 10% Fetal calf serum (FCS) (Gibco, ThermoFisher Scientific Inc., Waltham, MA) (R10). An HHD gene-transfected RMA cell line, RMA-HHD, was previously described [12]. RMA and RMA-HHD cells were transfected with a pEGFP-N3 vector (Clontech Laboratories, Inc. Mountain View, CA) by electroporation (Gene Pulser, Bio-Rad Laboratories, Hercules, CA) as described [16]. After 2 days, GFP-expressing RMA (RMA-GFP) and RMA-HHD (RMA-HHD-GFP) cells were sorted by FACSAria II (BD Biosciences, San Jose, CA), and cloned. RMA-HHD, RMA-GFP and RMA-HHD-GFP cells were maintained in R-10 containing 500 mg/ml of G418 (Nacalai Tesque). RMA-HHD cells expressing the GFP-fused FMP protein (RMA-HHD-FMP-GFP) was generated before [16], and maintained in R-10 containing 500 mg/ ml of G418 (Nacalai Tesque) and 500 mg/ml of Hygromycin B (Nacalai Tesque). AD-293 (Agilent Technologies) was maintained in Dulbecco's Modified Eagle Medium (Nacalai Tesque) with 10% FCS (Gibco).
2.5. Effector cells Mice were intraperitoneally immunized with 5 108 TCID50 of either Ad-FMP or Ad-WT. After 2 weeks, spleen cells were prepared and used as primary effector cells. In some experiments, primary effector cells were in vitro stimulated for 1 week with gammairradiated (30 Gy), syngeneic spleen cells pulsed with 10 mM of an appropriate peptide [15], and were used as peptide-stimulated effector cells. 2.6. FC-CTL assay RMA-GFP or RMA-HHD-GFP cells were pulsed with or without 10 mM of an appropriate peptide for 1 h (h), and used as target cells. RMA-HHD-FMP-GFP cells were also used as target cells loading endogenous FMP peptides. Target cells were plated in wells of a round-bottom 96-well plate at 5 104 cells/well with primary or peptide-stimulated effector cells at various effector to target (E/T) ratios in the presence of PI at a final concentration of 2.4 mg/ml, and were incubated at 37 C for 4 h. After washing, GFP positive target cells were analyzed for their PI staining on a FACSCanto II flow cytometer (BD Biosciences). Percent cell death (% CD) was calculated as (PIþ GFPþ target cells/total GFPþ target cells) 100. Percent specific cell death (% SCD) was calculated according to the formula: % SCD ¼ (sample % CD e spontaneous % CD)/(100 - spontaneous % CD) 100. Spontaneous % CD represents the % CD in the culture of target cells alone. 2.7.
51
Cr-release assay
51
Cr-release assays were performed as described before [16]. In brief, RMA-HHD cells were pulsed with or without 10 mM of FMP5851 66, followed by labelling with Na2 CrO4. Labelled target cells were cultured with primary or peptide-stimulated effector cells at various E/T ratios. After a 4-h incubation, supernatant of each well was harvested and the radioactivity was counted. Percent specific lysis was calculated according to the formula: % specific lysis ¼ (sample cpm-spontaneous cpm)/(maximal cpm-spontaneous cpm) 100. 2.8. Time-course analyses Cell culture in the FC-CTL assay was performed as described above. In brief, primary effector cells from Ad-FMP-immunized mice were incubated with RMA-HHD-GFP cells pulsed with or without FMP58-66 at an E/T ratio of 100 in the presence of PI for the indicated time periods (0.5e4 h). After washing, GFP positive target cells were analyzed for their PI staining by flow cytometry. % SCD at each time point was calculated according to the formula described above. Then, the value at each time point was calculated as the percent maximum response relative to the value at the 4th hour which was set as the maximum value: % maximum response ¼ (sample % SCD with a peptide e sample % SCD with no peptide)/(% SCD with a peptide at 4 h - % SCD with no peptide at 4 h) 100. Time-course analyses were also performed in the CD107a mobilization assay combined with ICS [15]. Briefly, primary effector cells from Ad-FMP-immunized mice were stained with FITC-antiCD107a monoclonal antibody (mAb) (BD Biosciences) during the stimulation with or without 10 mM of FMP58-66 for the indicated time periods in the presence of brefeldin A (BD Biosciences), followed by staining with PerCP-anti-CD8a mAb (BioLegend). Cells were then fixed, permeabilized, and stained with PE-anti-IFN-g mAb (BioLegend). After washing the cells, flow cytometric analyses were performed. Percent expression (% E) of CD107a or IFN-g was analyzed as (CD107aþCD8þ or IFN-gþCD8þ cells/total CD8þ
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
3
Fig. 1. Measurement of secondary CTL responses in the FC-CTL assay. (A) Peptide-stimulated effector cells from Ad-FMP- (upper & middle panels) or Ad-WT- (lower panels) immunized HHD mice were cultured with RMA-HHD-GFP cells pulsed with (upper & lower panels) or without (middle panels) FMP58-66 in the presence of PI at various E/T ratios. After a 4-h incubation, the frequency of PI-stained target cells was analyzed by flow cytometry. Numbers shown indicate percent cell death. Left panels show FSC/SSC dot plots of the entire populations at an E/T ratio of 10. (BeD) Percent specific cell death was measured by the FC-CTL assay. Peptide-stimulated effector cells were prepared from Ad-FMPimmunized HHD (B & C) or C57BL/6 (D) mice following stimulation with FMP58-66 (B & C) or FMP128-135 (D). Cells were then incubated with target cells for 4 hs at various E/T ratios in the presence of PI. Target cells used were: (B) RMA-HHD-GFP cells pulsed with (filled circle) or without (open circle) FMP58-66; (C) RMA-HHD-FMP-GFP (filled circle) and RMA-HHD-GFP (open circle) cells; (D) RMA-GFP cells pulsed with (filled circle) or without (open circle) FMP128-135. In each experiment, 3e5 mice per group were used, and spleen cells of mice per group were pooled. Results are shown as the mean ± SD of triplicate wells. Each experiment was repeated twice with similar results. *, p < 0.05; **, p < 0.01 compared to each of negative controls, Mann-Whitney U test.
cells) 100. Percent specific expression (% SE) of CD107a or IFN-g was calculated according to the formula: % SE ¼ (sample % E spontaneous % E)/(100 - spontaneous % E) 100. Spontaneous % E of CD107a or IFN-g represents the % E of CD107a or IFN-g on effector cells cultured with target cells at a time point of 0 h. The value of CD107a or IFN-g expression at each time point was calculated as the % maximum response relative to the value at the 4th hour as follows: % maximum response ¼ (sample % SE with a peptide e sample % SE with no peptide)/(% SE with a peptide at 4 h % SE with no peptide at 4 h) 100. Kinetics of PD-1 expression on primary effector cells cultured with tumor target cells was examined. Briefly, primary effector cells were cultured with RMA-HHD cells pulsed with or without FMP5866 at an E/T ratio of 10 for the indicated time periods, stained with both FITC-anti-PD-1 (eBioscience Inc., San Diego, CA) and PE-antiCD8a (BioLegend) mAbs, and were analyzed by flow cytometry. Percent expression (% E) of PD-1 was analyzed as (PD-1þCD8þ cells/ total CD8þ cells) 100. Percent specific expression (% SE) of PD-1 was calculated according to the formula: % SE ¼ (sample % E
spontaneous % E)/(100 - spontaneous % E) 100. Spontaneous % E of PD-1 represents the % E of PD-1 on effector cells cultured with target cells at a time point of 0 h. 2.9. Immunofluorescence microscopy Primary effector cells were incubated with RMA-HHD-GFP cells pulsed with or without FMP58-66 at an E/T ratio of 100 in the presence of PI for 1 h. Primary effector cells were also incubated with Alexa Fluor 647-anti-CD107a mAb (BioLegend) and brefeldin A with or without FMP58-66 for 1 h, followed by staining with FITCanti-CD8a mAb (BioLegend). Cells were analyzed using a Zeiss Axioplan 2 fluorescence microscope. 2.10. Statistical analyses Statistical analyses were performed with Mann-Whitney U test and one-way ANOVA. A value of p < 0.05 was considered statistically significant.
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
4
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
3. Results 3.1. Detection of antigen-specific cell-mediated cytotoxicity using flow cytometry Peptide-stimulated effector cells were co-cultured with RMAHHD-GFP target cells prepulsed with or without FMP58-66 at various E/T ratios in the presence of PI. After a 4-h incubation, the frequency of PI-stained dead target cells was analyzed by flow cytometry. As shown in Fig. 1A and B, percentages of PI-stained dead cells in GFP-positive target cells increased from 14.4 to 73.3% in an E/T ratio-dependent fashion when target cells were pulsed with FMP58-66. In contrast, much less increase of PI-positive target cells was observed in the co-culture of either unpulsed targets with effectors or peptide-pulsed targets with effectors derived from Ad-WT-injected mice (Fig. 1A and B). Effector cells from either Ad-FMP- or Ad-WT-immunized mice did not significantly lyse RMA-GFP target cells (data not shown). We also carried out the FCCTL assay using RMA-HHD-FMP-GFP target cells that express endogenously processed peptides of FMP. It was observed that the
FC-CTL assay demonstrated the epitope-specific cell death of RMAHHD-FMP-GFP target cells as well (Fig. 1C). Similar results were obtained with a different epitope for C57BL/6 mice. As shown in Fig. 1D, high percentages of specific cell death were detected in the culture of effector cells with FMP128-135-pulsed target cells, but not with unpulsed targets. These data show the practical utility of the FC-CTL assay to efficiently detect antigen-specific secondary CTL responses even at low E/T ratios ranging from 1 to 10.
3.2. Evaluation of antigen-specific primary CTL responses We next investigated whether the FC-CTL assay could be applied to the analysis of antigen-specific primary CTL responses. Primary effector cells were prepared and examined for their FMP58-66specific killing activities. As shown in Fig. 2A, a high percent cell death (87.2%) of peptide-pulsed target cells was induced at an E/T ratio of 100. Furthermore, 30.9% specific cell death was detected in peptide-pulsed target cells even at an E/T ratio as low as 10, whereas only 7.7% of dead cells were observed in unpulsed target cells at the same ratio. (Fig. 2A). In contrast, any significant primary
Fig. 2. Detection of primary CTL responses in the FC-CTL assay. (A) Primary effector cells from Ad-FMP-immunized HHD mice were cultured with RMA-HHD-GFP pulsed with (upper panels) or without (lower panels) FMP58-66 in the presence of PI for 4 hs at various E/T ratios. The frequency of PI-stained target cells was analyzed by flow cytometry. Numbers shown indicate percent cell death. Left panels show FSC/SSC dot plots of the entire populations at an E/T ratio of 100. (B & C) Comparison of the FC-CTL assay with the 51Cr-release assay in primary (B) and secondary (C) CTL responses. Primary (B) and secondary (C) effector cells were prepared from Ad-FMP-immunized HHD mice. In the FC-CTL assay (solid line), cells were incubated with RMA-HHD-GFP cells pulsed with (filled circle) or without (open circle) FMP58-66 for 4 h in the presence of PI, and percent specific cell death was analyzed by flow cytometry. In the 51Cr-release assay (dashed line), cells were incubated with 51Cr-labelled RMA-HHD cells pulsed with (filled square) or without (open square) FMP58-66 for 4 h, and percent specific lysis was measured. Three mice per group were used in each experiment. Results are shown as the mean ± SD. Each experiment was repeated twice with similar results. *, p < 0.05; **, p < 0.01 compared to negative controls (open symbol) of each assay, Mann-Whitney U test.
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
5
CTL response was not observed in the 51Cr release assay even at a high E/T ratio of 100 (Fig. 2B), although this assay worked well for the secondary CTL responses (Fig. 2C). These data indicate that the FC-CTL assay is much more sensitive than the 51Cr release assay. 3.3. Time-course analyses of the FC-CTL assay Having established that the FC-CTL assay can detect primary CTL responses, we next examined how the frequency of PI-stained target cells changed with the passage of time in primary CTL responses compared to the expression of surface CD107a and intracellular IFN-g in effector cells. Primary effector cells were cultured with RMA-HHD-GFP cells pulsed with or without FMP58-66 in the presence of PI for the indicated time points. Surprisingly, PI-stained target cells were observed as early as acquisition of CD107a surface expression on CD8þ effector cells after antigenic stimulation (Fig. 3AeC), suggesting that membrane damage of target cells occurs as soon as degranulation of effector cells begins after antigen stimulation. In contrast, intracellular expression of IFN-g in CD8þ effector cells was hardly detected at 1 h post-stimulation (Fig. 3A and B), indicating that the FC-CTL assay is more sensitive than ICS. We also examined the kinetics of PD-1 expression on CD8þ effector cells after antigen stimulation. It was found that the peptidespecific expression of PD-1 started to increase as early as 30 min to 1 h following stimulation with a specific peptide (Fig. 3A), suggesting that the inhibitory signal begins to be transmitted into CD8þ effector cells soon after antigen stimulation. We examined PD-1 expression up to 26 h after antigen stimulation, and the enhancement of PD-1 expression continued until that time (data not shown). 4. Discussion A critical issue for the accurate determination in the FC-CTL assay is to clearly differentiate targets from effectors because effector cells may also undergo apoptosis. For this reason, several groups stained target cells with a fluorescent dye like PKH-26 [9] or DiOC18(3) [7]. However, it is sometimes difficult to avoid the crosscontamination to the effector population. We found that high percentages of DiOC18(3)-stained dead cells were detected by PI staining in the co-culture of peptide-stimulated effector cells and DiOC18(3)-stained target cells with no peptide (data not shown). This non-specific killing presumably resulted from the crosscontamination of DiOC18(3) to dead effector cells. In the current study, we used GFP-expressing target cell lines which were clearly distinguished from effector cells by the flow cytometry (Figs. 1 and 2). Generally, many researchers prefer the CD107a mobilization assay, ICS or tetramer staining to evaluate CTL activities rather than the FC-CTL assay even though they do not measure cytotoxicity. This is partly because the preparation of labelled target cells is cumbersome. In our study, however, the usage of pre-established GFP-expressing target cell lines enabled us to skip this process, and therefore facilitated the procedure of the FC-CTL assay. On the other hand, our method is not likely to be suitable for the use of autologous human B-lymphoblastoid cell lines (BLCLs) or peripheral blood mononuclear cells (PBMCs) as targets, which is beneficial for clinical monitoring, because classical transfection techniques generally result in poor DNA transfer into them. In addition, our method lacks flexibility because target cells are restricted to the pre-established panel of transfected cells. However, a fluorescent dye such as PKH-26 can attain any given target cells and utilize them as target cells. The FC-CTL assay was much more sensitive than the 51Cr-release assay (Fig. 2). This is partly because the FC-CTL assay detects PIstained dead cells with weak membrane alteration, whereas the
Fig. 3. Time-course analyses. (A) Primary effector cells from Ad-FMP-immunized HHD mice were used for time-course analyses. In the FC-CTL assay, effector cells were cultured with RMA-HHD-GFP cells pulsed with or without FMP58-66 at an E/T ratio of 100 in the presence of PI for the indicated time periods. In the CD107a mobilization assay combined with ICS, effector cells were stained with FITC-anti-CD107a mAb during the peptide stimulation for the indicated periods, followed by staining of surface CD8 and intracellular IFN-g. After flow cytometry analyses, % maximum responses (solid line) regarding PI-stained target cells (filled circle), CD107a-positive effector cells (open circle), and IFN-g-positive effector cells (filled triangle) were calculated. Three to five mice per group were used, and results are shown as the mean ± SD of data of mice per group. Each experiment was repeated twice with similar results. For PD-1 expression (dashed line), effector cells were incubated with RMA-HHD cells pulsed with (filled square) or without FMP58-66 (open square) at an E/T ratio of 10 for the indicated time periods, and were stained with FITC-PD-1 and PE-CD8a mAbs. Results are shown as the mean of % specific expression of two mice per group. (B) Comparison of % maximum responses regarding PI-stained GFPþ target cells, CD107a-positive effector cells, and IFN-g-positive effector cells at a time point of 1 h. **, p < 0.01; NS, not significance, One-way ANOVA. (C) Immunofluorescence microscopy images at a time point of 1 h. Left, PIþ GFPþ targets (arrows); right, CO107aþ CD8þ effectors (arrows).
51 Cr-release assay measures the release of 51Cr-labelled cytosolic proteins which requires a more severe damage of cellular membrane. Thanks to the high sensitivity, the FC-CTL assay was capable of measuring primary CTL responses (Fig. 2), and thereby may be applied for immune monitoring of clinical samples. Degranulation in CTLs occurs immediately after triggering of the T cell receptor complex [6]. Therefore, surface expression of CD107a on effector cells can be detected earlier than intracellular IFN-g expression in effector cells (Fig. 3A and B). Surprisingly, PI-stained target cells were observed as early as CD107a expression of effector cells after
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
6
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
peptide stimulation (Fig. 3AeC). This data confirms that the PI/GFPbased FC-CTL assay is sufficiently sensitive to practically detect the early stages of cell death. Measurements of the cleavage of caspase 3 and annexin-V binding in target cells may be potential alternatives because they provide earlier information of apoptosis [17]. The advent of the immune checkpoint blockade has brought rays of hope to us in the cancer immunotherapy [1]. However, the exact relationship between the immune responses and tumor rejection has not fully been elucidated. Hence, a robust and appropriate method to analyze immune responses is required for the evaluation of immunotherapy. It is very unlikely that any single immunological parameter is sufficient to understand extremely complicated immune responses in the tumor microenvironment and a large amount of clinical data. Simultaneous monitoring of multiple immunological parameters is necessary and more advantageous to elucidate what types of immune responses effectively result in tumor rejection. In this respect, a multiparametric flow cytometry-based assay is a valid option in the clinical monitoring [17]. It is possible that the FC-CTL assay is combined with the 107a mobilization assay, tetramer labeling and ICS. Furthermore, immune checkpoint molecules such as PD-1 (Fig. 3A) and PD-L1 may be important parameters to measure cytotoxicity mediated by peptide-specific CTLs in the FC-CTL assay. Such comprehensive evaluation should be useful to define the correlation between tumor-specific CTL activities and clinical outcomes. In conclusion, we performed the FC-CTL assay to measure cellmediated cytotoxicity in comparison with other CTL assays. We showed that the FC-CTL assay may have a great potential for becoming a standard tool to measure CTL activity in the basic and clinical studies.
Conflict of interest None.
Acknowledgments The authors are grateful to Dr. F. Lemonnier for providing HHD II mice, RMA, and RMA-HHD cells. This work was supported by a Grant-in-Aid for Young Scientists (B) (No. 26860771) to A. T., a Grant-in-Aid for Scientific Research (C) (No. 26460560) to M. M. from Japan Society for the Promotion of Science, and grants from Saitama Medical University, Internal Grant (No. 16-B-1-17) to Y. H. and Maruki memorial special award (No. 16-A-1-01) to M. M.
Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.bbrc.2017.08.045.
References [1] I. Mellman, G. Coukos, G. Dranoff, Cancer immunotherapy comes of age, Nature 480 (2011) 480e489. [2] K.T. Brunner, J. Mauel, J.C. Cerottini, et al., Quantitative assay of the lytic action of immune lymphoid cells on 51-Cr-labelled allogeneic target cells in vitro; inhibition by isoantibody and by drugs, Immunology 14 (1968) 181e196. [3] Y. Miyahira, K. Murata, D. Rodriguez, et al., Quantification of antigen specific CD8þ T cells using an ELISPOT assay, J. Immunol. Methods 181 (1995) 45e54. [4] V.C. Maino, L.J. Picker, Identification of functional subsets by flow cytometry: intracellular detection of cytokine expression, Cytometry 34 (1998) 207e215. [5] J.D. Altman, P.A. Moss, P.J. Goulder, et al., Phenotypic analysis of antigenspecific T lymphocytes, Science 274 (1996) 94e96. [6] M.R. Betts, J.M. Brenchley, D.A. Price, et al., Sensitive and viable identification of antigen-specific CD8þ T cells by a flow cytometric assay for degranulation, J. Immunol. Methods 281 (2003) 65e78. [7] L. Piriou, S. Chilmonczyk, N. Genetet, et al., Design of a flow cytometric assay for the determination of natural killer and cytotoxic T-lymphocyte activity in human and in different animal species, Cytometry 41 (2000) 289e297. [8] B. Langhans, M. Ahrendt, J. Nattermann, et al., Comparative study of NK cellmediated cytotoxicity using radioactive and flow cytometric cytotoxicity assays, J. Immunol. Methods 306 (2005) 161e168. [9] A.E. Lee-MacAry, E.L. Ross, D. Davies, et al., Development of a novel flow cytometric cell-mediated cytotoxicity assay using the fluorophores PKH-26 and TO-PRO-3 iodide, J. Immunol. Methods 252 (2001) 83e92. [10] C.A. van Baalen, D. Kwa, E.J. Verschuren, et al., Fluorescent antigen-transfected target cell cytotoxic T lymphocyte assay for ex vivo detection of antigenspecific cell-mediated cytotoxicity, J. Infect. Dis. 192 (2005) 1183e1190. [11] K. Chen, L. Chen, P. Zhao, et al., FL-CTL assay: fluorolysometric determination of cell-mediated cytotoxicity using green fluorescent protein and red fluorescent protein expressing target cells, J. Immunol. Methods 300 (2005) 100e114. [12] S. Pascolo, N. Bervas, J.M. Ure, et al., HLA-A2.1-restricted education and cytolytic activity of CD8þ T lymphocytes from b2 microglobulin (b2m) HLAA2.1 monochain transgenic H-2Db b2m double knockout mice, J. Exp. Med. 185 (1997) 2043e2051. [13] F. Gotch, J. Rothbard, K. Howland, et al., Cytotoxic T lymphocytes recognize a fragment of influenza virus matrix protein in association with HLA-A2, Nature 326 (1987) 881e882. [14] A. Vitiello, L. Yuan, R.W. Chesnut, et al., Immunodominance analysis of CTL responses to influenza PR8 virus reveals two new dominant and subdominant Kb-restricted epitopes, J. Immunol. 157 (1996) 5555e5562. [15] T. Suda, M. Kawano, Y. Nogi, et al., The route of immunization with adenoviral vaccine influences the recruitment of cytotoxic T lymphocytes in the lung that provide potent protection from influenza A virus, Antivir. Res. 91 (2011) 252e258. [16] M. Matsui, M. Kawano, S. Matsushita, et al., Introduction of a point mutation into an HLA class I single-chain trimer induces enhancement of CTL priming and antitumor immunity, Mol. Ther. - Methods Clin. Dev. 1 (2014) 14027. [17] L. Zaritskaya, M.R. Shurin, T.J. Sayers, et al., New flow cytometric assays for monitoring cell-mediated cytotoxicity, Expert Rev. Vaccine 9 (2010) 601e616.
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes Akira Takagi, Yutaka Horiuchi, Masanori Matsui* Department of Microbiology, Faculty of Medicine, Saitama Medical University, Moroyama-cho, Iruma-gun, Saitama 350-0495, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 August 2017 Accepted 12 August 2017 Available online xxx
Several non-radioactive methods have widely been utilized to detect antigen-specific cytotoxic T lymphocyte (CTL) responses instead of the classical 51Cr-release assay. These methods include intracellular cytokine staining, major histocompatibility complex-class I tetramers, and the CD107a mobilization assay. However, they do not directly measure target-cell death. In contrast, several attempts have been made to develop the flow cytometric CTL (FC-CTL) assay for evaluation of cytotoxicity. However, further improvement is necessary for it to become standardized. Here, we evaluated the characteristics of the FCCTL assay based on the uptake of propidium iodide (PI) using target cell lines expressing the green fluorescent protein (GFP). The FC-CTL assay was found to be sensitive enough to detect primary CTL responses. The usage of a pre-established GFP-expressing target cell line facilitated the procedure of the assay, and enabled a clear discrimination between target and effector cells. Time-course analyses demonstrated that PI-stained target cells were detected as early as surface CD107a expression after antigenic stimulation. Thus, the PI/GFP-based FC-CTL assay is sufficiently sensitive to practically detect the early stages of target-cell death, and may have a great potential for becoming a standard tool to measure CTL activity. © 2017 Elsevier Inc. All rights reserved.
Keywords: Flow cytometry CTL assay Target-cell death Propidium iodide Green fluorescent protein Time-course analyses
1. Introduction After the unproductive history of the cancer immunotherapy, a novel approach of the immune checkpoint blockade has brought therapeutic benefits to some patients suffering from malignancies [1]. This strategy involves blockade with antibodies targeting immunosuppressive molecules to reactivate antitumor immune cells. Successful results in the clinical trials have confirmed that tumor-specific CD8þ cytotoxic T lymphocytes (CTLs) play a pivotal role for the clearance of tumor cells, and therefore, efficient and accurate evaluation of CTL function becomes more and more important for the development of the cancer immunotherapy. In the past, the 51Cr-release assay [2] was most often used to measure killing activities of antigen-specific CTLs because this was relatively simple and reproducible. However, this assay has several
Abbreviations: FC-CTL assay, flow cytometric CTL assay; ICS, intracellular cytokine staining; PI, propidium iodide; GFP, green fluorescent protein; FMP, influenza A virus matrix protein; Ad, adenovirus; E/T, effector to target; % CD, % cell death; % SCD, % specific cell death; % E, % expression; % SE, % specific expression. * Corresponding author. E-mail address: [email protected] (M. Matsui).
disadvantages including low sensitivity, radioisotope usage, and lack of information about effector T cells. Additionally, data provided by this assay show only the information regarding the end point of target-cell death. More recently, several non-radioactive methods have widely been utilized to assess antigen-specific CTL responses. These alternative methods including the interferon (IFN)-g enzyme-linked immunospot assay [3], intracellular cytokine staining (ICS) [4], major histocompatibility complex-class I tetramers [5], and the CD107a mobilization assay [6] have improved sensitivity and convenience to enumerate antigenspecific CTLs. However, they do not measure target-cell death that directly demonstrates the cytolytic function of CTLs. On the other hand, several attempts have been made to develop the flow cytometric CTL (FC-CTL) assay for the evaluation of target-cell death. Among them, the most popular strategies depend on the uptake of DNA intercalating fluorescent agents such as propidium iodide (PI) [7], 7-amino-actinomycin D [8] and TO-PRO-3 iodide [9] in target cells that have been labelled with fluorescent dyes [7,9], or transfected with a gene encoding a fluorescent protein [10,11]. These data demonstrated that the FC-CTL assay had various advantages over the 51Cr-release assay, since it was more sensitive, more informative and safer. However, further improvement seems
http://dx.doi.org/10.1016/j.bbrc.2017.08.045 0006-291X/© 2017 Elsevier Inc. All rights reserved.
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
2
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
to be necessary for this assay to become standardized in the basic and clinical studies. One of the most critical issues may be the cumbersome preparation of labelled target cells. In the current study, we performed the PI-based FC-CTL assay using tumor target cell lines that stably express the green fluorescent protein (GFP). The stable expression of GFP in target cells facilitated the procedure of the assay, and enabled a clear discrimination between target and effector cells. We evaluated the FC-CTL assay for sensitivity in comparison with the 51Cr-release assay. We also performed the time-course analyses of the FC-CTL assay along with other assays to examine how the frequency of apoptotic target cells changed after antigen stimulation. 2. Materials and methods 2.1. Mice HHD II mice express a transgenic HLA-A*0201 monochain, designated as HHD, in which human b2-microglobulin is covalently linked to a chimeric heavy chain composed of HLA-A*0201 and H2Db [12]. C57BL/6J mice were purchased from the Charles River Laboratories Japan, Inc. (Yokohama, Japan). Eight- to twelve-weekold mice were used for all experiments. Mice were housed in appropriate animal care facilities at Saitama Medical University, Saitama, Japan, and handled according to international guidelines for experiments with animals. 2.2. Synthetic peptides Influenza A virus matrix protein 1 (FMP)-derived peptides, FMP58-66 (GILFGVFTL) [13] and FMP128-135 (MGLIYNRM) [14], which were restricted by HLA-A*0201 and H-2Kb, respectively, were synthesized by Operon Biotechnologies (Tokyo, Japan). 2.3. Construction of recombinant adenovirus Recombinant adenovirus expressing the full-length FMP protein of influenza A virus (Ad-FMP) was previously generated using the AdEasy adenoviral vector system (Agilent Technologies, Santa Clara, CA) [15]. Virus titers were determined by calculating the 50% tissue culture infectious dose (TCID50) on AD-293 cells. Wild-type adenovirus (Ad-WT) was used as a control. 2.4. Cell lines A mouse lymphoma cell line, RMA (H-2b), was cultured in RPMI1640 medium (Nacalai Tesque, Kyoto, Japan) with 10% Fetal calf serum (FCS) (Gibco, ThermoFisher Scientific Inc., Waltham, MA) (R10). An HHD gene-transfected RMA cell line, RMA-HHD, was previously described [12]. RMA and RMA-HHD cells were transfected with a pEGFP-N3 vector (Clontech Laboratories, Inc. Mountain View, CA) by electroporation (Gene Pulser, Bio-Rad Laboratories, Hercules, CA) as described [16]. After 2 days, GFP-expressing RMA (RMA-GFP) and RMA-HHD (RMA-HHD-GFP) cells were sorted by FACSAria II (BD Biosciences, San Jose, CA), and cloned. RMA-HHD, RMA-GFP and RMA-HHD-GFP cells were maintained in R-10 containing 500 mg/ml of G418 (Nacalai Tesque). RMA-HHD cells expressing the GFP-fused FMP protein (RMA-HHD-FMP-GFP) was generated before [16], and maintained in R-10 containing 500 mg/ ml of G418 (Nacalai Tesque) and 500 mg/ml of Hygromycin B (Nacalai Tesque). AD-293 (Agilent Technologies) was maintained in Dulbecco's Modified Eagle Medium (Nacalai Tesque) with 10% FCS (Gibco).
2.5. Effector cells Mice were intraperitoneally immunized with 5 108 TCID50 of either Ad-FMP or Ad-WT. After 2 weeks, spleen cells were prepared and used as primary effector cells. In some experiments, primary effector cells were in vitro stimulated for 1 week with gammairradiated (30 Gy), syngeneic spleen cells pulsed with 10 mM of an appropriate peptide [15], and were used as peptide-stimulated effector cells. 2.6. FC-CTL assay RMA-GFP or RMA-HHD-GFP cells were pulsed with or without 10 mM of an appropriate peptide for 1 h (h), and used as target cells. RMA-HHD-FMP-GFP cells were also used as target cells loading endogenous FMP peptides. Target cells were plated in wells of a round-bottom 96-well plate at 5 104 cells/well with primary or peptide-stimulated effector cells at various effector to target (E/T) ratios in the presence of PI at a final concentration of 2.4 mg/ml, and were incubated at 37 C for 4 h. After washing, GFP positive target cells were analyzed for their PI staining on a FACSCanto II flow cytometer (BD Biosciences). Percent cell death (% CD) was calculated as (PIþ GFPþ target cells/total GFPþ target cells) 100. Percent specific cell death (% SCD) was calculated according to the formula: % SCD ¼ (sample % CD e spontaneous % CD)/(100 - spontaneous % CD) 100. Spontaneous % CD represents the % CD in the culture of target cells alone. 2.7.
51
Cr-release assay
51
Cr-release assays were performed as described before [16]. In brief, RMA-HHD cells were pulsed with or without 10 mM of FMP5851 66, followed by labelling with Na2 CrO4. Labelled target cells were cultured with primary or peptide-stimulated effector cells at various E/T ratios. After a 4-h incubation, supernatant of each well was harvested and the radioactivity was counted. Percent specific lysis was calculated according to the formula: % specific lysis ¼ (sample cpm-spontaneous cpm)/(maximal cpm-spontaneous cpm) 100. 2.8. Time-course analyses Cell culture in the FC-CTL assay was performed as described above. In brief, primary effector cells from Ad-FMP-immunized mice were incubated with RMA-HHD-GFP cells pulsed with or without FMP58-66 at an E/T ratio of 100 in the presence of PI for the indicated time periods (0.5e4 h). After washing, GFP positive target cells were analyzed for their PI staining by flow cytometry. % SCD at each time point was calculated according to the formula described above. Then, the value at each time point was calculated as the percent maximum response relative to the value at the 4th hour which was set as the maximum value: % maximum response ¼ (sample % SCD with a peptide e sample % SCD with no peptide)/(% SCD with a peptide at 4 h - % SCD with no peptide at 4 h) 100. Time-course analyses were also performed in the CD107a mobilization assay combined with ICS [15]. Briefly, primary effector cells from Ad-FMP-immunized mice were stained with FITC-antiCD107a monoclonal antibody (mAb) (BD Biosciences) during the stimulation with or without 10 mM of FMP58-66 for the indicated time periods in the presence of brefeldin A (BD Biosciences), followed by staining with PerCP-anti-CD8a mAb (BioLegend). Cells were then fixed, permeabilized, and stained with PE-anti-IFN-g mAb (BioLegend). After washing the cells, flow cytometric analyses were performed. Percent expression (% E) of CD107a or IFN-g was analyzed as (CD107aþCD8þ or IFN-gþCD8þ cells/total CD8þ
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
3
Fig. 1. Measurement of secondary CTL responses in the FC-CTL assay. (A) Peptide-stimulated effector cells from Ad-FMP- (upper & middle panels) or Ad-WT- (lower panels) immunized HHD mice were cultured with RMA-HHD-GFP cells pulsed with (upper & lower panels) or without (middle panels) FMP58-66 in the presence of PI at various E/T ratios. After a 4-h incubation, the frequency of PI-stained target cells was analyzed by flow cytometry. Numbers shown indicate percent cell death. Left panels show FSC/SSC dot plots of the entire populations at an E/T ratio of 10. (BeD) Percent specific cell death was measured by the FC-CTL assay. Peptide-stimulated effector cells were prepared from Ad-FMPimmunized HHD (B & C) or C57BL/6 (D) mice following stimulation with FMP58-66 (B & C) or FMP128-135 (D). Cells were then incubated with target cells for 4 hs at various E/T ratios in the presence of PI. Target cells used were: (B) RMA-HHD-GFP cells pulsed with (filled circle) or without (open circle) FMP58-66; (C) RMA-HHD-FMP-GFP (filled circle) and RMA-HHD-GFP (open circle) cells; (D) RMA-GFP cells pulsed with (filled circle) or without (open circle) FMP128-135. In each experiment, 3e5 mice per group were used, and spleen cells of mice per group were pooled. Results are shown as the mean ± SD of triplicate wells. Each experiment was repeated twice with similar results. *, p < 0.05; **, p < 0.01 compared to each of negative controls, Mann-Whitney U test.
cells) 100. Percent specific expression (% SE) of CD107a or IFN-g was calculated according to the formula: % SE ¼ (sample % E spontaneous % E)/(100 - spontaneous % E) 100. Spontaneous % E of CD107a or IFN-g represents the % E of CD107a or IFN-g on effector cells cultured with target cells at a time point of 0 h. The value of CD107a or IFN-g expression at each time point was calculated as the % maximum response relative to the value at the 4th hour as follows: % maximum response ¼ (sample % SE with a peptide e sample % SE with no peptide)/(% SE with a peptide at 4 h % SE with no peptide at 4 h) 100. Kinetics of PD-1 expression on primary effector cells cultured with tumor target cells was examined. Briefly, primary effector cells were cultured with RMA-HHD cells pulsed with or without FMP5866 at an E/T ratio of 10 for the indicated time periods, stained with both FITC-anti-PD-1 (eBioscience Inc., San Diego, CA) and PE-antiCD8a (BioLegend) mAbs, and were analyzed by flow cytometry. Percent expression (% E) of PD-1 was analyzed as (PD-1þCD8þ cells/ total CD8þ cells) 100. Percent specific expression (% SE) of PD-1 was calculated according to the formula: % SE ¼ (sample % E
spontaneous % E)/(100 - spontaneous % E) 100. Spontaneous % E of PD-1 represents the % E of PD-1 on effector cells cultured with target cells at a time point of 0 h. 2.9. Immunofluorescence microscopy Primary effector cells were incubated with RMA-HHD-GFP cells pulsed with or without FMP58-66 at an E/T ratio of 100 in the presence of PI for 1 h. Primary effector cells were also incubated with Alexa Fluor 647-anti-CD107a mAb (BioLegend) and brefeldin A with or without FMP58-66 for 1 h, followed by staining with FITCanti-CD8a mAb (BioLegend). Cells were analyzed using a Zeiss Axioplan 2 fluorescence microscope. 2.10. Statistical analyses Statistical analyses were performed with Mann-Whitney U test and one-way ANOVA. A value of p < 0.05 was considered statistically significant.
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
4
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
3. Results 3.1. Detection of antigen-specific cell-mediated cytotoxicity using flow cytometry Peptide-stimulated effector cells were co-cultured with RMAHHD-GFP target cells prepulsed with or without FMP58-66 at various E/T ratios in the presence of PI. After a 4-h incubation, the frequency of PI-stained dead target cells was analyzed by flow cytometry. As shown in Fig. 1A and B, percentages of PI-stained dead cells in GFP-positive target cells increased from 14.4 to 73.3% in an E/T ratio-dependent fashion when target cells were pulsed with FMP58-66. In contrast, much less increase of PI-positive target cells was observed in the co-culture of either unpulsed targets with effectors or peptide-pulsed targets with effectors derived from Ad-WT-injected mice (Fig. 1A and B). Effector cells from either Ad-FMP- or Ad-WT-immunized mice did not significantly lyse RMA-GFP target cells (data not shown). We also carried out the FCCTL assay using RMA-HHD-FMP-GFP target cells that express endogenously processed peptides of FMP. It was observed that the
FC-CTL assay demonstrated the epitope-specific cell death of RMAHHD-FMP-GFP target cells as well (Fig. 1C). Similar results were obtained with a different epitope for C57BL/6 mice. As shown in Fig. 1D, high percentages of specific cell death were detected in the culture of effector cells with FMP128-135-pulsed target cells, but not with unpulsed targets. These data show the practical utility of the FC-CTL assay to efficiently detect antigen-specific secondary CTL responses even at low E/T ratios ranging from 1 to 10.
3.2. Evaluation of antigen-specific primary CTL responses We next investigated whether the FC-CTL assay could be applied to the analysis of antigen-specific primary CTL responses. Primary effector cells were prepared and examined for their FMP58-66specific killing activities. As shown in Fig. 2A, a high percent cell death (87.2%) of peptide-pulsed target cells was induced at an E/T ratio of 100. Furthermore, 30.9% specific cell death was detected in peptide-pulsed target cells even at an E/T ratio as low as 10, whereas only 7.7% of dead cells were observed in unpulsed target cells at the same ratio. (Fig. 2A). In contrast, any significant primary
Fig. 2. Detection of primary CTL responses in the FC-CTL assay. (A) Primary effector cells from Ad-FMP-immunized HHD mice were cultured with RMA-HHD-GFP pulsed with (upper panels) or without (lower panels) FMP58-66 in the presence of PI for 4 hs at various E/T ratios. The frequency of PI-stained target cells was analyzed by flow cytometry. Numbers shown indicate percent cell death. Left panels show FSC/SSC dot plots of the entire populations at an E/T ratio of 100. (B & C) Comparison of the FC-CTL assay with the 51Cr-release assay in primary (B) and secondary (C) CTL responses. Primary (B) and secondary (C) effector cells were prepared from Ad-FMP-immunized HHD mice. In the FC-CTL assay (solid line), cells were incubated with RMA-HHD-GFP cells pulsed with (filled circle) or without (open circle) FMP58-66 for 4 h in the presence of PI, and percent specific cell death was analyzed by flow cytometry. In the 51Cr-release assay (dashed line), cells were incubated with 51Cr-labelled RMA-HHD cells pulsed with (filled square) or without (open square) FMP58-66 for 4 h, and percent specific lysis was measured. Three mice per group were used in each experiment. Results are shown as the mean ± SD. Each experiment was repeated twice with similar results. *, p < 0.05; **, p < 0.01 compared to negative controls (open symbol) of each assay, Mann-Whitney U test.
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
5
CTL response was not observed in the 51Cr release assay even at a high E/T ratio of 100 (Fig. 2B), although this assay worked well for the secondary CTL responses (Fig. 2C). These data indicate that the FC-CTL assay is much more sensitive than the 51Cr release assay. 3.3. Time-course analyses of the FC-CTL assay Having established that the FC-CTL assay can detect primary CTL responses, we next examined how the frequency of PI-stained target cells changed with the passage of time in primary CTL responses compared to the expression of surface CD107a and intracellular IFN-g in effector cells. Primary effector cells were cultured with RMA-HHD-GFP cells pulsed with or without FMP58-66 in the presence of PI for the indicated time points. Surprisingly, PI-stained target cells were observed as early as acquisition of CD107a surface expression on CD8þ effector cells after antigenic stimulation (Fig. 3AeC), suggesting that membrane damage of target cells occurs as soon as degranulation of effector cells begins after antigen stimulation. In contrast, intracellular expression of IFN-g in CD8þ effector cells was hardly detected at 1 h post-stimulation (Fig. 3A and B), indicating that the FC-CTL assay is more sensitive than ICS. We also examined the kinetics of PD-1 expression on CD8þ effector cells after antigen stimulation. It was found that the peptidespecific expression of PD-1 started to increase as early as 30 min to 1 h following stimulation with a specific peptide (Fig. 3A), suggesting that the inhibitory signal begins to be transmitted into CD8þ effector cells soon after antigen stimulation. We examined PD-1 expression up to 26 h after antigen stimulation, and the enhancement of PD-1 expression continued until that time (data not shown). 4. Discussion A critical issue for the accurate determination in the FC-CTL assay is to clearly differentiate targets from effectors because effector cells may also undergo apoptosis. For this reason, several groups stained target cells with a fluorescent dye like PKH-26 [9] or DiOC18(3) [7]. However, it is sometimes difficult to avoid the crosscontamination to the effector population. We found that high percentages of DiOC18(3)-stained dead cells were detected by PI staining in the co-culture of peptide-stimulated effector cells and DiOC18(3)-stained target cells with no peptide (data not shown). This non-specific killing presumably resulted from the crosscontamination of DiOC18(3) to dead effector cells. In the current study, we used GFP-expressing target cell lines which were clearly distinguished from effector cells by the flow cytometry (Figs. 1 and 2). Generally, many researchers prefer the CD107a mobilization assay, ICS or tetramer staining to evaluate CTL activities rather than the FC-CTL assay even though they do not measure cytotoxicity. This is partly because the preparation of labelled target cells is cumbersome. In our study, however, the usage of pre-established GFP-expressing target cell lines enabled us to skip this process, and therefore facilitated the procedure of the FC-CTL assay. On the other hand, our method is not likely to be suitable for the use of autologous human B-lymphoblastoid cell lines (BLCLs) or peripheral blood mononuclear cells (PBMCs) as targets, which is beneficial for clinical monitoring, because classical transfection techniques generally result in poor DNA transfer into them. In addition, our method lacks flexibility because target cells are restricted to the pre-established panel of transfected cells. However, a fluorescent dye such as PKH-26 can attain any given target cells and utilize them as target cells. The FC-CTL assay was much more sensitive than the 51Cr-release assay (Fig. 2). This is partly because the FC-CTL assay detects PIstained dead cells with weak membrane alteration, whereas the
Fig. 3. Time-course analyses. (A) Primary effector cells from Ad-FMP-immunized HHD mice were used for time-course analyses. In the FC-CTL assay, effector cells were cultured with RMA-HHD-GFP cells pulsed with or without FMP58-66 at an E/T ratio of 100 in the presence of PI for the indicated time periods. In the CD107a mobilization assay combined with ICS, effector cells were stained with FITC-anti-CD107a mAb during the peptide stimulation for the indicated periods, followed by staining of surface CD8 and intracellular IFN-g. After flow cytometry analyses, % maximum responses (solid line) regarding PI-stained target cells (filled circle), CD107a-positive effector cells (open circle), and IFN-g-positive effector cells (filled triangle) were calculated. Three to five mice per group were used, and results are shown as the mean ± SD of data of mice per group. Each experiment was repeated twice with similar results. For PD-1 expression (dashed line), effector cells were incubated with RMA-HHD cells pulsed with (filled square) or without FMP58-66 (open square) at an E/T ratio of 10 for the indicated time periods, and were stained with FITC-PD-1 and PE-CD8a mAbs. Results are shown as the mean of % specific expression of two mice per group. (B) Comparison of % maximum responses regarding PI-stained GFPþ target cells, CD107a-positive effector cells, and IFN-g-positive effector cells at a time point of 1 h. **, p < 0.01; NS, not significance, One-way ANOVA. (C) Immunofluorescence microscopy images at a time point of 1 h. Left, PIþ GFPþ targets (arrows); right, CO107aþ CD8þ effectors (arrows).
51 Cr-release assay measures the release of 51Cr-labelled cytosolic proteins which requires a more severe damage of cellular membrane. Thanks to the high sensitivity, the FC-CTL assay was capable of measuring primary CTL responses (Fig. 2), and thereby may be applied for immune monitoring of clinical samples. Degranulation in CTLs occurs immediately after triggering of the T cell receptor complex [6]. Therefore, surface expression of CD107a on effector cells can be detected earlier than intracellular IFN-g expression in effector cells (Fig. 3A and B). Surprisingly, PI-stained target cells were observed as early as CD107a expression of effector cells after
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045
6
A. Takagi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
peptide stimulation (Fig. 3AeC). This data confirms that the PI/GFPbased FC-CTL assay is sufficiently sensitive to practically detect the early stages of cell death. Measurements of the cleavage of caspase 3 and annexin-V binding in target cells may be potential alternatives because they provide earlier information of apoptosis [17]. The advent of the immune checkpoint blockade has brought rays of hope to us in the cancer immunotherapy [1]. However, the exact relationship between the immune responses and tumor rejection has not fully been elucidated. Hence, a robust and appropriate method to analyze immune responses is required for the evaluation of immunotherapy. It is very unlikely that any single immunological parameter is sufficient to understand extremely complicated immune responses in the tumor microenvironment and a large amount of clinical data. Simultaneous monitoring of multiple immunological parameters is necessary and more advantageous to elucidate what types of immune responses effectively result in tumor rejection. In this respect, a multiparametric flow cytometry-based assay is a valid option in the clinical monitoring [17]. It is possible that the FC-CTL assay is combined with the 107a mobilization assay, tetramer labeling and ICS. Furthermore, immune checkpoint molecules such as PD-1 (Fig. 3A) and PD-L1 may be important parameters to measure cytotoxicity mediated by peptide-specific CTLs in the FC-CTL assay. Such comprehensive evaluation should be useful to define the correlation between tumor-specific CTL activities and clinical outcomes. In conclusion, we performed the FC-CTL assay to measure cellmediated cytotoxicity in comparison with other CTL assays. We showed that the FC-CTL assay may have a great potential for becoming a standard tool to measure CTL activity in the basic and clinical studies.
Conflict of interest None.
Acknowledgments The authors are grateful to Dr. F. Lemonnier for providing HHD II mice, RMA, and RMA-HHD cells. This work was supported by a Grant-in-Aid for Young Scientists (B) (No. 26860771) to A. T., a Grant-in-Aid for Scientific Research (C) (No. 26460560) to M. M. from Japan Society for the Promotion of Science, and grants from Saitama Medical University, Internal Grant (No. 16-B-1-17) to Y. H. and Maruki memorial special award (No. 16-A-1-01) to M. M.
Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.bbrc.2017.08.045.
References [1] I. Mellman, G. Coukos, G. Dranoff, Cancer immunotherapy comes of age, Nature 480 (2011) 480e489. [2] K.T. Brunner, J. Mauel, J.C. Cerottini, et al., Quantitative assay of the lytic action of immune lymphoid cells on 51-Cr-labelled allogeneic target cells in vitro; inhibition by isoantibody and by drugs, Immunology 14 (1968) 181e196. [3] Y. Miyahira, K. Murata, D. Rodriguez, et al., Quantification of antigen specific CD8þ T cells using an ELISPOT assay, J. Immunol. Methods 181 (1995) 45e54. [4] V.C. Maino, L.J. Picker, Identification of functional subsets by flow cytometry: intracellular detection of cytokine expression, Cytometry 34 (1998) 207e215. [5] J.D. Altman, P.A. Moss, P.J. Goulder, et al., Phenotypic analysis of antigenspecific T lymphocytes, Science 274 (1996) 94e96. [6] M.R. Betts, J.M. Brenchley, D.A. Price, et al., Sensitive and viable identification of antigen-specific CD8þ T cells by a flow cytometric assay for degranulation, J. Immunol. Methods 281 (2003) 65e78. [7] L. Piriou, S. Chilmonczyk, N. Genetet, et al., Design of a flow cytometric assay for the determination of natural killer and cytotoxic T-lymphocyte activity in human and in different animal species, Cytometry 41 (2000) 289e297. [8] B. Langhans, M. Ahrendt, J. Nattermann, et al., Comparative study of NK cellmediated cytotoxicity using radioactive and flow cytometric cytotoxicity assays, J. Immunol. Methods 306 (2005) 161e168. [9] A.E. Lee-MacAry, E.L. Ross, D. Davies, et al., Development of a novel flow cytometric cell-mediated cytotoxicity assay using the fluorophores PKH-26 and TO-PRO-3 iodide, J. Immunol. Methods 252 (2001) 83e92. [10] C.A. van Baalen, D. Kwa, E.J. Verschuren, et al., Fluorescent antigen-transfected target cell cytotoxic T lymphocyte assay for ex vivo detection of antigenspecific cell-mediated cytotoxicity, J. Infect. Dis. 192 (2005) 1183e1190. [11] K. Chen, L. Chen, P. Zhao, et al., FL-CTL assay: fluorolysometric determination of cell-mediated cytotoxicity using green fluorescent protein and red fluorescent protein expressing target cells, J. Immunol. Methods 300 (2005) 100e114. [12] S. Pascolo, N. Bervas, J.M. Ure, et al., HLA-A2.1-restricted education and cytolytic activity of CD8þ T lymphocytes from b2 microglobulin (b2m) HLAA2.1 monochain transgenic H-2Db b2m double knockout mice, J. Exp. Med. 185 (1997) 2043e2051. [13] F. Gotch, J. Rothbard, K. Howland, et al., Cytotoxic T lymphocytes recognize a fragment of influenza virus matrix protein in association with HLA-A2, Nature 326 (1987) 881e882. [14] A. Vitiello, L. Yuan, R.W. Chesnut, et al., Immunodominance analysis of CTL responses to influenza PR8 virus reveals two new dominant and subdominant Kb-restricted epitopes, J. Immunol. 157 (1996) 5555e5562. [15] T. Suda, M. Kawano, Y. Nogi, et al., The route of immunization with adenoviral vaccine influences the recruitment of cytotoxic T lymphocytes in the lung that provide potent protection from influenza A virus, Antivir. Res. 91 (2011) 252e258. [16] M. Matsui, M. Kawano, S. Matsushita, et al., Introduction of a point mutation into an HLA class I single-chain trimer induces enhancement of CTL priming and antitumor immunity, Mol. Ther. - Methods Clin. Dev. 1 (2014) 14027. [17] L. Zaritskaya, M.R. Shurin, T.J. Sayers, et al., New flow cytometric assays for monitoring cell-mediated cytotoxicity, Expert Rev. Vaccine 9 (2010) 601e616.
Please cite this article in press as: A. Takagi, et al., Characterization of the flow cytometric assay for ex vivo monitoring of cytotoxicity mediated by antigen-specific cytotoxic T lymphocytes, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/ j.bbrc.2017.08.045