JC-1, a sensitive probe for a simultaneous detection of P-glycoprotein activity and apoptosis in leukemic cells

Authors


Abstract

Background

JC-1 probe has been successfully used for the analysis of either apoptosis or P-glycoprotein (P-gp) activity. Therefore, we wanted to see if JC-1 could also simultaneously assess both, P-gp activity and apoptosis, in acute myeloid leukemia (AML) cells.

Methods

P-gp activity was measured using JC-1 and compared to the results of the Rhodamine 123 (Rh 123) assay in P-gp negative and P-gp positive cell lines, and 12 AML samples. For apoptosis, spontaneous apoptosis, as well as, apoptosis induced by Cytosine Arabinosine and Homoharringtonine were analyzed. Both mitochondrial red fluorescence and cytoplasmic green fluorescence of JC-1 with and without a P-gp inhibitor (Cyclosporine A : CsA) were used for the identification of apoptotic cells, and this was compared to Annexin V/PI staining.

Results

(1) We found a good correlation between JC-1 and Rh 123 in viable cells. Even in a small population of viable cells, P-gp positive cells emitting low red fluorescence, gained on red fluorescence after P-gp inhibition with CsA permitting an evaluation of P-gp activity. (2) We found a good correlation between the Annexin V/PI staining and JC-1 (P < 0.0001) in the assessment of apoptotic cells. Most importantly, the apoptotic cells could be distinguished by the loss of red fluorescence and the increase of green fluorescence without any change after P-gp inhibition with CsA.

Conclusions

JC-1 can simultaneously evaluate two important parameters involved in drug resistance in AML cells, P-gp activity and apoptosis. © 2006 International Society for Analytical Cytology

JC-1 (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolocarbocyanine iodide) has been initially used as a probe for cytofluorimetric analysis of the mitochondrial transmembrane potential (MTP), and characterized as the most suitable for probing mitochondria in living cells (1). In living cells, JC-1 enters the cytoplasm as a monomer, emitting at 527 nm (green fluorescence) after excitation at 490 nm, but taken up by the mitochondria, the lipophilic cationic dye accumulates within the negatively charged, hyperpolarized, mitochondrial matrix, and forms J-aggregates that emit at 590 nm (orange–red fluorescence) (2). The interchange of the two forms depends upon the cytoplasmic uptake of JC-1 and the mitochondrial accumulation as a function of MTP, resulting in the formation of J-aggregates and the shift of its emission spectrum from green to red.

JC-1 proved to be a reliable probe for the analysis of MTP changes occurring very early in apoptosis, therefore it is now also applied for the quantification of cells undergoing apoptotic cell death. Apoptotic cells were identified by the increase of green fluorescence and the loss of red fluorescence (3, 4).

On the other hand, one of the best-characterized resistance mechanisms in AML is the drug extrusion mediated by P-glycoprotein (P-gp) (5). A drug-efflux-independent role of P-gp in the increase of the leukemia cells survival was also described recently, associating P-gp with the modulation of the sphingomyelin–ceramid pathway (6), and the protection of cells from caspase dependent cell death (7). We have shown recently that JC-1 is a substrate of P-gp and that the concentrations of JC-1 depend on the activity of this transmembrane efflux protein. With JC-1, the sensitive cell lines display both green and red fluorescence. In resistant cells, with the increase of P-gp activity the green fluorescence of JC-1 decreased, and the red fluorescence was lost completely (8). Our study thus proved JC-1 to be a very sensitive probe, more sensitive than Rhodamine 123 (Rh 123), for the detection of low-level P-gp activity in adult patients with acute myeloid leukemia (AML). P-gp is an independent prognostic factor in AML patients regarding the achievement of complete remission, overall survival, and disease free survival (9, 10).

Resistance to chemotherapy can be related to both P-gp activity and the antiapoptotic mechanisms (6, 7). Recently, Van der Pol et al. (11) demonstrated that Syto 16 could be used for simultaneous analysis of P-gp activity and apoptosis. However, since P-gp function is lost in apoptotic cells, a complex gating strategy using 7-AAD positive cells and Syto 16 low staining exclusion is required to gate out dead cells and apoptotic cells so that an accurate evaluation of P-gp activity could also be performed.

Since JC-1 has proven to be a very sensitive probe for the evaluation of apoptotic cell death and P-gp activity separately, in our study we tried to use JC-1 for simultaneous evaluation of both the drug extrusion mediated by P-gp and apoptosis in AML cells without the need of a complex gating strategy.

MATERIALS AND METHODS

Cell Lines and Patients

We used five cell lines with different level of P-gp expression (12). U937, a P-gp negative cell line was used (human myelomonocytic leukemia cells). K562S (human erythroleukemia), HL60S (human promyelocytic leukemia cells), and their P-gp positive sublines K562/HHT300 and HL60DNR cell lines were also used in our studies. K562/HHT300 and HL60DNR cell lines strongly express P-gp. P-gp expression and chemotherapy resistance were maintained by adding every 2 weeks 300 ng/ml of Homoharringtonine (HHT) for the K562/HHT300 cell line, and 10−6 M of Daunorubicine (DNR) for the HL60DNR cell line (13). Twelve adult AML fresh leukemia cell samples were obtained from peripheral blood (PB) samples of AML patients after they signed the informed consent for this study. Five samples were collected at the time of diagnosis and seven were from relapsed patients. All samples selected were from patients with high PB blast counts (>80%), providing enough cells for the study of P-gp expression, activity and apoptosis at different times, in treated [Cytosine Arabinosine (Aracytine) and Homoharringtonine (HHT)] and untreated cells. As previously described (9), in AML patients with low blast level, blast cells gating was performed using the CD34 antibody (FL3 Channel; HPCA2 clone Becton Dickinson, France) or according to the physical characteristics of the blast cells when the cells did not express the characteristic markers. Mononuclear cells were isolated using the Ficoll density gradient.

Cell Cultures

Cell lines and patient cells were grown in RPMI 1640 with 10% fetal calf serum (FCS), 2 mM L-glutamine, 50 U/ml penicillin, and 50 μg/ml streptomycin and incubated at 37°C in a humidified atmosphere containing 5% CO2. The viability of cells was assessed using trypan blue dye exclusion. For the induction of apoptosis, the cells were treated with Aracytine (10−5 M) and HHT (1500 ng/ml) for 48 h. Such high drug dosage was required due to the high P-gp expression in these cells.

P-gp Expression

Fresh viable cells (5 × 105) (cell lines and patient cells) were incubated with the monoclonal antibody UIC2 (1 μg/ml; Immunotech, France), and then labeled with a secondary antibody IgG2a conjugated with phycoerythrin, as previously described (14). Fluorescence was analyzed on a flow cytometer [at the beginning of the study on a FACSort flow cytometer (Becton Dickinson, France) and more recently with EPICS Altra flow cytometer (Beckman Coulter, France)]. For each sample, 10,000 events were collected. UIC2 staining of leukemia cells was compared to the control staining with only IgG2a using the Kolmogorov–Smirnov test (12). This test generates a D value from 0 to 1 between the UIC2 and the control staining. P-gp expression was considered as positive if D > 0.20 (15).

P-gp Acitivity Detection with Rh 123

As previously described (9, 12, 16), fresh cells (5 × 105) were stained with 200 ng/ml Rh 123 for 20 min at 37°C in RPMI medium. The cells were washed twice in PBS at 4°C and resuspended in Rh 123-free medium for 1 h at 37°C with and without the P-gp inhibitor (Cyclosporine A (CsA), 2 μM) (14). Fluorescence was analyzed on a flow cytometer. In accordance with previous studies, a threshold of positivity of 0.20 (D value) was used to assess a positive activity of P-gp (9).

Apoptosis Detection Using Annexin V/Propidium Iodide

Apoptosis was assessed using the Annexin V/Propidium Iodide (PI) kit according to the manufacturer's instructions (Roche, France). Cells (5 × 105) were washed twice with PBS and suspended in Annexin V and PI solution for 15 min at room temperature. Cells were immediately analyzed on a flow cytometer.

P-gp Activity Analysis and Apoptosis Detection Using JC-1

Cells (5 × 105) were washed twice with PBS and suspended in PBS containing 0.1 μM JC-1 (Molecular Probe, France) monomer with and without CsA (from 2.10−6 to 10−5 M in high P-gp positive cells) for 1 h at 37°C. P-gp activity and apoptosis were simultaneously analyzed. Mitochondrial red fluorescence and cytoplasmic green fluorescence were analyzed on a flow cytometer. Green and red fluorescence were analyzed on the FL-1 (530-nm band-pass filter) and FL-2 (585-nm band-pass filter) channels, respectively (2, 3). P-gp activity was evaluated comparing the fluorescence with and without CsA using Kolmogorov–Smirnov test. P-gp activity was considered as positive if D > 0.20 (8, 9).

Statistical Analysis

Statview software (version 5.0) was used for the statistical analysis (SAS Institute, CA). A value of P ≤ 0.05 was considered as significant. For P-gp activity, a correlation between JC-1 and Rh 123 was assessed using the chi2 test. Regarding apoptosis, a correlation between Annexin V/PI and JC-1 was analyzed using the Spearman test.

RESULTS

JC-1 Can Detect P-gp Activity in Nonapoptotic Samples

We carried out the P-gp activity testing in leukemic cells using JC-1 with and without CsA. To elude the JC-1 fluorescence changes related to apoptosis, first we used the P-gp positive and negative untreated cell lines (Fig. 1).

Figure 1.

Flow cytometry analysis of P-gp activity with JC-1 and Rh 123 in a P-gp negative cell line [HL60S (A)] and a P-gp positive cell line [HL60DNR (B)]. A: Cells displayed both green and red fluorescence (upper left quadrant, UL) (A1), but there was no change in red and green fluorescences after the inhibition of the P-gp activity by CsA (A2) (D = 0.11 for both green (A3) and red (A4) fluorescence of JC-1 and D = 0 for Rh 123 (A5)). B: Cells with P-gp activity showed a decrease in green and loss their red fluorescence (lower left quadrant, LL) (B1). When P-gp activity was inhibited by CsA (B2), cells gained both on red and green fluorescence (UL), and the change of fluorescence (D) according to Kolmogorov–Smirnov test was 0.76 for the green (B3) and 0.79 for the red fluorescence (B4). The same high P-gp activity was found with Rh 123 (D = 0.96) (B5). The dotted and solid lines represent fluorescence of cells incubated with JC-1 alone and JC-1 in the presence of CsA, respectively.

The sensitive HL60S cells did not express P-gp (D = 0), and using JC-1, no P-gp activity was observed (D = 0.11 using both red and green fluorescence of JC-1) (Fig. 1). The resistant HL60DNR cells were highly positive for P-gp as measured both with UIC2 (D = 0.89) and with JC-1 (D = 0.76 for green and D = 0.79 for red fluorescence) (Fig. 1). Comparable results were found with Rh 123. In fact, no activity was found in HL60S cell line (D = 0) and a high P-gp activity was found in HL60DNR cell line (D = 0.96) (Fig. 1).

JC-1 Can Also Detect Apoptosis

After the evaluation of P-gp activity, we evaluated the apoptosis in cell lines and in leukemic patient cells. In Figure 2, we report the result found in a P-gp negative patient, first to demonstrate the reproducibility of this technique in leukemic patient cells and to elude the JC-1 fluorescence changes related to P-gp activity. Results were compared to the Annexin V/PI staining. Apoptosis observed in Figure 2 is the result of a spontaneous apoptosis frequently observed in P-gp negative AML patients at day 2.

Figure 2.

Flow cytometry for simultaneous analysis of P-gp activity and apoptosis with JC-1 in a P-gp negative AML patient (patient no. 5). A (Annexin-V kit): FL-1, Annexin-V FITC; FL-2, Propidium iodide (PI) fluorescence. Lower left (LL) quadrant: viable cells; lower right (LR) quadrant: early apoptotic cells; upper right (UR) quadrant: late apoptotic cells. B1 (JC-1 without CsA): Viable P-gp negative cells displayed high red fluorescence (UL quadrant). Apoptotic cells had lost red fluorescence and gained on green fluorescence (LR quadrant). B2 (JC-1 with CsA): No significant gain of both green and red fluorescence was observed in viable cells compared to B1. C1 and C2: Kolmogorov–Smirnov test confirmed the P-gp negative activity using red (C1) and green fluorescence (C2). The dotted and solid lines represent fluorescence of cells incubated with JC-1 alone and JC-1 in the presence of CsA, respectively.

On the Annexin V/PI staining, the viable cells were Annexin V and PI negative (lower left quadrant, LL). The early apoptotic cells were only Annexin V positive (lower right quadrant, LR), and the late apoptotic cells were both Annexin V and PI positive (upper right quadrant, UR).

On the JC-1 staining, the viable cells (Annexin V−, PI−) emitted strongly within the red spectrum (upper left quadrant, UL). However, the apoptotic cells [early apoptotic cells (Annexin V+, PI−) and late apoptotic cells (Annexin V+, PI+)] had gained on green fluorescence and lost on red fluorescence (LR). In the UR quadrant, we observed a population that had gained on green fluorescence but hadn't yet lost on red fluorescence. We hypothesized that this population corresponds to early apoptotic cells.

We found the same fluorescence changes in all P-gp negative AML patients. Thus, all apoptotic cells could be easily discerned from viable cells with the gain of green fluorescence.

JC-1 Can Analyze P-gp Activity and Apoptosis Simultaneously in P-gp Negative and Positive AML Patients

So, in a P-gp negative patient (Fig. 2), on staining with JC-1, viable cells were observed in UL quadrant and apoptotic cells in LR quadrant. At the same time, JC-1 being also a P-gp substrate, we looked at the apoptotic changes with CsA-inhibited P-gp during the evaluation of P-gp activity. The addition of CsA did not change any fluorescence of either viable cells (UR) or apoptotic cells (LR), which confirmed the absence of P-gp activity.

In P-gp positive patients (Fig. 3), we observed a low level of spontaneous apoptosis at day 2. Therefore, the AML fresh cells were coincubated with HHT (1500 ng/ml) and the analysis performed at day 2. In Figure 3, we report on HHT-induced apoptosis.

Figure 3.

Flow cytometry for simultaneous analysis of P-gp activity and apoptosis with JC-1 in a P-gp positive AML patient (patient no. 11). Two different levels of apoptosis were reported: spontaneous apoptosis (A, A1, A2) and Homoharringtonine-induced apoptosis (B, B1, B2). A and B (Annexin-V kit): Viable cells in LL quadrant; early apoptotic cells in LR quadrant; late apoptotic cells in UR quadrant. A1 and B1 (JC-1 without CsA): UL quadrant, viable P-gp negative cells; LL quadrant, viable P-gp positive cells discharging JC-1 and leading to a low red fluorescence staining; LR quadrant, apoptotic cells. A2 and B2 (JC-1 with CsA): Comparing to A1 and B1, viable P-gp positive cells gained red fluorescence and are now present in UL quadrant. Apoptotic cells remained in LR quadrant. Comparison of red fluorescence of JC-1 +/− CsA using Kolmogorov–Smirnov test shows a high P-gp activity: D = 0.61. In contrast, green fluorescence appeared less sensitive to evaluate P-gp activity: D = 0.10.

We observed the same changes in Annexin V/PI staining as already described for P-gp negative cells (Fig. 2).

When stained with JC-1, some of the viable cells (Annexin V−, PI−) emitted strongly within the red spectrum (UL), while other (also Annexin V−, PI−) had lost their red fluorescence to a large extent (LL). The cells undergoing apoptosis [early apoptotic cells (Annexin V+, PI−) and late apoptotic cells (Annexin V+, PI+)] had gained on green fluorescence and lost on red fluorescence (LR). A distinct population in UR quadrant was observed that only gained on green fluorescence, but so far did not loose on red fluorescence (UR). This population probably corresponds to early apoptotic cells.

On incorporation of JC-1 within the CsA-inhibited P-gp cells, all the viable cells (Annexin V−, PI−) previously emitting weakly within the red fluorescence spectrum (LL), gained on red fluorescence, and emitted strongly within the red spectrum (UL). The change of fluorescence (D) of JC-1 within the red spectrum without and with CsA was D = 0.61, which demonstrated a high P-gp activity in this sample. In these viable cells, the green fluorescence of JC-1 was not sensitive enough to evaluate the activity of P-gp. At the same time no change was observed in the fluorescence pattern of apoptotic cells [early apoptotic cells (Annexin V+, PI−) and late apoptotic cells (Annexin V+, PI+)] in either green or red fluorescence of JC-1 (LR).

So, P-gp activity and apoptosis could be evaluated also in P-gp positive samples. Actually, the green fluorescence could discern viable cells from apoptotic cells. At the same time, the P-gp positive cells presented with weak red fluorescence. Following the addition of CsA, the cells showed a gain of red fluorescence, which permitted simultaneous evaluation of P-gp activity within the red fluorescence of JC-1.

Correlation Between JC-1 and Rh 123 in the Assessment of P-gp Activity in Treated and Untreated Cells

Comparison between UIC2 positivity (P-gp expression) and JC-1 positivity (P-gp activity) is reported in Table 1. The results were concordant except for three patients (P = 0.02). As previously described, patient no. 2 with a diagnosis of AML5 did express P-gp but no P-gp activity was found (10). Patients 8 and 10 showed a kind of P-gp-like activity, but no tangible expression of P-gp was found.

Table 1. P-gp Expression and Activity in Cell Lines and 12 Samples from AML Patients
 Untreated cellsaTreated cellsb
Percentage of apoptotic cellsUIC2Rh 123Red fluorescenceGreen fluorescencePercentage of apoptotic cellsRed fluorescenceGreen fluorescence
  •  NA: data not available. Values reported correspond to D values calculated using Kolmogorov–Smirnov test.

  •  Percentage of apoptotic cells reported correspond to Annexin and Propidium iodide positive cells. In the group of treated cells, apoptosis was induced by Aracytine or HHT and we report cases with more than 50% apoptotic cells.

  •  Expression of P-gp was assessed with UIC2 monoclonal antibody.

  •  P-gp activity was evaluated with Rh 123 in untreated cells and with JC-1 in untreated and treated cells. For JC-1, results of green and red fluorescence were reported.

  • a

    P-gp activity and expression in untreated cells (spontaneous apoptosis).

  • b

    P-gp activity in treated cells.

U9377<0.20<0.20<0.20<0.2081<0.20<0.20
HL60S15<0.20<0.20<0.20<0.2059<0.20<0.20
K562S8<0.20<0.20<0.20<0.2051<0.20<0.20
HL60DNR7>0.20>0.20>0.20>0.2051>0.20>0.20
K562/HHT30030>0.20>0.20>0.20>0.2080>0.20>0.20
Patient no. 120<0.20<0.20<0.20<0.2081<0.20<0.20
Patient no. 254>0.20<0.20<0.20<0.2060<0.20<0.20
Patient no. 318<0.20<0.20<0.20<0.2077<0.20<0.20
Patient no. 425>0.20>0.20>0.20>0.2084>0.20<0.20
Patient no. 535<0.20<0.20<0.20<0.2063<0.20<0.20
Patient no. 614>0.20>0.20>0.20>0.2081>0.20>0.20
Patient no. 735NA>0.20>0.20>0.2074>0.20>0.20
Patient no. 831<0.20>0.20>0.20<0.2062>0.20<0.20
Patient no. 932<0.20NA<0.20<0.2050>0.20<0.20
Patient no. 1015<0.20NA>0.20<0.2059>0.20<0.20
Patient no. 1113>0.20NA>0.20>0.2045>0.20<0.20
Patient no. 1226<0.20>0.20<0.20<0.2088>0.20<0.20

In untreated samples of all cell lines and nine AML samples, the P-gp activity assessed with JC-1 was compared to the Rh 123 assay. A good correlation was observed between Rh 123 and both JC-1 red fluorescence (P = 0.005) and JC-1 green fluorescence (P = 0.05) (Table 1).

In treated samples with more than 50% of apoptotic cells, even in a small population of viable cells, P-gp activity could be evaluated within the red fluorescence of JC-1. In fact, the correlation between Rh 123 and the red fluorescence of JC-1 was good (P = 0.01) (Table 1). In contrast, we did not found a significant correlation between Rh 123 and the green fluorescence of JC-1 (P = 0.5) (Table 1). So, green fluorescence of JC-1 appeared to be less sensitive for the evaluation of P-gp activity in treated cells. In patients 9 and 12, we observed low P-gp activity in treated cells while there was no P-gp activity in untreated cells. The mechanisms involved are probably the same as those observed in drug resistant selected cell lines (13).

Correlation Between JC-1 and Annexin V/PI in the assessment of apoptosis

Correlation between Annexin V/PI staining and JC-1 was analyzed. A total of 51 measurements were obtained from 17 samples [fresh cells obtained from 12 patients and five cell lines (U937, K562S, HL60S, HL60DNR, and K562/HHT300)]. For each sample, we evaluated spontaneous apoptosis and apoptosis induced by Aracytine (10−5 M) and HHT (1500 ng/ml) at day 2. A good correlation was found between the JC-1 (UR and LR quadrants) and Annexin-V+/PI+ measurement of apoptotic cells (P < 0.0001, R2 = 0.81) (Fig. 4). However, the median level of apoptotic cells in the 51 samples was higher when evaluated with JC-1 than on Annexin-V/PI staining, 64.5% [range from 3 to 93] and 54.5% [range from 7 to 93], respectively.

Figure 4.

JC-1 apoptotic cells and Annexin V/IP apoptotic cells correlation. We compared percentage of Annexin V+/IP+ apoptotic cells to percentage of JC-1 apoptotic cells identified by a high green fluorescence (UR quadrant + LR quadrant).

DISCUSSION

Previously, we have shown that JC-1, using both red and green fluorescence, was more sensitive than Rh 123 in the assessment of P-gp activity in AML (9). In this study, we demonstrate that JC-1 can detect P-gp activity even in a small population of viable cells, and that the red fluorescence of JC-1 appears to be more sensitive for the evaluation of P-gp activity than the green fluorescence. Kuhnel et al. (8) have also found a higher sensitivity of the red fluorescence of JC-1 for the changes in P-gp activity. Several other P-gp substrates are being used for the evaluation of P-gp activity: Rh 123, Calcein-AM (Ca-AM), Syto 16 (9, 12, 17). So far none of them has been described as a substrate specific for P-gp. Rh 123 is a P-gp substrate, but its also a MRP1 and BCRP substrate (18, 19), and Ca-AM is a P-gp and MRP substrate (12). Until now, JC-1 and Syto 16 haven't been described as either a MRP or a BCRP substrate. Still, using drug-selected resistant cell lines (K562/HHT300, HL60DNR), we couldn't rule out the involvement of MRP, BCRP, or any other ABC protein. The discrepancy observed in patient nos. 8 and 10 who have a P-gp-like activity without P-gp expression suggests expression of another ABC protein. However, using K562 cell lines with increasing levels of resistance, we demonstrated that the level of JC-1 accumulation correlated with the level of P-gp expression as detected by UIC2 antibodies (8). Also, in AML patients, JC-1 was shown to be able to discern between three groups of patients: resistant, intermediate, and sensitive (9). In contrast, the Rh 123 assay could detect only the “resistant” group, as defined by JC-1. The “intermediate” group had an intermediate expression of P-gp compared to resistant and sensitive patients (9). Since we couldn't assess the specificity of JC-1 as a P-gp substrate, we used the combination of JC-1, a specific P-gp antibody (UIC2 immunostaining) and CsA as a P-gp inhibitor, to improve the specificity of our results. Further studies are warranted to assess if JC-1 is a substrate of MRP and/or BCRP, and/or other ABC transporter proteins.

Regarding apoptosis, JC-1 has been shown to be a much more sensitive probe than DIOC6 and Rh 123 for the analysis of the MTP (3). The studies done so far analyzing JC-1 as an apoptotic probe, performed in reference to the staining with Annexin V/PI, never attempted to establish whether it is possible to analyze simultaneously the apoptosis and P-gp activity. Including P-gp positive and P-gp negative patients, we were able to confirm the recent study demonstrating a good correlation between Annexin V/PI and JC-1 staining of apoptotic cells (4). The loss of red fluorescence observed in apoptotic cells due to the loss of mitochondrial membrane potential (Δψm) could also be the result of the activity of the P-gp discharging JC-1 in P-gp positive cells. So, red fluorescence alone is not sufficient for the identification of apoptotic cells in P-gp positive samples. In contrast, green fluorescence increased in apoptotic cells and decreased in P-gp positive cells, which allowed for an easy identification of apoptotic cells in P-gp positive samples. The median percentage of apoptotic cells when assessed by JC-1 was higher than that observed on staining with Annexin V/PI. The reason for this could be the fact that during apoptosis, the MTP changes assessed by JC-1 precede the externalization of the phosphatidylserine molecules on the membrane labeled with Annexin V/FITC (20).

A simultaneous analysis of P-gp activity and apoptosis is possible with JC-1. Compared to Syto 16 (11), the combination of JC-1 green and red fluorescence offers a better, 2-dimensional, separation of the two parameters evaluated (apoptosis and P-gp activity). Green fluorescence discriminates apoptotic cells from viable cells. Red fluorescence appears to be more sensitive for the evaluation of P-gp activity, when performed using the Kolmogorov–Smirnof test (9). This test quantified the gain of red and green fluorescence when we added a P-gp inhibitor (CsA). Thus, JC-1 identifies viable P-gp negative cells in the UL quadrant, viable P-gp positive cells in the LL quadrant (these cells “move” to the UL quadrant when the P-gp action is inhibited), and apoptotic cells in the LR and UR quadrants.

Some potential limitations of this study have to be considered when interpreting the results. The first is that JC-1 could not discriminate between the dead and the apoptotic cells, which constitutes a disadvantage compared to Syto 16 staining (11). Second, we couldn't clarify why the apoptotic cells found in UR quadrant did not lose the red fluorescence yet. We hypothesized that these cells may correspond to cells in early apoptosis. Further studies are required. Finally, since we selected only patients with high level of circulating leukemic cells, we did not use specific markers for the selection of blast cells. However, we recommend the use of antiCD34 which can easily be combined with JC-1; or, in CD34 negative cells, the use of the CD45/SSC gating strategy which identifies blasts cells by low CD45, low SSC characteristics.

P-gp activity remains a major prognostic factor in AML. Besides the well-described role of P-gp as an efflux protein, extruding drugs, and conferring resistance, recent research has suggested an additional antiapoptotic role of P-gp. Therefore, JC-1 could prove to be very useful as the probe able to monitor simultaneously both, P-gp activity and the apoptosis of leukemic cells.

Ancillary