SEARCH

SEARCH BY CITATION

Keywords:

  • NF-κB;
  • leukemia;
  • ImageStream;
  • p65;
  • confocal;
  • western blot;
  • quantitative imaging

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The nuclear factor kappa B (NF-κB) pathway, which regulates many cellular processes including proliferation, apoptosis, and survival, has emerged as an important therapeutic target in cancer. Activation of the NF-κB transcription factor is associated with nuclear translocation of the p65 component of the complex. Conventional methods employed to determine nuclear translocation of NF-κB either lack statistical robustness (microscopy) or the ability to discern heterogeneity within the sampled populations (Western blotting and Gel Shift assays). The ImageStream platform combines the high image content information of microscopy with the high throughput and multiparameter analysis of flow cytometry which overcomes the aforementioned limitations of conventional assays. It is demonstrated that ImageStream assessment of receptor-mediated (TNFα) and drug (Daunorubicin, DNR)-induced NF-κB translocation in leukemic cell lines correlates well with microscopy analysis and Western blot analysis. It is further demonstrated that ImageStream cytometry enables quantitative assessment of p65 translocation in immunophenotypically defined subpopulations; and that this assessment is highly reproducible. It is also demonstrated that, quantitatively, the DNR-induced nuclear translocation of NF-κB correlates well with a biological response (apoptosis). We conclude that the ImageStream has the potential to be a powerful tool to evaluate NF-κB /p65 activity as a determinant of response to therapies designed to target aberrant NF-κB signaling activities. © 2011 International Society for Advancement of Cytometry

The increasing mechanistic knowledge of how dys-regulated signal transduction pathways contribute to the genesis of various diseases including cancer has lead to the development of therapies specifically targeting the identified abnormalities. Among the main signal transduction pathways dys-regulated in cancer is the NF-κB pathway (1).

The Nuclear Factor kappa B (NF-κB) transcription factor plays a central role in regulating many key processes in mammalian cells, including proliferation, drug resistance, and survival (2). The NF-κB transcription factor complex is held in an inactive state in the cytoplasm by binding to its inhibitor IκB. Upon activation, IκB is targeted for ubiquitination, allowing phosphorylation of the NF-κB complex, and facilitating its shuttling into the nucleus where it can bind to promoter sites to regulate transcription of target genes (3).

The NF-κB proteins include five structurally related monomers. These proteins form a variety of homo- and heterodimers in different activation states, though the NF-κB complex commonly refers to the heterodimer of p65/RelA and p50 as it is the most prevalent dimer found in cells (3). NF-κB has been found to be constitutively active in a variety of hematopoietic and solid cancers (4–9). With the development of therapies targeting NF-κB, for example with proteasome inhibitors (10), it is essential to have accurate and reproducible techniques that measure its activation and inhibition as parameters of response. Since the activation of this pathway is associated with the intracellular localization of the NF-κB complex, the nuclear localization is commonly used as a parameter of activation. Nuclear localization can be determined quantitatively by molecular approaches such as Western blotting and Gel Shift assays or qualitatively by microscopic evaluation of immuno-labeled cells (11–15). A major drawback of Western blotting and Gel Shift approaches is that they do not provide information regarding heterogeneity within a sample. For example, if a 50% decrease in signal intensity is observed, it will be impossible to tell if this is due to a 50% reduction in 100% of the population, or a 100% reduction in 50% of the population. Consequently, these molecular approaches are also not very sensitive in detecting rare events. These shortcomings are partially overcome by microscopy. An advantage of microscopic evaluation is the ability to determine heterogeneity within a sample, but quantitative visual analysis of the degree of translocation in each individual cell can be subject to operator bias. In addition, the limited number of cells that are commonly evaluated (100-200) also makes this method relatively insensitive to detecting rare events. Thus, analyzing cell images with microscopy provides detailed information but commonly lacks statistical strength. Laser Scanning Cytometry (LSC) is a microscopy-based imaging technology that improves on conventional microscopy in that it can analyse thousands of cells per slide in an automated way (16). This technique has been used to quantify NF-κB translocation (17) but since it is microscopy-based, it is particularly well-suited to study adherent cell lines and tissue sections rather than cells in suspension such as those in leukemic cell lines and blood or bone marrow.

Flow cytometry analysis provides great statistical strength by collecting data on a large number of cells, but it is unable to achieve detailed analysis of the intracellular origin of a fluorescent signal. In the study of the NF-κB pathway, in order to obviate the requirement for determining nuclear localization of the complex as a parameter of activity, flow cytometry approaches rely on the detection of phospho-specific signaling intermediaries (e.g. P-p65), the appearance of which are necessary for translocation to occur (18). A potential issue with this approach is that it is assumed that the phosphorylation is notonly necessary but also sufficient to induce nuclear translocation.

The ImageStream platform is a novel flow-cytometric technology that is able to capture high-resolution images of cells in flow at rates of up to 1,000 cells per second (19). By combining the high-throughput power of flow cytometry with its imaging capabilities, ImageStream cytometry allows for acquisition of statistically robust data on the detailed structure of cells and the location of intracellular components, effectively combining the strengths of microscopy and flow cytometry to achieve large scale detailed cell analysis (20). The results from this study validate the ability of ImageStream cytometry to quantitatively assess the cellular localization of NF-κB (p65) when compared with other established methods for determining the intracellular location of p65.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Cell Culture and NF-κB Activation

HL-60 human promyelocytic leukemia cells were cultured in Iscoves media supplemented with 20% fetal bovine serum (PAA Laboratories Pty Ltd, Queensland, Australia), 2 mM L-glutamine, 20 U/mL penicillin, and 20 μg/mL streptomycin (Mediatech Inc., Manassas, VA). ML-1 human myeloblastic leukemic cells were cultured in RPMI-1640 media supplemented with 10% fetal bovine serum, L-glutamine, penicillin, and streptomycin, as above. The cells were maintained at exponential growth at 37°C in a fully humidified atmosphere of 5% CO2 in air.

For drug treatments, cell densities were adjusted to 1 × 106 cells/mL. NF-κB translocation was activated by exposure to 10 ng/mL TNFα (Invitrogen, Carlsbad, CA) for 30 min at 37°C, unless otherwise stated. Following activation, cells were fixed for 10 min in 4% methanol-free paraformaldehyde (PFA) (Polysciences Inc, Warrington, PA) and stained as outlined below.

Antibodies and Staining

The p65 sub-unit of NF-κB was visualized by indirect labeling. Antibodies were diluted in permeabilization wash buffer (PWB) consisting of 3% fetal bovine serum, and 0.1% Triton X-100 (Sigma, St. Louis, MO) in PBS. The primary Rabbit polyclonal NF-κB/p65 antibody (SantaCruz Biotechnology Inc, Santa Cruz, CA) was diluted 1:50 in PWB and added to 1 × 106 PFA-fixed cells. Samples were incubated for 30 min at room temp. Primary antibody was removed and 1:200 dilution of secondary FITC conjugated F(ab′)2 fragment donkey anti rabbit IgG antibody (Jackson ImmunoResearch Laboratories Inc., West Grove, PA) was added and incubated at room temp in the dark for 45 min. Secondary antibody was removed and cells resuspended in 100 μL PBS. Samples that were additionally labeled with an anti-CD33 PE conjugated antibody (Immunotech, Marsalle, France) were surface stained before PFA-fixation. The antibody was diluted 1:200 in PWB, added to 1 × 106 cells, and incubated on ice for 10 min in the dark. Antibody was removed and the CD33-labeled cells underwent the same procedure for NF-κB staining as described above. Just prior to running on the ImageStream, all samples had DRAQ5 (Cell Signaling, Danvers, MA) added (20 μM final concentration), to visualize the nucleus.

ImageStream (IS) Analysis

Cells were probed for p65 using the protocol outlined above. All cells in this study were stained for anti-NF-κB/p65-FITC and DRAQ5 for nuclear imaging, unless otherwise stated. Totally, 5,000 events were collected for all samples on an ImageStream IS100 using 488 nm laser excitation. Cell populations were hierarchically gated for single cells that were in focus and were positive for both DRAQ5 and p65 as described previously by George et al., (20). After the gates were applied, 70–80% of the total number of acquired images were incorporated into the final analysis. Following data acquisition, the spatial relationship between the NF-κB and nuclear images was measured using the “Similarity” feature in the IDEAS® software package (20). The “Similarity Score” (SS), a log-transformed Pearson's correlation coefficient between the pixel values of two image pairs, provides a measure of the degree of nuclear localization of NF-κB/p65 by measuring the pixel intensity correlation between the anti- NF-κB/p65-FITC and DRAQ5 images. Cells with low similarity scores exhibit no correlation between the images (corresponding with a predominant cytoplasmic distribution of NF-κB), whereas cells with high SS scores exhibit a positive correlation between the images (corresponding with a predominant nuclear distribution of NF-κB). The relative shift in this distribution between two populations (e.g., control versus treated cells) was calculated using the Fisher's Discriminant ratio (Rd value).

The hallmark nuclear condensation of apoptotic cells results in a relative decrease of the nuclear staining area. Furthermore, due to cell shrinkage, apoptotic cells also have a higher image contrast in the bright field image. Therefore, for determination of apoptosis, cells were gated on a scatterplot of nuclear area, vs. contrast of the brightfield image (21).

Immunoblotting

Nuclear fractions were isolated using the BioVision Nuclear/Cytosol Fractionation Kit (Mountain View, CA). Proteins quantified by Lowry Assay were separated by 10% SDS-PAGE, transferred onto a nitrocellulose membrane, and blocked overnight in blocking solution (10% 10xTBS pH7.6, 0.1% Tween-20, and 5% w/v of nonfat dry milk). The membrane was then incubated with 1:200 dilution of Rabbit polyclonal anti-NF-κB/p65 antibody (SantaCruz Biotechnology Inc, Santa Cruz, CA) and 10 μL Rabbit polyclonal anti-PCNA antibody (ab2426, Abcam, Cambridge, MA) as loading control, in blocking solution for 1 h at room temp. The nitrocellulose membrane was then incubated with a 1:1,000 dilution of HRP conjugated anti-rabbit secondary antibody in blocking solution for 1 h at room temp. Proteins were detected by chemilluminescence and autoradiography. Band density was measured using a ChemiDoc XRES (Bio-Rad Laboratories, Hercules, CA) and analyzed using Quantity One software (Bio-Rad).

Confocal Imaging

Cells were probed for p65 using the protocol outlined above. Once stained, a 10 μL aliquot of the resuspended cell pellet for both a control and treated sample was placed on frosted bev-l-edge microslide (Propper Manufacturing Company, Long Island City, NY) slides in a drop of vectashield mounting medium for fluorescence (Vector Laboratories, Burlingame, CA). Micro cover glasses (VWR Scientific) were then carefully placed on top of the mixture. Slides were left stationary for 30 min to let the cells settle. Images were obtained with a TSC SP2 Leica Confocal Microscope, where the FITC labeled antibody was excited with a 488 nm line argon laser, and 200 cells for each sample were visually assessed at random.

Visual Analysis of p65 Translocation

Visual analysis of labeled NF-κB/p65 images acquired on the ImageStream was assessed by five operators. Each operator assessed 200 blinded cell images, 100 from a control sample (mean SS = −0.410) and 100 from a TNFα-treated sample (mean SS = 1.52). Cell images were scored as untranslocated (0), mid-translocation (1), or translocated (2).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

ImageStream Assessment of Concentration- and Time-Dependent Receptor-Mediated Translocation of NF-κB

The concentration-dependent effect of nuclear NF-κB/p65 translocation following 30 min activation with a TNFα dose range from 0.1 to 10 ng/mL was determined and it was found that concentrations of 5 ng/mL or higher achieved maximum NF-κB/p65 translocation (Fig. 1A). From these experiments, a concentration of 10ng/mL TNFα was determined to be sufficient to achieve maximum translocation. Time-dependent analysis of NF-κB translocation using this concentration of TNFα from 5 to 60 min indicated a rapid translocation maximizing at 30 min (Fig. 1B). Of note, a decline of NF-κB/p65 translocation at time-points greater than 30 min was also observed.

thumbnail image

Figure 1. Concentration and Time kinetics of TNFα-induced nuclear p65 translocation quantified by ImageStream cytometry. TNFα induces translocation in the ML1 cell line in a dose dependent manner with 5–10 ng/mL TNFα showing greatest translocation (A). ML1 cells incubated with 10ng/mL TNFα for 5–60 min show optimal translocation at 30 min (B). Results represent four to five independent experiments.

Download figure to PowerPoint

ImageStream Analysis Enables Quantifying Nuclear Translocation of NF-κB in Heterogenous Cell Populations

Heterogeneity is unpredictable. However, in order to demonstrate the ability of the ImageStream approach to quantify nuclear NF-κB translocation differentially within a heterogenous cell population, a predictable model system is required, particularly in order to demonstrate sensitivity. Varying percentages of 5–90% admixtures of unstimulated ML1 cells and cells stimulated by 10 ng/mL TNFα were prepared. Both the unstimulated and stimulated cells were stained for NF-κB/p65 before admixture, and the stimulated cells were additionally labeled with PE-CD33 to allow specific gating. If the total population is evaluated without discriminating the unstimulated from the stimulated cells, the percent change in Rd Value shows a strong linear correlation from 0 to 100% of treated cells with an R2 value of 0.9946 (Fig. 2A). The shift in Rd Values for each titer was normalized so that the negative control (100% unstimulated cells) represented baseline NF-κB activity, and the positive control (100% TNFα stimulated cells) represented full activation of NF-κB. The actual percentage of stimulated cells within a sample was determined by gating the CD33 positive population, and the shift in Rd Value for each titration was assessed, with the negative control set at 0% and positive control set at 100%. Therefore, the percent shift in Rd Value measured by the ImageStream correlated well with the percentage of cells in which NF-κB/p65 was translocated in the population. However, this “population” approach, which is similar to the approach used in western blot analysis, obviously does not recognize heterogeneity within a population. Specificity of the ImageStream analysis approach was determined by comparing the similarity score of TNFα-treated and untreated cells in these same admixtures (Fig. 2B), enabled by the presence of the PE-CD33 label on the treated cells. It is demonstrated that in the 5–90% range, the differences between the similarity scores of the treated and untreated cells are relatively constant, regardless of the abundance of the populations in the admixture.

thumbnail image

Figure 2. ImageStream cytometry enables quantitative assessment of nuclear p65 translocation in immunophenotypically-defined subpopulations. Unstimulated ML1 cells spiked with known concentrations of ML1 cells stimulated with 10 ng/mL TNFα, and stained additionally with CD33-PE, show a linear relationship (R2:0.9946) in change of Rd value for titrations from 0 to 100% of stimulated cells vs. control (A), and, the similarity score of the stimulated and unstimulated cells in the 5–90% admixtures is constant (B). Results represent two to four independent experiments.

Download figure to PowerPoint

Data Acquisition and Analysis of NF-κB Translocation by the ImageStream Is Highly Reproducible

To determine reproducibility of data acquisition and subsequent analysis of NF-κB translocation by the ImageStream, the similarity score variability was assessed in five independent test scenarios (Figs. 3A and 3B). First, ML1 cells were stained with NF-κB/p65-FITC and DRAQ5 and analyzed on three independent occasions. Next, ML1 cells were stained and data acquisition of 5,000 events was performed three times, then each was analyzed independently. Then, ML1 cells were stained with NF-κB/p65-FITC and DRAQ5 on three independent occasions and analyzed separately. Analysis of a single stained sample was then performed by three independent operators familiar with the ImageStream IDEAS® analysis software. Last, a single stained sample was analyzed three times by the same operator. These tests show that the Mean Similarity score measured by the ImageStream remains constant during multiple acquisitions and analyses, and that operator bias is negligible, indicating the high sensitivity and reproducibility of the ImageStream analysis approach.

thumbnail image

Figure 3. Reproducibility of quantifying NF-κB translocation by ImageStream Cytometry. Individuals familiar with the ImageStream software ran samples under the following conditions and results were averaged. Bars show from left to right; Single source of cells, stained and analyzed on three separate occasions; Single sample run three consecutive times and analyzed independently; Three individual cell sources, stained, and analyzed separately; (A); a single sample analyzed by three independent operators; Single sample analyzed three times by one operator (B).

Download figure to PowerPoint

Correlation of ImageStream Analysis and Visual Analysis

To validate the ability of the ImageStream to visually assess translocation of NF-κB/p65, a sample of control and TNFα-treated cells were split and analyzed with both confocal microscopy and ImageStream analysis (Fig. 4A). Samples acquired with the ImageStream were assessed by determining Similarity Scores and the Rd Value between the control and treated populations. Since the confocal microscope only captures the fluorescence signal from the focal plane and eliminates out-of-focus fluorescence, the images produced with confocal microscopy have lower background fluorescence than the ImageStream images. However, the visualization of NF-κB in cytoplasmic or nuclear areas is obvious in images obtained with both techniques.

thumbnail image

Figure 4. Correlation of ImageStream analysis with Visual analysis. Unstimulated ML1 cells and cells stimulated with 10 ng/mL TNFα were visualized by Confocal microscopy and by ImageStream and representative cells are shown (BF = brightfield image, FITC = p65, DRAQ5 = nucleus) (A). Scale bars in confocal images and ImageStream images represent 10 μm. The Similarity score and resultant Rd value from the ImageStream analysis is included below its images. Five individuals familiar with visual analysis of translocation events classified many images of the ML1 cell line into untranslocated, mid-translocation, or fully translocated, in complete agreement with each other, and the ImageStream (B). As summarized by the cumulative density plot, the probability that cells would be scored untranslocated or fully translocated by both the individuals and the ImageStream was highly statistically significant (ANOVA P-value = 1.62e-39) (C).

Download figure to PowerPoint

The Similarity Scores determined after assessing control and TNFα-treated cells with the ImageStream were then calculated. The mean Similarity Score for the control population was found to be −0.0177, showing the dissimilarity of FITC (NF-κB/p65) and DRAQ5 images, and therefore the absence of translocation events. The mean Similarity Score for the treated population was 1.6161, indicating a high similarity between the FITC (NF-κB/p65) and DRAQ5 images, and therefore that translocation of NF-κB/p65 to the nucleus had occurred.

To test the validity of the Similarity Score calculated with IDEAS® software to quantify translocation of NF-κB/p65, the calculated similarity score for each of 200 individual cells was compared to the visual interpretation of the corresponding ImageStream image following blinded visual analysis, and both were plotted in a probability density plot (Fig. 4B). The probability density plots were estimated using kernel density estimation; a nonparametric way of estimating the density or relative likelihood at a given point. In short, the location of the maximum in a probability density plot represents the most likely (probable) observations. The probability density of mean SS scores is studied according to operator cohorts based on agreement in operator response; the density of SS scores for the slides unanimously scored 0 is in black, unanimous 1 cohort in red, unanimous 2 cohort in green, and the disagreement cohort in blue. Figure 4B shows that there is a high probability that the visual analysis and similarity score would classify cells in a similar way. The dashed lines in Figure 4B represent the mean SS score for each unanimous operator group cohort (unanimous 0-black, unanimous 1-red, unanimous 2-green, disagreement-blue). From Figure 4B, we see the expected trend, the mean SS score for the unanimous 0 group is smallest, while the SS score for the unanimous 2 group is the largest, the distribution of the SS score for the disagreement group is centered between these extremes, but nearer to the unanimous 0 group. Therefore, we speculate that it was less likely to see user agreement in identifying an untranslocated cell as opposed to translocated cells. The range of Similarity Scores for images of cells in which user disagreement occurred, points to the issue of operator bias associated with visual analysis.

Figure 4C shows the same data summarized in a cumulative probability plot. In a cumulative probability plot we summarize the probability that the SS score will be observed at a value less than or equal to a given value on the x axis. This plot summarizes only the data for those cells/images for which the five operators unanimously agreed that they were translocated or not, and shows the probability of cells from a given category (0 =not translocated and 2 =translocated) to be scored a given similarity score. The separation (measured on the x-axis) indicated the unanimous 0 group and unanimous 2 operator group yielded significantly distinct SS scores. A statistical significance analysis of the differences between the 2 operator groups was performed (One Way ANOVA, P= 1.62e-39) which, ultimately, demonstrated the highly significant correlation between visual analysis and ImageStream analysis for these categories.

The cells in which there was user disagreement as to the translocation of NF-κB/p65 showed bias towards lower Similarity Score values, therefore it was less likely to see user agreement in identifying an untranslocated cell as opposed to translocated cells. The range of Similarity Scores for images of cells in which user disagreement occurred, points to the issue of operator bias associated with visual analysis.

Correlation of ImageStream Analysis with Western Blot

To compare Western blot analysis of NF-κB translocation events with ImageStream, titrations of TNFα treated ML-1 cells into an untreated population were prepared and analyzed for translocation by assessing the amount of NF-κB in the nucleus, and in parallel, by ImageStream similarity score analysis (Fig. 5). With increasing fractions of TNFα-treated cells, the western blot analysis qualitatively and quantitatively demonstrates the expected increase in the nuclear NF-κB/p65 (Figs. 5A and 5B). PCNA was used as a loading control for nuclear extracts and density of the NF-κB/p65 bands was corrected for the density of PCNA in Figure 5B. To compare western blot results with ImageStream analysis, a parallel analysis of the same cell admixtures was performed by the ImageStream. Since western blots provide the average results of the whole sample analyzed without the ability to discern heterogeneity, the ImageStream mean similarity scores were also determined for the whole population without taking advantage of the ability to discern a heterogeneous response. This “population” approach to the ImageStream analysis yielded a linear correlation with the percent treated cells in the admixture (Fig. 5C) that correlated well with the densitometry analysis of the western blot, yielding an R2 of 0.9136 (Fig. 5D).

thumbnail image

Figure 5. Correlation of ImageStream analysis with Western blotting. Unstimulated ML1 cells spiked with known concentrations of ML1 cells stimulated with 10 ng/mL TNFα were lysed and nuclear fractions isolated. Western blotting performed for NF-κB/p65, with PCNA as equal loading control (A). Increase in NF-κB/p65 in nuclear fraction is seen with increasing % of stimulated cells, and quantified by densitometry (B). This corresponds to increase in mean similarity score in NF-κB translocation with increasing % of stimulated cells as measured by the ImageStream (C). There is a positive correlation between band density measured by densitometry and mean similarity score from the ImageStream (R2 = 0.9136) (D).

Download figure to PowerPoint

Correlation Between ImageStream Analysis of DNR-Induced Nuclear Translocation of NF-κB/p65 and Induction of Apoptosis

Daunorubicin (DNR) is known to induce apoptosis in leukemia cells (22). HL-60 cells were exposed to DNR at 0.1–1 μM for 4 h. In addition to the DNR exposure the HL60 cells were also exposed to 10 ng/mL TNFα for 30 min as a positive control for each experiment. DNR in this dose-range induces a dose-dependent increase of nuclear translocation of NF-κB, albeit not as effectively as TNFα which could be related to their different mechanisms of action (Fig. 6A). The data in Figure 6 were normalized to the negative and positive control values for each individual data set. A nonlinear regression of the data set (five replicates) resulted in an R2 = 0.8796.

thumbnail image

Figure 6. Correlation between nuclear translocation of p65 assessed by ImageStream cytometry and biological response. The HL-60 cell line was treated with 0.1–1 μM DNR, a known NF-κB activator. Analysis shows a dose-dependent increase in NF-κB/p65 translocation measured by the ImageStream, relative to a positive control (10 ng/mL TNFα) (A). In addition to measuring NF-κB translocation, the same software can be used to analyze the biological effect of DNR: apoptosis, using the nuclear DRAQ5 stain. Analysis shows that as % apoptosis increases, so does similarity score i.e., nuclear translocation of p65 (B). Images and gating on nuclear intensity can distinguish healthy cells (C) and cells undergoing apoptosis with 1 μM DNR (D). Scale bars in ImageStream images represent 10 μm. Results of four independent experiments are shown.

Download figure to PowerPoint

In order to correlate the ImageStream analysis of nuclear NF-κB translocation with a biological response to the drug exposure, the % apoptotic cells induced following 48 h additional culture of the cells was assessed (Fig. 6B). The % of apoptotic cells was determined by quantifying the number of cells with condensed, fragmented nuclear images with the ImageStream technology. Results demonstrate a positive correlation between NF-κB/p65 translocation (Similarity Score) and % apoptosis with nonlinear regression of the data resulting in an R2 = 0.9342. Images of individual cells and corresponding scatter plots from 500 cells are shown for both the control and cells treated with 1μM DNR in Figures 6C and 6D. Apoptotic cells can clearly be imaged and gated in treated cells.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The nuclear-cytoplasmic transport of signaling proteins involved in signal transduction pathways is an important rate-limiting step in gene transcription. Thus cellular localization of these proteins can be used not only as a measure of base-line pathway activation, but also of the effectiveness of therapy aiming to restore the pathway activity in disease. The current study demonstrates that the ImageStream exhibits sensitivity and specificity in the measurement of NF-κB localization, that this measurement is highly reproducible and can distinguish heterogeneity amongst immunophenotypically defined subpopulations. Notably, this approach can identify, in parallel, a biological response associated with the translocation event, shown here by the association of NF-κB translocation and apoptosis in DNR treated cells. Importantly, analysis of NF-κB translocation by ImageStream cytometry correlates well with conventional approaches (microscopy and western blot) and overcomes many of the limitations associated with these conventional techniques.

The combination of quantitative image analysis on the individual cell level and the ability to perform this on statistically robust cell numbers has made the ImageStream platform a powerful research tool (20, 23). The data presented herein show the potential of this approach for correlative studies in conjunction with clinical trials using targeted therapies (in this case targeting NFkB) and to determine if the desired response is achieved in the (immunophenotypically defined) target cells (20). Flow cytometry is routinely used to identify malignant subpopulations in Acute Myeloid Leukemia (AML) (24, 25).

Increasing evidence suggests that targeting of specific pathways found to be overactive in cancer cells could spare healthy cells from the destructive properties of chemotherapeutic drugs (26). The NF-κB signaling pathway has shown to be over activated in leukemic stem cells (LSCs) of AML and not normal hematopoietic stem cells, and is thought to be associated with neoplastic transformation as well as resistance to chemotherapeutic intervention (27). Furthermore, since standard chemotherapeutic regimens have been shown to activate NF-κB, inhibition of NF-κB can inhibit this method of drug resistance (28). NF-κB targeted therapies are known to circumvent MDR protein-related drug resistance since they operate independent of MDR. The combination of drugs targeting NF-κB signaling pathways and standard chemotherapy has been found to increase response to therapy (29). Bortezomib is one NF-κB targeted therapy, currently in Phase II clinical trials (CALGB 10502) (30), which has been shown to increase cell death when combined with standard chemotherapy (31–33). In this study, we have shown that the ImageStream can accurately measure the relationship between translocation and apoptosis using DNR, and this approach could easily be extended to other chemotherapeutics.

ImageStream multispectral flow cytometry offers a way to accurately quantify the activity of NF-κB within immunophenotypically defined populations of cells with less bias and more statistical strength than can be achieved by visual analysis. Since novel drugs that target the NF-κB pathway are being tested in the treatment of AML, assessment of its activity can potentially be used as a parameter of response to these therapies.

The presented approach is not limited to NF-κB translocation but could be used for any factor where regulation is associated with intracellular localization. ImageStream would prove particularly useful in measuring the activity of signaling proteins of other major signal transduction pathways such as MAPK, STAT proteins and NFAT where conventional molecular and microscopy methods are currently being used (34–36).

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The authors wish to acknowledge Dr Thaddeus George (Amnis Corp) for his continuous application support with regards to the ImageStream data acquisition and analysis.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED
  • 1
    Staudt LM. Oncogenic activation of NF-kappaB. Cold Spring Harb Perspect Biol 2010; 2: a000109.
  • 2
    Prasad S,Phromnoi K,Yadav VR,Chaturvedi MM,Aggarwal BB. Targeting inflammatory pathways by flavonoids for prevention and treatment of cancer. Planta Med 2010; 76: 10441063.
  • 3
    Hoffmann A,Baltimore D. Circuitry of nuclear factor kappaB signaling. Immunol Rev 2006; 210: 171186.
  • 4
    Bargou RC,Emmerich F,Krappmann D,Bommert K,Mapara MY,Arnold W,Royer HD,Grinstein E,Greiner A,Scheidereit C, et al. Constitutive nuclear factor-kappaB-RelA activation is required for proliferation and survival of Hodgkin's disease tumor cells. J Clin Invest 1997; 100: 29612969.
  • 5
    Hussain AR,Ahmed M,Al-Jomah NA,Khan AS,Manogaran P,Sultana M,Abubaker J,Platanias LC,Al-Kuraya KS,Uddin S. Curcumin suppresses constitutive activation of nuclear factor-kappa B and requires functional Bax to induce apoptosis in Burkitt's lymphoma cell lines. Mol Cancer Ther 2008; 7: 33183329.
  • 6
    Kordes U,Krappmann D,Heissmeyer V,Ludwig WD,Scheidereit C. Transcription factor NF-kappaB is constitutively activated in acute lymphoblastic leukemia cells. Leukemia 2000; 14: 399402.
  • 7
    Mathas S,Johrens K,Joos S,Lietz A,Hummel F,Janz M,Jundt F,Anagnostopoulos I,Bommert K,Lichter P, et al. Elevated NF-kappaB p50 complex formation and Bcl-3 expression in classical Hodgkin, anaplastic large-cell, and other peripheral T-cell lymphomas. Blood 2005; 106: 42874293.
  • 8
    Nakshatri H,Bhat-Nakshatri P,Martin DA,Goulet RJ,Jr.,Sledge GWJr. Constitutive activation of NF-kappaB during progression of breast cancer to hormone-independent growth. Mol Cell Biol 1997; 17: 36293639.
  • 9
    Sandur SK,Deorukhkar A,Pandey MK,Pabon AM,Shentu S,Guha S,Aggarwal BB,Krishnan S. Curcumin modulates the radiosensitivity of colorectal cancer cells by suppressing constitutive and inducible NF-kappaB activity. Int J Radiat Oncol Biol Phys 2009; 75: 534542.
  • 10
    Reikvam H,Olsnes AM,Gjertsen BT,Ersvar E,Bruserud O. Nuclear factor-kappaB signaling: A contributor in leukemogenesis and a target for pharmacological intervention in human acute myelogenous leukemia. Crit Rev Oncog 2009; 15: 141.
  • 11
    Hu Y,Gu X,Li R,Luo Q,Xu Y. Glycogen synthase kinase-3beta inhibition induces nuclear factor-kappaB-mediated apoptosis in pediatric acute lymphocyte leukemia cells. J Exp Clin Cancer Res 2010; 29: 154.
  • 12
    Liao X,Che X,Zhao W,Zhang D,Long H,Chaudhary P,Li H. Effects of propranolol in combination with radiation on apoptosis and survival of gastric cancer cells in vitro. Radiat Oncol 2010; 5: 98.
  • 13
    Hsieh HL,Wang HH,Wu WB,Chu PJ,Yang CM. Transforming growth factor-beta1 induces matrix metalloproteinase-9 and cell migration in astrocytes: roles of ROS-dependent ERK- and JNK-NF-kappaB pathways. J Neuroinflammation 2010; 7: 88.
  • 14
    Yan M,Xu Q,Zhang P,Zhou XJ,Zhang ZY,Chen WT. Correlation of NF-kappaB signal pathway with tumor metastasis of human head and neck squamous cell carcinoma. BMC Cancer 2010; 10: 437.
  • 15
    Singh K,Chaturvedi R,Barry DP,Coburn LA,Asim M,Lewis ND,Piazuelo MB,Washington MK,Vitek MP,Wilson KT. The apolipoprotein E-mimetic peptide COG112 inhibits NF-{kappa}B signaling, proinflammatory cytokine expression, and disease activity in murine models of colitis. J Biol Chem 2008;283:16752–16761.
  • 16
    Darzynkiewicz Z,Bedner E,Li X,Gorczyca W,Melamed MR. Laser-scanning cytometry: A new instrumentation with many applications. Exp Cell Res 1999; 249: 112.
  • 17
    Deptala A,Bedner E,Gorczyca W,Darzynkiewicz Z. Activation of nuclear factor kappa B (NF-kappaB) assayed by laser scanning cytometry (LSC). Cytometry 1998; 33: 376382.
  • 18
    Armstrong L,Hughes O,Yung S,Hyslop L,Stewart R,Wappler I,Peters H,Walter T,Stojkovic P,Evans J, et al. The role of PI3K/AKT, MAPK/ERK and NFkappabeta signalling in the maintenance of human embryonic stem cell pluripotency and viability highlighted by transcriptional profiling and functional analysis. Hum Mol Genet 2006; 15: 18941913.
  • 19
    Ortyn WE,Perry DJ,Venkatachalam V,Liang L,Hall BE,Frost K,Basiji DA. Extended depth of field imaging for high speed cell analysis. Cytometry A 2007; 71A: 215231.
  • 20
    George TC,Fanning SL,Fitzgerald-Bocarsly P,Medeiros RB,Highfill S,Shimizu Y,Hall BE,Frost K,Basiji D,Ortyn WE, et al. Quantitative measurement of nuclear translocation events using similarity analysis of multispectral cellular images obtained in flow. J Immunol Methods 2006; 311: 117129.
  • 21
    Henery S,George T,Hall B,Basiji D,Ortyn W,Morrissey P. Quantitative image based apoptotic index measurement using multispectral imaging flow cytometry: A comparison with standard photometric methods. Apoptosis 2008; 13: 10541063.
  • 22
    Allouche M,Charrad RS,Bettaieb A,Greenland C,Grignon C,Smadja-Joffe F. Ligation of the CD44 adhesion molecule inhibits drug-induced apoptosis in human myeloid leukemia cells. Blood 2000; 96: 11871190.
  • 23
    Barrajon-Catalan E,Menendez-Gutierrez MP,Falco A,Carrato A,Saceda M,Micol V. Selective death of human breast cancer cells by lytic immunoliposomes: Correlation with their HER2 expression level. Cancer Lett 2010; 290: 192203.
  • 24
    Bacher U,Schnittger S,Haferlach C,Haferlach T. Molecular diagnostics in acute leukemias. Clin Chem Lab Med 2009; 47: 13331341.
  • 25
    Grimwade D,Vyas P,Freeman S. Assessment of minimal residual disease in acute myeloid leukemia. Curr Opin Oncol 2010; 22: 656663.
  • 26
    Darnell JEJr. Transcription factors as targets for cancer therapy. Nat Rev Cancer 2002; 2: 740749.
  • 27
    Guzman ML,Swiderski CF,Howard DS,Grimes BA,Rossi RM,Szilvassy SJ,Jordan CT. Preferential induction of apoptosis for primary human leukemic stem cells. Proc Natl Acad Sci USA 2002; 99: 1622016225.
  • 28
    Jani TS,DeVecchio J,Mazumdar T,Agyeman A,Houghton JA. Inhibition of NF-kappaB signaling by quinacrine is cytotoxic to human colon carcinoma cell lines and is synergistic in combination with tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) or oxaliplatin. J Biol Chem 2010; 285: 1916219172.
  • 29
    Romano MF,Avellino R,Petrella A,Bisogni R,Romano S,Venuta S. Rapamycin inhibits doxorubicin-induced NF-kappaB/Rel nuclear activity and enhances the apoptosis of melanoma cells. Eur J Cancer 2004; 40: 28292836.
  • 30
    Blum KA,Johnson JL,Niedzwiecki D,Canellos GP,Cheson BD,Bartlett NL. Single agent bortezomib in the treatment of relapsed and refractory Hodgkin lymphoma: Cancer and leukemia Group B protocol 50206. Leuk Lymphoma 2007; 48: 13131319.
  • 31
    Bai L,Xu S,Chen W,Li Z,Wang X,Tang H,Lin Y. Blocking NF-kappaB and Akt by Hsp90 inhibition sensitizes Smac mimetic compound 3-induced extrinsic apoptosis pathway and results in synergistic cancer cell death. Apoptosis 2011; 16: 4554.
  • 32
    Walsby EJ,Pratt G,Hewamana S,Crooks PA,Burnett AK,Fegan C,Pepper C. The NF-kappaB inhibitor LC-1 has single agent activity in multiple myeloma cells and synergizes with bortezomib. Mol Cancer Ther 2010; 9: 15741582.
  • 33
    Satou Y,Nosaka K,Koya Y,Yasunaga JI,Toyokuni S,Matsuoka M. Proteasome inhibitor, bortezomib, potently inhibits the growth of adult T-cell leukemia cells both in vivo and in vitro. Leukemia 2004; 18: 13571363.
  • 34
    Song YA,Park YL,Yoon SH,Kim KY,Cho SB,Lee WS,Chung IJ,Joo YE. Black tea polyphenol theaflavin suppresses LPS-induced ICAM-1 and VCAM-1 expression via blockage of NF-kappaB and JNK activation in intestinal epithelial cells. Inflamm Res 2011;60:493500.
  • 35
    Behera R,Kumar V,Lohite K,Karnik S,Kundu GC. Activation of JAK2/STAT3 signaling by osteopontin promotes tumor growth in human breast cancer cells. Carcinogenesis 2010; 31: 192200.
  • 36
    Pham LV,Tamayo AT,Yoshimura LC,Lin-Lee YC,Ford RJ. Constitutive NF-kappaB and NFAT activation in aggressive B-cell lymphomas synergistically activates the CD154 gene and maintains lymphoma cell survival. Blood 2005; 106: 39403947.