SEARCH

SEARCH BY CITATION

Keywords:

  • cell signaling;
  • bladder cancer;
  • multiparameter analysis;
  • flow cytometry;
  • single cell network profiling;
  • specimen source;
  • solid tumor

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Literature Cited
  8. Supporting Information

The aim of this study was to assess the feasibility of applying the single cell network profiling (SCNP) assay to the examination of signaling networks in epithelial cancer cells, using bladder washings from 29 bladder cancer (BC) and 15 nonbladder cancer (NC) subjects. This report describes the methods we developed to detect rare epithelial cells (within the cells we collected from bladder washings), distinguish cancer cells from normal epithelial cells, and reproducibly quantify signaling within these low frequency cancer cells. Specifically, antibodies against CD45, cytokeratin, EpCAM, and cleaved-PARP (cPARP) were used to differentiate nonapoptotic epithelial cells from leukocytes, while measurements of DNA content to determine aneuploidy (DAPI stain) allowed for distinction between tumor and normal epithelial cells. Signaling activity in the PI3K and MAPK pathways was assessed by measuring intracellular levels of p-AKT and p-ERK at baseline and in response to pathway modulation; 66% (N = 19) of BC samples and 27% (N = 4) of NC samples met the “evaluable” criteria, i.e., at least 400,000 total cells available upon sample receipt with >2% of cells showing an epithelial phenotype. The majority of epithelial cells detected in BC samples were nonapoptotic and all signaling data were generated from identified cPARP negative cells. In four of 19 BC samples but in none of the NC specimens, SCNP assay identified epithelial cancer cells with a quantifiable increase in epidermal growth factor-induced p-AKT and p-ERK levels. Furthermore, preincubation with the PI3K inhibitor GDC-0941 reduced or completely inhibited basal and epidermal growth factor-induced p-AKT but, as expected, had no effect on p-ERK levels. This study demonstrates the feasibility of applying SCNP assay using multiparametric flow cytometry to the functional characterization of rare, bladder cancer cells collected from bladder washing. Following assay standardization, this method could potentially serve as a tool for disease characterization and drug development in bladder cancer and other solid tumors. © 2013 International Society for Advancement of Cytometry

Cancer displays biologic and clinical heterogeneity due to a complex range of cytogenetic and molecular aberrations that result in downstream effects on gene expression, protein function, and cell signal transduction pathways, ultimately affecting proliferation, survival, and cellular differentiation (1, 2). Elucidation of the relationship between malignancy and its underlying molecular mechanisms are essential to improve understanding of the tumorigenesis and the mechanisms of drug efficacy and resistance. Since chromosomal, genetic, epigenetic, and other molecular alterations converge at the level of protein function and cell signaling pathways, it is anticipated that tools assessing the activity of these pathways will be highly predictive of the natural history of the disease and the response to therapy.

Single cell network profiling (SCNP) assay is a multiparametric flow cytometry-based assay that simultaneously provides measurements, at the single cell level, of extracellular surface markers and quantitative changes in the activation levels of intracellular signaling proteins in response to extracellular modulators (3–7). This approach interrogates the physiology of signaling pathways and networks by measuring properties beyond those detected in resting cells, which reveals otherwise unseen functional heterogeneity in apparently morphologically and molecularly homogeneous disease groups.

Studies in hematologic malignancies have shown the value of quantitatively measuring single cell signaling networks under modulated conditions as a basis for the development of prognostic and predictive tests. Although the utility of SCNP assay has been well demonstrated in hematologic malignancies (5, 8, 9), to date, its application to solid tumors has not been widely studied, with the exception of a limited number of studies focused on lung cancer (10). The primary reason for this discrepancy is essentially preanalytic. Specifically, in hematologic malignancies, disease samples are represented primarily by aliquots of peripheral blood and/or bone marrow containing discrete numbers of cells that are live, functional, and in suspension, all critical parameters in the application of SCNP analysis by flow cytometry, which are usually absent in the standard collection of solid tumor samples. In particular, SCNP by flow cytometry requires cells in suspension and any assay performed on solid tumors would require some form of mechanical or enzymatic dissociation of the tumor into a single cell suspension. It is still unclear how this manipulation of the solid tumor would affect downstream signaling cascades. Ultimately, performing SCNP analysis on isolated circulating epithelial tumor cells from peripheral blood or bodily fluids, such as bladder washes, could provide an alternative relevant sampling source for generating functional data using SCNP in solid tumors. These data would then be applicable to understanding the biology of the disease and eventually aid in guiding clinical treatments.

The overall aim of this study was to assess the feasibility of applying the SCNP assay to the examination of signaling networks (and their modulation by activation or inhibition) in low numbers of epithelial cells present in otherwise cellularly heterogeneous samples. These data demonstrate that the SCNP assay can be applied reproducibly even on nonconventional samples such as bladder washings, allowing for the identification and characterization of epithelial cells and their response to targeted signaling modulators and inhibitors.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Literature Cited
  8. Supporting Information

Flow Sample and Specimen Description

The study was approved by the Institutional Review Board at the North Shore-Long Island Jewish (NSLIJ) Health System. In accordance with the Declaration of Helsinki, all patients provided written informed consent for the collection and use of their samples for research purposes. Clinical data were deidentified in compliance with Health Insurance Portability and Accountability Act regulations.

The study consisted of two patient groups: a bladder cancer group (N = 29) and a noncancer group (N = 15). The bladder cancer group included patients with a suspected or confirmed diagnosis of transitional cell carcinoma scheduled to undergo transurethral resection of the tumor or cystectomy (Supporting Information Table 1). The noncancer group included “control” patients with no evidence of bladder malignancy that were scheduled to undergo a diagnostic or surveillance cystoscopy. The study excluded patients younger than 18 years at time of enrollment and patients with the history of bladder cancer treated with prior local or systemic chemotherapy or immunotherapy.

Bladder washing and urine cytology specimens were collected as part of a planned surgical procedure. As part of the planned procedure, after the cystoscope was inserted through the urethra into the bladder, saline was irrigated within the bladder and 40–90 mL of the resultant saline bladder wash was collected into specimen containers. Each fluid specimen was transferred into 50-mL plastic conical sample tubes prefilled with 5 mL fetal bovine serum. Deidentified specimens were shipped on ice via overnight courier to Nodality.

Sample Processing

The sample processing and workflow is illustrated schematically in Supporting Information Figure 1.

thumbnail image

Figure 1. Gating scheme for epithelial cell identification in peripheral blood samples. HT-1376 bladder cancer cells were spiked at different concentrations ranging from 50,000 to 50 in PB samples. Epithelial cells were identified as (A) intact cells as defined by linear FSC and log SSC characteristics, (B) nucleated cells as defined by DAPI staining, (C) CD45 negative for discrimination from WBCs, and (D) positive for cytokeratin and EpCAM staining. Cell color scheme: epithelial cells (red), lymphocytes (light blue), and myeloid (dark blue).

Download figure to PowerPoint

Cell line controls

HT-1376 bladder carcinoma cell line (ATCC, Manassas, VA), a cytokeratin+ EpCAM+ and CD45low cell line known to be aneuploid, was used as a tool to develop the gating scheme for identifying rare epithelial bladder cancer cells in complex tissue by spiking whole blood or PBMC samples with decreasing concentrations of HT-1376 cells. An aliquot of HT-1376 cells was thawed and processed in parallel with fresh bladder wash samples for each SCNP assay conducted.

SCNP assay

All samples were processed upon arrival at the Nodality laboratory (1 day after collection). Bladder washes and HT-1376 cell line (used as positive control) were pelleted via centrifugation, washed with RPMI-1640 medium (Mediatech, Manassas, VA) supplemented with 1% heat-inactivated fetal bovine serum (ThermoFisher, Waltham, MA), 100 IU penicillin, 100 μg/mL streptomycin, 2 mM L-glutamine, and 25 mM HEPES (all from Mediatech) and passed through a 70-μm cell strainer (BD Biosciences, San Jose, CA). Cell counts were obtained using AcT-10 cell counter (Beckman Coulter, Brea, CA). Epidermal growth factor (EGF) stimulation was performed using 200 ng/mL EGF (PeproTech, Rocky Hill, NJ) at 37°C for 5 min. PI3K activity was inhibited with 200 nM GDC-0941 (ChemieTek, Indianapolis, IN) for 1 h at 37°C with 5% CO2. After the first 10 specimens had been analyzed, samples containing less than 4.0 × 105 cells were found to have an insufficient number of events per well to allow meaningful analysis of cytometer-acquired data (data not shown). Subsequently, samples containing fewer than 4.0 × 105 viable cells were not analyzed further.

Fixation and permeabilization

Samples were fixed by addition of either PFA to a final concentration of 1.6% (Electron Microscopy Sciences, Hatfield, PA) or PhosFlow Lyse/Fix Buffer (if red cells were visible) (BD Biosciences) to a final concentration of 1× and incubated at 37°C for an additional 10 min. After centrifugation, samples were permeabilized by addition of ice-cold 100% methanol (Sigma-Aldrich, St. Louis, MO) dropwise while vortexing. Samples were stored in methanol at −80°C for up to 30 days until further processing, consistent with previously published methods (11).

Staining and flow cytometry acquisition

Samples were removed from −80°C storage and washed twice using PBS (Mediatech) supplemented with 5 mg/mL bovine serum albumin (ThermoFisher) and 0.05% sodium azide (Sigma-Aldrich). Samples were incubated with the following directly conjugated antibodies: cytokeratin FITC, ERK1/2 (pT202/pY204) PE, EpCAM PerCP-Cy5.5, cleaved-PARP Alexa Fluor 700 (all from BD Biosciences), CD45 PE-Cy7 (Invitrogen, Carlsbad, CA), and AKT (pS473) Alexa Fluor 647 (Cell Signaling Technology, Danvers, MA). After incubation in the dark for 16 h at 4°C, samples were additionally stained with 0.5 μg/mL of 4′,6-diamidino-2-phenylindole (DAPI, Sigma-Aldrich) for 15 min at 4°C. Samples were washed twice using 1× PBS, 0.5% bovine serum albumin, and 0.5% sodium azide solution. Samples were subsequently analyzed on a LSRII flow cytometer (BD Biosciences) with three diode lasers (405, 488, and 640 nM). Data were analyzed utilizing BD Biosciences FACS Diva software version 6.1.3. Daily QC of the LSRII cytometers was performed as previously described (12).

Data analysis details

Data analysis was performed with FACSDiva software (BD Biosciences version 6.1.3) or WinList 3D software (Verify Software House version 7.0). Standard median fluorescence intensity (MFI) and the percentage of positively or negatively responding cells were used to quantify responses to modulators and inhibitors, respectively. Due to the low numbers of isolated epithelial cells and heterogeneous responses, the percent positive or negative response provided the most robust metric for quantifying results in the patient samples tested. Ploidy analysis of the epithelial cells was determined by computing the ratio of DNA staining intensity for the epithelial cells relative to the DNA staining intensity of healthy lymphocytes within the same specimen. A DNA index greater than 1.2 represented an operational cutoff for aneuploidy and was consistent with other reported studies as previously described (13, 14).

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Literature Cited
  8. Supporting Information

Identification of Rare Epithelial Cells in Cellularly Heterogeneous Tissues

The first step in the study was the definition of a reproducible, sequential gating scheme for identification of rare epithelial cancer cells in cellularly heterogeneous tissues using whole blood spiked with different numbers of HT-1376 cells. As shown in Figure 1, the gating scheme to identify nonapoptotic epithelial cells in whole blood employed a Boolean gating strategy for sequential identification of: (1) intact cells by scatter properties (Fig. 1A), (2) nucleated cells by DAPI staining (Fig. 1B), (3) med-low CD45 positive events (Fig. 1C); and (4) an “epithelial” gate, which included cleaved-PARP (cPARP) negative, EpCAM+ and cytokeratin+ cells (Fig. 1D). cPARP+ cells were excluded based on previous experience that cells with elevated levels of cPARP are nonfunctional, i.e., lack basal and induced signaling (15). Gating on cPARP negative cells (as shown in Fig. 5) removed noise and increased the robustness of the signaling metrics in particular when small numbers of cells were analyzed.

To assess the sensitivity (i.e., lower limit of epithelial cell detection) and robustness (i.e., effect of cell frequency on signaling measurements), experiments using the HT-1376 control cell line spiked into peripheral blood samples were performed. Signaling in the PI3K and MAPK pathways in response to EGF modulation was monitored by measuring the phosphorylation levels of AKT or ERK, respectively (Fig. 2A), and to demonstrate specificity of intracellular phosphorylation reagents, a PI3K specific inhibitor (GDC-0941) was also added (Fig. 2B). As expected, the PI3K inhibitor blocked induction of p-AKT in response to EGF modulation but had no effect on induction of p-ERK.

thumbnail image

Figure 2. EGF signaling pathway and EGF-modulated response in HT-1376 bladder cancer cells. A: Representative schematic of the evaluated EGF signaling pathway. EGF-induced signaling can be monitored by measuring the levels of phosphorylation of AKT and ERK as represented by the green- and red-labeled antibodies, respectively. EGF-induced phosphorylation of p-AKT is inhibited by the PI3K specific inhibitor GDC-0941. B: EGF-induced signaling in HT-1376 bladder cancer cells. p-AKT (top panel) or p-ERK (bottom panel) levels were measured in response to EGF signaling in the presence or absence of the PI3K inhibitor GDC-0941. C: EGF-induced signaling intensity for p-AKT (y-axis) measured as a factor of decreasing concentrations of gated HT-1376 bladder cancer cells (x-axis). (▪), EGF-induced p-AKT levels; ( equation image), unmodulated p-AKT levels.

Download figure to PowerPoint

As shown in Figure 2C, we were able to identify the HT-1376 cells spiked in PB samples at concentrations as low as 0.01% of total intact cells. Furthermore, the signaling metrics did not change as the spiked epithelial cell numbers decreased, demonstrating the ability to detect epithelial cells and signaling responses over a wide range of epithelial cell frequency down to as low as 0.01% epithelial cells. Taken together, the above results showed the establishment of methods to detect rare epithelial cells within heterogeneous tissue samples and reproducibly quantify signaling within these low frequency cancer cells.

Identification of Malignant Aneuploid Phenotype Within Normal Epithelial Cells

Following the development of the gating scheme for the identification of nonapoptotic epithelial cells and measurement of signaling response to EGF as described above, we assessed different methods for identification of a tumor phenotype, which could be used to identify malignant epithelial cells within a normal epithelial cell population (a phenomenon likely to occur in bladder washings from bladder cancer patients but also generalizable to other clinical situations such as pleural/peritoneal effusions). Based on the fact that ∼60% of bladder cancer washing specimens show aneuploidy (16), measurement of DNA aneuploidy by flow cytometry was chosen as a tool to distinguish tumor from normal epithelial cells. Due to compatibility with methanol-treated permeabilized cells, a DAPI stain was used. Specifically, the ratio of the MFIs between epithelial G1 and lymphocyte G1 (the latter used as internal negative control for aneuploidy), herein defined as DNA index, was measured for each sample. A DNA index above 1.2 was chosen as an operational cutoff for aneuploidy based on data from literature and data from foundational experiments (13, 14). As shown in Figure 3, the DNA index for the control aneuploidy cell line HT-1376 spiked into whole blood was 2.61. Thus, these results show that this approach provides a clear method for identifying aneuploid epithelial (presumed to be tumor derived) cells within a background of diploid epithelial and nonepithelial cells. However, it should be noted that epithelial cells identified as diploid from patients with bladder cancer may be normal epithelial or tumor-derived cells (since 40% of bladder cancer tumor do not show aneuploidy).

thumbnail image

Figure 3. Identifying DNA content within epithelial cells. DNA content of healthy lymphocytes (control) identified as blue DNA profile in top histogram. DNA content of spiked HT-1376 bladder cancer cells (test sample) identified as red DNA profile in bottom histogram. DNA index calculated as a ratio of DAPI intensity of test sample compared to DAPI intensity of healthy lymphocytes. To generate DNA index of identified epithelial cells, the G1 peak was initially set at Channel 50 for all lymphocytes identified in healthy donor and bladder cancer patient samples. Gating scheme was the same as that outlined in Figure 1.

Download figure to PowerPoint

Application of SCNP Assay to Bladder Cancer Specimens

The analytical tools described above were then applied to the analysis of primary cells from bladder washings from 29 bladder cancer patients and 15 nonbladder cancer subjects. The sample disposition is illustrated in Figure 4. Nineteen of the 29 (∼70%) bladder cancer samples and only four of the 15 (27%) nonbladder cancer samples met the “evaluable” criteria, e.g., at least 4 × 105 total cells upon sample receipt and >2% of cells with an epithelial phenotype (DAPI+, CD45 low, CK+, and EpCAM+). The majority of epithelial cells detected in the bladder cancer samples were nonapoptotic, as determined by cPARP negative cells, thus showing the feasibility of overnight sample shipment for SCNP analysis of viable epithelial cells from bladder washings. Supporting Information Figure 1 schematically illustrates the processing procedure of bladder cancer washes shipped overnight for next day processing.

thumbnail image

Figure 4. Sample disposition. Summary of control and bladder cancer samples processed during study.

Download figure to PowerPoint

In nine of 19 evaluable bladder cancer samples (and zero of the 15 noncancer samples), a DNA index (MFIE pithelial/MFIL ymphoid) ≥ 1.20 was observed by DAPI staining, suggesting aneuploidy. This frequency of occurrence is close to what was expected from the literature (16). Figure 5 shows a representative example of a bladder wash specimen with an epithelial cell subset that is aneuploid (Patient 27, Fig. 5B). Note that the lymphoid population has DAPI staining intensity set at Channel 50 and the epithelial cell population has a DAPI staining intensity between Channels 100 and 150, indicating an aneuploid phenotype. Importantly, no aneuploidy was observed in the specimens from nonbladder cancer subjects with a representative control DNA histogram also shown in Figure 5 (control patient Nodo-0D, Fig. 5A). Note that the DAPI staining for the epithelial cells from the control donor (green histogram peak) overlaps with the lymphocytes (small, blue histogram peak) at Channel 50 (Fig. 5A). The small peak for the lymphocyte population represents a lower overall number of lymphocytes when compared to the total number of epithelial cells in the same specimen. Interestingly, there was a significantly greater proportion of lymphocytes in the bladder cancer patient samples compared to the healthy donors, possibly due to an immune response and lymphocytic infiltration to the tumor site (data not shown).

thumbnail image

Figure 5. DNA content analysis of nonbladder cancer and bladder cancer donor samples. Representative control sample Nodo-0D (panel A) and bladder cancer sample Nodo-27 (panel B) with DAPI staining (DNA content) of lymphocytes and identified epithelial cells overlayed in far right histogram for each panel. Lymphocytes identified as blue peak and epithelial cells identified as green peak in far right histogram for each panel. Red arrow identifies blue lymphocyte peak in control sample. Peak height is reflective of cell numbers present. Gating scheme was the same as that outlined in Figure 1.

Download figure to PowerPoint

In four of the 19 evaluable bladder cancer samples, a quantifiable increase in EGF-induced p-AKT and p-ERK levels was identified in a subset of epithelial cells, showing functional heterogeneity within phenotypically homogeneous cell populations. Figure 6A shows a representative example (Patient 27, middle panel) beside a representative control sample (Donor Nodo-0D, top panel) showing no induced signaling for p-AKT or p-ERK. In all four positive signaling bladder cancer cases, the increase in p-AKT levels was inhibited by preincubation with GDC-0941 (Fig. 6B), a potent and specific inhibitor of PI3K. Interestingly, preincubation with GDC-0941 had differential effects on the induced levels of p-ERK in the four identified responders as shown in Figure 6B. Patient samples 16, 24, and 39 showed a minimal increase or decrease in EGF induced p-ERK, whereas Patient 27 showed a 50% reduction in p-ERK signaling response when preincubated with GDC-0941. While the inhibition of induced p-ERK signaling was unexpected, the percentage of induced p-ERK signaling in the presence of GDC-0941 was still threefold greater than the uninduced p-ERK levels for patient sample 27. Additionally, no inhibition of p-ERK was observed to the same level as p-AKT inhibition where all induced p-AKT levels were inhibited to, or below, the basal phosphorylation state (Fig. 6B). The identified heterogeneity of signaling or signaling pathway crosstalk could be potential reasons for unexpected observations with recent studies showing that inhibition of the PI3K signaling pathway affects the MEK signaling pathway and subsequent ERK phosphorylation (17, 18). Two additional bladder cancer samples were identified as nonresponders to EGF but did exhibit a reduction in basal p-AKT levels following inhibition of PI3K with GDC-0941. A representative example, Nodo-18, is shown as the bottom panel in Figure 6A. Note the high percentage of positive p-AKT cells for the first column (basal state) versus the percent positive in the second column (GDC-0941 inhibited). These values are unchanged for the third and fourth columns (EGF treated), indicating a high basal p-AKT level that can be inhibited by GDC-0941 but not induced to higher levels with EGF. The lack of induced p-AKT observed in this sample may be due to a constitutively active EGF receptor that cannot be stimulated to a higher degree by addition of the specific ligand or, alternatively, to the absence or low level expression of the EGF receptor by the tumor cells. Combined with the EGF responders, our data have identified different functional phenotypes within this set of bladder cancer samples tested, indicating a heterogeneous disease state that may require functional characterization to better understand clinical treatment and outcome.

thumbnail image

Figure 6. EGF signaling in representative control and bladder cancer samples. A: Measured EGF signaling response as determined by increase in intracellular p-AKT and p-ERK levels. In all cases, samples were preincubated at 37°C for 1 h with the PI3K inhibitor GDC-0931 where indicated and then stimulated by EGF for 5 min where indicated. Top panel, control Nodo-0D; middle panel, bladder cancer sample Nodo-27; bottom panel, bladder cancer sample Nodo-18. Black arrow in the middle panel, third dot plot indicates EGF responsive epithelial cell population. Asterisk in bottom panel first dot plot indicates the high basal p-AKT levels in a subset of epithelial cells. B: Percentage of p-AKT and p-ERK positive cells in response to the following sample treatments: (▪), unmodulated; ( equation image), GDC-0941 alone; (equation image), EGF modulated; and (equation image), EGF + GDC-0941. Quad-stat analysis regions were based on the GDC-0941 treated sample (Column 2, 6A) so that all positive regions (p-AKT+, p-ERK+, or p-AKT+/p-ERK+) were < 2% of the total population.

Download figure to PowerPoint

Of the four evaluable samples exhibiting response to EGF modulation, two were identified as aneuploid and two were identified as diploid. Based upon the current phenotyping markers used in conjunction with DAPI staining, definitive identification of the two diploid epithelial samples as normal or tumor epithelial cells is not possible. However, the lack of signaling in the majority of the healthy bladder washing samples is a strong indication that these two bladder cancer donors fall into the 40–50% of bladder cancers that are diploid.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Literature Cited
  8. Supporting Information

While the application of SCNP assay in hematologic malignancies has been shown in multiple studies to be feasible, robust, reproducible, and prognostically informative, the application of the same technology to solid tumors has yet to be realized. This is due, at least in part, to the preanalytical challenges associated with solid tumor sampling, which are intrinsically not permissive for a functional, flow cytometry-based analysis. Theoretically, if this limitation can be overcome, then the application of SCNP assay to solid tumors should be straightforward due to the largely overlapping nature of the intracellular pathways found in hematologic and solid tumor malignancies. Hematologic malignancies naturally occur for the most part as single cell suspensions, and tissues such as peripheral blood and bone marrow provide a rich source of live tumor cells for flow-based analysis. Solid tumor samples are rarely collected in live form, since rapid tissue fixation to preserve tissue structure alterations is usually needed for pathology analysis. Additionally, some of the accepted methodologies for studying solid tumors such as immunohistochemistry or genomic sequencing do not allow for functional analysis at the single cell level. While it is possible to quantitatively assess protein expression by immunohistochemistry or identify gene mutations by sequencing, these methodologies do not necessarily reflect biochemical pathway alterations and in the case of genomic sequencing do not take into account possible epigenetic factors such as DNA methylation or histone modification. Furthermore, as targeted therapies become more specific, it is critical to understand the distinct protein and biochemical pathway alterations and functional changes related to a patient's disease in order to guide a clinician in delivering the most effective treatment possible. SCNP is ideally suited to providing this type of information and allows potential testing of drugs on the tumor sample not possible with the other methodologies.

An alternative to solid tumor sampling that is more suitable to SCNP assays would be the use of samples from fluids, which contains tumor cells in suspensions such as pleural effusions and other pathologic fluids or peripheral blood [the latter also known as circulating tumor cells (CTCs)]. The utility of enumerating CTCs as a clinical prognostic tool has been demonstrated in several studies in breast, prostate, and colon cancer; yet, the current methodology does not provide any functional information regarding the biology of the identified tumor cell (19–23). This inability to evaluate relevant functional biology within CTCs is primarily due to the sample collection methodology that requires cell fixation and to the very low tumor cell recovery within the peripheral blood cell repertoire. An assay platform such as SCNP that could be used postcell enrichment to functionally evaluate identified rare CTCs is crucial for further understanding the biology and clinical relevance of CTCs. While peripheral blood is the most common source of CTCs studied to date, other pathological fluids such as pleural effusions or bladder washes may be adaptable to this platform and could be used for studying the biology of those tumor cells.

In this feasibility study, we demonstrated the applicability of the SCNP technology to solid tumors using bladder washes from bladder cancer patients as sample source. Patients with bladder cancer who have completed a cystoscopic procedure have usually low frequency of epithelial cells in the residual bladder wash fluid. Thus, using this sample source allowed us to establish methods for identification of nonapoptotic and aneuploid epithelial cells, measure basal and induced signaling (simultaneously in two parallel pathways, i.e., PI3K and MAPK pathways) in these rare cells, and detect and quantify specific signaling inhibition using targeted inhibitors. To the best of our knowledge, this is the first application of a flow cytometric-based technology platform for functional, quantitative measurement of basal and modulated signaling in bladder washings.

A novel component of this study is the simultaneous use of static and functional measurements consisting of a combination of cell lineage markers, apoptosis, DNA content, and signaling readouts to both define and functionally characterize the biology of malignant epithelial cells in the urine. Importantly, our studies to date demonstrate the ability to detect a consistent and reproducible response across a wide range of cells even at very low numbers. For practical purposes, these experiments were performed by spiking cell lines into surrogate specimens, but the robustness of the assay indicates that a similar approach should be applicable to other tumor types. We have also utilized DNA content analysis to determine if the identified epithelial cell is of tumor origin. While this approach does identify aneuploid cells as being of tumor origin, it does not allow the identification of diploid cells as normal or cancerous. Additional tumor-specific markers will need to be developed to clearly identify tumor from nontumor cells in pathological fluids such as pleural effusions or bladder washes where healthy epithelial cells may be present.

One area for assay enhancement is identification and utilization of markers for more accurately capturing epithelial cells in pathologic fluids as well as peripheral blood. Since it has been reported that a significant number of those tumor cells, in particular CTCs, do not express EpCAM and thus would be excluded from analysis when using EpCAM as a positive selection marker, alternative methodologies are necessary(24–26). One approach that has already shown success in peripheral blood is to enrich for CTCs by removing normal cells rather than positive cell selection using anti-EpCAM or anticytokeratin antibodies (27–29). This approach prevents loss of CTCs that no longer express the prototypical epithelial cell markers cytokeratin and EpCAM possibly due to epithelial to mesenchymal transition or EMT (30–32) To quantify these cells and evaluate signaling pathways via SCNP, additional phenotypic markers are required. Inclusion of mesenchymal cell markers such as vimentin and N-cadherin in conjunction with epithelial cell markers has already shown promise in identifying CTCs in multiple tumor types such as breast, prostate, and head and neck cancers when compared to positive selection of EpCAM alone (28, 33, 34). Combined use of these mesenchymal markers in the described SCNP assay could greatly enhance the cellular detection level (∼30% of samples in this pilot study were not analyzed due to low cell numbers) and allow for the quantification of distinct signaling profiles in phenotypically heterogeneous subsets of tumor cells, information with potential prognostic and predictive value (i.e., use of targeted therapies such as tyrosine kinase inhibitors, etc.).

Although significant progress has been made in the characterization of the molecular pathophysiology of bladder cancer tumorigenesis, these advancements have not resulted in the development of better drug treatments or prediction of tumor behavior. As a functional assay of tumor behavior, this technology has far reaching potential in the treatment of bladder cancer. The ability to measure basal (noninduced) and potentiated signaling levels with SCNP assay may be able to predict platinum sensitivity in patients considering neoadjuvant systemic therapy. Identification of specific pathway activation may lead to incorporation of personalized molecular targeted therapy into treatment regimens. Furthermore, improved disease characterization could aid in the identification of which patients will respond to available therapies, i.e., patient stratification. The assay development concepts shown here could also be informative with other body fluids including the potential identification and characterization of CTCs.

In conclusion, our results demonstrate the initial feasibility of applying SCNP assay to the functional characterization of bladder cancer epithelial cells. Furthermore, our study supports the potential of applying functional pathway analysis using SCNP to other solid tumors. Future development of this technology may allow for the correlation of in vitro network signaling profiles with clinical outcomes, enabling not only a prognostic tool but also furthering progress toward individualized treatment strategies.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Literature Cited
  8. Supporting Information

Conflict of Interest Disclosure: T.M.C., M.W., M.G., C.M., R.H., and A.C. are employees of and/or stockholders in Nodality, Inc.

Literature Cited

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Literature Cited
  8. Supporting Information
  • 1
    Black PC, Brown GA, Dinney CP. Molecular markers of urothelial cancer and their use in the monitoring of superficial urothelial cancer. J Clin Oncol 2006; 24: 55285535.
  • 2
    Mitra AP, Datar RH, Cote RJ. Molecular pathways in invasive bladder cancer: new insights into mechanisms, progression, and target identification. J Clin Oncol 2006; 24: 55525564.
  • 3
    Irish JM, Hovland R, Krutzik PO, Perez OD, Bruserud O, Gjertsen BT, Nolan GP. Single cell profiling of potentiated phospho-protein networks in cancer cells. Cell 2004; 118: 217228.
  • 4
    Irish JM, Kotecha N, Nolan GP. Mapping normal and cancer cell signalling networks: Towards single-cell proteomics. Nat Rev Cancer 2006; 6: 146155.
  • 5
    Kornblau SM, Minden MD, Rosen DB, Putta S, Cohen A, Covey T, Spellmeyer DC, Fantl WJ, Gayko U, Cesano A. Dynamic single-cell network profiles in acute myelogenous leukemia are associated with patient response to standard induction therapy. Clin Cancer Res 2010; 16: 37213733.
  • 6
    Krutzik PO, Nolan GP. Fluorescent cell barcoding in flow cytometry allows high-throughput drug screening and signaling profiling. Nat Methods 2006; 3: 361368.
  • 7
    Sachs K, Perez O, Pe'er D, Lauffenburger DA, Nolan GP. Causal protein-signaling networks derived from multiparameter single-cell data. Science 2005; 308: 523529.
  • 8
    Kotecha N, Flores NJ, Irish JM, Simonds EF, Sakai DS, Archambeault S, Diaz-Flores E, Coram M, Shannon KM, Nolan GP, et al. Single-cell profiling identifies aberrant STAT5 activation in myeloid malignancies with specific clinical and biologic correlates. Cancer Cell 2008; 14: 335343.
  • 9
    Nolan GP. Deeper insights into hematological oncology disorders via single-cell phospho-signaling analysis. Hematology Am Soc Hematol Educ Program 2006: 123–127, 509.
  • 10
    Lin CC, Huang WL, Su WP, Chen HH, Lai WW, Yan JJ, Su WC. Single cell phospho-specific flow cytometry can detect dynamic changes of phospho-Stat1 level in lung cancer cells. Cytometry Part A 2010; 77A: 10081019.
  • 11
    Krutzik PO, Nolan GP. Intracellular phospho-protein staining techniques for flow cytometry: Monitoring single cell signaling events. Cytometry Part A 2003; 55A: 6170.
  • 12
    Purvis N, Stelzer G. Multi-platform, multi-site instrumentation and reagent standardization. Cytometry 1998; 33: 156165.
  • 13
    Decensi A, Torrisi R, Bruno S, Costantini M, Curotto A, Nicolo G, Malcangi B, Baglietto L, Bruttini GP, Gatteschi B, et al Randomized trial of fenretinide in superficial bladder cancer using DNA flow cytometry as an intermediate end point. Cancer Epidemiol Biomarkers Prev 2000; 9: 10711078.
  • 14
    Cai T, Margallo E, Nesi G, Giubilei G, Rizzo M, Bartoletti R. Prognostic value of static cytometry in transitional cell carcinoma of the bladder: Recurrence rate and survival in a group of patients at 10 years follow-up. Oncol Rep 2006; 15: 213219.
  • 15
    Cesano A, Rosen DB, O'Meara P, Putta S, Gayko U, Spellmeyer DC, Cripe LD, Sun Z, Uno H, Litzow MR, et al. Functional pathway analysis in acute myeloid leukemia using single cell network profiling assay: Effect of specimen source (bone marrow or peripheral blood) on assay readouts. Cytometry B Clin Cytom 2012; 82B: 158172.
  • 16
    Puntoni M, Zanardi S, Branchi D, Bruno S, Curotto A, Varaldo M, Bruzzi P, Decensi A. Prognostic effect of DNA aneuploidy from bladder washings in superficial bladder cancer. Cancer Epidemiol Biomarkers Prev 2007; 16: 979983.
  • 17
    Aksamitiene E, Kholodenko BN, Kolch W, Hoek JB, Kiyatkin A. PI3K/Akt-sensitive MEK-independent compensatory circuit of ERK activation in ER-positive PI3K-mutant T47D breast cancer cells. Cell Signal 2010; 22: 13691378.
  • 18
    Serra V, Scaltriti M, Prudkin L, Eichhorn PJ, Ibrahim YH, Chandarlapaty S, Markman B, Rodriguez O, Guzman M, Rodriguez S, et al. PI3K inhibition results in enhanced HER signaling and acquired ERK dependency in HER2-overexpressing breast cancer. Oncogene 2011; 30: 25472557.
  • 19
    Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, Matera J, Miller MC, Reuben JM, Doyle GV, Allard WJ, Terstappen LW, et al. Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N Engl J Med 2004; 351: 781791.
  • 20
    Cristofanilli M, Hayes DF, Budd GT, Ellis MJ, Stopeck A, Reuben JM, Doyle GV, Matera J, Allard WJ, Miller MC, et al. Circulating tumor cells: A novel prognostic factor for newly diagnosed metastatic breast cancer. J Clin Oncol 2005; 23: 14201430.
  • 21
    Danila DC, Heller G, Gignac GA, Gonzalez-Espinoza R, Anand A, Tanaka E, Lilja H, Schwartz L, Larson S, Fleisher M, et al. Circulating tumor cell number and prognosis in progressive castration-resistant prostate cancer. Clin Cancer Res 2007; 13: 70537058.
  • 22
    de Bono JS, Scher HI, Montgomery RB, Parker C, Miller MC, Tissing H, Doyle GV, Terstappen LW, Pienta KJ, Raghavan D. Circulating tumor cells predict survival benefit from treatment in metastatic castration-resistant prostate cancer. Clin Cancer Res 2008; 14: 63026309.
  • 23
    Cohen SJ, Punt CJ, Iannotti N, Saidman BH, Sabbath KD, Gabrail NY, Picus J, Morse M, Mitchell E, Miller MC, et al. Relationship of circulating tumor cells to tumor response, progression-free survival, and overall survival in patients with metastatic colorectal cancer. J Clin Oncol 2008; 26: 32133221.
  • 24
    Deng G, Herrler M, Burgess D, Manna E, Krag D, Burke JF. Enrichment with anti-cytokeratin alone or combined with anti-EpCAM antibodies significantly increases the sensitivity for circulating tumor cell detection in metastatic breast cancer patients. Breast Cancer Res 2008; 10: R69.
  • 25
    Tarin D, Thompson EW, Newgreen DF. The fallacy of epithelial mesenchymal transition in neoplasia. Cancer Res. 2005; 65: 59966000; discussion 6000–6001.
  • 26
    Mikolajczyk SD, Millar LS, Tsinberg P, Coutts SM, Zomorrodi M, Pham T, Bischoff FZ, Pircher TJ. Detection of EpCAM-negative and cytokeratin-negative circulating tumor cells in peripheral blood. J Oncol 2011; 2011: 252361.
  • 27
    Yang L, Lang JC, Balasubramanian P, Jatana KR, Schuller D, Agrawal A, Zborowski M, Chalmers JJ. Optimization of an enrichment process for circulating tumor cells from the blood of head and neck cancer patients through depletion of normal cells. Biotechnol Bioeng 2009; 102: 521534.
  • 28
    Balasubramanian P, Lang JC, Jatana KR, Miller B, Ozer E, Old M, Schuller DE, Agrawal A, Teknos TN, Summers TAJr, et al. Multiparameter analysis, including EMT markers, on negatively enriched blood samples from patients with squamous cell carcinoma of the head and neck. PLoS One 2012; 7: e42048.
  • 29
    Liu Z, Fusi A, Klopocki E, Schmittel A, Tinhofer I, Nonnenmacher A, Keilholz U. Negative enrichment by immunomagnetic nanobeads for unbiased characterization of circulating tumor cells from peripheral blood of cancer patients. J Transl Med 2011; 9: 70.
  • 30
    Lee JM, Dedhar S, Kalluri R, Thompson EW. The epithelial-mesenchymal transition: New insights in signaling, development, and disease. J Cell Biol 2006; 172: 973981.
  • 31
    Hugo H, Ackland ML, Blick T, Lawrence MG, Clements JA, Williams ED, Thompson EW. Epithelial–mesenchymal and mesenchymal–epithelial transitions in carcinoma progression. J Cell Physiol 2007; 213: 374383.
  • 32
    Moustakas A, Heldin CH. Signaling networks guiding epithelial-mesenchymal transitions during embryogenesis and cancer progression. Cancer Sci 2007; 98: 15121520.
  • 33
    Kallergi G, Papadaki MA, Politaki E, Mavroudis D, Georgoulias V, Agelaki S. Epithelial to mesenchymal transition markers expressed in circulating tumour cells of early and metastatic breast cancer patients. Breast Cancer Res 2011; 13: R59.
  • 34
    Armstrong AJ, Marengo MS, Oltean S, Kemeny G, Bitting RL, Turnbull JD, Herold CI, Marcom PK, George DJ, Garcia-Blanco MA. Circulating tumor cells from patients with advanced prostate and breast cancer display both epithelial and mesenchymal markers. Mol Cancer Res 2011; 9: 9971007.

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Literature Cited
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
CYTO_22244_sm_SuppFig1.tiff653KSupporting Information Figure 1.
CYTO_22244_sm_SuppTab1.tiff484KSupporting Information Table 1.
MIFlowCyt-Checklist.doc58KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.