Software compensation improves the analysis of heterogeneous tumor samples stained for multiparameter DNA flow cytometry

FACS-Based Proteomics Enables Profiling of Proteins in Rare Cell Populations

Understanding disease pathology often does not require an overall proteomic analysis of clinical samples but rather the analysis of different, often rare, subpopulations of cells in a heterogeneous mixture of cell types. For the isolation of pre-specified cellular subtypes, fluorescence activated cell sorting (FACS) is commonly used for its ability to isolate the required cell populations with high purity, even of scarce cell types. The proteomic analysis of a limited number of FACS-sorted cells, however, is very challenging as both sample preparation inefficiencies and limits in terms of instrument sensitivity are present. In this study, we used CD14+CD15+ immune cells sorted out of peripheral blood mononuclear cells isolated from whole blood to improve and evaluate FACS-based proteomics. To optimize both the protein extraction protocol and the mass spectrometry (MS) data acquisition method, PBMCs as well as commercialized HeLa digest were used. To reflect the limited number of sorted cells in some clinical samples, different numbers of sorted cells (1000, 5000, 10,000, or 50,000) were used. This allowed comparing protein profiles across samples with limited protein material and provided further insights in the benefits and limitations of using a very limited numbers of cells.

Data-driven compensation for flow cytometry of solid tissues

Propidium Iodide is a fluorochrome that is used to measure the DNA content of individual cells, taken from solid tissues, with a flow cytometer. Compensation for spectral cross-over of this fluorochrome still leads to compensation results that are depending on operator experience. We present a data-driven compensation (DDC) algorithm that is designed to automatically compensate combined DNA phenotype flow cytometry acquisitions. The generated compensation values of the DDC algorithm are validated by comparison with manually determined compensation values. The results show that (1) compensation of two-color flow cytometry leads to comparable results using either manual compensation or the DDC method; (2) DDC can calculate sample-specific compensation trace lines; (3) the effects of two different approaches to calculate compensation values can be visualized within one sample. We conclude that the DDC algorithm contributes to the standardization of compensation for spectral cross-over in flow cytometry of solid tissues.

Filter selection for five color flow cytometric analysis with a single laser

Flow cytometry has evolved from single- and two-color analysis to the current use of 11-16 colors. The relatively bright excitation spectra of most fluorochromes have made color compensation a challenge especially when performed manually. We describe how by choosing filters with narrower bandwidths results in the color compensation values between FITC, PE, PE-TxR (ECD), PE-Cy5, and PE-Cy7 that range from 0 % to 50% depending on the combination of fluorochromes. Peripheral blood mononuclear cells were stained with alpha-CD4-FITC, alpha-CD27-PE, alpha-CD62L-ECD, alpha-CD45RA-PE-Cy5 and alpha-CD3-PE-Cy7. The samples were acquired on a MO Flo. The initial (first) and second filter sets for our experiments consisted of 530/30 or 519/20 for FITC, 580/30 or 575/20for PE, 630/30 or 630/22 for PE-TxR (ECD), 670/30 or 675/20 for PE-Cy5 and 740LP or 780/40 for PE-Cy7. Nonstained cells were used to adjust the threshold values of detection for each photo multiplier tube (PMT) for each filter set. The mean fluorescent intensity (MFI) of each fluorochrome was not reduced to any great extent by either filter set. However, the compensation value between PE and PE-TxR (ECD) with the first filter selection ranged from 84% to 89% and with the second set of filters it was 25-36%. In addition, the compensation between PE-TxR (ECD) and PE-Cy5 were reduced to 30.2% from 44.2% with the second filter set. The reduction of filter bandwidths that results in minimizing spectral overlaps without lost of signal provides a method by which discrimination of signals between PE containing fluorochromes can be achieved.

High-resolution multi-parameter DNA flow cytometry enables detection of tumour and stromal cell subpopulations in paraffin-embedded tissues

The accuracy of DNA ploidy measurements of paraffin-embedded tissues is limited by the lack of resolution and the inability to identify the DNA diploid population unequivocally in bimodal DNA histograms. A multi-parameter DNA flow cytometric method has been developed that enables the simultaneous detection of neoplastic and stromal cells in samples from dewaxed 50 microm sections or 2 mm diameter punches of archival tissue blocks. The method combines heat pretreatment in sodium citrate buffer and subsequent enzymatic dissociation with a collagenase/dispase mixture. Cells were simultaneously stained for keratin (FITC), vimentin (R-PE), and DNA (PI) before flow cytometric analysis. The method was applied to 12 paraffin-embedded cervical carcinomas and four colorectal carcinomas. In all cervical cancers, distinct keratin-positive and vimentin-positive cell populations were observed. While the exclusive vimentin-positive cell fractions always yielded unimodal DNA content distributions, bimodal distributions were observed for the keratin-positive cell fractions in nine cervical carcinomas, whereas one cervical carcinoma showed three distinct G0G1 populations. Coefficients of variation of the G0G1 peaks ranged from 1.70% to 4.79%. Average background, aggregate, and debris values were 14.7% (vimentin-positive fraction) and 33.8% (keratin-positive fraction). Flow sorting confirmed that the exclusively vimentin-positive cell fractions represent different normal stromal and infiltrate cells that can serve as an internal ploidy reference enabling discrimination between DNA hypo-diploid and DNA hyper-diploid tumour cell subpopulations. The neoplastic origin of the keratin-vimentin co-expressing cells from two cervical carcinomas was confirmed by genotyping of flow-sorted samples revealing loss of heterozygosity (LOH) of 6p. This improved method obviates the need for fresh/frozen tumour tissue for high-resolution DNA ploidy measurements and enables the isolation of highly purified tumour subpopulations for subsequent genotyping.