Low-level expression of HER2 and CK19 in normal peripheral blood mononuclear cells: relevance for detection of circulating tumor cells

Abstract

Background: Detection of circulating tumor cells (CTC) in the blood of cancer patients may have prognostic and predictive significance. However, background expression of 'tumor specific markers' in peripheral blood mononuclear cells (PBMC) may confound these studies. The goal of this study was to identify the origin of Cytokeratin 19 (CK19) and HER-2 signal in PBMC and suggest an approach to enhance techniques involved in detection of CTC in breast cancer patients.

Methods: PBMC from healthy donors were isolated and fractionated into monocytes, lymphocytes, natural killer cells/granulocytes and epithelial populations using immunomagnetic selection and fluorescent cell-sorting for each cell type. RNA isolated from each fraction was analyzed for CK19, HER2 and Beta 2 microglobulin (B2M) using real-time qRT-PCR. Positive selection for epithelial cells and negative selection for NK/granulocytes were used in an attempt to reduce background expression of CK19 and HER2 markers.

Results: In normal PBMC, CK19 was expressed in the lymphocyte population while HER-2 expression was highest in the NK/granulocyte population. Immunomagnetic selection for epithelial cells reduced background CK19 signal to a frequency of <5% in normal donors. Using negative selection, the majority (74-98%) of HER2 signal could be removed from PBMC. Positive selection methods are variably effective at reducing these background signals.

Conclusion: We present a novel method to improve the specificity of the traditional method of detecting CTC by identifying the source of the background signals and reducing them by negative immunoselection. Further studies are warranted to improve sensitivity and specificity of methods of detecting CTC will prove to be useful tools for clinicians in determining prognosis and monitoring treatment responses of breast cancer patients.

Figure 1

Detection of Cytokeratin 19 (CK19) by LCx in metastatic breast cancer patients. Semi-quantitative RT-PCR for CK19 and Beta2 microglobulin (B2M) was performed on PBMC after BerEP4 immunomagnetic selection for malignant epithelial cells from 10 patients with untreated metastatic breast cancer. CK19 assays were run in duplicate on two separate occasions with the average from each sample shown using error bars. B2M assays were run once per sample as previous experiments have shown the CV of duplicates to be <2%. The change in fluorescent energy serves as the reported value expressed in counts/sec/sec (c/s/s).

Figure 2

Titration of cell lines for HER2 signal. To test the sensitivity of HER2 detection in blood, real time quantitative RT-PCR was performed on RNA isolated from mock positive controls of 8 ml of leukopak cells spiked with serial dilutions of breast cancer cell lines with varying levels of HER2 expression. The control samples were immunomagnetically selected with BerEP4 antibody prior to RNA isolation. Error bars represent the standard deviation of duplicate PCR reactions. If a sample never crossed the threshold, it is plotted as zero on this graph.

Figure 3

HER2 signal after BerEP4 selection in normal donors, patient samples and SKBR3 spiked normal blood samples. Blood samples from 36 metastatic breast cancer patients, 23 normal donors and three normal donor samples spiked with 10 SKBR3 human breast cancer cells were subjected to BerEP4 immunomagnetic enrichment for epithelial cells. After RNA extraction from the positively selected cellular fraction, HER2 and B2M expression were quantitated by real time qRT-PCR. The standard curve was obtained by serial dilution of DNA from a HER2 positive breast cancer cell line (BT474). HER2 relative expression per sample was calculated by obtaining the HER2 value/B2M value ratio for each sample and then normalizing against the HER2/B2M ratio of Normal Control Sample N4, as it represented the median value for HER2 in normal samples. Error bars represent the standard deviation of triplicate reactions. Normal controls are depicted by cross hatched mark. Samples from breast cancer patients are depicted in grey bars (HER2 negative tumors) and black bars (HER2 positive tumors).

Figure 4

CK19 and HER2 RNA expression from serial immunomagnetic selection of peripheral blood mononuclear cells (PBMC) from 4 normal subjects. Subpopulations of PBMC from 4 normal subjects (C, D, E, F) were isolated by serial immunomagnetic selection with CD14 (monocytes), CD3/CD19 (lymphocytes), and CD16 (natural killer cells/granulocytes). An additional PBMC sample from each subject underwent immunomagnetic selection with BerEP4 (epithelial cells) as a negative control. MDA-MB-361 RNA was used for standard curves when detecting CK19 and HER2. RNA from normal leukocytes was used for standard curve for B2M detection. End cells: cells remaining after the serial selection. Panel A. CK19 signal detected using quantitative real-time RT-PCR and expressed as cell equivalents of MDA-MB-361 from epithelial cell and mononuclear cell subfractions. Panel B. HER2 signal detected using quantitative real-time RT-PCR and expressed as cell equivalents of MDA-MB-361 from epithelial cell and mononuclear cell subfractions.

Figure 5

A-D Flow cytometry sorting of peripheral blood mononuclear cells (PBMC) subpopulations: CD3/CD19, CD14, and CD16. PBMC were incubated with an antibody cocktail of CD3-Cy5, CD19-APC, CD16-FITC, and CD14-PE, then placed in the cell sorter. This method showed 93.7%, 96.4% and 96.5% purity for CD16, CD14 and CD3/19 fractions respectively. E-H. CK19 and HER2 signal was detected in sorted subpopulations of Peripheral Blood Mononuclear Cells. RNA from subpopulations of flow cytometry sorted cells was isolated and subjected to quantitative RT-PCR for CK19, HER2, and B2M. HER2 and CK19 expression/cell was calculated as the ratio of HER2/CK19 cell equivalents over the total number of PBL cells within the sample (determined using B2M expression). E. CK19 expression in sorted normal PBL. F. CK19 expression/cell in sorted normal PBL. G. HER2 expression in sorted normal PBL. H. HER2 expression/cell in sorted normal PBL

Figure 6

Increasing volume of CD16 beads resulted in HER2 signal depletion. Negative selection was performed on 3 PBL samples per 3 normal subjects with increasing amounts of CD16-labelled immunomagnetic beads. HER2 RNA expression was measured in both the CD16 selected cells (beads) and unselected cells (supernatant) by Real-time qRT-PCR. Data is expressed as the percent HER2 signal in each subfraction (bead or supernatant) over total HER2 signal for that subject.