Article pubs.acs.org/ac
Label-Free Electric Monitoring of Human Cancer Cells as a Potential Diagnostic Tool Clarisse Vaillier,†,‡ Thibault Honegger,*,†,‡ Frédérique Kermarrec,†,§,∥ Xavier Gidrol,†,§,∥ and David Peyrade†,‡ †
Université Grenoble Alpes, F-38000 Grenoble, France CNRS, LTM, F-38000 Grenoble, France § CEA, iRTSV, Biologie à Grande Echelle, F-38054 Grenoble, France ∥ INSERM, U1038, F-38054 Grenoble, France ‡
S Supporting Information *
ABSTRACT: Dielectrophoresis is widely used for cell characterization, and the exerted force on cells depends on the difference of polarizability between the latter and the surrounding medium. This physical phenomenon is translated by the real part of the Clausius−Mossotti factor. It is mostly modeled from the imaginary part, measured by electrorotation. The method described here measures experimentally the real part of the Clausius−Mossotti factor. It relies on the cell velocity when submitted to pure dielectrophoresis, and it was conducted on several human cell lines, at different times. A variety of cell lines was evaluated, from different organs or representative of different stages of cancer, with promising findings for early cancer detection. (Re[CMF] = 0). The first crossover frequency, f x0, occurs generally in the kHz range and depends on the medium conductivity and intrinsic cell properties, such as the cell size and the membrane capacitance.10 The second one, called f hx0, occurs at higher frequencies (MHz range) and depends mostly on the cytoplasmic conductivity.13 The first crossover frequency can be experimentally observed when the cell experiences neither pDEP nor nDEP,14 or more objectively measured using contactless dielectrophoresis.15 The cell electrophysiological properties can be inferred from electrorotation spectrum, and the Re[CMF] can be predicted (and so the crossover frequency).16 Recently, it has been pinpointed that such models do not completely agree with the Re[CMF] experimentally measured.17 Techniques to experimentally measure the Re[CMF] are rare in the literature but bring important information about the cell electrical properties. Most of them are based on competition between DEP and hydrodynamical forces, such as the voltage capture spectrum,18 the dielectrophoretic spring,19 or the wall-effect DEP.20 We present an efficient and robust method to measure the real part of the Clausius−Mossotti factor, based only on pure dielectrophoretic motion. This method was applied elsewhere on inorganic particles21 and was adapted here for human cell
T
he emergence of label-free methods for cell characterization and separation avoids the preliminary step of cell labeling, by taking advantage of intrinsic properties of cells. These methods rely on cell size,1 density,2 deformability,3 magnetic susceptibility,4 optical properties,5,6 or electrical polarizability.7 Dielectrophoresis (DEP) offers a contactless and rapid measurement of dielectric properties of cells and enables characterization, separation, and/or sorting of different types of cells. The spectrum of applications can be found in the literature.8−10 The application of DEP to the study and separation of biological cells was initiated by Pohl11 and widely used in microsystems for the past three decades. It occurs when a polarizable particle, such as a cell, is placed in a nonuniform electric field. The induced charges create a dipole which interacts with the spatial gradient of the electric field, and the unequal distribution of Columbic forces on each side results in a net force applied to the cell. The direction and magnitude of cell motion depend on the medium and cells’ properties, the electrode configuration, and the applied electric field and are expressed through the Clausius−Mossotti factor (Re[CMF]).12 According to whether the cell is more or less polarizable than the medium, the cell moves toward (positive DEP, Re[CMF] > 0) or against (nDEP, Re[CMF] < 0) the direction of high field strength regions, respectively. The frequency where the polarizability of the cell and the medium are equal and where the cell remains stationary is called the crossover frequency © 2016 American Chemical Society
Received: April 26, 2016 Accepted: August 26, 2016 Published: August 26, 2016 9022
DOI: 10.1021/acs.analchem.6b01648 Anal. Chem. 2016, 88, 9022−9028
Article
Analytical Chemistry
dioxide. PC3, JURKAT, and LnCap lines grew in RPMI standard medium, HEK in DMEM standard medium, both supplemented with 5% of fetal calf serum and penicillin− streptomycin mix (1%). The RWPE-1 cell line was obtained from ATCC (CRL-11609). Epithelial cells from the peripheral zone of a histologically normal adult human prostate were originally transfected with a plasmid carrying one copy of the human papilloma virus 18 (HPV-18) genome to establish the RWPE-1 cell line. WPE1NA22 cells (ATCC CRL-2849), WPE1-NB11 (ATCC CRL2851) and WPE1-NB26 cells (ATCC CRL-2852) cell lines were derived by exposure of RWPE-1 to N-methyl-Nnitrosourea (MNU), selected and cloned in vivo and in vitro, and characterized by prostatic epithelial and differentiation markers, karyotype analysis, anchorage-independent growth, invasiveness, tumorigenicity, and pathology of the derived tumors25. These cell lines represent increasing invasive stages of neoplastic transformation, WPE1-NA22 cells being the least malignant, and WPE1-NB26 cells the most malignant. All these lines are commercially available. The base medium for RWPE-1 and WPE-1 derivatives (MNU cells) is the Keratinocyte Serum Free Medium (KSFM), supplied with bovine pituitary extract (0.05 mg/mL) and human recombinant epidermal growth factor (5 ng/mL). PC3, LnCaP, and RWPE1 were tested and authenticated by LGC Standard, U.K. The method was the LOCI identification based on ATCC loci (PC3, LnCaP and RWPE1). The RWPE1 derivatives were tested on their capacity to form mostly spheroids instead of acini in Matrigel 3D cultures. Cells were tested at the end of February 2014 by LGC standard, U.K. The RWPE1 line was tested in 3D cultures in our laboratory in early 2015. Each sample was defrosted and grown for 1 to 2 weeks (passage 150 kV/m) or at frequencies below 1 kHz. It also has been showed that the cell viability did not decrease significantly in such low conductivity medium, even after the cells are submitted to the electric field in these conditions of field magnitude and medium conductivity.27 Harmful conditions were avoided during our experiments (20 < |E| < 150 kV/m; 1 kHz < f < 10 MHz; duration ∼30 min). We have conducted a cell viability experiment for RWPE1 cells by trypan blue staining and an automatic cell counter (EVE 1.0.3). Our results show a
