284142 Investigating the Electrical Properties of Prostate Cancer Cell Lines Using Contactless Dielectrophoresis

Tuesday, October 30, 2012
Hall B (Convention Center )
Alireza Salmanzadeh1,2, Mohammad Bonakdar1,3, Michael B. Sano1, Lina Romero4, Rafael V. Davalos1 and Scott D. Cramer4, (1)Biomedical Engineering and Sciences, Virginia Tech, Blacksburg, VA, (2)Engineering Science and Mechanics, Virginia Tech, Blacksburg, VA, (3)Mechanical Engineering, Virginia Tech, Blacksburg, VA, (4)Department of Pharmacology, University of Colorado Denver, Aurora, CO

Prostate cancer is the most common malignancy in men and the second cause of cancer death among men in the United States. An early diagnosis of this cancer is crucial to decrease the mortality rate due to this cancer. Investigating the electrical and mechanical properties of cancer cells may lead to finding novel techniques to diagnose cancer in early stages and new cancer treatment methods. Dielectrophoresis (DEP), the polarity induced motion of particles in a non-uniform electric field, is a successful technique for characterization and separation of cells. The non-uniform electric field is traditionally generated by patterning metal electrodes onto the bottom of a microfluidic channel. While this technique has been tremendously successful, electrode delamination and fouling are a challenge. We have developed a relatively new technique, known as contactless dielectrophoresis (cDEP), in which metal electrodes are exchanged for conductive fluid electrode channels. These fluid electrode channels are isolated from a main sample channel by means of a thin insulating membrane. The absence of direct contact with metal electrodes provides a permissive environment for the cells and allows for studying their electrical properties.

We previously characterized the dielectrophoretic response of prostate tumor initiating cells (TICs) of a prostate cancer cell line (PC3) utilizing cDEP. An electric field generated within the cDEP microdevice using applied voltages between 0 and 300 VRMS with frequencies between 200 and 600 kHz. We have shown that the required voltage to completely trap TICs is different from that of non-TICs. This data was then used to find the optimal parameters to sort then culture enriched TIC populations. We observed that the TIC-identified cells, sorted using cDEP produced significantly more spheroids than control. Additionally, the average size of the cDEP enriched TIC spheroids was about 4 times larger than control, and about 17 times larger than cDEP enriched non-TICs spheroids.

In this study, we investigated the dielectrophoretic properties of three different prostate cancer cell lines which have previously been evaluated in the study of prostate cancer progression:  PC3, DU145, and LNCaP. The LNCaP cell line is established from human prostate adenocarcinoma and derived from the lymph node metastasis. The PC3 cell line has a high metastatic potential and was derived from the bone metastasized prostate and the DU145 cell line was derived from the brain metastasized prostate cancer.

A low frequency cDEP device was used to find the crossover frequency of these cell lines.  A silicon master stamp was first fabricated using typical photolithography techniques and deep reactive ion etching. The device was then cast in PDMS using a 10:1 ratio of monomers to curing agent. The polymer replicates were bonded to glass slides using air plasma after access holes were punched. The cells from three lines were independently suspended in a low conductivity buffer and driven through the device using a syringe pump at a flow rate of 0.005 mL/hour. A 200 VRMS signal was then applied across the device for frequencies between 5 and 100 kHz. The distribution of cells within the device was recorded and processed to determine the first crossover frequency.

We found that the crossover frequency for PC3, DU145, and LNCaP are 23.23 ± 0.51, 22.68 ± 0.64, and 31.22 ± 1.30 kHz, respectively. Crossover frequencies of PC3 and DU145 were not statistically different but both of them were significantly different from LNCaP (p<0.001 and p<0.005, respectively). We hypothesize that being derived from different types of metastasis could be the main reason for difference in their DEP response.

Extended Abstract: File Not Uploaded