A novel circular invasion assay mimics in vivo invasive behavior of cancer cell lines and distinguishes single-cell motility in vitro

Yoonseok Kam^*¹ , Cherise Guess^*¹^,2 , Lourdes Estrada¹ , Brandy Weidow¹ and Vito Quaranta¹

¹Department of Cancer Biology, Vanderbilt University School of Medicine, Nashville, TN, USA

²Department of Microbiology, Meharry Medical College, Nashville, TN, USA

author email corresponding author email* Contributed equally

BMC Cancer 2008, 8:198doi:10.1186/1471-2407-8-198


Published:	14 July 2008

Abstract

Background

Classical in vitro wound-healing assays and other techniques designed to study cell migration and invasion have been used for many years to elucidate the various mechanisms associated with metastasis. However, many of these methods are limited in their ability to achieve reproducible, quantitative results that translate well in vivo. Such techniques are also commonly unable to elucidate single-cell motility mechanisms, an important factor to be considered when studying dissemination. Therefore, we developed and applied a novel in vitro circular invasion assay (CIA) in order to bridge the translational gap between in vitro and in vivo findings, and to distinguish between different modes of invasion.

Method

Our method is a modified version of a standard circular wound-healing assay with an added matrix barrier component (Matrigel™), which better mimics those physiological conditions present in vivo. We examined 3 cancer cell lines (MCF-7, SCOV-3, and MDA-MB-231), each with a different established degree of aggressiveness, to test our assay's ability to detect diverse levels of invasiveness. Percent wound closure (or invasion) was measured using time-lapse microscopy and advanced image analysis techniques. We also applied the CIA technique to DLD-1 cells in the presence of lysophosphatidic acid (LPA), a bioactive lipid that was recently shown to stimulate cancer cell colony dispersal into single migratory cells, in order to validate our method's ability to detect collective and individual motility.

Results

CIA method was found to be highly reproducible, with negligible levels of variance measured. It successfully detected the anticipated low, moderate, and high levels of invasion that correspond to in vivo findings for cell lines tested. It also captured that DLD-1 cells exhibit individual migration upon LPA stimulation, and collective behavior in its absence.

Conclusion

Given its ability to both determine pseudo-realistic invasive cell behavior in vitro and capture subtle differences in cell motility, we propose that our CIA method may shed some light on the cellular mechanisms underlying cancer invasion and deserves inclusion in further studies. The broad implication of this work is the development of a reproducible, quantifiable, high-resolution method that can be applied to various models, to include an unlimited number of parameters and/or agents that may influence invasion.