Introduction

In different physio-pathological situations, such as embryogenesis, wound repair, and tumor invasion, isolated cells or cell populations exhibit changes to their normal behavior. Studying the spatio-temporal behavior of cell populations in cultures provides a way to assess alterations in their functions in relation to the mechanisms leading to pathology. This behavior is generally complex and involves different processes such as migration, aggregation, proliferation, adhesion, and spreading.

Here, we will concentrate on the first two of these aspects: aggregation and migration. In the general context of tumor metastasis, we are involved in the study of the specific behavior of invasive cells compared with that of noninvasive cells. We have previously shown that the sociological behavior of these two populations of cells is fundamentally different (1,2,3). Starting from a random seeding, noninvasive cells have a tendency to form clusters, while invasive cells tend to remain isolated. From a practical point-of-view, the quantitative characterization of these different properties (formation of clusters and migration) is usually very demanding.

To quantify the agglomerative behavior implies that the coordinates of all cells must be known. Since the automatic detection of cells in a field-of-view containing one or several hundred cells is not obvious, this often means that the user has to pick these cell coordinates interactively with a pointing device.

To quantify the migration characteristics of cells is even more demanding, since the knowledge of cell coordinates in all the images of a temporal series is not sufficient. What is necessary is the coordinates of each cell as a function of time. In principle, this data could be obtained by tracking procedures. However, again, automatic tracking procedures are not reliably able to cope with dense populations of cells located very close to each other, especially when non-directed migration is involved. Thus, we often rely on interactive tracking of cells rather than on automatic tracking. This is a very tedious, error-prone procedure.

The aim of the work presented here is to find an alternative to the tracking of individual cells in order to quantify the migration characteristics of cell populations in culture. The main idea is to consider the whole set of cells in the field-of-view of a microscope at once rather than considering each cell individually, and to perform statistics while computing parameters rather than applying statistical analysis after computing these parameters.

The method we propose is an extension to cell populations of a method originally proposed for the characterization of biomolecules. This method, image correlation spectroscopy (ICS), is itself an extension of another method called fluorescence correlation spectroscopy (FCS). FCS was developed to estimate macromolecule transport and concentration properties via the auto-correlation analysis of a temporal signal—the fluorescence amplitude variation arising from spontaneous fluctuations in molecular occupation number within a small volume (4,5,6).

ICS has the same goal as FCS, but relies on spatial fluctuations recorded through a series of images (7,8,9,10). The acronym STICS, for spatio-temporal image-correlation spectroscopy, has also been used (11,12). Since the acronym ICS was not very well adapted to the technique it represents, a new acronym, ICM (image-correlation microscopy), was suggested (13). Here, we adopt this terminology. It should also be mentioned that the idea of using image cross-correlation to characterize migration of molecules was described previously (14,15).

Materials and Methods

Cell Cultures

The human bronchial cell lines used in this study, 16HBE14o- (16HBE) and BZR, were derived from normal human bronchial cells immortalized after transfection with the simian virus 40 (SV40) large T-antigen gene. The BZR cell line was also infected with the v-Ha-ras oncogene. The human mammary epithelial cell line MCF-7 was obtained from ATCC (accession no. HTB-22; Manassas, VA, USA). These cell lines display different invasive potentials in vitro, as well as tumorigenicity and metastatic ability in athymic nude mice. Cells were cultured in a 5% CO₂ fully humidified atmosphere at 37°C in Dulbecco's modified Eagle's medium (DMEM) (Gibco®; Invitrogen, Grand Island, NY, USA) supplemented with penicillin, streptomycin, and 10% fetal calf serum. Into each culture dish (diameter 3.5 cm), 2×10⁵ cells/mL were plated.