Cell chemotaxis is frequently required in normal or pathological situations such as invasion, metastasis, and tumor angiogenesis and may involve many different cell types. At present, no device can simultaneously (i) make morphological observations, (ii) quantify cell migration, (iii) test multiple chemoattracting gradients, and (iv) analyze cell-cell interactions. We developed an agarose-based assay to address these questions. Two glass molds were designed, around which agarose gel could be poured to form specific well shapes. Using a vital nuclear stain (Hoechst 33258), we characterized the migration profile of adherent or suspension cells. Cells could be observed during the entire migration process. We were able to follow cells moving toward chemoattractants or being repulsed by other molecules, and we could estimate average migration speed. Using this inexpensive assay, we were able to obtain precise, reproducible results concerning the chemotactic behavior of different cell types. The resulting data differentiated between chemokinetic and chemotactic movement. Chemotactic potencies could be compared using different criteria, such as the number of attracted cells, induced speed, and morphological aspect. This improved agarose assay appears to be a reliable and inexpensive alternative to other available chemotaxis study tools.

Introduction

Mural cell motility, for cells such as pericytes and vascular smooth muscle cells (VSMCs) is frequently required for pathological processes such as tumor angiogenesis or atherosclerosis (1,2,3,4). In order to relate experimental in vitro results to in vivo conditions, it is crucial to use the appropriate experimental approach to elucidate the mechanisms driving cell migration. Among the variety of systems used to assess chemotaxis, the most commonly used is the modified Boyden chamber assay and its variations (5,6,7,8). It consists of two superposed cell culture chambers separated by a filter that allows cells to pass through. Once the cells placed in the upper chamber have migrated through pores in the filter to the lower one, they are collected and counted. It is a prevailing method to assess the number of cells attracted by chemoattractant (9), and it is easy to differentiate chemokinetic and chemotactic movements (10). The Zigmond chamber is a glass slide in which two wells are separated by a bridge where cells are seeded (11). It was designed for direct viewing of small amounts of cells (12). Similar to the Zigmond chamber, the Dunn chamber was developed to address chemoattractant gradient prediction and experimental reliability for slowly migrating cells (13,14). Since the late 1970s, different under-agarose assays have been developed to assess chemotaxis (15). With this technique, cells are placed in wells cut in the agarose gel or on the surface of the gel (16,17). The under-agarose assay has mainly been used for studying blood cells such as neutrophils (18,19,20). Chemotaxis studies of other cell types using the agarose assay are rare (21,22) mainly because of the difficulty of data analysis due to the agarose gel, which makes cell observation difficult.

In our case, we needed a tool that enabled us to understand and to precisely describe VSMC chemotactic behavior. For this reason, our experimental approach had to fulfill the following expectations. We wanted to be able to (i) observe the cells at any time; (ii) quantify induced migration; (iii) compare the effect of different factors; and (iv) analyze cellular interactions. Although the modified Boyden chamber is still the most popular chemotaxis study tool, as it is available in multi-well plates, giving ready-to-use and reproducible experimental conditions, it is difficult to follow cells during their migration (23) and to evaluate the influence of gravity since it employs vertical displacement. Moreover, cells cannot be observed microscopically or immunostained during a particular step of the migration. In the case of the Zigmond or Dunn chambers, one chamber is needed for each experimental condition, which makes this a long and expensive procedure when many conditions are tested.

Agarose gels are simple to make, can be modified to specific experimental needs, and are quite inexpensive compared to other commercial devices. Therefore, we developed a strategy that uses glass molds to create particular agarose patterns that allow for the study of adherent or nonadherent cell migration and cell attraction by another cell type. With this assay, we observed the behavior of different cell types toward growth factors, such as Platelet-derived growth factor (PDGF-BB) and basic fibroblast growth factor (bFGF), or toward secreted factors contained in the supernatant of bovine capillary endothelial cells (EJG) and tumor cells (1547 cells). We show that this assay permits differentiation between chemokinetic and chemotactic movements of different cell types and comparison of the chemotactic potencies of growth factors using criteria such as the number of attracted cells and migration speed.

Materials and Methods

Cell Culture and Chemicals

PDGF-BB was purchased from R&D Systems (Lille, France). Sphingosine-1-phosphate and FGF2 were purchased from Sigma-Aldrich (Lyon, France). Human erythroleukemia cells (HEL), osteosarcoma cells (1547), and bovine endothelial cells (EJG) were obtained from ATCC (Manassas, VA, USA). T lymphocytes were isolated from total human blood by Ficoll® gradient density centrifugation (histopaque 1077; Sigma-Aldrich) and the Pan T cell isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany). VSMCs were obtained by explants of thoracic aorta from 3-week-old male Sprague-Dawley rats using a method adapted from Sachinidis et al. (24). After four passages, cells were immunostained with an anti-α-smooth muscle actin antibody (anti-αSMA; Sigma-Aldrich) to characterize the cells as smooth muscle cells. Rat fibroblasts were isolated from explants of thoracic lymphoid duct and labeled with an anti-fibroblast surface protein antibody (Sigma-Aldrich). Primary cultured cells, VSMCs, and fibroblasts were used for our experiments between passages 5–15. All cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS; BioWhittaker, Walkersville, MD, USA). These cultures were maintained at 37°C in 5% CO₂, passaged prior to confluence, and fed twice weekly.