Peter B. Noble

Lymphocyte locomotion in three-dimensional collagen gels comparison of three quantitative methods for analysing cell trajectories

We evaluated three different quantitative evaluation methods for lymphocyte locomotion in three-dimensional collagen gels: (1) the length of the two-dimensional migration path (distance migrated) was compared to (2) the resulting average displacement from the starting to the end point and (3) the displacement of the furthest migrating population (cells with high displacement). Locomotion of immunomagnetically isolated human CD4+ and CD8+ peripheral blood lymphocytes suspended in type I collagen gels was recorded using time-lapse videomicroscopy. Paths of randomly selected locomoting cells were digitized, reconstructed and quantitatively analysed. For spontaneously locomoting CD4+ and CD8+ lymphocytes (90 min observation period) the mean total distance migrated was 10.0 +/- 3.7 microns/min (CD4+; n = 114 cells) and 5.6 +/- 3.3 microns/min (CD8+; n = 90 cells). The mean displacement from the individual starting point amounted to 1.3 +/- 0.7 micron/min for CD4+ and 1.1 +/- 0.7 micron/min for CD8+ cells, thus representing only 5-25% of the total migration path (index range displacement/distance migrated: 0.13-50%). Incubation with interleukin-8 and/or receptor blocking by monoclonal antibodies against VLA-2 (Gi9) or VLA-4 (HP2/1) integrins significantly altered the mean length of the migration paths for six out of ten different experimental conditions. Average displacement or displacement of the most active cells detected significant changes in two and three out of ten samples. Whereas the interleukin-8 induced locomotory changes were correctly represented by end point determination, relatively slight but significant modulation in lymphocyte behaviour by anti-integrin antibodies was revealed solely by analysis of the complete cell trajectory. In conclusion, the cell trajectory may represent a sensitive method for evaluating induced subtle changes in lymphocyte locomotory characteristics.

A Rule-Based System for Characterizing Blood Cell Motion

The main function of a blood cell’s surface is to receive information from the environment. Recently, experiments have indicated that the cell membrane plays a vital role in the life, development, and regulation of cells. However, there is no existing method to quantify the observable changes in membrane shape that occur in cell locomotion. The main goal of this research is to develop an image interpretation system capable of analyzing the structural changes in the morphology of cells from a sequence of pictures. This paper describes a scene analysis system for understanding cell motion. The system has been successfully used to analyze the pseudopod kinetics of the cell and relate it to the overall motion patterns.

Understanding blood cell motion

The main function of the surface of a blood cell is to receive information from the environment. Experiments have indicated that the cell membrane plays a vital role in the life, development, and regulation of cells. There is, however, no existing method to quantify the observable changes in membrane shape that occur in locomotion. The main goal of this research is to develop an image interpretation system capable of analyzing the structural changes in the morphology of cells from a sequence of pictures, using automatic techniques of image processing. A model for a general dynamic scene analysis system is described. It consists of three basic entities: dynamic data, static data, and a collection of analysis processes. Based on this model, a rule-based image interpretation system for moving cells has been implemented. The system consists of different cooperating computational processes, which interact with two common memories, a short term memory (STM) and a long term memory (LTM), The STM contains a dynamic record of the instantaneous cell motion, shape, and structural changes, as well as the current global description of the cell behavior. The LTM data arc static, and are implemented as rules. These describe the general model of the morphology of the cells under analysis, as well as control information pertinent to the computational processes. The latter are activated by the control rules throughout the three hierarchical analysis stages: static, incremental, and global. They interact through the STM using the information stored in the LTM, until a complete description of the dynamic cell motion and morphology is obtained. The system has been successfully employed to analyze and study the pseudopod kinetics of white blood cells.

The quantification of blood cell motion by a method of automatic digital picture processing

In order to facilitate the study of cell movement, the paper introduces a new automatic picture processing method for tracking and quantifying the dynamics of blood cell motion. The input to the program is a 16 mm cine film of the cell culture viewed in a steady condition. Under the circumstances, the global directional movement tendencies of the group of cells under consideration can be characterized by a Markov chain model. By observing and quantifying the cell paths, it is possible, using this model, to determine the probability that the cell population is moving in a particular direction. This information might be of interest in the study of the effect of substances on cell movement, defects in white blood cell migration, and the general interaction among cells.

Chemotaxis in Vitro: Quantitation of Human Granulocyte Movement Using a Stochastic Differential Equation

The quantitation of human granulocyte movement using a stochastic differential equation is described. The method has the potential to distinguish both positive and negative chemotaxis. Analysis and information concerning cell movements can be obtained for any point in time and distance for the duration of the experiment.

Time-based changes in fibroblast three-dimensional locomotory characteristics and phenotypes.

Three-dimensional locomotory trajectories have been determined for an embryonic fibroblast population and also for these cells after a 20-day period in tissue culture (C-20 cells). Differences in locomotory characteristics between these two cell populations are reported. Applying factor analysis to the distribution of angle changes between vectors in a cells trajectory reveals the possibility that different locomotory phenotypes exist. Of the three phenotypes detected in the embryonic population only one continues in the C-20 population while a new phenotype appears in the latter. These results document changes in locomotory characteristics and phenotypes in an embryonic population with time.

Analysis of Cell Three-Dimensional Locomotory Vectors

A method is presented for analysing the vectors of cells locomoting within three-dimensional collagen gels. The method detected differences in locomotory vector patterns between cells locomoting in two different gel formats. The potential of this analytical method for determining the role of the extracellular matrix in modulating cell locomotory behaviour is discussed.

Methodology for detection of heterogeneity of cell locomotory phenotypes in three-dimensional gels

A computer-driven, three-dimensional, optical, cell-tracking system was used to quantify the locomotory phenotypes of Mos-11 cells in a collagen gel. Factor analysis was used as a statistical means to make the complex variables generated by the system more tractable to analysis. Heterogeneity in locomotory behavior (two independent locomotory phenotypes) was detected among a sample of 54 cells.

Images of Cells Changing Shape: Pseudopods, Skeletons and Motile Behaviour

Cell locomotion is of importance in embryological development, wound healing and in the invasiveness and metastasis of tumour cells. The advent of microcomputers and image processing systems has greatly facilitated the study of cell locomotion. This report comments on some of the methods that have been developed to capture and process images of cells in both two and three-dimensional environments and presents ways of quantifying and analysing the locomotory behaviours of cells.

Invasion and migration of coordinated cell clusters in rhabdomyosarcoma and other head and neck tumors in a 3-D collagen matrix

EXPRESSION OF NM23-H1 AND NM23-H2 IN HUMAN EPITHELIOID SARCOMA CELL LINES OF DIFFERENT INVASIVE POTENTIAL IN VITRO T. Heymer, R. Engers, C.D. Gerharz, R. Krause-Paulus, N. EIBadawy, H.E. Gabbert Objective: Although epithelioid sarcoma is a very aggressive tumor with high metastatic potential in vivo, no investigations of metastasis associated molecules have been reported yet. Therefore, three clonal subpopulations (GRU-1A, GRU-1B, GRU-1C), derived from the same human epithelioid sarcoma cell line (GRU-1), were characterized with regard to their invasive potential in vitro and the expression of the two candidate metastasis suppressor genes nm23-H1 and nm23-H2. Methods: In vitro invasivaness was determined by the Matrigel invasion assay and by adhesion assays to laminin, fibronectin, collagen IV, hyaluronic acid and plastic. Nm23-H1 and nm23-H2 expression were investigated by RT-PCR and subsequent isoform specific oligonucleotide-hybridisation. Results: In the Matrigel invasion assay GRU-1A exhibited the highest invasive potential while GRU-1B and GRU-1C were significantly less invasive (GRU1A: 112_+12; GRU-1B: 12+_2; GRU1C: 23+_3 transmigrated tumor cells/counting area). Adhesion assays revealed major differences in the adhesive capacity for laminin (GRU-1A: 60_+7; GRU-1B: 18_+9; GRU-1C: 29_+ 7 adherent cells) and plastic (GRU-1A: 115-+1; GRU-1B: 106+_2; GRU-1C: 50 9 +5 adherent cells), while no significant differences could be detected for fibronectin, collagen IV and hyaluronic acid. These results were paralleled by differences in the expression of nm23-H1 and nm23-H2. While in the highly invasive cell line GRU-IA only nm23-H1 was detected, the cell lines of low invasive potential GRU-1B and GRU-1C both expressed nm23-H2. Furthermore, in GRU-1B also nm23-Hl was detected. Conclusions: In our tumor model the expression of nm23-H2 is inversely correlated with the invasive phenotype in vitro and the adhesive capacity to laminin. These results suggest a possibly inhibitory function of nm23-H2 in tumor invasion. Transfection experiments are under progress in order to examine a causative role of nm23-H2.