Analyzing the topography of the CCR5 coreceptor using electron microscopy images of cell surface replicas with nanometer-scale electron-dense probes

Supervisors:
Prof Mark Marsh, MRC LMCB & Department of Developmental Biology, UCL
Dr Lewis Griffin, Computer Science Department & CoMPLEX, UCL

In this essay I analyzed the topography of the CCR5 coreceptor on the cell surface using electron microscopy images of cell surface replicas which were acquired during previous studies at the Medical Research Council Laboratory for Molecular Cell Biology (MRC LMCB), University College London. RBL and CHO cell lines transfected to express human CCR5 coreceptor (treated with RANTES or untreated) were labeled with an anti-CCR5 antibody and protein A conjugated to 15nm electrondense gold probes. Whole mount replicas of the labeled cells were imaged by transmission electron microscopy (TEM). I used previously developed software (ImageJ) to extract the spatial coordinates of these probes and applied spatial statistics methodology in order to affirm their clustering by implementing MATLAB code. CCR5 receptor is used by HIV virus particles to induce endocytosis and so to get internalised in the cell. The degree of clustering of the CCR5 coreceptor is suggested to affect crucially the success of this procedure.

Clustering Analysis. TEM image of CHO cell membrane treated with RANTES (left). Arrangement on the plane of the localized and the randomly added probes in white areas which correspond to small protrusions and villi (center). Depiction of the clustering separation by the clustering algorithm using cut off distance of 50 pixels (<62nm) (right). The probes are separated in clusters shown in different colors. If clusters contain less than 4 members it is chosen not to be shown. (implemented in matlab)