Pair cross-correlation analysis for assessing protein co-localization.

Abstract

Measuring co-localization of different types of molecules is essential to understand molecular organization in biological systems. The pair cross-correlation (PCC) function computed from two-color microscopy images provides a measure of co-localization between differently labeled molecules. Here, we compute a theoretical expression for the PCC function between two molecules using two-dimensional Gaussian distributions as the effective point-spread functions for single molecules. Through our analytical calculations, we provide a quantitative description of PCC in the case of multiple signal pairs. By fitting our analytical solutions to simulated images, we can estimate both small and large separation distances. We then apply this method to malaria-infected red blood cells (RBCs) imaged by stimulated emission depletion (STED) microscopy. We cross-correlate the signal for the knob-associated histidine-rich protein, which the parasite uses to remodel the spectrin-actin network of RBCs, with different signals from the RBCs and find that its average separation from the ankyrin junctions increases from 40 nm to 120 nm during the 48 h of the infectious cycle.