Abstract
A large number of nuclear-encoded proteins are targeted to the organelles of endosymbiotic origin, namely mitochondria and plastids. To determine the targeting specificity of these proteins, fluorescent protein tagging is a popular approach. However, ectopic expression of fluorescent protein fusions commonly results in considerable background signals and often suffers from the large size and robust folding of the reporter protein, which may perturb membrane transport. Among the alternative approaches that have been developed in recent years, the self-assembling split-fluorescent protein (sasplit-FP) technology appears particularly promising to analyze protein targeting specificity in vivo. Here, we have improved this technology with respect to sensitivity and systematically evaluated its utilization to determine protein targeting to plastids and mitochondria. Furthermore, to facilitate high throughput screening of candidate proteins we have developed a Golden Gate-based vector toolkit, named PlaMiNGo (Plastid and/or Mitochondria targeted proteins N-terminally fused to GFP11 tags via Golden Gate cloning). As a result of these improvements, dual targeting could be detected for a number of proteins, which had earlier been characterized as being targeted to a single organelle only. These results were independently confirmed with a plant phenotype complementation approach thus demonstrating the sensitivity and robustness of the sasplit-FP-based method to analyze the targeting specificity of nuclear-encoded proteins.