Explanations that cannot be eliminated at this stage include 3,4,5-Trimethoxyphenylacetic acid possible differences in translation efficiency or protein stability. Loss of either Tic22 isoform, individually, does not have a major impact on chloroplast development in Arabidopsis. However, when both genes are knocked out, an important role in chloroplast development is clearly revealed, particularly in very young plants. However, as the Tulathromycin B plants grow older, the severity of the double mutant phenotypes decreases markedly, and can be quite difficult to detect in mature plants. This does not seem to reflect expression differences, as neither of the genes is expressed at much higher levels during early development compared with later stages. A plausible explanation is that the role of Tic22 is only critical during early stages of chloroplast biogenesis, when protein import rates are especially high as the photosynthetic apparatus is being established. Declining phenotype severity might also reflect the possibility that Tic22’s role in import is to increase the efficiency of a process that can nonetheless proceed in its absence. Thus, chloroplast biogenesis might proceed without Tic22, but at a slower rate, so that eventually such plants are able to “catch-up” with the wild type. One possible role for Tic22 that might conform to the above criteria is the chaperone function that was proposed in relation to the cyanobacterial and apicomplexan homologues. In this regard, it is interesting to note that levels of the Omp85-related protein, Toc75, are noticeably reduced in the tic22 double mutants. This is consistent with the notion that Tic22 plays a chaperone-like role in the passage of proteins such as Toc75 through the inter-membrane space. However, it seems unlikely that the phenotype of the tic22 double mutants can be attributed entirely to Toc75 deficiency, as the toc75-III-3 mutant is visibly greener than the tic22 mutants during early development. Moreover, chloroplast protein levels are more severely reduced in the tic22 mutants than in the toc75III-3 mutant. Thus, the data are consistent with a more general chaperoning role for plant Tic22 in the transport of a range of different preproteins. Previous work showed that the chloroplast import of Tic22 depends on its N-terminal presequence. However, processing of preTic22 was seen to be a slow event in the import process, with large amounts of imported preTic22 remaining uncleaved in envelope. It was proposed that Tic22 is processed in the intermembrane space by an unknown peptidase, and that its targeting pathway is different from other known routes. These observations may account for our failure to observe efficient chloroplast targeting of a range of different full-length and truncated Tic22 fusions to YFP in transfected protoplasts. It is conceivable that this unusual import pathway is unable to accommodate such heterologous passenger proteins, or that the C-terminal addition of a large tag like YFP somehow disrupts the targeting signals of the Tic22 protein. Regardless, there is little doubt that the Arabidopsis Tic22 proteins are chloroplast localized, as this has been shown by in vitro import, protease treatment of isolated chloroplasts, and by the proteomic analysis of purified chloroplast fractions. The data presented in this report complement those of Rudolf et al. in the following ways: we conducted a comprehensive analysis of the evolution of the Tic22 gene family; we identified and characterized a tic22-III single mutant phenotype; we identified and characterized more than one tic22 double-mutant genotype; we analysed the accumulation of a range of different chloroplast proteins in the tic22 mutants by immunoblotting; we conducted a quantitative analysis of chloroplasts using electron microscopy in single as well as double-mutant plants.