Dual localization of translocon subunit Tic56 to chloroplasts and mitochondria modulates rRNA accumulation in both organelles

Mitochondria and chloroplasts, descendants of endosymbiotic events, retain their own genomes but rely heavily on nuclear-encoded proteins for their function. In plants, c. 1000 mitochondrial and 3000–4000 chloroplast proteins are synthesized in the cytosol and imported into their respective organelles (Cline Duncan et al., 2013; Murcha et al., 2014; Ghifari et al., 2018; Nakai, 2018; Thomson et al., 2020; Xu et al., 2021). This import process is mediated by organelle-specific targeting signals and dedicated protein import machinery. Despite their distinct evolutionary origins and functional specializations, mitochondria and chloroplasts share a remarkable exception to strict organelle specificity: the existence of dual-targeted proteins capable of localizing to both organelles (Millar et al., 2006; Carrie Sharma et al., 2018). The N-terminal cleavable transit peptides (TPs) are the primary targeting signals for most chloroplast and mitochondrial precursor proteins in plants (Murcha et al., 2014; Jeong et al., 2021). These TPs are generally necessary and sufficient for organelle-specific targeting, even when fused to heterologous proteins. In plants, dual targeting occurs through two primary mechanisms: (1) twin TPs, generated by alternative in-frame translation initiation codons (Danpure, 1995; Silva-Filho, 2003); and (2) ambiguous dual-TPs, in which a single peptide directs dual localization (Peeters Carrie Xing et al., 2025). Tic56 is a well-characterized component of the TIC complex (Kikuchi et al., 2013; Köhler et al., 2015, 2016; Jin et al., 2022; Liu et al., 2023; Liang et al., 2024). Interestingly, beyond its role in protein import, Tic56 also participates in chloroplast rRNA processing and ribosome assembly (Köhler et al., 2016), a function apparently distinct from canonical import activity. In this study, we demonstrate that Tic56 is dually targeted to chloroplasts and mitochondria in a cell-type-specific manner, employing distinct targeting mechanisms. Furthermore, mitochondrial-localized Tic56 is involved in rRNA accumulation, paralleling its known function in chloroplasts. Notably, the mitochondrial targeting signal of Tic56 acts as a dominant signal that can override the chloroplast-targeting activity of the canonical Tic20 transit peptide. These findings reveal a novel regulated mechanism of organelle-specific protein targeting and underscore the functional versatility of translocon components in coordinating interorganellar proteome dynamics. The tic56 mutant used in this study is a base substitution mutant of the TIC56 gene on the Arabidopsis thaliana Columbia-0 (Col-0) background (the 731th G was changed into nucleotide A). Arabidopsis seeds were stratified at 4°C in darkness for 48 h, followed by surface sterilization in 10% (v/v) sodium hypochlorite with agitation (10 min). Seeds were rinsed five times with sterile distilled water and sown on Murashige and Skoog (MS) medium containing 2% (w/v) sucrose. Plants were grown vertically on plates for 14 d under controlled conditions: 22°C with a 12 h : 12 h, light : dark photoperiod (100 μmol m−2 s−1 photosynthetic photon flux density). For soil-grown seedlings, identical light and temperature regimes were maintained in growth chambers. For transient expression assays, coding sequences of FRO1, Tic20, Tic56, and its truncation variants were cloned into either pUC18-35S-sGFP or 163-mCherry vectors to generate C-terminal fluorescent protein fusions (Zou et al., 2020; Zhang et al., 2021). For stable transformation, full-length or truncated Tic56 sequences fused with the Tic20 transit peptide were inserted into the binary vector pRI101-GFP downstream of the CaMV 35S promoter. The 1486 base pair upstream of the start codon was used as the native promoter of the TIC56 gene. All constructs were verified by Sanger sequencing before transformation. The binary constructs were electroporated into Agrobacterium tumefaciens strain GV3101 and introduced into wild-type Arabidopsis (Col-0) via the floral dip method (Clough HT801-01; TransGen Biotech, Beijing, China). Three independent transgenic lines showing stable GFP fusion protein expression were advanced to the F3 generation through successive rounds of selection. Homozygous lines confirmed by Mendelian segregation analysis were used for subsequent experiments. Transient expression in Arabidopsis mesophyll protoplasts was performed via polyethylene glycol-mediated transformation (Kovtun et al., 2000). Following 14–18 h incubation at 22°C under dim light, fluorescent fusion proteins were imaged using a Zeiss LSM980 confocal microscope equipped with Airyscan 2 detector. Imaging parameters were standardized as follows: Chl autofluorescence: 488-nm excitation (argon laser), 627- to 726-nm emission collection; GFP/YFP signals: 488/514-nm excitation, 525- to 546-nm bandpass emission detection. Mitochondria were isolated from 30-d-old hydroponically grown A. thaliana roots using a mitochondrial extraction kit (R24019; Yuanye, Shanghai, China) as described (Lyu et al., 2018). The mitochondrial pellet was resuspended in lysis buffer (500 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, 10% glycerol, 0.3% Tween-20, 10 mM Na₂S₂O₅, pH 7.5) and incubated on ice for 20–30 min with intermittent vortexing. After centrifugation at 16 000 g (4°C, 20 min), the supernatant was collected as the mitochondrial protein extract. Protein concentration was quantified before immediate use or storage at −20°C. Proteins were separated by SDS-PAGE, transferred to nitrocellulose membranes, and probed with the indicated primary antibodies. Blots were developed using Horseradish Peroxidase (HRP)-conjugated secondary antibodies and the Pro-Light HRP Chemiluminescent Kit (Tiangen). Total RNA was extracted from frozen leaves of 14-d-old plants using the RNeasy Plant Mini Kit (Qiagen, Beijing, China). RNA samples were separated on formaldehyde-denaturing 1.2% (w/v) agarose gels and transferred onto Hybond-N+ membranes by capillary transfer. Gene-specific probes for chloroplast and mitochondrial rRNAs were generated by PCR, purified from agarose gels, and labeled with biotin via random priming. To assess functional divergence among chloroplast translocon components, we analyzed the localization of Arabidopsis TIC subunits using GFP fusions. When expressed transiently in mesophyll protoplasts, Tic20-GFP showed the expected ring-like pattern around chloroplasts, consistent with inner envelope localization (Shang et al., 2010), whereas Tic56-GFP displayed a punctate distribution resembling the mitochondrial marker FRO1-GFP (Supporting Information Fig. S1a). Colocalization with FRO1-YFP confirmed mitochondrial targeting of Tic56 in this system (Fig. S1a). Considering the possible limitation of transient expression system, we further examined its localization in stable transgenic lines expressing Tic56-GFP. In mesophyll cells, Tic56-GFP fluorescence overlapped precisely with Chl autofluorescence (Fig. 1a), validating its chloroplast localization in planta. However, in roots, Tic56-GFP appeared as discrete puncta that partially colocalized with MitoTracker Red, indicating mitochondrial targeting (Fig. 1a). The remaining nonoverlapping signals may reflect import into nongreen plastids, as reported for other chloroplast proteins (Wan et al., 1996; Yan et al., 2006), although other possibilities could not be excluded. This pattern persisted when Tic56-GFP was expressed under its native promoter, excluding an artefact of ectopic overexpression (Fig. 1a). To further corroborate organelle-specific targeting, we co-expressed Tic56-GFP with the mitochondrial marker FRO1-YFP in transgenic plants. In mesophyll cells, Tic56-GFP did not colocalize with FRO1-YFP, confirming exclusive chloroplast targeting. By contrast, root cells showed clear overlap between Tic56-GFP and FRO1-YFP signals, demonstrating mitochondrial localization (Fig. 2b,c). Together, these results establish that Tic56 undergoes cell-type-specific dual targeting: to chloroplasts in photosynthetic tissues and to mitochondria in roots. We next asked whether this dual targeting is evolutionarily conserved. Transient expression in Arabidopsis protoplasts showed that soybean Tic56 localized to chloroplasts, while Chlamydomonas Tic56 did not (Fig. 1b). However, in stable transgenic plants, both orthologues localized to chloroplasts in leaves and to mitochondria in roots (Fig. 1d). Thus, cell-type-specific dual targeting of Tic56 appears to be a conserved feature among green plants. Using Tic20 as a control, we examined whether dual localization is a general feature of TIC complex components. In transgenic plants expressing Tic20-GFP, fluorescence was exclusively chloroplast-localized in leaves and remained nonmitochondrial in roots, showing no overlap with MitoTracker (Fig. S2). These results demonstrate that dual targeting is specific to Tic56 and not a general property of TIC proteins. We first asked whether mitochondrial-localized Tic56 functions in mitochondrial protein import as well as the chloroplast-localized version. To this end, we performed the immunoblotting on the total extracts from a point mutant carrying a mutation at amino acid 244 of TIC56 protein. In contrast to the previously reported Tic56 knockout lines, which exhibit an albino phenotype and require sucrose supplementation for survival (Kikuchi et al., 2013; Köhler et al., 2016), our mutant displays a relatively weak pale-green phenotype (Fig. S3a,b). We therefore named it tic56-4 to distinguish it from the three previously described alleles. However, the accumulation of the nuclear-encoded mitochondrial protein AOX (Alternative Oxidase) was unchanged in the tic56-4 mutant, unlike the nuclear-encoded chloroplast proteins, which were markedly reduced (Fig. 1e). Thus, Tic56 does not appear to act as a general mitochondrial import factor. Unexpectedly, levels of the mitochondrially encoded proteins COX1 and COX2 were decreased in the mutant (Fig. 1e), suggesting that mitochondrial translation might be affected. Because chloroplast Tic56 is required for chloroplast rRNA accumulation (Köhler et al., 2016), we hypothesized that mitochondrial Tic56 might similarly affect mitochondrial rRNA. RNA blotting confirmed a clear decline in chloroplast 23S and 16S rRNAs in tic56-4 mutants, as reported (Köhler et al., 2016). In mitochondria, 26S rRNA abundance was notably reduced, while 18S rRNA showed only a modest decrease (Fig. 1f). These results indicate that dual-targeted Tic56 functions in rRNA accumulation in both organelles, although the underlying molecular mechanisms remain to be defined. Interestingly, we did observe delayed root growth and development in the tic56 mutant (Fig. S3b,c), which may hint at a possible contribution of mitochondrial-localized TIC56 to root development. However, impaired chloroplast or nongreen plastid function can indirectly affect root development. Therefore, we cannot entirely exclude the possibility that the observed root phenotypes are attributable, at least in part, to plastid dysfunction. Tic56 carries an N-terminal 48-amino acid region predicted as a chloroplast TP (Kikuchi et al., 2013). To test whether this region also directs mitochondrial import, we expressed a series of N-terminal Tic56 truncations fused to GFP in mesophyll protoplasts (Fig. S4). The predicted TP (Tic561–48 aa) did not target GFP to chloroplasts or mitochondria, and remained cytosolic. Extending the region to 200 aa still conferred no organellar targeting. By contrast, a 300-aa N-terminal fragment (Tic561-300 aa) efficiently localized GFP to mitochondria and could redirect mature Tic20 (lacking its native TP, Tic2066-274aa) to mitochondria (Fig. S4). These results indicate that mitochondrial import of Tic56 depends on an unusually long N-terminal segment, pointing to a noncanonical targeting mechanism distinct from typical TP-mediated pathways. To validate these findings in a stable expression system, we generated transgenic plants expressing the putative TP of Tic56 (Tic561-48 aa) fused to GFP (Fig. S5). In mesophyll cells, Tic561-48aa-GFP colocalized with Chl autofluorescence, confirming its functionality in chloroplast targeting. However, unlike Tic20 TP (will be discussed later), no GFP signal was detected in root cells (Fig. S5), indicating that this N-terminal region is insufficient for localization to nongreen plastids. This localization pattern suggests that distinct import mechanisms or stabilizing factors operate in different cell types of leaves and roots. Differential import into distinct plastid types is often mediated by tissue-specific translocon components and TP motif interactions (Chu Chu et al., 2020). To determine whether Tic56 uses a single ambiguous transit peptide for dual targeting (Peeters Carrie Fig. S7), yet none of these alone could direct GFP to mitochondria (Fig. S4). In Arabidopsis, an N-terminal helix is essential for targeting, but this feature is absent in Chlamydomonas Tic56 (Fig. S7). Moreover, deleting the N-terminal 48 aa abolished mitochondrial import, suggesting that overall structural integrity, rather than a short linear signal, governs mitochondrial entry. Notably, like yeast Ups proteins that lack canonical targeting signals and instead rely on lipid-mediated, affinity-driven import via Mdm35p (Tamura et al., 2010; Potting et al., 2010), Tic56 is also soluble, associates with membrane lipids, and interacts through its N terminus with partners such as Tic100/Ctap3 (Jin et al., 2022; Liu et al., 2023). We thus propose that Tic56 may enter mitochondria through a similar lipid-assisted, affinity-dependent pathway. Among the known TIC complex components, Tic56 is unique in lacking transmembrane domains. Interestingly, among the broader group of dual-targeted soluble proteins, there is a notable bias toward proteins involved in core processes such as nucleotide metabolism, DNA replication and repair, tRNA biogenesis, and translation (Carrie & Small, 2013). Our findings appear consistent with this trend. Generally, dual-targeted proteins are indiscriminately delivered to both mitochondria and chloroplasts. However, our study demonstrated that Tic56 is specifically targeted to chloroplasts, not mitochondria, in mesophyll cells. This uneven distribution may result from the higher affinity between Tic56's chloroplast transit peptide and chloroplast import receptors than between its mitochondrial signals and mitochondrial import receptors. Thus, Tic56 is targeted to mitochondria only when chloroplasts are absent. Alternatively, there may be a cell-specific mitochondrial import receptor and/or accessory factor for Tic56 present only in root cells, not mesophyll cells. Considering that mitochondria and chloroplasts likely have different needs for dual-targeted proteins in various cell types and developmental stages, the cell-type-dependent distribution of dual-targeted proteins such as Tic56 could be crucial for coordinating protein targeting with cellular physiology. In yeast, disruption of the Guided Entry of Tail-anchored protein system causes certain tail-anchored proteins to mislocalize to mitochondria for degradation (Chen et al., 2014). By analogy, TIC56 in root cells, which lack chloroplasts, might also be mistargeted to mitochondria for processing or degradation. While we cannot entirely exclude this possibility, several observations suggest it is unlikely to be the primary mechanism. Unlike canonical tail-anchored proteins, TIC56 lacks a predicted transmembrane domain or membrane helix (Fig. S7), and whether it associates with mitochondrial membranes remains unclear. Moreover, if TIC56 were predominantly destined for mitochondrial degradation, its steady-state accumulation in mitochondria would likely be extremely low under normal conditions. Our detection of notable TIC56 levels in mitochondria thus argues that such even if is not the for its mitochondrial localization. We are to of for on Tic56 We would like to from for with This was by the and the of and the of and the and performed the experiments. and analyzed the the and to this All data are the and in data from this can be in the Arabidopsis or under the Arabidopsis thaliana Chlamydomonas Fig. of Tic56 by the transient expression Fig. localization of Tic20-GFP fusion in stable transgenic plants. Fig. and phenotype of the tic56-4 Fig. of Tic56 targeting signals in Arabidopsis mesophyll Fig. localization of localization of Tic561-48aa-GFP fusion proteins. Fig. between Tic20 TP and Tic56 targeting Fig. of Tic56 from photosynthetic is not for the or functionality of Information by the than be to the The is not for the or functionality of by the than be to the for the The remains with to in and in