Update your browser to view this website correctly. Update my browser now
TOM22 is a core component of the mitochondrial protein import complex, spanning the outer mitochondrial membrane. As a tail-anchored protein, TOM22 is inserted into mitochondria post-translationally. The sequence of TOM22, however, allows it to be inserted into the endoplasmic reticulum, if the C-terminus is masked by another sequence or if there is a mutation in the transmembrane domain. Here we report that C- and N- terminal fluorescent protein fusions of TOM22 preferentially localize to the endoplasmic reticulum in mammalian cells. Our finding has relevance for cell biological studies that use TOM22 as a mitochondrial marker, and for attempts to understand the mechanism of TOM22 mitochondrial insertion.
Compartmentalization allows cells to explore a remarkable range of functions. Correct targeting of membrane proteins is essential to the proper function of membrane compartments. Membrane proteins are inserted in a variable set of topologies: C- and N- terminal anchoring of single transmembrane domain proteins, spanning of the membrane with multiple helical TMDs, or beta-barrel insertion. Tail-anchored or C-terminally anchored proteins have a Nout-Cin orientation and have a short, usually basic sequence after the hydrophobic transmembrane domain (TMD) that anchors it in the membrane. The topology of tail-anchored proteins necessitates post-translational membrane insertion and proceeds via competing mechanisms in mitochondria and the endoplasmic reticulum (ER), to the point where dual localization is possible. For example, cytochrome b5 exists in two isoforms which are targeted to mitochondria and the ER.
It is becoming clear that there are subtle sequence differences between ER and mitochondria-targeted tail-anchored proteins, which function to ensure correct localization. These sequence differences are mainly thought to consist of hydrophobicity in the TMD and of basic residues in the C-terminus.
TOM22 is a small outer mitochondrial membrane protein, with one TMD, which is part of the TOM40 protein import complex and a receptor for pro-apoptotic protein Bax. TOM22 is a tail-anchored protein, and hence its targeting depends on the composition of the TMD and a short C-terminal basic sequence, additionally, a small sequence upstream of TMD is implicated in TOM22 targeting mitochondria. Change in the hydrophobicity of TOM22 TMD by replacement of residues with Valines results in its re-localization to the ER. Additionally, it has been demonstrated that increasing the length of the TOM22 C-terminus in yeast by flanking regions of other tailed anchored proteins masks the mitochondrial localization sequence resulting in re-localization to other membrane compartments including ER, while N-terminal fusion doesn’t interfere with TOM22 mitochondrial localization. Here we report that both C- and N-terminal fusions of TOM22 with GFP are preferentially localized to the ER in human cell lines.
Fusions of small mitochondrial proteins are widely used as mitochondrial markers (e.g. COX4). While constructing mitochondrial markers for live-cell imaging we noticed that TOM22 fusions do not localize to mitochondria. The objective of this study is to characterize the cellular localization of TOM22 fluorescent protein fusions in human cells.
The C-termini of tailed anchored membrane proteins govern correct targeting, and therefore masking the C-terminus with a C-terminal fusion has been shown to result in sub-cellular mislocalization in yeast. We examined the localization of N-terminal fusions of the mitochondrial protein TOM22, in order to construct better mitochondrial membrane markers since N-terminal fusions were correctly localized to mitochondria in yeast. Both C- and N- terminal fusions did not insert into the mitochondrial membrane (Fig. 1A, D) in a human cell line, preferentially localizing to an endomembrane system resembling the ER. We, therefore, confirmed that the TOM22 fusion localizes to the ER and found significant colocalization (Pearson correlation coefficient 0.93; Fig. 1B-C).
To confirm that the localization is not specific to the SH-SY5Y human cell line, we tested it in another human cell line, HEK293T. Both C- and N- terminal fusions also localized to the ER (Fig. 1E-F). Next, we verified that endogenous TOM22 is not detectable in the ER in these cell lines (Fig. 1E). Although TOM22 fusions preferentially localized to the ER, they are also subtly detectable on the mitochondrial membrane in part of the cell population, indicating that the fusion is capable of being inserted into the mitochondrial membrane. The C-terminal GFP fusion of TOM22 showed an additional increase in diffuse cytoplasmic localization, confirming that C-terminal masking interferes with membrane targeting. The N-terminal fusion preferentially localizes to the ER (Fig. 1G). The N-terminus of TOM22 is part of the TOM40 import complex, hence it is possible that complex interactions are necessary to retain TOM22 in the mitochondrial membrane and that adding an N-terminal tag may mask a short import sequence on the N- terminus. Additionally, dual protein targeting is influenced by the folding process, which can be disrupted by fusion proteins.
In this study we observe TOM22 fusions localizing preferentially to the ER, rather than to mitochondria, in human cells. This is surprising, but consistent with findings indicating the possibility of dual targeting for tail-anchored proteins. TOM22 is still detectable on mitochondria in some cells in the population, but less prominently than in the ER. Our findings recommend caution in using tagged TOM22 constructs as a mitochondrial marker in biochemical and cell biological studies.
In this study, we tested fluorescent protein fusions with both termini of TOM22. We have not demonstrated how the length and amino acid composition of the tag affect TOM22 localization. It would be interesting and relevant for biochemical studies to determine whether various affinity tags (e.g. FLAG) similarly interfere with TOM22 cellular localization. Additionally, it would be interesting to examine whether overexpression of TOM22, without the addition of tags, is sufficient to interfere with its localization (for example due to competition for insertion machinery).
HEK293T cells were maintained in high glucose DMEM supplemented with 10% fetal bovine serum (FBS), 1% penicillin/streptomycin, at 37°C/5% CO2, SH-SY5Y cells were maintained in high glucose 1:1 F12/DMEM media supplemented with 10% FBS, 1% penicillin/streptomycin at 37°C/5% CO2. Cells were transfected using ROTI®Fect (Carl Roth) according to manufacturer instructions.
Chemicals and antibodies
Mito Red 569/594 (Sigma), Hoechst (Sigma), Rhodamine800 (Sigma), anti-Tom22 (sc58308, Santa Cruz Biotechnology).
All plasmids were constructed using Escherichia coli strain DH5α. Total mRNA was extracted from cells using. cDNA synthesis was performed. TOM22 sequence was obtained from human cDNA from HEK293 cells using a first-strand cDNA synthesis kit (NEB) on total mRNA extracted using TRI Reagent (Sigma). TOM22 was cloned into the pEGFP-C1 plasmid using SalI, BamHI restriction enzymes to obtain the GFP-TOM22 construct. TOM22 and GFP were cloned into pcDNA3.1 with NheI, SalI, and XhoI, NotI respectively to obtain the TOM22-GFP construct. ER marker was cloned by adding 1-63 nucleotides of CALR to N-terminus of mCherry and KDEL sequence to C-terminus to obtain pcDNA3.1 CALR(21)-mCHkdel.
For live-cell imaging, we used 4-well microscope glass-bottom plates (IBIDI). Confocal images were acquired using a dual point-scanning Nikon A1R-si microscope equipped with a PInano Piezo stage (MCL), temperature, and CO2 incubator, using a 60X PlanApo VC oil objective NA 1.40. We used 406 nm, 488 nm, 561 nm, and 640 nm lasers (Coherent, OBIS). Image processing was performed using NIS-Elements software.
Three independent experiments were performed to obtain the data. p values were calculated by two-tailed Student t-test or one-way ANOVA.