Simple prerequisite of presequence for mitochondrial protein import in the unicellular red alga Cyanidioschyzon merolae

Abstract

Mitochondrial biogenesis relies on hundreds of proteins that are derived from genes encoded in the nucleus. According to the characteristic properties of N-terminal targeting peptides (TPs) and multi-step authentication by the protein translocase called the TOM complex, nascent polypeptides satisfying the requirements are imported into mitochondria. However, it is unknown whether eukaryotic cells with a single mitochondrion per cell have a similar complexity of presequence requirements for mitochondrial protein import compared to other eukaryotes with multiple mitochondria. Based on putative mitochondrial TP sequences in the unicellular red alga Cyanidioschyzon merolae, we designed synthetic TPs and showed that functional TPs must have at least one basic residue and a specific amino acid composition, although their physicochemical properties are not strictly determined. Combined with the simple composition of the TOM complex in C. merolae, our results suggest that a regional positive charge in TPs is verified solely by TOM22 for mitochondrial protein import in C. merolae. The simple authentication mechanism indicates that the monomitochondrial C. merolae does not need to increase the cryptographic complexity of the lock-and-key mechanism for mitochondrial protein import.

Keywords: Mitochondrial presequence; Mitochondrial protein import; Monomitochondrial eukaryote; Unicellular red alga.

Fig. 1.

Genomic information of mitochondrial and chloroplast proteins in Cyanidioschyzon merolae. (A) Fluorescence images of a C. merolae cell. The ACTIN knockout cell was used for the imaging (Tanaka et al., 2021). The nucleus, mitochondrion, chloroplast and peroxisome were visualized by Cas9–Venus, mScarlet, chlorophyll autofluorescence and mCerulean3, respectively. Images are representative of more than three independent experiments. (B) The mitochondrial translocater of the outer mitochondrial membrane (TOM) complex and the inner mitochondrial membrane (TIM) complex. The TOM complex is composed of the β-barrel protein TOM40, α-helical membrane-integrated receptors TOM20, TOM22 and TOM70, and the regulators TOM5, TOM6 and TOM7. Proteins which are not identified in the C. merolae protein-coding genes are illustrated with dashed lines. OM, outer membrane; IMS, intermembrane space; IM, inner membrane. (C,D) Scatterplot comparisons of mitochondrial targeting peptide (mTP) and chloroplast targeting peptide (cTP) scores for all ORFs (4803 proteins) (C) and for well-characterized mitochondrial (113 proteins) and chloroplast proteins (97 proteins) (D). Prediction scores for all ORFs are shown in Table S1 and the list of mitochondrial and chloroplast proteins is given in Tables S2 and S3. (E,F) Histograms of the length of presequences for the mitochondrion (E) and the chloroplast (F). (G) Venn diagram showing the classification of mitochondrial presequences containing α-helices (magenta), β-sheets (blue), both α-helices and β-sheets (purple), and no α-helices or β-sheets. See also Figs S1 and S2.

Fig. 2.

Synthetic mitochondrial presequence. (A) Scheme of the 1–33 amino acid sequence of aspartate aminotransferase (AAT, CMC148C). (B) Fluorescence images of AAT 1–33 and AAT full-length fused with mVenus. Fluorescence signals for mVenus are shown in green and chlorophyll autofluorescence is shown in red. See also Fig. S3. Images are representative of three independent experiments. (C) Comparisons of amino acid compositions in all ORFs and α-helices in mitochondrial presequences. The α-helices in mitochondrial presequences are identified by structural simulation using the AlphaFold program. See also Table S4 for sequences. (D) Distribution of amino acids in all ORFs (individual bars on the left) and α-helices in presequences (individual bars on the right). (E) An amino acid sequence logo of α-helices in presequences. Asterisks indicate the arginine residues that were adopted in the synthetic presequence. (F) Sequence of the synthetic presequence of 24 amino acids. (G) Structural prediction score for α-helix (left) and local distance difference test (LDDT) score (right) of the synthetic presequence. (H) A simulated structure of the synthetic presequence in lateral and front views.

Fig. 3.

Effects of the number of basic residues in the synthetic presequence. The helical wheel diagrams for amino acid sequences of the synthetic presequences containing three, two, one or zero arginine residues and three lysine residues. Representative images of cells from three independent experiments are shown below each helical wheel.

Fig. 4.

Arginine scanning analysis and physicochemical properties of the modified synthetic presequences. (A) Arginine scanning of the modified synthetic presequence. (B) Fluorescence images of each modified synTP^1R–mVenus. Images are representative of three independent experiments. (C–E) Physicochemical properties of each synthetic presequence. See also Table S5.

Fig. 5.

Fluorescent reporter assay for putative presequences of FMNL and RPSA. (A) Sequences of the first 24 amino acids of formin-like protein (FMNL) and r-protein S1 (RPSA). (B) Helical wheels for FMNL 1–24 and RPSA 1–24. (C) Fluorescence images of the N-terminal 1–24 peptide of FMNL fused with mVenus. Mutated FMNLs (E2A, E18A and E2A/E18A) fused with mVenus are also shown. (D) The Bayesian tree of chloroplast, mitochondrial and bacterial RPSA proteins (see Fig. S4 for the unabbreviated tree). Numbers on the left and right near branches indicate posterior probabilities of Bayesian inference and bootstrap values of the maximum likelihood method, respectively. Branch lengths are proportional to the evolutionary distances indicated by the scale bar. The C. merolae RPSA (CMM019C) is shown as the black circle. (E) Fluorescence images of RPSA 1–24 or RPSA full-length fused with mVenus. Images are representative of three independent experiments.

Fig. 6.

Schematic representation of mitochondrial protein import in C. merolae. (A) Comparison of the N-terminal domains of human TOM22 and C. merolae TOM22. Human TOM22 helix domains and C. merolae TOM22 helix domains are shown in boxes. Helical regions for C. merolae TOM22 were computationally predicted by NetSurfP2.0. Red boxes indicate negatively charged helices. (B) Protein architectures of TOM22 and TOM40. Human TOM22 and TOM40 are depicted using protein structural data (PDB: 7VC4). The structure for C. merolae TOM22 was computationally simulated. Detailed structures of the boxed areas are shown on the right. (C) Schematic models for mitochondrial protein import in animal and fungi (left) and in C. merolae (right). In animal and fungi, a mitochondrial presequence, which is an amphiphilic helix, is recognized by TOM20 and TOM22 as a multi-step authentication process. In contrast, a presequence α-helix containing a few basic residues would be recognized by the acidic part of the α-helix in TOM22 via electrostatic residue–residue interaction as a single-step authentication in C. merolae. During the process, basic residues on the presequence work as the key and the acidic part on TOM22 functions as the lock. After the authentication, a precursor protein is imported into TOM40 and drawn in by the function of the TIM complex. OM, outer membrane; IMS, intermembrane space; IM, inner membrane.