Method for designing RNA-binding proteins using PPR motifs and use thereof

Abstract

To provide a composition comprising a nucleic acid encoding a PPR protein, the composition being used for treating a disease mediated by any modification selected from the group consisting of translation, splicing, and stability of an RNA having a target sequence to which the PPR protein can bind.SOLUTION: The protein to be used is a protein containing one or more (preferably 2 to 14) PPR motifs consisting of a polypeptide of 30 to 38 amino acids in length represented by formula 1 (where: Helix A is a portion of 12 amino acids in length that can form an α-helical structure, represented by formula 2, and A1 to A12 each independently represent an amino acid in the formula 2; X is absent or is a portion consisting of 1 to 9 amino acids in length; Helix B is a portion of 11 to 13 amino acids in length that can form an α-helical structure; L is a portion of 2 to 7 amino acids in length that is represented by formula 3; in the formula 3, each amino acid is numbered from a C-terminus side as "i" (-1), "ii" (-2), except that Liii to Lvii may not exist). The combination of three amino acids A1, A4, and Lii, or the combination of two amino acids A4 and Lii, is determined according to a target RNA base or a base sequence.SELECTED DRAWING: None

Description

[Technical field]

[0001] The present invention relates to a protein capable of selectively or specifically binding to an intended RNA base or RNA sequence. In the present invention, a pentatricopeptide repeat (PPR) motif is utilized. The present invention can be used for identifying and designing RNA-binding proteins, identifying target RNAs of PPR proteins, and controlling RNA functions. The present invention is useful in the fields of medicine, agriculture, etc. [Background technology]

[0002] In recent years, a technology has been established and put to use that uses nucleic acid-binding protein factors that have been revealed through various analyses to bind to a desired sequence. By utilizing this sequence-specific binding, it is becoming possible to analyze the intracellular location of a target nucleic acid (DNA or RNA), remove a target DNA sequence, or control (activate or inactivate) the expression of a protein-coding gene located downstream of the target DNA sequence.

[0003] Research and development has been conducted on zinc finger proteins (Non-Patent Document 1) and TAL effectors (Non-Patent Document 2, Patent Document 1) as protein engineering materials acting on DNA, but the development of protein factors acting specifically on RNA is still very limited. This is because the affinity of the amino acid sequence constituting a protein with RNA and the rules of binding to the RNA sequence have not been clarified, or no rules have been found. As an exception, it has been shown that one puf motif binds to one base of RNA for the pumilio protein, which is composed of multiple repeats of a puf motif consisting of 38 amino acids (Non-Patent Document 3), and a protein with novel RNA binding properties and a technique for modifying RNA binding properties using the pumilio protein have been attempted (Non-Patent Document 4). However, the puf motif is highly conserved and exists in small numbers. Therefore, it has only been used to create protein factors acting on limited RNA sequences.

[0004] On the other hand, from genome sequence information, PPR proteins (proteins with pentatricopeptide repeat (PPR) motifs), which form a large family of as many as 500 proteins in plants alone, have been identified (Non-Patent Document 5). PPR proteins are nuclear-coded, but act in a gene-specific manner exclusively on the control, cleavage, translation, splicing, RNA editing, and RNA stability at the RNA level of organelles (chloroplasts and mitochondria). PPR proteins typically have a structure in which about 10 consecutive PPR motifs, i.e., 35 amino acid motifs with low conservation, are arranged, and it is believed that the combination of PPR motifs is responsible for sequence-selective binding to RNA. Most PPR proteins are composed of only about 10 repeats of PPR motifs, and in many cases, the domain required for catalytic activity cannot be found. Therefore, the actual entity of this PPR protein is thought to be an RNA adaptor (Non-Patent Document 6).

[0005] The present inventors have proposed a method for modifying an RNA-binding protein by utilizing this PPR motif (Patent Document 2). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] WO2011 / 072246 [Patent Document 2] WO2011 / 111829 [Non-patent literature]

[0007] [Non-Patent Document 1] Maeder , ML , Thibodeau-Beganny , S , Osiak , A , Wright , DA , Anthony , RM , Eichtinger , M , Jiang , T , Foley , JE , Winfrey , RJ , Townsend , JA , et al. (2008). Rapid "open-source" engineering of customized zinc-finger nucleases for highly efficient gene modification. Mol. Cell 31, 294–301.

Outdoor Tool2

Outdoor Tools3

Outdoor Tools 4

Direct Environment 5

[0008] The properties of PPR proteins as RNA adaptors are expected to be determined by the properties of each PPR motif that constitutes the PPR protein, and by the combination of multiple PPR motifs. However, the correlation between the amino acid composition and function is largely unclear. If the amino acids that function when PPR motifs exert their RNA binding properties are identified and the relationship between the structure of the PPR motif and the target base is clarified, it may be possible to construct proteins that can bind to RNA of any sequence and length by artificially manipulating the structure of the PPR motif or the combination of multiple motifs. [Means for solving the problem]

[0009] In order to solve the above problems, the present inventors have studied genetically analyzed PPR proteins, particularly PPR proteins involved in RNA editing (modification of genetic information at the RNA level, particularly conversion of cytosine (hereinafter C) to uracil (hereinafter U)) and their target RNA sequences. Using computational science techniques, the inventors have clarified that three amino acids (amino acids 1, 4, and "ii" (-2)) in the PPR motif contain information that controls binding to a specific RNA base. More specifically, the present inventors have found that the selectivity (also called specificity) of the binding RNA base of the PPR motif is determined by three amino acids: the first and fourth amino acids in the first helix of the two α-helix structures constituting the motif, and the second amino acid from the back (C-terminal side) in the part that can form a loop structure after the second helix ("ii"; -2), and have completed the present invention.

[0010] The present invention provides the following: [1] A method for designing a protein capable of binding to an RNA base selectively or specifically to an RNA base sequence, comprising the steps of: The protein contains one or more (preferably 2 to 14) PPR motifs consisting of a 30-38 amino acid long polypeptide represented by formula 1.

[0011] [ka] (In the formula: Helix A is a 12 amino acid long portion capable of forming an α-helical structure and is represented by formula 2:

[0012] [ka] In formula 2, A1~A 12 each independently represents an amino acid; X is absent or a moiety consisting of 1 to 9 amino acids in length; Helix B is a portion that can form an α-helical structure and is 11 to 13 amino acids long; L is a moiety represented by formula 3 that is 2 to 7 amino acids in length;

[0013] [ka] In formula 3, each amino acid is numbered from the C-terminus as "i"(-1), "ii"(-2), etc. However, L iii ~L vii may not exist.) A1, A4, and L ii A combination of three amino acids, or A4, L ii The combination of the two amino acids is determined according to a target RNA base or base sequence. [2] A1, A4, and L ii The method according to [1], wherein the combination of the three amino acids is determined based on an RNA base or a target base sequence, and the combination of the amino acids is determined based on any one of the following: (3-1) A1, A4, and L ii When the three amino acids are, in order, valine, asparagine, and aspartic acid, the PPR motif can selectively bind to U (uracil); (3-2) A1, A4, and L ii When the three amino acids are, in order, valine, threonine, and asparagine, the PPR motif can selectively bind to A (adenine); (3-3) A1, A4, and L ii When the three amino acids are, in order, valine, asparagine, asparagine, the PPR motif can selectively bind to C (cytosine); (3-4) A1, A4, and L ii When the three amino acids are, in order, glutamic acid, glycine, and aspartic acid, the PPR motif can selectively bind to G (guanine); (3-5) A1, A4, and L ii when the three amino acids are, in order, isoleucine, asparagine, asparagine, the PPR motif can selectively bind to C or U; (3-6) A1, A4, and L ii when the three amino acids are, in order, valine, threonine, and aspartic acid, the PPR motif can selectively bind to G; (3-7) A1, A4, and L ii When the three amino acids are, in order, lysine, threonine, and aspartic acid, the PPR motif can selectively bind to G; (3-8) A1, A4, and L ii are, in order, phenylalanine, serine, and asparagine, the PPR motif can selectively bind to A; (3-9) A1, A4, and L ii are, in order, valine, asparagine, and serine, the PPR motif can selectively bind to C; (3-10) A1, A4, and L ii when the three amino acids are, in order, phenylalanine, threonine, and asparagine, the PPR motif can selectively bind to A; (3-11) A1, A4, and L ii are, in order, isoleucine, asparagine, and aspartic acid, the PPR motif can selectively bind to U or A; (3-12) A1, A4, and L ii when the three amino acids are, in order, threonine, threonine, and asparagine, the PPR motif can selectively bind to A; (3-13) A1, A4, and L ii When the three amino acids are, in order, isoleucine, methionine, and aspartic acid, the PPR motif can selectively bind to U or C; (3-14) A1, A4, and L ii When the three amino acids are, in order, phenylalanine, proline, and aspartic acid, the PPR motif can selectively bind to U; (3-15) A1, A4, and L iiWhen the three amino acids are, in order, tyrosine, proline, and aspartic acid, the PPR motif can selectively bind to U; (3-16) A1, A4, and L ii When the three amino acids are, in order, leucine, threonine, and aspartic acid, the PPR motif can selectively bind to G. [3] A4 and L ii The method according to [1], wherein the combination of the two amino acids is determined based on an RNA base or a target base sequence, and the combination of the amino acids is determined based on any one of the following: (2-1) A4, L ii is, in order, asparagine, aspartic acid, the motif can selectively bind to U; (2-2) A4, L ii is, in the order, asparagine, asparagine, the motif can selectively bind to C; (2-3) A4, L ii is, in order, threonine, asparagine, the motif can selectively bind to A; (2-4) A4, L ii is, in order, threonine, aspartic acid, the motif can selectively bind G; (2-5) A4, L ii is, in order, serine, asparagine, the motif can selectively bind to A; (2-6) A4, L ii are, in order, glycine, aspartic acid, the motif can selectively bind to G; (2-7) A4, L ii are, in order, asparagine and serine, the motif can selectively bind to C; (2-8) A4, L ii is, in order, proline, aspartic acid, the motif can selectively bind to U; (2-9) A4, L ii is, in order, glycine, asparagine, the motif can selectively bind to A; (2-10) A4, L ii are, in order, methionine, aspartic acid, the motif can selectively bind U; (2-11) A4, L ii are, in order, leucine, aspartic acid, the motif can selectively bind to C; (2-12) A4, L ii are, in order, valine and threonine, the motif can selectively bind to U. [4] A method for identifying a base or a base sequence that is a target of an RNA-binding protein and contains one or more (preferably 2 to 14) PPR motifs defined in [1], comprising the steps of: The identification is based on any one of (3-1) to (3-16) described in 2 or any one of (2-1) to (2-12) described in 3, and includes PPR motifs A1, A4, and L ii A combination of three amino acids, or A4, L ii The method is carried out by identifying the presence or absence of a base according to the combination of two amino acids. [5] A method for identifying a PPR protein capable of binding to a target RNA base or a target RNA having a specific base sequence and containing one or more (preferably 2 to 14) PPR motifs defined in [1], comprising the steps of: The identification is based on any one of (3-1) to (3-16) described in 2 or any one of (2-1) to (2-12) described in 3, and includes A1, A4, and L of the PPR motif according to the target RNA base or a specific base constituting the target RNA. ii The method is carried out by identifying the presence or absence of a combination of three amino acids. [6] A method for controlling RNA function using a protein designed by the method described in [1]. [7] A complex comprising a region of a protein designed by the method described in [1] and a functional region linked together. [8] A method for modifying the genetic material of a cell, comprising the steps of: providing a cell containing RNA having a target sequence; and [7] A method in which the complex described in [7] is introduced into a cell, whereby the protein domain of the complex binds to an RNA having a target sequence, and thus the functional domain modifies the RNA having the target sequence. [9] detecting amino acid polymorphisms in a PPR protein gene that acts as a fertility restorer for cytoplasmic male sterility, found among various varieties; identifying an association between polymorphisms in the gene and fertility; identifying the base sequence of the PPR protein gene obtained from the test sample and determining the fertility of the test sample; A method for determining the fertility of a PPR protein gene, comprising:

[10] The method according to [9], wherein the PPR protein is a protein containing one or more (preferably 2 to 16) PPR motifs consisting of a 30-38 amino acid long polypeptide represented by formula 1 defined in [1].

[11] The method according to [9] or

[10] , characterized in that an amino acid polymorphism is identified as a polymorphism for each PPR motif.

[12] The polymorphism of the PPR motif is A1, A4, and L of the motif of formula 1. ii A combination of three amino acids, or A4, L ii The method according to any one of [9] to

[11] , wherein the sequence is identified by a combination of two amino acids.

[13] The method according to claim 12, wherein the polymorphism of the PPR motif is identified by a polymorphism of the fourth amino acid (A4) of the motif of formula 1.

[14] The method described in claim 13, wherein fertility is indicated when the fourth amino acid of all PPR motifs on the PPR protein gene is identical to the fourth amino acid of all corresponding PPR motifs of Enko B.

[15] The method according to any one of claims 9 to 14, wherein the PPR protein gene is an orf687-like gene (i.e., a family gene located at a locus homologous to the "687 gene" encoding Enko B, a gene having 90% or more amino acid sequence identity to Enko B, a gene having 90% or more nucleotide sequence identity to the "ORF687 gene" encoding Enko B).

[16] The method according to any one of claims 9 to 15, wherein the protein encoded by the orf687-like gene of various varieties is any one of SEQ ID NOs: 576 to 578, 585 to 591. Effect of the Invention

[0014] According to the present invention, a PPR motif capable of binding to a target RNA base and a protein containing the same can be provided. By arranging multiple PPR motifs, a protein capable of binding to a target RNA having any sequence or length can be provided.

[0015] According to the present invention, the target RNA of any PPR protein can be predicted and identified, and conversely, the PPR protein that binds to any RNA can be predicted and identified. By predicting the target RNA sequence, the genetic substance can be clarified and the possibilities for its use can be expanded. For example, when fertility is considered as the function of the PPR protein in the present invention, the functionality of homologous genes having various amino acid polymorphisms can be examined from the difference in the target RNA sequence for an industrially useful PPR protein gene that acts as a restorer of cytoplasmic male sterility.

[0016] Furthermore, a complex can be prepared by binding a functional group to the PPR motif or PPR protein provided by the present invention.

[0017] Furthermore, the present invention can be used to deliver the above-mentioned complex into a living body and make it function, or to create transformants using nucleic acid sequences (DNA, RNA) that code for the proteins obtained by the present invention, and to specifically modify, control, and impart functions to living organisms (cells, tissues, individuals) in various situations. [Brief description of the drawings]

[0018] [Figure 1]FIG. 1 shows the conserved sequences and amino acid numbers of the PPR motif. (A) The amino acids constituting the PPR motif defined in the present invention and their amino acid numbers are shown. (B) The positions of three amino acids (1, 4, and "ii" (-2)) that control the binding base selectivity on the predicted structure are shown. (C) The positions of the amino acids on the predicted structure. When the predicted structures were analyzed using the PHYRE (http: / / www.sbg.bio.ic.ac.uk / phyre / ) program with all amino acids of Arabidopsis thaliana CRR4 (SEQ ID NO: 6) and CRR21 (SEQ ID NO: 3) as query sequences, structures were predicted with high scores (4.3e-17 and 4.7e-16; for CRR4 and CRR21) using O-GlucNAc transferase (1w3b) as a template. Among them, the fifth PPR motif of CRR4 (left figure) and the eighth PPR motif of CRR21 (right figure) are shown. Numbers 1, 4, and “ii” (-2) are shown as magenta (dark gray in black and white) sticks. [Diagram 2] FIG. 2 shows the RNA-editing PPR proteins that have been analyzed so far and their target RNA-editing sites. [Figure 3-1] FIG. 3 shows the PPR motif sequences and amino acid numbers of Arabidopsis RNA-editing PPR proteins. [Figure 3-2] Figure 3-2 shows a continuation of Figure 3-1. [Figure 3-3] Figure 3-3 shows a continuation of Figure 3-2. [Diagram 3-4] Figure 3-4 shows a continuation of Figure 3-3. [Figure 4]Figure 4 shows the amino acids in the PPR motif involved in RNA recognition. (A) Identification of amino acids with binding base specification ability in the PPR motif. The PPR motif of the RNA-editing PPR protein was aligned in various arrangements with the sequence upstream of the RNA editing site. The alignment was performed by arranging the motif and base one-to-one in a continuous manner. Alignment P1 was set to correspond to the last PPR motif, which corresponds to the base immediately before the edited C. Alignments P2 to P6 were obtained by shifting the base sequence one base at a time to the right. The squares indicate the PPR motifs, and the diamonds indicate the additional motifs (E, E+, DYW) on the C-terminus side. If an amino acid at a specific position in the motif (e.g., an amino acid in the motif in green (dark gray in black and white display)) is responsible for RNA base recognition, low randomness can be expected between the corresponding base in a specific alignment (lower right). If not, high randomness is expected (upper right). (B) Binding RNA base specification ability of amino acids 1, 4, and “ii” (-2). The low randomness of the amino acid and base in each alignment is shown by P value. (C) The ability of the 1st, 4th and “ii”(-2) amino acids to specify the RNA base in various nucleic acid classifications. Same as (B). However, the nucleic acids are classified into purine or pyrimidine (RY; A&G or U&C) and hydrogen-bonding group (WS; A&U, G&C). (D) Further detailed analysis of the ability of the RNA recognition amino acids in the PPR motif shown in (C) above to specify the binding base showed that in addition to mainly distinguishing the purine / pyrimidine (RY) base to which the 4th amino acid binds, the “ii”(-2) amino acid also distinguishes between amino (A and C) and keto (G and U) (MK) types of bases (Fig. 4D). (E) Examples of the RNA recognition codes (PPR codes) of several PPR motifs. The white background indicates the type of the 1st, 4th and “ii”(-2) amino acids. The frequency of occurrence of each code is shown in No., and the frequency of occurrence of the corresponding nucleic acid is shown in Nucleotide frequency. [Diagram 5]Figure 5 shows an example of the identification of amino acids in the PPR motif involved in RNA recognition. Amino acids involved in RNA recognition were searched for using the data set of RNA bases corresponding to the PPR motif in each alignment. For example, the RNA base binding ability of the 4th and 5th amino acids was analyzed using the data of the RNA bases corresponding to the PPR motif in alignment P4. For each alignment, the data was first sorted by amino acid type and the number of RNA bases contained was calculated (upper left). Next, a theoretical value was created based on the median frequency of occurrence of all RNAs contained in the data set (upper right). A chi-square test was performed using these two data to calculate the P value. The upper figure shows the analysis results for the 4th amino acid in alignment P4, where a significant P value was obtained, and the lower figure shows the analysis results for the 5th amino acid in alignment 4, where a significant P value was not obtained. [Figure 6] Figure 6 shows the search for amino acids that are responsible for RNA base specification ability. (A) The low randomness between the type of amino acid and the frequency of occurrence of the base was calculated for the amino acids at all positions in the alignment of P1 to P6. Amino acids that showed a significant P value (P<0.01) are shown in magenta (dark gray in black and white display). The cyan line (horizontal line in the graph; dark gray in black and white display) indicates a P value = 0.01. (B) Summary of low randomness in each alignment. The product of the P values ​​for the amino acids at each position shown in (A) is shown as the overall value of low randomness in that alignment. [Figure 7] Figure 7 shows the ability of two amino acids to specify the base of a bound RNA. The ability of two amino acids in different combinations (1&4, 1&“ii”, 4&“ii” amino acids) to specify the base of a bound RNA was analyzed in the same manner as in Figure 4, with low randomness between the amino acids and the corresponding bases. [Figure 8] FIG. 8 shows the RNA recognition code of the PPR motif extracted from Arabidopsis thaliana. [Figure 9]Figure 9 shows the sequences of the Physcomitrella patens RNA-editing PPR proteins and the RNA-editing site where each protein acts. The amino acid sequences of 1, 4, and "ii" (-2) in each PPR motif are shown together with the protein motif structure. Magenta and cyan (both dark gray in black and white display) letters indicate amino acid combinations that are homologous to the triPPR or diPPR codes extracted from Arabidopsis thaliana, respectively. Additional motifs (E, E+, DYW) on the C-terminus are also shown. The sequences of the RNA-editing site where each protein acts (upstream sequences including the edited C) are shown at the position of aligned P4 in Figure 4. [Figure 10] Figure 10 shows a flowchart of the method for calculating the compatibility score between PPR proteins and RNA editing site RNA sequences. Obtain a protein PPR model from the Uniprot or PROSITE database, and assign each amino acid number according to Figure 1. Extract the 1st, 4th, and "ii" amino acids. As an example, the moss PPR protein, PpPPR71, is shown. Next, the combination of matching amino acids is converted into a triPPR code matrix. Moths that could not be converted into triPPR codes are then converted into diPPR code matrices. In parallel, convert the 30nt RNA editing site (the last C is edited) into a formula matrix. As an example, the ccmFCeU122SF sequence, which the PpPPR71 protein acts on, is shown. Next, calculate the product of the corresponding squares of the protein code matrix and the RNA formula matrix, and calculate the compatibility score from the sum. The last row of the protein code matrix should be matched to the row corresponding to the four bases before the edited C. This calculation is performed for each protein code matrix created from the triPPR code and the diPPR code. Using a standard distribution curve created from the match values ​​with multiple RNA sequences, provisional P values ​​for each RNA sequence are calculated for the triPPR code and the diPPR code. The final match value (P value) is calculated as the product of the provisional P values ​​for the triPPR and diPPR codes. [Figure 11]Figure 11 shows the prediction of the target RNA sequence of PPR protein using the PPR code. (A) The 1st, 4th, and "ii" (-2) amino acids were extracted from the moss PPR protein and converted to triPPR or diPPR code as shown in Figure 10. The match value with the RNA editing site was calculated and shown as a P value. 13 RNA editing sites in moss were used as the RNA editing sites, and 34 RNA editing sites in Arabidopsis chloroplasts were used as the reference sequence. The figure shows only the match value with the 13 RNA editing sites in moss. Diamonds indicate the match value of each protein for each editing site. The correct editing site is shown in magenta (solid gray in black and white display). (B) The P values ​​shown in (A) are shown in a table. [Figure 12] Figure 12 shows the verification of the accuracy of RNA editing site prediction using Arabidopsis RNA editing proteins. The prediction accuracy was verified using Arabidopsis PPR proteins used for code extraction. (A) Prediction of RNA editing sites for 13 known PR proteins for all 34 chloroplast RNA editing sites. Each diamond indicates the match score between the protein and the RNA editing site sequence. Correct RNA editing sites are shown in magenta (solid gray in black and white display). (B) Prediction of RNA editing sites for 11 known PR proteins for all 488 mitochondrial RNA editing sites. [Figure 13]Figure 13 shows the predicted target RNA editing site of the Arabidopsis PPR protein AHG11 and experimental verification. (A) Motif structure of AHG11. It has a typical structure of an RNA editing PPR protein, consisting of 12 PPR motifs and additional motifs (E, E+, DYW) on the C-terminus side. In the Ahg11 mutant, a new translation termination codon is found in the coding region due to a point mutation at the position indicated by an asterisk (Trp 295). (B) Target RNA editing site prediction using all RNA editing sites in Arabidopsis chloroplasts and mitochondria. The top 10 editing sites with the highest P value are shown. The presence or absence of RNA editing in the wild type and mutant strains was experimentally verified and shown as editing status. Sites where editing was detected in both the wild type and mutant strains are indicated as E, and sites where RNA editing was not found only in the mutant strain are indicated as Un. (C) The predicted results are shown in a graph. (D) Experimental verification of the target RNA editing site of AHG11. The results of sequence analysis of the region containing mitochondrial nad4 are shown. RNA was extracted from the wild type and the ahg11 mutant strain, cDNA was prepared by reverse transcription, and the base sequence was analyzed. There are two RNA editing sites (nsd4362 and 376) in this region. The edited sites are indicated with black arrows, and the non-edited sites with white arrows. [Figure 14] Figure 14 shows the prediction of target sites from the chloroplast genome sequence. Using six PPR proteins, the target sites were predicted from the Arabidopsis chloroplast whole genome sequence (154,478 bp). For the prediction, the code extracted from Arabidopsis (At code) or the code extracted from Arabidopsis and moss (At+Pp code) was used. [Figure 15] FIG. 15 shows the RNA recognition codes of PPR motifs extracted from Arabidopsis thaliana and Physcomitrella patens. [Figure 16-1] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 16-2] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 16-3] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 16-4] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 16-5] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 16-6] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 16-7] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 16-8] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 16-9] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 16-10] FIG. 16 shows amino acid sequences or nucleotide sequences related to the present invention. [Figure 17] FIG. 17 shows the binding analysis of Enko B protein with RNA containing the cytoplasmic male sterility (CMS) gene. [Figure 18] FIG. 18 shows the binding of ORF687-like proteins to RNA. [Figure 19] FIG. 19 shows predicted binding sequences of fertility restorers that act on Ogura-type cytoplasm. [Figure 20] FIG. 20 shows the secondary structure and structural changes of the candidate binding RNA region of the ORF687-like protein. [Figure 21-1] FIG. 21 shows an alignment of ORF687-like proteins. [Figure 21-2] FIG. 21 shows an alignment of ORF687-like proteins. [Figure 22] FIG. 22 shows a list of amino acid base designations of ORF687-like proteins contained in various radish varieties. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] [PPR motifs and PPR proteins] In the present invention, the term "PPR motif" refers to a polypeptide consisting of 30 to 38 amino acids, with an amino acid sequence in which the E value obtained by analyzing the amino acid sequence with a protein domain search program on the Web is equal to or less than a predetermined value (preferably E-03) for PF01535 in Pfam and PS51375 in Prosite. The position number of the amino acid constituting the PPR motif defined in the present invention is almost synonymous with PF01535, but corresponds to the number obtained by subtracting 2 from the position of the amino acid in PS51375 (e.g., from No. 1 in the present invention to No. 3 in PS51375). However, when referring to an amino acid at No. "ii" (-2), the amino acid refers to the second amino acid from the end (C-terminal side) of the amino acid constituting the PPR motif, or the amino acid two amino acids N-terminal to the No. 1 amino acid of the next PPR motif, that is, the -2 amino acid (Figure 1). When the next PPR motif is not clearly identified, the amino acid two amino acids before the first amino acid of the next helix structure is designated as "ii". For Pfam, see http: / / pfam.sanger.ac.uk / ; for Prosite, see http: / / www.expasy.org / prosite / .

[0020] The conserved amino acid sequence of the PPR motif is not highly conserved at the amino acid level, but the two α-helices in the secondary structure are well conserved. A typical PPR motif consists of 35 amino acids, but its length varies from 30 to 38 amino acids.

[0021] More specifically, the PPR motif in the present invention consists of a polypeptide having a length of 30 to 38 amino acids and represented by formula 1.

[0022] [ka] During the ceremony: Helix A is a 12 amino acid long portion capable of forming an α-helical structure and is represented by formula 2:

[0023] [ka] In formula 2, A1~A 12 each independently represents an amino acid; X is absent or a moiety consisting of 1 to 9 amino acids in length; Helix B is a portion that can form an α-helical structure and is 11 to 13 amino acids long; L is a moiety represented by formula 3 that is 2 to 7 amino acids in length;

[0024] [ka] In formula 3, each amino acid is numbered from the C-terminus as "i" (-1), "ii" (-2), etc. However, L iii ~L vii may not exist.

[0025] In the present invention, the term "PPR protein" refers to a PPR protein having one or more, preferably two or more, of the above-mentioned PPR motifs, unless otherwise specified. In this specification, the term "protein" refers to a substance consisting of a polypeptide (a chain in which a plurality of amino acids are peptide-bonded), in general, unless otherwise specified, and includes those consisting of relatively low molecular weight polypeptides. In the present invention, the term "amino acid" may refer to a normal amino acid molecule, as well as to an amino acid residue constituting a peptide chain. It will be clear to one of ordinary skill in the art from the context which is being referred to.

[0026] PPR proteins are abundant in plants, with 500 proteins and approximately 5,000 motifs found in Arabidopsis. Many land plants, such as rice, poplar, and Selaginella, also have PPR motifs and PPR proteins with diverse amino acid sequences. Some PPR proteins are known to be important genes for obtaining F1 seeds for hybrid vigor as fertility restorer factors that act on pollen formation (male gametes). Similar to fertility restorer, some PPR proteins have been shown to act on speciation. Most PPR proteins are also known to act on RNA in mitochondria or chloroplasts.

[0027] In animals, abnormalities in the PPR protein identified as LRPPRC are known to cause Leigh syndrome French Canadian (LSFC; subacute necrotizing encephalomyelopathy).

[0028] In the present invention, the term "selective" with respect to the binding of a PPR motif to an RNA base means that the binding activity to any one of the RNA bases is higher than the binding activity to the other bases, unless otherwise specified. This selectivity can be confirmed by a person skilled in the art through experiments, or can be calculated as disclosed in the examples of the present specification.

[0029] In the present invention, the term "RNA base" refers to the base of a ribonucleotide constituting RNA, unless otherwise specified, and specifically refers to any of adenine (A), guanine (G), cytosine (C), or uracil (U). Note that although PPR proteins may have selectivity for bases in RNA, they do not bind to nucleic acid monomers.

[0030] Prior to the present invention, a method for searching sequences for conserved amino acids as PPR motifs had been established, but no rules concerning selective binding with RNA bases had been discovered.

[0031] The present invention provides the following findings.

[0032] (I) Information on the locations of amino acids important for selective binding. Specifically, the combination of three amino acids at positions 1, 4, and “ii”(-1) of the PPR motif (A1, A4, L ii ), or a combination of two amino acids at positions 4, “ii”(-1) (A4, L ii ) are important for selective binding to RNA bases, and the combination of these determines which RNA base will bind.

[0033] The present invention relates to A1, A4, and L iiand / or A4, and L ii It is based on knowledge of the combination of two amino acids.

[0034] (II) A 1 、A 4 , and L ii Information on the correspondence between the three amino acid combinations and the NA bases. Specifically, it is as follows: (3-1) A1, A4, and L ii When the combination of these three amino acids is, in order, valine, asparagine, and aspartic acid, the PPR motif has selective RNA base binding ability, binding strongly to U, then to C, and then to A or G. (3-2) A1, A4, and L ii When the three amino acid combinations are, in order, valine, threonine, and asparagine, the PPR motif has selective RNA base-binding ability, binding strongly to A, then to G, then to C, but not to U. (3-3) A1, A4, and L ii When the combination of these three amino acids is, in order, valine, asparagine, asparagine, the PPR motif has selective RNA base binding ability, binding strongly to C, then to A or U, but not to G. (3-4) A1, A4, and L ii When the combination of these three amino acids is, in order, glutamic acid, glycine, and aspartic acid, the PPR motif has selective RNA base-binding ability, binding strongly to G but not to A, U, or C. (3-5) A1, A4, and L ii When the combination of these three amino acids is, in order, isoleucine, asparagine, asparagine, the PPR motif has selective RNA base-binding ability, binding strongly to C, then to U, then to A, but not to G. (3-6) A1, A4, and L iiWhen the three amino acid combinations are, in order, valine, threonine, and aspartic acid, the PPR motif has selective RNA base-binding ability, binding strongly to G, then to U, but not to A or C. (3-7) A1, A4, and L ii When the combination of these three amino acids is, in order, lysine, threonine, and aspartic acid, the PPR motif has selective RNA base-binding ability, binding strongly to G, then to A, but not to U or C. (3-8) A1, A4, and L ii When the combination of these three amino acids is, in order, phenylalanine, serine, and asparagine, the PPR motif has selective RNA base-binding ability, binding strongly to A, then to C, and then to G and U. (3-9) A1, A4, and L ii When the combination of these three amino acids is, in order, valine, asparagine, and serine, the PPR motif has selective RNA base-binding ability, binding strongly to C, then to U, but not to A or G. (3-10) A1, A4, and L ii When the combination of these three amino acids is, in order, phenylalanine, threonine, and asparagine, the PPR motif has selective RNA base binding ability, binding strongly to A but not to G, U, or C. (3-11) A1, A4, and L ii When the combination of these three amino acids is, in order, isoleucine, asparagine, and aspartic acid, the PPR motif has selective RNA base-binding ability, binding strongly to U, then to A, but not to G or C. (3-12) A1, A4, and L ii When the combination of these three amino acids is, in order, threonine, threonine, and asparagine, the PPR motif has selective RNA base binding ability, binding strongly to A but not to G, U, or C. (3-13) A1, A4, and Lii When the combination of these three amino acids is, in order, isoleucine, methionine, and aspartic acid, the PPR motif has selective RNA base-binding ability, binding strongly to U, then to C, but not to A or G. (3-14) A1, A4, and L ii When the three amino acid combinations are, in order, phenylalanine, proline, and aspartic acid, the motif is called PPR, and has selective RNA base-binding ability, binding strongly to U, followed by C, but not A or G. (3-15) A1, A4, and L ii When the combination of these three amino acids is, in that order, tyrosine, proline, and aspartic acid, the PPR motif has selective RNA base-binding ability, binding strongly to U but not to A, G, or C. (3-16) A1, A4, and L ii When the combination of these three amino acids is, in order, leucine, threonine, and aspartic acid, the PPR motif has selective RNA base binding ability, binding strongly to G but not to A, U, or C.

[0035] (II) A 4 , and L ii Information on the correspondence between the combination of two amino acids and the NA base. Specifically, it is as follows: (2-1) A4, L ii However, in the case of asparagine and aspartic acid, respectively, the PPR motif has selective RNA base binding ability, binding strongly to U, then C, and then A and G. (2-2) A4, L ii However, in the case of asparagine, the PPR motif has selective RNA base binding ability, binding strongly to C, then U, and then A and G. (2-3) A4, L iiHowever, in the case of threonine and asparagine, respectively, the PPR motif has selective RNA base binding ability, binding strongly to A, followed by weaker binding to G, U and C. (2-4) A4, L ii However, in the case of threonine and aspartic acid, respectively, the PPR motif has selective RNA base binding ability, binding strongly to G, followed by weaker binding to A, U and C. (2-5) A4, L ii However, in the case of serine and asparagine, respectively, the PPR motif has selective RNA base binding ability, binding strongly to A, followed by binding to G, U and C. (2-6) A4, L ii However, in the case of glycine and aspartic acid, respectively, the PPR motif has selective RNA base binding ability, binding strongly to G, then U, then A, but not to C. (2-7) A4, L ii However, in the case of asparagine and serine, respectively, the PPR motif has selective RNA base binding ability, binding strongly to C, then U, and then A and G. (2-8) A4, L ii However, in the case of proline and aspartic acid, respectively, the PPR motif has selective RNA base binding ability, binding strongly to U, followed by G, C and C, but not A. (2-9) A4, L ii However, in the case of glycine and asparagine, respectively, the PPR motif has selective RNA base binding ability, binding strongly to A, then to G, but not to C or U. (2-10) A4, L ii However, in the case of methionine and aspartic acid, respectively, the PPR motif has selective RNA base binding ability, binding strongly to U, followed by weaker binding to A, G, and C. (2-11) A4, L iiHowever, in the case of leucine and aspartic acid, respectively, the PPR motif has selective RNA base binding ability, binding strongly to C, then to U, but not to A or G. (2-12) A4, L ii However, in the case of valine and threonine, respectively, the PPR motif has selective RNA base binding ability, binding strongly to U, then to A, but not to G or C.

[0036] In the examples of the present specification, the above findings are obtained by further analyzing the binding between a protein that has been partially analyzed genetically or molecular biologically and its potential RNA target sequence computationally. More specifically, the binding between a protein and RNA or its selective binding is analyzed using a P value (probability) as an index. In the present invention, when the P value, which is a general significance level, is 0.05 or less (5% or less chance of occurrence), preferably when the P value is 0.01 or less (1% or less chance of occurrence), and more preferably when a more significant P value is calculated, it is evaluated that the probability of the protein binding to RNA is sufficiently high. Such a judgment based on the P value can be fully understood by those skilled in the art.

[0037] The binding between a specific combination of amino acids at specific positions and an RNA base can be confirmed by an experiment. Experiments for such a purpose include preparation of a protein containing a PPR motif or multiple PPR motifs, preparation of a substrate RNA, and a binding test (e.g., a gel shift method). Each experiment is well known to those skilled in the art, and for more specific procedures and conditions, refer to, for example, Patent Document 2.

[0038] [Utilization of PPR motifs and PPR proteins] Identification and Design: A PPR motif can recognize a specific base of RNA. According to the present invention, by appropriately selecting amino acids at specific positions, PPR motifs selective for A, U, G, and C can be selected or designed, and a protein containing an appropriate series of such PPR motifs can recognize a corresponding specific sequence. Therefore, according to the present invention, it is possible to predict and identify a natural PPR protein that selectively binds to an RNA having a specific base sequence, and conversely, it is possible to predict and identify an RNA that is a target of binding of a PPR protein. Predicting and identifying a target is useful in clarifying genetic entities and is also useful in expanding the availability of targets.

[0039] Furthermore, according to the present invention, it is possible to design a PPR motif capable of selectively binding to a desired RNA base, and a protein having a plurality of PPR motifs capable of sequence-specifically binding to a desired RNA. When designing, the sequence information of a natural PPR motif can be used as a reference for the portion other than the amino acid at the important position in the PPR motif. Alternatively, the design may be performed by using a natural type as a whole and substituting only the amino acid at the relevant position. The number of repetitions of the PPR motif can be appropriately determined depending on the target sequence, and can be, for example, 2 or more, and can be 2 to 20.

[0040] In designing, consideration may be given to a combination of amino acids other than 1, 4, and "ii" or 4 and "ii". For example, consideration of amino acids 8 and 12 described in the above-mentioned Patent Document 2 may be important in order to exhibit RNA binding activity. According to the studies of the present inventors, the combination of A8 of a certain PPR motif and A of the same PPR motif 12 and may cooperate in RNA binding. A8 can be a basic amino acid, preferably lysine, or an acidic amino acid, preferably aspartic acid; A 12 can be a basic amino acid, a neutral amino acid, or a hydrophobic amino acid.

[0041] The designed motif or protein can be prepared by a method well known to those skilled in the art. That is, the present invention provides a PPR motif that selectively binds to a specific RNA base focusing on a combination of amino acids 1, 4, and "ii", or a combination of amino acids 4 and "ii", and a PPR protein that specifically binds to an RNA having a specific sequence. In particular, when considering the effect on fertility as the function of the PPR protein, it was found that the 4th amino acid (A4) and the "ii" amino acid are effective in either the above combination of three amino acids or the combination of two amino acids. Such motifs and proteins can be prepared in relatively large quantities by a method well known to those skilled in the art, and such a method may include determining a nucleic acid sequence encoding the desired motif or protein from the amino acid sequence of the motif or protein, cloning it, and preparing a transformant that produces the desired motif or protein.

[0042] Preparation of the complex and its use: The PPR motif or PPR protein provided by the present invention can be linked to a functional region to form a complex. A functional region refers to a portion having a specific biological function in a living body or cell, such as an enzyme function, a catalytic function, an inhibitory function, or an enhancing function, or a portion having a function as a label. Such a region is, for example, a protein, a peptide, a nucleic acid, a physiologically active substance, or a drug. An example of a functional region that is a protein is ribonuclease (RNase). Examples of RNases are RNase A (e.g., bovine pancreatic ribonuclease A: PDB 2AAS) and RNase H. Such a complex does not exist in nature and is novel.

[0043] Furthermore, the complex obtained by the present invention may be capable of delivering a functional region in an RNA sequence-specific manner in vivo or in cells and making it function. This may enable RNA sequence-specific modification or destruction in vivo or in cells, similar to zinc finger proteins (Non-Patent Document 1, supra) and TAL effectors (Non-Patent Document 2, supra, Patent Document 1, supra), and may also enable new functions to be imparted. Furthermore, it may enable drugs to be delivered in an RNA sequence-specific manner. Thus, the present invention also provides a method for delivering a functional substance in an RNA sequence-specific manner.

[0044] Some PPR proteins are known to be important in obtaining F1 seeds for hybrid vigor as fertility restorers that act in pollen formation (male gamete) formation. The present invention is expected to identify as yet unidentified fertility restorers and develop techniques for highly utilizing the factors. For example, in a case revealed in the examples of the present application, by detecting amino acid polymorphisms in a specific PPR motif in a PPR protein gene that acts as a fertility restorer for cytoplasmic male sterility, it is possible to determine whether the PPR protein gene of a test sample is a genotype related to fertility or a genotype related to sterility based on the association between the polymorphism in the gene and fertility. In this case, the PPR protein genes to be detected for polymorphism include, for example, in the case of radish, family genes located at a locus homologous to the "ORF687 gene" encoding the ORF687 protein of Kosena (named Enko B), genes having 90% or more amino acid sequence identity with Enko B, and genes having 90% or more nucleotide sequence identity with the "ORF687 gene" encoding Enko B. Here, the family genes located at a locus homologous to the "ORF687 gene" encoding the ORF687 protein of Kosena (named Enko B) include the genes shown in Figures 21 and 22 (Kosena B, Comet B, Enko A, Comet A, Icicle CA, rrORF690-1, rrORF690-2, PC PPR-A, PC PPR-BL) are included, but are not limited to these. In addition, a gene having 90% or more amino acid sequence identity with Enko B, and a gene having 90% or more base sequence identity with the "ORF687 gene" encoding Enko B can be obtained by searching a gene database, and the species of origin is not limited to radish. In addition, the PPR motif is a PPR motif consisting of a polypeptide having a length of 30 to 38 amino acids and represented by the above formula 1, and the PPR protein may be characterized by containing one or more such PPR motifs (preferably 2 to 16). In the case of polymorphism in this PPR motif, a polymorphism based on a combination of amino acids 1, 4, and "ii", or a combination of amino acids 4 and "ii", which have been revealed in the present invention to be responsible for binding to RNA in each PPR motif, can be used. As shown by the P value calculated in Figure 4B or 4D, among the combinations of amino acids responsible for binding to RNA in such PPR motifs, amino acid 4 plays the most important role, followed by amino acid ii. It was also revealed that, in comparison with the PPR protein of Enko B, it is important for the function as a fertility restorer that the 4th amino acid of all PPR motifs in the protein encoded by the gene to be tested is identical to that of Enko B, or that the "ii" amino acid of all corresponding PPR motifs is identical to that of Enko B. It has also been revealed that, similar to fertility restoration, some PPR proteins act on speciation. Identification and modification of the target RNA of the PPR protein raises the possibility of interspecies hybridization, which was previously impossible. Furthermore, since most PPR proteins act on RNA in mitochondria and chloroplasts, the novel PPR protein provided by the present invention will contribute to the modification and improvement of functions related to photosynthesis, respiration, and synthesis of useful metabolites.

[0045] On the other hand, in animals, abnormalities in the PPR protein identified as LRPPRC are known to cause Leigh syndrome French Canadian (LSFC; subacute necrotizing encephalomyelopathy). The present invention can contribute to the treatment (prevention, treatment, and inhibition of progression) of LSFC.

[0046] In addition, PPR proteins are involved in all steps of RNA processing found in organelles, including cleavage, RNA editing, translation, splicing, and RNA stabilization. The present invention is expected to modify the expression of a desired RNA by modifying the binding base selectivity of the PPR motif.

[0047] The PPR proteins used in the present invention function exclusively to specify editing sites in RNA editing (conversion of genetic information on RNA; in many cases, C → U) (see References 2 and 3 below). This type of PPR protein has an additional motif on the C-terminus that is suggested to interact with an RNA-mutagenase. It is expected that PPR proteins with such a structure will be able to introduce nucleotide polymorphisms or treat diseases or conditions caused by nucleotide polymorphisms.

[0048] Some PPR proteins also have an RNA cleavage enzyme attached to the C-terminus. By modifying the binding RNA base selectivity of the N-terminus PPR motif, an RNA sequence-specific RNA cleavage enzyme can be constructed. In addition, a complex linked to a labeling moiety such as GFP can be used to visualize the desired RNA in vivo.

[0049] On the other hand, some of the existing PPR proteins act on DNA. One is reported to be a transcription activator of mitochondrial genes, and the other is a transcription activator localized in the nucleus. Therefore, based on the knowledge obtained in this invention, it may be possible to design a protein factor that binds to a desired DNA sequence. EXAMPLES

[0050] Example 1: Collection of PPR proteins involved in RNA editing and their target sequences With reference to the information shown in Figure 2, the PPR proteins involved in Arabidopsis RNA editing that have been analyzed so far (SEQ ID NOs: 2 to 24) were listed in the Arabidopsis Genome Information Database (MATDB: http: / / mips.gsf.de / proj / thal / db / index.html), and the surrounding sequences of the target RNA editing sites (SEQ ID NOs: 48, 50, 53, 55, 57, 59, 60, 61, 62, 63, 64, 65, 68, 69, 70, 71, 73, 74, 76, 78, 80, 122, 206, 228, 232, 252, 284, 316, 338, 339, 358, 430, 433, 455, 552, 563) were listed in the RNA editing database (http: / / biologia.unical.it / py The RNA sequences were collected from the 31 bases upstream of the edited C (cytosine) residue. All the collected proteins and the RNA editing sites corresponding to each protein are shown in Figure 2.

[0051] The PPR motif structures in the proteins were assigned amino acid numbers defined in the present invention along with information from the Uniprot database (http: / / www.uniprot.org / ). The PPR motifs and their amino acid numbers contained in the 24 Arabidopsis thaliana PPR proteins (SEQ ID NOs: 2 to 25) used in the experiment are shown in FIG. 3.

[0052] Example 2: Identification of Amino Acids That Confer Binding Base Selectivity Previous studies have shown that PPR proteins involved in RNA editing have motifs (E, E+, and DYW motifs, although DYW is often absent) with specific conserved amino acid sequences at their C-terminus. It has been suggested that the dozen or so amino acids in the E+ motif are necessary for the conversion of C (cytosine) to U (uracil) rather than for selective binding to RNA (Reference 3). In addition, a past non-patent paper has suggested that the information necessary for recognizing the edited C is contained in 20 bases upstream and 5 bases downstream. In other words, it can be predicted that the multiple PPR motifs in the PPR protein recognize "somewhere" in the upstream sequence of the edited C, and that the E+ motif is located in the vicinity of the edited C. Furthermore, it is possible that the PPR motif recognizes the RNA residues in the upstream sequence to which it binds (Figure 4A).

[0053] This possibility was verified using the 24 RNA-editing PPR proteins of Arabidopsis thaliana described in Example 1 and their target RNA sequences. First, the last PPR motif in the protein was placed at the first base of the C to be edited, and all PPR motifs were aligned with the RNA residues in a one-to-one correspondence and linear continuity (FIG. 4A, alignment P1). Next, the RNA sequence was shifted one base at a time to the right to obtain alignments P2 to P6. In the data set of alignments P1 to P6, information on the RNA residues corresponding to each PPR motif was collected.

[0054] For PPR proteins that work on one RNA editing site, 1 point was assigned to the RNA residue (A, U, G or C) that appeared. For PPR proteins that work on RNA editing on two or three sites, 0.5 or 0.3 points were assigned to the RNA residue that appeared, respectively. Next, the amino acid types were rearranged for each amino acid number in the PPR motif. Usually, the types of amino acids and the types of RNA residues that appear can be predicted to be random (high-randomness or high-entropy) (e.g., upper right diagram of Figure 4A). However, if an amino acid at a specific site has the ability to select a binding RNA base, in the correct alignment (P1 to P6 above), the corresponding RNA base is predicted to converge to one type or a limited type (low randomness or low entropy; e.g., lower right diagram of Figure 4A).

[0055] The above low randomness was calculated for all amino acid numbers of the PPR motif for the data set of the alignments P1 to P6 created above. Low randomness was calculated by a chi-square test against the theoretical value (average of all base occurrence frequencies) (e.g., Figure 5).

[0056] As a result, the significance value P<0.01 (probability of 1% or less) was calculated for the 1st, 4th, and “ii” (-2) amino acids in the alignment of P4 (Figure 4B). This indicates that the last PPR motif in the RNA-editing PPR protein is located four bases before the edited C, and that three amino acids (1st, 4th, and “ii”) are responsible for the binding RNA base selection. In addition, the fact that no significant P value was calculated for the alignments of P3 and P5 indicates that there is no interference from the PPR motifs on both sides, that is, one PPR motif recognizes one RNA residue and is independent of the motif composition. No significant P value was obtained for the other amino acids in the alignment P4, and for all amino acids in the other alignments (Figure 6). In addition, when the RNA bases were classified into purines (A and G) or pyrimidines (C and U) (RY) and the same calculation was performed, only the 4th amino acid was obtained with a highly significant P value (P<0.01) (Figure 4C). This indicates that the 4th amino acid mainly distinguishes between purine and pyrimidine of the RNA base to which it binds. We further analyzed the ability of the RNA recognition amino acids in the PPR motif shown in Figure 4C to specify the binding base in detail. As a result, we found that in addition to mainly distinguishing between purine and pyrimidine (RY) of the base to which the 4th amino acid binds, the "ii" (-2) amino acid also functions to distinguish between amino (A and C) and keto (G and U) (MK) types of bases (Figure 4D).

[0057] Combinations of three amino acids (1, 4, and "ii") that are used three or more times were defined as triPPR codes among the RNA recognition codes of PPR motifs, and their P values ​​were calculated to determine their ability to specify RNA bases. Part of the identified triPPR codes is shown in Figure 4E.

[0058] Since the amino acids at these three positions are highly diverse, the binding RNA base specification ability of two amino acids (1&4, 1&"ii" or 4&"ii") was calculated, and a significant P value was calculated for the combination of amino acids 4&"ii" (Figure 7). Therefore, the combination of amino acids 4&"ii" that is used three or more times was designated as a diPPR code, one of the RNA recognition codes of the PPR motif. The identified triPPR code and diPPR code are shown in Figure 8.

[0059] Example 3: Verification of the identified RNA recognition code The RNA recognition code of the PPR motif identified was verified using the RNA-editing PPR protein of Arabidopsis thaliana. For the verification, the RNA-editing PPR protein of Physcomitrella patens was used. It has already been revealed that RNA editing takes place at a total of 13 sites (11 sites in mitochondria and 2 sites in chloroplasts; SEQ ID NOs: 32 to 44) in Physcomitrella patens (hereinafter referred to as moss). Furthermore, it has been found that RNA editing takes place at 6 PPR proteins (PpPPR 56, 71, 77, 78, 79, and 91) have been shown to function in RNA editing at nine sites. The proteins and the corresponding RNA editing sites are shown in Figure 9.

[0060] The verification was performed as shown in FIG. 10. First, the amino acid sequence information of the moss PPR protein was obtained from a non-patent paper (SEQ ID NO: 26 to 31; FIG. 2 and FIG. 9), and three amino acids (1, 4, and "ii") were extracted from each PPR motif according to the PPR motif model defined in FIG. 1. If the combination of the three extracted amino acids matches the triPPR code identified from Arabidopsis thaliana, it was replaced with the binding base score matrix presented by that code. Next, among the PPR motifs that could not be converted by the triPPR code, if it matches the diPPR code, the motif was replaced with the binding base score matrix of the diPPR code. In parallel, the sequence around the RNA editing site (31-mer sequence with the edited C at the 3' end) was obtained from a non-patent paper (SEQ ID NO: 32 to 44; FIG. 2, FIG. 9, and FIG. 16), and replaced with the numerical matrix of the RNA sequence as shown in FIG. 10. Next, following the above alignment P4 (corresponding to the 4 bases before C where the last PPR motif is edited), the protein binding base scoring matrix and the number matrix of the RNA sequence were multiplied by each square, and the sum of the obtained values ​​was calculated as the matching score between the protein and the RNA sequence. This calculation was performed using the triPPR code, diPPR code, and each PPR binding base scoring matrix (PPR scoring matrix) This was carried out at.

[0061] This calculation was performed for all moss RNA editing sites (13 sites) for one protein. In addition, similar calculations were performed for 34 RNA sequences of the RNA editing sites in Arabidopsis chloroplasts (FIG. 16, SEQ ID NOs: 45 to 78) as reference sequences for sequences surrounding the RNA editing sites.

[0062] Next, a normal distribution curve was drawn from the protein matching values ​​for each RNA sequence, and a provisional P value for the matching value for each RNA sequence was calculated for each triPPR code and diPPR code.

[0063] The final P value (match between protein and RNA sequence) was calculated as the product of the provisional P values ​​of the triPPR code and the diPPR code.

[0064] The matching scores between each moss PPR protein and the 13 moss RNA editing sites are shown in Figure 11. As a result of the analysis, six of the seven proteins were computationally identified as having the correct RNA editing site. In other words, this analysis means that three amino acids (1, 4, and "ii") contain all the information regarding the binding RNA base designation of the PPR motif. In other words, it shows that it is possible to search for PPR proteins that bind to the intended RNA sequence by referring to the information of the combination of three or two amino acids (triPPR, diPPR code) shown in Figure 8. At the same time, it shows that it is possible to synthesize an artificial protein that binds to the intended RNA sequence by using or linking a PPR motif that has the amino acid information.

[0065] Example 4: Identification of target molecules of uncharacterized RNA-editing PPR proteins Next, we conducted an analysis using Arabidopsis thaliana, which contains more RNA editing sites than moss (34 sites in the chloroplast genome (sequence numbers 45 to 78), 488 sites in the mitochondrial genome (sequence numbers 79 to 566), see Figure 6). To verify the prediction accuracy, we predicted the RNA variant sites of the 24 PPR proteins used in the code extraction. As a result, for chloroplast-localized PPR proteins, 10 out of 13 predicted at least one correct RNA editing site with the highest P value. For mitochondrial-localized PPR proteins, 8 out of 11 predicted correct RNA editing sites within the top 20 (Figure 12). Based on this prediction accuracy verification, we predicted the target RNA editing site of a PPR protein with unknown function. The AHG11 mutant is a mutant that causes abnormalities in the abscisic acid pathway, and the protein encoded by its gene (ahg11, at2g44880) has a typical RNA editing PPR protein-like motif structure (Figure 13; sequence number 1). We predicted RNA editing sites and experimentally verified the top 20 RNA editing sites in 405 mitochondria and 30 chloroplasts. As a result, the mitochondrial nad4 predicted with the seventh highest P value was It was revealed that only 376 RNA editing was abnormal in the mutant (Figure 13).

[0066] Next, we sequenced the entire organelle genome, i.e., about 3 × 10 5 We attempted to identify target RNA sequences from the RNA sequence dataset of . For this analysis, we used the probability matrix of the PPR code shown in Figure 8. In addition, background frequencies were applied to motifs with amino acid combinations that did not match the diPPR or triPPR codes. The probability matrix of the protein we created was used together with the complete chloroplast sequence of Arabidopsis thaliana (AP000423) through FIMO analysis of the MEME suite (http: / / meme.nbcr.net / meme423). 6 1 / fimo-intro.html).

[0067] As a result, we were able to accurately predict the target RNA sequences of CRR4 and CRR21. In addition, we improved the codes by extracting PPR codes from moss PPR proteins (Figure 15), and the prediction accuracy was significantly improved for some proteins.

[0068] These results show that by using the identified PPR code, it is possible to identify one correct target sequence from hundreds of thousands of RNA sequence patterns. Conversely, by searching for a PPR motif that has amino acids in accordance with the code at the relevant positions (1, 4, "ii"), it is possible to identify a protein that binds to a desired useful RNA sequence. Alternatively, it is shown that by linking PPR motifs, it is possible to create an artificial RNA-binding protein with high sequence selectivity. It is also understandable to those skilled in the art that it is possible to obtain the desired RNA-binding selectivity by introducing a mutation to combine amino acids at the relevant positions in accordance with the PPR code.

[0069] In Figure 15, the RNA binding base selectivity of triPPR code and diPPR code was evaluated by P value. PPR codes that showed significant P values ​​(P < 0.05) can be inferred to have high RNA binding base selectivity.

[0070] Example 5: Prediction of target RNA sequences of radish Rf Next, based on the findings of the present invention, the function of the PPR protein that acts as a fertility restorer for cytoplasmic male sterility was evaluated (Examples 5 to 9).

[0071] Cytoplasmic male sterility (CMS) is a trait in which male gametes do not function normally due to mutations in the cytoplasmic genome, especially the mitochondrial genome. This trait is often counteracted by the Restorer of Fertility (Rf) gene present in the nucleus, which restores male gametes to normal. It is used in hybrid breeding and is one of the agriculturally important traits. In this CMS-Rf system, the Rf gene is known to often code for a PPR protein.

[0072] The Ogura type (also known as Kosena type) cytoplasm used in the hybrid breeding method for radish and rapeseed is derived from the expression of the orf125 gene in the mitochondrial genome, and the presence of the nuclear-coded orf687 gene relieves sterility and makes the plant fertile. The orf687 gene product is a PPR protein, and it is thought that it acts on RNA containing orf125 to inactivate its expression, thereby relieving sterility.

[0073] However, previous breeding analyses have revealed that orf687-like genes in various radish lines have amino acid polymorphisms, and that these amino acid polymorphisms affect the functionality of the genes as fertility restorers. However, no method has been established to infer the functionality of the genes from their amino acid sequences.

[0074] Therefore, we first identified the PPR motif from the amino acids of the ORF687 protein (named Enko B) of the radish cultivar Enko, which is known to function as a dominant Rf, extracted the amino acids (1, 4, ii) that control the base specification ability, and converted them into a PPR code. After that, we performed target RNA sequence prediction for transcripts containing mitochondrial orf125 (Figure 19).

[0075] In parallel, we biochemically analyzed the characteristics of three ORF687-like proteins: the ORF687 protein (named Enko B) from the radish cultivar, Sonobeni, which is known to function as a dominant Rf; an ORF687-like protein (named enko A) also contained in Sonobeni and which is very similar to ORF687 but functions as a recessive gene; and a gene (named kosena B; a recessive gene) homologous to Sonobeni ORF687 present in the genome of a different radish cultivar, Kosena.

[0076] (5-1) Preparation of genomic DNA from radish Radish was cultured in Murashige and Skoog medium (containing 2% sucrose and 0.5% Gellangam) for 3 weeks. Green leaves (0.5 g) of the cultured plants were extracted with phenol / chloroform, and then ethanol was added to insolubilize the DNA. The recovered DNA was dissolved in 100 μl of TE solution (10 mM Tris-HCl (pH 8.0), 1 mM EDTA), and 10 units of RNase A (DNase-free, Takara Bio) was added and reacted at 37°C for 30 minutes. The reaction solution was then extracted again with phenol / chloroform, and the DNA was recovered by ethanol precipitation. 10 μg of DNA was obtained.

[0077] (5-2) Cloning of the gene encoding the ORF687-like protein Radish genomic DNA was used as a template, and Enko B was used as an oligonucleotide primer (Enko BF primer and Enko BR primers; set forth in SEQ ID NOs: 567 and 568, respectively), and Kosena B is an oligonucleotide primer (Kosena BF primer and kosena BR primer; set forth in SEQ ID NOs: 569 and 570, respectively), Enko A is an oligonucleotide primer (Enko AF Primer and Enko Each was amplified by PCR using AR primers (seen in SEQ ID NOs: 571 and 572, respectively) and 50 μl of the reaction solution at 95° C. for 30 seconds, 60° C. for 30 seconds, and 72° C. for 30 seconds for 25 cycles, using KOD-FX (TOYOBO) as a DNA elongation enzyme.

[0078] The obtained DNA fragment was cloned into pBAD / Thio-TOPO vector (Invitrogen) according to the protocol attached to the product. The DNA sequence was determined and it was confirmed that the sequence was homologous to the target DNA sequence (Enko B, SEQ ID NO: 573; kosena B, SEQ ID NO: 574; Enko A, SEQ ID NO: 575).

[0079] (5-3) Preparation of recombinant ORF687-like protein The plasmid obtained above was transformed into Escherichia coli TOP10 strain (Invitrogen). This E. coli was cultured at 37°C in 300 ml of LB medium (1 L Erlenmeyer flask containing 300 ml of medium) containing ampicillin at a concentration of 100 μg / ml. When the turbidity of the culture liquid reached an absorbance of 0.5 at a wavelength of 600 nm, the inducer L-arabinose was added to a final concentration of 0.2%, and the culture was continued for another 4 hours.

[0080] After harvesting by centrifugation, the cells were suspended in 200 ml of buffer A (50 mM Tris·HCl pH 8.0, 500 mM KCl, 2 mM imidazole, 10 mM MgCl2, 0.5% Triton X100, 10% glycerol) containing 1 mg / ml lysozyme, and disrupted by sonication and freeze-thawing. After centrifugation at 15,000×g for 20 min, the supernatant was collected as a crude extract.

[0081] This crude extract was applied to a column packed with nickel column resin (ProBond A, Invitrogen) equilibrated with buffer A.

[0082] Column chromatography was performed using a two-step concentration gradient in which the target protein was eluted with Buffer A containing 200 mM imidazole after thorough washing with Buffer A containing 20 mM imidazole. The obtained protein is a fusion protein having the amino acid sequence described in SEQ ID NO: (Enko B, SEQ ID NO: 576; Kosena B, SEQ ID NO: 577; Enko A, SEQ ID NO: 578), as well as the amino acid sequence of thioredoxin at the N-terminus to increase solubility, and a histidine tag sequence at the C-terminus. 100 μl of the purified fraction was dialyzed against 500 mL of Buffer E (20 mM Tris·HCl pH 7.9, 60 mM KCl, 12.5 mM MgCl2, 0.1 mM EDTA, 17% glycerol, 2 mM DTT) to obtain a purified preparation.

[0083] (5-4) Preparation of substrate RNA As substrate RNAs, three kinds of RNAs, RNAa, RNAb, and RNAc, containing the mitochondrial DNA sequence of Ogura-type cytoplasmic radish, were used.

[0084] RNAa was amplified by using oligonucleotide primers AF and AR (SEQ ID NOs: 579 and 580, respectively), RNAb was amplified by using oligonucleotide primers BF and BR (SEQ ID NOs: 581 and 582, respectively), and RNAC was amplified by using oligonucleotide primers CF and CR (SEQ ID NOs: 583 and 584), and 50 μl of the reaction solution containing 10 ng of the above-mentioned Ogura-type cytoplasmic radish DNA as template DNA was amplified by PCR using KOD FX (TOYOBO) as a DNA elongation enzyme for 25 cycles of 95°C for 30 seconds, 60°C for 30 seconds, and 72°C for 30 seconds. A T7 promoter sequence was added to each forward primer (-F) to synthesize substrate RNA in vitro.

[0085] The resulting DNA fragment was purified by running it on an agarose gel and then cutting it out from the gel. The purified DNA fragment was used as a template and NTP mix (10 nmol GTP, CTP, ATP, 0.5 nmol UTP), 4 μl [ 32 P] α-UTP (GE Healthcare, 3000 Ci / mmol) and T7 RNA polymerase (Takara Bio) were added to 20 μl of the reaction solution, which was then reacted at 37° C. for 60 minutes to synthesize substrate RNA.

[0086] The substrate RNA was extracted with phenol / chloroform and precipitated with ethanol. The entire amount was then subjected to electrophoresis on a denaturing 6% polyacrylamide gel containing 6 M urea, and exposed to X-ray film for 60 seconds. 32 P-labeled RNA was detected.

[0087] next, 32 P-labeled RNA was excised from the gel and immersed in 200 μl of gel elution solution (0.3 M sodium acetate, 2.5 mM EDTA, 0.01% SDS) at 4°C for 12 hours to elute the RNA from the gel. The radioactivity of 1 μl of the RNA was measured, and the total amount of synthesized RNA was calculated. After ethanol precipitation, the RNA was dissolved in ultrapure water to a concentration of 2500 cpm / μl (1 fmol / μl). This preparation method usually yielded approximately 100 μl of RNA at 2500 cpm / μl.

[0088] (5-5) Protein-RNA binding experiment Recombinant proteins of Enko B (Rf), Kosena B (rf), and Enko A (rf; an ORF687-like protein present in Kosena cultivar) were constructed and their RNA-binding activities were examined.

[0089] The RNA binding activity of the recombinant proteins (Enko B (SEQ ID NO: 576), Kosena B (SEQ ID NO: 577), Enko A (SEQ ID NO: 578)) was analyzed by gel shift analysis. 375 pM (7.5 fmol / 20μL) of the above substrate RNA (BD120) and 0-2500 nM of the recombinant protein were mixed in 20μL of reaction solution (10 mM Tris-HCl pH 7.9, 30 mM KCl, 6 mM MgCl2, 2 mM DTT, 8% glycerol, 0.0067% Triton X-100) and reacted at 25℃ for 15 minutes. After that, 4μL of 80% glycerol solution was added to the reaction solution, and 10μL was developed on a 10% native polyacrylamide gel containing 1×TBE (89 mM Tris-HCl, 89 mM Boric acid, 2 mM EDTA), and the gel was dried after electrophoresis.

[0090] The radioactivity of the RNA in the gel was measured using a bioimaging analyzer BAS2000 (Fuji Film).

[0091] Example 6: RNA binding experiments using recombinant proteins FIG. 17 shows the binding analysis of Enko B protein with RNA containing the cytoplasmic male sterility (CMS) gene. FIG. 17A shows a schematic diagram of the vicinity of mitochondrial orf125, and also shows the regions of RNA a, RNA bc, RNA b, and RNA c used in the binding experiment. FIG. 17B shows the RNA binding of Enko B protein. Enko B protein (1.4 nmol) and 32 P-labeled RNA bc (0.1 ng) was reacted with unlabeled RNA a, RNA bc, RNA b, and RNA c (used as competitive inhibitors, x5 and x10 w / w relative to RNA bc) in a 20 μL reaction solution, and a gel shift competition experiment was performed. Complex (▽) on the left of the figure indicates a complex of protein and RNA, and Free (▼) indicates RNA only.

[0092] As shown in these figures, the binding of proteins and RNA is 32 This appears as a difference in the mobility of P-labeled RNA. 32The molecular weight of the P-labeled RNA-protein complex is 32 This is because its molecular weight is larger than that of the P-labeled RNA alone, and its mobility during electrophoresis is slower. In this experiment, recombinant EnkoB protein was prepared, and its binding to mitochondrial RNA containing orf125 was verified by competitive gel shift analysis. RI-labeled RNAb was mixed with the protein, and then unlabeled RNA was added. In other words, the band with the greater decrease in signal intensity at the position indicated by Complex indicates that the RNA added as competitor binds to the protein, i.e., the RNA region to which EnkoB binds with high affinity. As a result, it was revealed that EnkoB binds strongly to the region of RNAb.

[0093] The candidate sequence of No. 208 showed the most significant P value in the predicted binding sequence shown in Figure 19, and is located exactly at the 3' end of tRNA methionine. However, previous analyses have shown that there is no difference in the amount of tRNA and the shape of RNA containing orf125 (presence or absence of truncation) between sterile and restored lines, and in an in vitro binding experiment (Figure 17B), the RNAa sequence containing No. 208 does not bind to Enko B, so we determined that this region is not related to the fertility / sterility of Ogura-type cytoplasm.

[0094] Therefore, we focused on the regions of No. 316, 352, and 373 contained in RNAb and continued our analysis. RNAb consists of 125b. We attempted to narrow down the binding region to 20b units by scanning mutation, but were unable to narrow it down to a single site (data not shown). Therefore, it was thought that there may be multiple binding sites for Enko B in RNAb.

[0095] Example 7: RNA binding activity of Rf-like proteins FIG. 18 shows the binding of ORF687-like proteins to RNA. Of these, FIG. 18A shows the results of analyzing the binding of Enko B (Rf), Kosena B (rf), and Enko A (rf) to RNAb by gel shift analysis with respect to the RNA binding properties of ORF687-like proteins. FIG. 18B is a graph of the results of (A), and the dissociation constant (KD), which represents the RNA binding ability of each protein, was calculated from this graph. FIG. 18C shows the compatibility values ​​of Enko B (Rf), Kosena B (rf), and Enko A (rf) with potential binding regions calculated in the same manner as in FIG. 19.

[0096] As a result, in the non-competitive state, all three proteins (Enko B, Kosena B, Enko A) bound to RNA with high affinity to RNAb. The RNA binding activity of Kosena B in the competitive state was analyzed, but no clear difference was observed between it and Enko B (Figures 18A and 18B).

[0097] Kosena B often shows slightly lower RNA binding activity than Enko B (KD is about 2-fold lower). However, the activity of general RNA binding is often more than 10-fold higher, and this difference cannot be considered significant.

[0098] Even in predictions based on the PPR code, there was no clear difference between the match scores for the corresponding regions between the proteins (Figure 18C). Based on this, we decided to examine the possibility that the difference between EnkoB and kosenaB was not simply a difference in RNA binding affinity, but rather a difference in the action after binding.

[0099] Furthermore, the predicted binding sequence of the fertility restorer acting on Ogura-type cytoplasm is shown in Figure 19. Here, Figure 19A shows the predicted binding sequence of Enko B protein using the PPR code, and the lower diagram of Figure 19A shows the structure of RNA containing the CMS gene orf125. For the regions RNAa to RNAc in Figure 19A, see Figure 17. In Figure 19A, we focused on No. 208, 230, 316, 352, and 373 among the regions that showed significantly high P values ​​(Figure 19A).

[0100] Next, Figure 19B shows the logo representation of the target RNA sequence predicted from the ORF687 protein sequence (the sequence of the region (No. 208, 316, 352, 373) that showed a significant P value), the candidate binding RNA sequence, and the logo representation of the target RNA sequence predicted from the ORF687-like protein (Kosena B) sequence of Kosena, a radish cultivar with recessive rf. The predicted binding base of Kosena B, which is a recessive rf, is also shown.

[0101] EnkoB and KosenaB were found to have different designated bases due to amino acid polymorphisms in the second and third PPR motifs (Rf is UA, rf is GC). This difference was predicted to be directly related to the functional difference between Rf and rf.

[0102] Example 8: RNA structure prediction and analysis Computer prediction and in vitro RNA binding experiments suggested that Rf may bind to regions of RNAb, especially to nos. 316, 352, and 373. In vitro analysis suggested that there may be multiple binding sites in RNAb. Therefore, we predicted the secondary structure of the RNAb sequence and focused on the relevant regions.

[0103] The results are shown in Figure 20. Here, Figure 20 shows the secondary structure and structural changes of the candidate binding RNA region of ORF687-like protein. Figure 20A shows the secondary structure of the region containing No. 306 and the predicted binding site of ORF687-like protein, with each PPR motif shown in a box with the corresponding base. The second and third PPR motifs, which are significantly different between EnkoB (Rf) and Kosena B (rf), are highlighted. Figure 20B shows the secondary structure of the region containing No. 352 and 373 and the predicted binding site of ORF687-like protein. Figure 20C shows the structural change of RNAb by Enko B, in which RNAb and Enko B protein were mixed and then a double-strand selective RNA degrading enzyme (Rnase V1) was added.

[0104] As a result, it was revealed that the No. 316 region corresponds to the stem-loop structure immediately downstream of the start codon of orf125 (Fig. 20A). In addition, the 2nd and 3rd PPR motifs, which are polymorphic between Enko B and Kosena B, were located in the double strand at the base of the stem-loop. In particular, the corresponding base of the 3rd PPR motif is A in Enko B, whereas it is C in Kosena B (see Fig. 19B). Based on these findings, a working hypothesis was proposed that EnkoB binds to the region, promotes the formation of the stem-loop structure, and inhibits the translation of orf125.

[0105] A double-stranded structure was predicted for the No. 352 and 373 regions, and it was thought that the Rf protein would bind to both sides (Figure 20B). However, in this case, Rf binding would be predicted to destroy the structure (promote single-strand formation). Furthermore, the difference between Rf and rf, that is, the difference in the bases and structure corresponding to the 2nd and 3rd PPR motifs, was not considered, and a specific molecular mechanism could not be predicted.

[0106] Therefore, the internally labeled RNA was mixed with protein, and RNase V1 was added to the mixture to perform limited degradation of the labeled RNA. RNase V1 is an RNase that selectively cleaves only the double-stranded region of RNA. As a result, it was shown that the substrate RNA was degraded quickly in the presence of protein, that is, the formation of double-stranded RNA was promoted in the presence of Rf (Enko B) (Figure 20C). In other words, it was thought that the translation inhibition due to the formation of double-stranded RNA of orf125 mRNA by Rf was the main cause of fertility restoration in Ogura-type cytoplasmic male sterility.

[0107] Example 9: Functional assessment of fertility restoration ability of ORF687-like genes ORF687-like genes have been isolated from various radish varieties, and their functionality as Rf has been inferred through crossbreeding experiments. However, the amino acid sequences of each gene are very similar, and it is not possible to determine their functionality as Rf from the overall amino acid conservation.

[0108] In this example, first, sequence analysis of the ORF687-like protein was performed. Specifically, sequence analysis was performed as a PPR protein using the protein sequences shown in SEQ ID NOs: 576 to 578 and 585 to 591 as materials. All sequences were used as query sequences to obtain sequence alignments using CLUSTALW (http: / / www.genome.jp / tools / clustalw / ). Web-based domain analysis software, Pfam (http: / / pfam.sanger.ac.uk / ), InterProScan (http: / / www.ebi.ac.uk / Tools / InterProScan / ), Prosite(http: / / www.expasy.org / prosite / ), Using this method, we created an alignment of ORF687-like proteins and analyzed the PPR motif structure of each protein. The results are shown in Figure 21. All ORF687-like genes are composed of 16 PPR motifs (Figure 21).

[0109] According to the obtained PPR motif model and the amino acid numbering shown in Non-Patent Document 5, the amino acids 1, 2, and “ii” (−2) were extracted and used to evaluate the function of the fertility restoration ability of the ORF-like protein.

[0110] Therefore, the functions of nine Rf-like genes were evaluated using the PPR code. As with EnkoB described above, the amino acids (1, 4, ii) that control the base specification ability were extracted and converted into the PPR code, and the amino acid species were used as the RNA binding window to evaluate the functionality (Figure 22). Enko B and Kosena B show 99.4% overall homology, but there are two amino acid polymorphisms in the RNA binding window, which are thought to be deeply involved in the dominance / recession of the ORF687-like gene in fertility restoration (Non-patent paper 4). On the other hand, in the Comet variety, the gene Comet B, which is located at a locus homologous to Enko B, shows 98.0% homology with Enko B and has exactly the same RNA binding window. This confirmed the finding that CometB is a dominant gene obtained from past mating tests. In addition, Enko A is a duplicated gene located near Enko B, but it was suggested that it is a recessive gene from the perspective of RNA recognition. These data suggest that the dominant / recessive nature of ORF687-like genes in fertility restoration depends on the amino acids (1, 4, ii) that control the base specification ability being identical in all the corresponding PPR motifs in ORF687-like genes, especially the same 4th amino acid (A4) or the same "ii" amino acid. Of these, it is considered to be especially important to have the same 4th amino acid (A4). From this point of view, it is considered that the genes located at the locus homologous to Enko B, rrORF690-1, rrORF690-2, icicle, and rORF690-3, which are located in various radish lines with unknown information on fertility, are important for the dominant / recessive nature of ORF687-like genes in fertility restoration. pprCA, P.C. PPR-A, PC PPR-BL has an RNA binding window different from that of the dominant gene Enko B, and these genes were also considered to be recessive rf genes.

[0111] The above results show that the PPR code shown in this invention can speed up the functional determination of industrially useful PPR proteins that act as fertility restorers. When applying a new line to a F1 hybrid breeding method using the CMS-Rf system, this technology makes it possible to determine whether or not the line has fertility restoration ability from the sequence of the candidate Rf gene. The inventors have performed functional determination of ORF687-like genes in 21 new radish varieties, and have succeeded in determining the dominant / recessive fertility restoration ability of 19 ORF-like genes (data not published). This technology is not limited to radishes with Ogura-type cytoplasm, but can be applied to various cytoplasms and plant species that use PPR proteins as Rf.

[0112] [Papers cited in the examples] Reference 1: Small, ID, and Peeters, N. (2000). The PPR motif - a TPR-related motif prevalent in plant organellar proteins. Trends Biochem. Sci. 25, 46-47. Reference 2: Lurin, C., Andres, C., Aubourg, S., Bellaoui, M., Bitton, F., Bruyere, C., Caboche, M., Debast, C., Gualberto, J., Hoffmann, B., et al. (2004). Genome-wide analysis of Arabidopsis pentatricopeptide proteins repeat reveals their essential role in organelle biogenesis. Plant Cell 16, 2089-2103. Reference 3: Okuda, K., Myouga, F., Motohashi, R., Shinozaki, K., and Shikanai, T. (2007). Conserved domain structure of pentatricopeptide repeat proteins involved in chloroplast RNA editing. Proc Natl Acad Sci USA 104, 8178-8183. Reference 4: Koizuka N, Imai R, Fujimoto H, Hayakawa T, Kimura Y, et al. (2003) Genetic characterization of a pentatricopeptide repeat protein gene, orf687, that restores fertility in the cytoplasmic male-sterile Kosena radish. Plant J 34: 407-415. Reference 5: Nakamura T, Yagi Y, Kobayashi K (2012) Mechanistic insight into pentatricopeptide repeat proteins as sequence-specific RNA-binding proteins for organellar RNAs in plants. Plant & Cell Physiology 53: 1171-1179

Claims

1. A composition comprising a nucleic acid encoding a PPR protein, The composition is for use in treating a disease mediated by any modification selected from the group consisting of translation, splicing, and stability of an RNA having a target sequence to which the PPR protein can bind, The PPR protein contains 2 to 20 PPR motifs, which are polypeptides having a length of 30 to 38 amino acids and are represented by the following formula 1: 【Chemistry 1】 (In formula 1: Helix A is a 12 amino acid long portion capable of forming an α-helical structure, Formula 2: 【Chemistry 2】 (In formula 2, A 1 ~A 12 each independently represents an amino acid), In formula 1, X is absent or a moiety consisting of 1 to 9 amino acids in length; In formula 1, Helix B is a portion capable of forming an α-helix structure, which is 11 to 13 amino acids long. In formula 1, L is a 2-7 amino acid long molecule of formula 3: 【Chemistry 3】 (In formula 3, each amino acid is numbered from the C-terminus side as "i" (-1), "ii" (-2), etc. However, L iii ~L vii may not exist.) A PPR motif in the PPR protein that selectively binds to one RNA base in the RNA having the target sequence 1 , A 4 , and L ii A composition characterized in that the combination of the three amino acids is any of the following: (3-1) A, which constitutes the PPR motif that selectively binds to U (uracil) 1 , A 4 , and L ii The combination of these three amino acids is, in order, valine, asparagine, and aspartic acid; (3-2) A, which constitutes the PPR motif that selectively binds to A (adenine) 1 , A 4 , and L ii The combination of these three amino acids is, in order, valine, threonine, and asparagine; (3-3) A, which constitutes the PPR motif that selectively binds to C (cytosine) 1 , A 4 , and L ii A combination of the three amino acids, in the order of valine, asparagine, and asparagine; (3-4) A, which constitutes the PPR motif that selectively binds to G (guanine) 1 , A 4 , and L ii The combination of these three amino acids is, in order, glutamic acid, glycine, and aspartic acid; (3-5) A, which constitutes a PPR motif that selectively binds to C or U 1 , A 4 , and L ii The combination of these three amino acids is, in order, isoleucine, asparagine, asparagine; (3-6) A, which constitutes the PPR motif that selectively binds to G 1 , A 4 , and L ii The combination of these three amino acids is, in order, valine, threonine, and aspartic acid; (3-7) A, which constitutes the PPR motif that selectively binds to G 1 , A 4 , and L ii The combination of these three amino acids is, in order, lysine, threonine, and aspartic acid; (3-8) A, which constitutes the PPR motif that selectively binds to A 1 , A 4 , and L ii The combination of these three amino acids is, in order, phenylalanine, serine, and asparagine; (3-9) A, which constitutes the PPR motif that selectively binds to C 1 , A 4 , and L ii The combination of the three amino acids is, in order, valine, asparagine, and serine; (3-10) A, which constitutes a PPR motif that selectively binds to A 1 , A 4 , and L ii The combination of these three amino acids is, in order, phenylalanine, threonine, and asparagine; (3-11) A, which constitutes a PPR motif that selectively binds to U or A 1 , A 4 , and L ii The combination of these three amino acids is, in order, isoleucine, asparagine, and aspartic acid; (3-12) A, which constitutes a PPR motif that selectively binds to A 1 , A 4 , and L ii The combination of these three amino acids is, in order, threonine, threonine, asparagine; (3-13) A, which constitutes a PPR motif that selectively binds to U or C 1 , A 4 , and L ii The combination of these three amino acids is, in order, isoleucine, methionine, and aspartic acid; (3-14) A, which constitutes the PPR motif that selectively binds to U 1 , A 4 , and L ii The combination of these three amino acids is, in order, phenylalanine, proline, and aspartic acid; (3-15) A, which constitutes the PPR motif that selectively binds to U 1 , A 4 , and L ii The combination of these three amino acids is, in order, tyrosine, proline, and aspartic acid; (3-16) A, which constitutes the PPR motif that selectively binds to G 1 , A 4 , and L ii The combination of these three amino acids is, respectively, leucine, threonine, and aspartic acid.

2. A composition comprising a nucleic acid encoding a PPR protein, The composition is for use in treating a disease mediated by any modification selected from the group consisting of translation, splicing, and stability of an RNA having a target sequence to which the PPR protein can bind, The PPR protein comprises 2 to 20 PPR motifs, which are polypeptides having a length of 30 to 38 amino acids and represented by formula 1 as defined in claim 1; A PPR motif in the PPR protein that selectively binds to one RNA base in the RNA having the target sequence 4 , and L ii A composition, characterized in that the combination of the two amino acids is any of the following: (2-1) A, which constitutes the PPR motif that selectively binds to U 4 , and L ii The combination of these two amino acids is, in order, asparagine and aspartic acid; (2-2) A, which constitutes the PPR motif that selectively binds to C 4 , and L ii A combination of two amino acids, (2-3) A, which constitutes the PPR motif that selectively binds to A 4 , and L ii The combination of these two amino acids is, in order, threonine and asparagine; (2-4) A, which constitutes the PPR motif that selectively binds to G 4 , and L ii The combination of these two amino acids is, in order, threonine and aspartic acid; (2-5) A, which constitutes the PPR motif that selectively binds to A 4 , and L ii The combination of these two amino acids is, in order, serine and asparagine; (2-6) A, which constitutes the PPR motif that selectively binds to G 4 , and L ii The combination of these two amino acids is, in order, glycine and aspartic acid; (2-7) A, which constitutes the PPR motif that selectively binds to C 4 , and L ii The combination of these two amino acids is, in order, asparagine and serine; (2-8) A, which constitutes the PPR motif that selectively binds to U 4 , and L ii The combination of these two amino acids is, in order, proline and aspartic acid; (2-9) A, which constitutes a PPR motif that selectively binds to A 4 , and L ii The combination of these two amino acids is, in order, glycine and asparagine; (2-10) A, which constitutes the PPR motif that selectively binds to U 4 , and L ii The combination of these two amino acids is, in order, methionine and aspartic acid; (2-11) A, which constitutes a PPR motif that selectively binds to C 4 , and L ii The combination of these two amino acids is, in order, leucine and aspartic acid; (2-12) A, which constitutes the PPR motif that selectively binds to U 4 , and L ii The combination of these two amino acids is, in order, valine and threonine.

3. The composition according to claim 1 or 2, wherein the PPR protein comprises 2 to 16 PPR motifs represented by formula 1.

4. A composition comprising a nucleic acid encoding a PPR protein, The composition is for use in treating a disease mediated by any modification selected from the group consisting of translation, splicing, and stability of an RNA having a target sequence to which the PPR protein can bind, The PPR protein comprises 2 to 20 PPR motifs, which are polypeptides having a length of 30 to 38 amino acids and represented by formula 1 as defined in claim 1; A 1 , A 4 , and L ii The composition is characterized in that the combination of the three amino acids is any one of (3-1) to (3-16) defined in claim 1 depending on one RNA base in the RNA having the target sequence.

5. A composition comprising a nucleic acid encoding a PPR protein, The composition is for use in treating a disease mediated by any modification selected from the group consisting of translation, splicing, and stability of an RNA having a target sequence to which the PPR protein can bind, The PPR protein comprises 2 to 20 PPR motifs, which are polypeptides having a length of 30 to 38 amino acids and represented by formula 1 as defined in claim 1; A 4 , and L ii The composition is characterized in that the combination of the two amino acids is any one of (2-1) to (2-12) defined in claim 2, depending on one RNA base in the RNA having the target sequence.

6. The composition according to claim 4 or 5, wherein the PPR protein comprises 2 to 16 PPR motifs represented by formula I.

Images