JP3179055B2 - Protein identification method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a method for identifying a protein. More specifically, the present invention relates to a new method for easily and reliably identifying an unknown protein isolated from cells of an organism, and a search list used for the identification.

[0002]

2. Description of the Related Art In biological phenomena such as ontogenesis, tissue regeneration, and carcinogenesis, various changes occur in individual cells constituting an individual. At present, attempts to analyze such changes at the genetic level are being actively conducted.
It is not possible to understand all the changes occurring in cells by analyzing only at the genetic level. This is because the proteins that form the actual cells are modified after translation from the gene sequence (genomic gene) encoded in genomic DNA on the chromosome (post-translati
onal modification), which results in a mature cell protein (see FIG. 1). Therefore, it is not possible to know what post-translational modification has occurred in a protein only by analyzing the genomic gene. Further, even if a protein exists independently at the genomic gene level, it may be phosphorylated by another protein at the stage when it is expressed in a cell, or the protein may form a complex. And, in fact, in the cell, many proteins interact functionally and structurally with each other and play their respective roles, thereby performing the function of the cell.

[0003] Under these circumstances, in recent years, attempts have been made to comprehensively grasp changes in proteins in cells as patterns, and the concept of "genome", which means a set of genes in cells, has been developed. The concept of "proteome", which means a set of proteins in cells, has been proposed (Science 270: 369-370, 1995). However, comprehensive analysis of intracellular protein changes has not necessarily progressed smoothly compared to analysis at the genetic level. This is because the protein constituting one cell is estimated to be about 5,000 to 7000, and the work of individually identifying these requires much labor and time compared to the identification of the gene.

[0004] As a method for identifying an unknown protein expressed in cells, a partial amino acid sequence of an isolated protein is determined by a known method (for example, Edman degradation method using a protein sequencer), and an existing protein database is determined. A method for identifying an unknown protein by specifying a known protein containing the partial amino acid sequence is known. However, this method requires a lot of cost and time to identify one protein, and is not practical as a means for identifying thousands of proteins as in proteome analysis.

On the other hand, in recent years, a method called peptide mass fingerprinting (Current Biol. 3: 327-332, 1993) has attracted attention as a new protein identification method. This method has been made possible by the appearance of a mass spectrometer capable of measuring a slight difference in the mass of amino acids constituting a protein and the appearance of an amino acid sequence database for various types of proteins. That is, as shown in FIG. 2, the protein sample is first separated by two-dimensional electrophoresis or the like.
An unknown protein cut out of a single spot is digested with an arbitrary protease or the like, and the mass of the digested fragment is measured by a mass spectrometer (for example, a commercially available device such as MALDI-TOF-MS), and the protease digested fragment of the unknown protein is analyzed. To obtain the mass pattern of By the way, since any protease always recognizes a predetermined amino acid and cleaves the protein, the protein digested with the protease shows a unique digestion fragment pattern (fingerprint). Therefore, the theoretical mass pattern of a known protein can be estimated by searching the amino acid sequence database for a sequence to be cleaved by the same protease and calculating its mass. Therefore, by comparing the theoretical mass value of a digested fragment of a known protein in the amino acid sequence database with the actual measured value of the actual digested fragment of an unknown protein one after another, the unknown protein is identified using the agreement of each other's mass as an index. It becomes possible. In fact, the mass measurement by the mass spectrometer can be performed within 5 minutes, and the search of the amino acid sequence database by the computer can obtain the result in a few seconds.

[0006]

The peptide mass fingerprinting method has a feature that a trace amount of protein can be identified in a short time as compared with a conventional method of directly comparing amino acid sequences, and the like. It is expected as a very effective means in proteome analysis and the like in which many proteins need to be identified.

[0007] However, in the case of the conventional peptide mass fingerprinting method, even if the proteins are the same, the theoretical value of the mass calculated based on the amino acid sequence and the mass value actually measured are different. Often they did not match, making accurate identification difficult. One of the reasons is that, as described above, proteins produced in cells are subjected to various post-translational modifications, and the measured mass of each digested fragment varies depending on the type and degree of modification. It is considered that this does not match the theoretical mass value uniquely estimated from the sequence.

[0008] The present invention has been made in view of the above circumstances, and solves the problems of the conventional peptide mass fingerprinting method. It is an object of the present invention to provide an identification method that can be easily and highly accurately identified.

[0009]

Means for Solving the Problems The present invention solves the above-mentioned problems by a method of identifying an unknown protein by using a search list as a search list of a specific protein group whose mass measurement value of a digested fragment by a proteolytic reagent is known. So,
(A) the unknown protein separated from the cells is digested and fragmented with the same proteolytic reagent as described above, (b) the mass of each digested fragment is measured, and (c) the number of digested fragments whose mass was measured ( The number of fragments (z) having the same mass value between x) and the number of digested fragments (y) of each protein in the search list was measured, and the coincidence rates RF1 and RF2 below (d): RF1 = z / x RF2 = z / y is calculated, and (e) identifying the protein in the search list in which each of RF1 and RF2 is 0.3 or more and the maximum is identified as the same protein as the unknown protein. (Claim 1).

[0010] Further, in this protein identification method, it is preferable to identify a protein in the search list in which one of RF1 and RF2 is 0.5 or more as the same protein as the unknown protein (Claim 2). It is an aspect. Further, the invention provides a method of mass measured value of the digested fragments by proteolytic reagent to build for protein identification search list comprising a specific group of proteins known, separated from the cell A ₁ in (a) Species A (B) digesting and fragmenting each protein with a proteolytic reagent, (c) measuring the mass of the digested fragment of each protein, and (d) measuring the cell A ₁
A search list consisting of the types of proteins produced by the above and the mass values of each digested fragment is created, and (e) the unknown protein separated from another cell A ₂ of the species A is subjected to the same proteolysis as described above. After digestion and fragmentation with the reagent, the mass of each digested fragment was measured, and (f) the number of digested fragments of unknown protein (x) and the number of digested fragments of each protein in the search list created in (d) above The number of fragments (z) whose mass value matched with (y) was measured, and the matching rates RF1 and RF2 of (g) or less: RF1 = z / x RF2 = z / y were calculated, and (h) RF1 And RF2 are both 0.3
As novel protein of unknown protein is less than, after identifying the type, and added to the search list with the mass values for each digest fragment, (g) or less, wherein the cell A ₃ for cell A _n species A (e ) To (h) are sequentially repeated to create a search list including the types of all the proteins constituting the species A and the mass values of each digested fragment, thereby providing a search for protein identification. A method for creating a list (claim 3) is provided.

[0011] In this search list creation method, unknown proteins in which one of RF1 and RF2 is less than 0.3 and the other is less than 0.5 are added to the search list as new proteins ( Claim 4) is also a preferred embodiment. The present invention further provides a protein identification search list (claim 5) created by the method described above.

Hereinafter, embodiments of the present invention will be described in detail.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS First, a method of constructing a search list used for the identification method of the present invention will be described. (1) any cell A ₁ in any species A preparation of the protein that constitutes the search list is collected and total protein samples are produced, it separated into individual proteins. As such a separation means, two-dimensional electrophoresis similar to the conventional method can be used, but any other method may be used as long as it can separate proteins individually. Next, the types of individually separated proteins are specified. This protein can be specified, for example, by determining the partial amino acid sequence of each protein by the Edman degradation method or the like, and comparing each with an existing amino acid sequence database. And
The specified protein is digested with a proteolytic reagent and fragmented. Any kind of proteolytic reagent can be used as long as it always cleaves the same portion of the protein. Protease (endopeptidase) isolated and purified from microorganisms and protein such as CNBr It can be appropriately selected from decomposition chemical reagents and the like. Further, in the conventional peptide mass fingerprinting method, in order to estimate a digested fragment from an amino acid sequence database, it was necessary to use a protease or the like whose amino acid portion to be cleaved is known. Proteolytic reagents with unknown amino acid moieties can also be used. (2) Preparation of primary search list Each protein digested with a proteolytic reagent is applied to a mass spectrometer to measure a mass pattern and a mass value of each digested fragment. The obtained mass value is recorded for each protein, and a primary search list is created.

The primary search list prepared in this manner includes the measured mass values of each digested fragment for all proteins produced by the cell A _{1 of the} species A. Accordingly, the protein identification method described later, as long as the unknown protein of interest identification is derived from a cell A ₁ species A, it is possible to identify with this primary search list. However, since different types of cells produce about 30% different proteins,
The primary search list alone cannot identify all proteins of species A. Therefore, a higher-order search list is created by the following procedure. (3) any unknown protein from another cell A ₂ Creating species A high-order search list separated, fragmented by digestion of the protein with the same proteolytic reagent, measuring the mass of each fragment I do.

Next, the number of fragments (z) having the same mass value between the number of digested fragments (x) of the unknown protein and the number of digested fragments (y) of each protein in the primary search list is measured, The following coincidence rates RF1 and RF2 are calculated: RF1 = z / x RF2 = z / y

[0016] Here, the unknown protein and the same protein isolated from the cell A ₂ has if present primary search list is obtained RF1 and RF2 are both high values. On the other hand, if the unknown protein does not exist in the primary search list (theoretically, about 30% of the protein does not exist),
As shown in the examples described later, RF1 and RF2 are both less than 0.3. Therefore, proteins whose RF1 and RF2 are less than 0.3 are identified as new proteins, and after identifying their types, the mass values of the digested fragments are added to the primary search list. To be more accurate,
It is preferable that one of RF1 and RF2 is less than 0.3 and the other is less than 0.5 as a novel protein.

The reason for using two types of RF1 and RF2 as the coincidence rate is that even if the same protein is digested with the same digestion reagent, the number of digested fragments may be different, and the search list contains If the number of registered fragments (y) and the number of fragments obtained from an unknown protein (x) are different, the coincidence rate greatly differs depending on which reference is used. Thus, by employing such two types of coincidence rates, it is possible to determine with high accuracy whether the proteins are the same or different.

Since a protein having a large molecular weight has many digestion sites, many digested fragments can be obtained. Therefore, when comparing a large number of digested fragments obtained from high-molecular-weight proteins, even if the proteins are different,
There are several fragments whose mass values coincide with each other, and there is a risk that they may be erroneously identified as the same protein. RF1 and RF2 are also effective to avoid such erroneous identification due to coincidence. That is, RF1 is an index for avoiding that the unknown protein is a high molecular weight protein, so that a small number of data of the heterologous proteins in the search list match many, and as a result, both are erroneously identified as the same protein. It is. On the other hand, RF2 is an index for avoiding erroneous identification due to the fact that most of the unknown protein data is included in the high molecular weight protein data in the search list.

Further, as another embodiment of the present invention, when the unknown protein is determined to be the same as the protein in the search list, and there is a fragment whose mass does not match in both proteins, Fragment data can also be added to the corresponding protein data in the search list. As a result, when the same protein is identified again, an even higher matching rate is obtained especially for RF1, and the accuracy of the search list is improved. Alternatively, the data of the unmatched fragment in the search list may be deleted from the search list. This is because the mismatched fragment is likely to be a fragment generated by various fluctuation factors. In this case, when the same protein is tested again, RF2 shows a high matching rate.

[0020] By performing the above operation for all the proteins of the cells A _2, second search list that covers both cellular protein A ₁ and cell A ₂ is created. Similarly, by repeating the same operation for the cells A _n from the cell A _3, can be finally create a search list covering all proteins that comprise the species A. Further, after a search list for the species A is created, proteins of species B, C, D,... Can be sequentially added.

The search list created as described above can be incorporated in, for example, a computer system and used as a database for protein identification. Next, the protein identification method of the present invention will be described.
The identification method of the present invention is a method characterized by using the search list constructed by the above method as, for example, a computer database and identifying an unknown protein by the following procedure. First, an unknown protein separated from cells is digested and fragmented with the same proteolytic reagent used to create the search list, and the mass of each digested fragment is measured. Next, the number of fragments (z) having the same mass value between the number of digested fragments (x) of the unknown protein and the number of digested fragments (y) of each protein in the search list was measured, and the following matching rates RF1 and RF2 were obtained. : RF1 = z / x RF2 = z / y is calculated.

As described above, when the same protein as the unknown protein is present in the search list, the matching rates RF1 and RF2 between the two are both higher than the matching rates between the other search proteins and the unknown protein. To indicate the numerical value, the unknown protein is identified as homologous to the search protein. However, as described above, when the maximum value of each of RF1 and RF2 is less than 0.3, there is an extremely high risk that they are different types. Therefore, the determination of protein identity is that RF1 and RF2 are the largest and both are at least 0.3, more preferably
Either one is 0.5 or more. Note that the coincidence rate is 0.
If it is less than 3, that is, if it is determined that the unknown protein is a new protein not in the search list, the new protein is identified by another means, and the mass data of the digested fragment is added to the search list. In addition, the search list can be expanded.

Hereinafter, the present invention will be described in more detail and specifically with reference to Examples, but the present invention is not limited to the following Examples.

[0024]

【Example】

Example 1 A search list was created for proteins produced by rat papilla cells. First, the culture solution was removed from the culture dish of rat papilla cells, and a lysate containing urea (9 M u
rea, 3.2% NP-40, 1.8% ampholytes: pH 3-10, 120mM DT
T, 0.06% SDS, 10 mM Tris) was added to solubilize the protein of the cells, and the protein concentration was measured to obtain a sample for two-dimensional electrophoresis. Electrophoresis can be performed by a known method (B. B.
According to iol. Chem. 250: 4007-4021, 1975), isoelectric focusing was performed in the first dimension, and SDS-polyacrylamide electrophoresis was performed in the second dimension. The gel after the electrophoresis was stained with Coomassie Blue G-250, affixed to a filter paper with a gel dryer, and dried. FIG. 3 is a computer analysis image of a two-dimensional electrophoresis gel of papillary cells obtained as described above,
More than 100 protein spots were obtained.

Next, spots of each protein separated on the gel were cut out and transferred to a PVDF membrane, and PVP-
After blocking with 40, trypsin solution (50 μg / ml
modified trypsin, 0.1M Tris-HCl: pH8.5, 2mM calciu
(37 ° C., 17 hours). The solution containing the digested protein fragment (peptide) is collected, dried in a tube, and then subjected to reverse phase HPLC (SMART system,
μRPC C2 / C18 SC 2.1 / 10, 2.1mm ID x 100mm, Pharmaci
a)). The amino acid sequences of these digested fragments were determined with a puttein sequencer, and the existing amino acid sequence database was searched using these amino acid sequences as indices to identify individual proteins.

The specific results of the protein based on such an amino acid sequence show that 186 as shown in Tables 1 and 2.
A variety of proteins have been identified. Of these proteins, about one-third are proteins of the same species but separated as separate spots in two-dimensional electrophoresis. For example, as shown in FIG.
n, peptidylprolyl, and trisphate isomerase were separated on the gel as four, two, and two spots, respectively. These separate spots are protein isoforms generated by splicing or phosphorylation, and the conventional method of calculating a theoretical mass based on amino acid sequence data requires that these isoforms be identified in a search list. Will not be listed.

[0027]

[Table 1]

[0028]

[Table 2]

Finally, each of the 15 proteins separated as 42 spots was digested with the same trypsin solution as described above, and the mass of each digested fragment was determined using a mass spectrometer (M
(ALDI-TOF-MS). Then, the actual mass values were sorted for each type of protein to create a search list, which was input to a computer system and used as a protein identification database. In this database, when identifying an unknown protein, the matching rate R
It was programmed to calculate F1 and RF2. Example 2 Using the database created in Example 1, unknown proteins isolated from rat papilla cells were identified. Separation of the unknown protein, trypsin digestion, and mass measurement of the digested fragment were performed in the same manner as in Example 1.

As a control, an existing amino acid sequence database was used to estimate a tryptic digested fragment of the same protein used in the search list created in Example 1, and the respective mass theoretical values were calculated to obtain the target search list. And the above-mentioned unknown protein was identified. The results are as shown in Tables 3 and 4. Tables 3 and 4 show the number of fragments having matching masses between the proteins in the search list and the unknown proteins, and the matching rate (RF1
And RF2) with the search list of the present invention (left column)
Each of the conventional search lists (center column) is shown. ** attached to the number of fragments indicates a case where the identification was not performed correctly. The right column shows the maximum number of matching fragments and the matching rate when a correct identification result was not obtained.

[0031]

[Table 3]

[0032]

[Table 4]

As is clear from the results of Tables 3 and 4, when the search list of the present invention comprising the measured mass data is used, the unknown protein is identified as any of the proteins in the search list. Was done. On the other hand, when a search list consisting of theoretical mass data is used, 4
Out of the two search targets, only 20 were correctly identified.

Also, among the 41 search objects, the search list of the present invention showed a higher RF value than the control, and the other one showed the same value. Furthermore, in both search lists, if correctly identified, RF1 and RF2 show 0.3 or more, while if incorrectly identified, both values are less than 0.3 (or both Is less than 0.5). Therefore, it was confirmed that it is possible to identify an unknown protein as a protein in the search list or to determine it as a novel protein that does not exist in the search list based on such an RF value.

[0035]

As described above in detail, according to the present invention, there is provided a method capable of easily and accurately identifying many proteins in a cell. This allows
Proteome analysis that comprehensively captures changes in proteins in cells has become a reality, enabling the elucidation of various biological phenomena, the search for new proteins, and the development of new reagents using these new proteins. Becomes

[Brief description of the drawings]

FIG. 1 is a schematic diagram showing a process in which a protein is produced in a cell.

FIG. 2 is a schematic view showing the principle of a conventional peptide mass fingerprinting method.

FIG. 3 is a computer analysis image of a two-dimensional electrophoresis gel of papilla cells.

FIG. 4 is a photograph of an electrophoresis gel illustrating the same type of protein separated as a plurality of spots on a two-dimensional electrophoresis gel.

──────────────────────────────────────────────────続 き Continued on the front page (56) References WO 95/25281 (WO, A1) CURRENT BIOLOGY, VOL. 3, NO. 6, (1993), pp. 327-332 (58) Fields investigated (Int. Cl. ⁷ , DB name) G01N 33/68 G01N 27/62 G01N 33/48

Claims (5)

(57) [Claims] 1. A method for identifying an unknown protein using a specific protein group whose mass measured value of a digested fragment by a proteolytic reagent is known as a search list, wherein (a) the unknown protein separated from cells is Digestion and fragmentation with the same proteolytic reagent, (b) measuring the mass of each digested fragment, (c) the number of digested fragments (x) whose mass was measured, and the number of digested fragments of each protein in the search list The number of fragments (z) whose mass value matched with (y) was measured, and the following matching ratios (d) were calculated: RF1 and RF2: RF1 = z / x RF2 = z / y, and (e) RF1 And a protein in the search list in which each of RF2 and RF2 is equal to or greater than 0.3 and the maximum is identified as the same protein as the unknown protein. 2. One of RF1 and RF2 is 0.5
The protein identification method according to claim 1, wherein the protein in the search list is identified as the same protein as the unknown protein. 3. A method for constructing a search list for protein identification comprising a specific protein group whose actual measured mass value of a fragment digested by a proteolytic reagent is known, wherein (a)
Identifies the type of each protein isolated from the cell A ₁ species A, the (b) each protein was fragmented by digestion with a proteolytic reagent, and measuring the mass of the digested fragment of (c) each protein ( d) the type of proteins the cell a ₁ is produced, to create a search list of the mass value of each of the digested fragments, the unknown protein that has been separated from other cells a ₂ in (e) species a, the After digestion and fragmentation with the same proteolytic reagent as above, measure the mass of each digested fragment,
(F) Count the number of fragments (z) having the same mass value between the number of digested fragments (x) of the unknown protein and the number of digested fragments (y) of each protein in the search list created in (d). (G) Calculate the following coincidence rates RF1 and RF2: RF1 = z / x RF2 = z / y, and (h) both RF1 and RF2 are 0.3
As novel protein of unknown protein is less than, after identifying the type, and added to the search list with the mass values for each digest fragment, (g) or less, wherein the cell A ₃ for cell A _n species A (e ) To (h) are sequentially repeated to create a search list including the types of all the proteins constituting the species A and the mass values of each digested fragment, thereby providing a search for protein identification. How to create a list. 4. The search list for protein identification according to claim 3, wherein an unknown protein in which one of RF1 and RF2 is less than 0.3 and the other is less than 0.5 is added to the search list as a new protein. How to make. 5. A search list for protein identification constructed by the method according to claim 3 or 4.