WO2017073600A1 - mRNA DESIGN METHOD - Google Patents

Abstract

The present invention is a method for designing mRNA which can be identified with a high degree of accuracy while the cell is still alive. The method is for designing mRNA comprising at least two miRNA response elements, wherein the mRNA comprises at least two miRNA response elements and marker gene elements functionally bonded to the at least two miRNA response elements. The method comprises the following steps: (1) a step in which the translation inhibiting effects of mRNA comprising one miRNA response element are measured in at least two target cells; (2) a step in which, on the basis of the measurement results from step (1), the translation inhibiting effects of mRNA comprising at least two miRNA response elements are calculated in each target cell; and (3) a step in which, on the basis of the values calculated in step (2), the mRNA comprising at least two miRNA response elements and exhibiting the largest difference, in terms of translation inhibiting effects, between the at least two target cells is selected.

Description

mRNA design method

The present invention relates to a method for designing mRNA.

The tissues and organs of multicellular organisms are composed of many types of cells. Humans are composed of as many as 60 trillion (6 × 10 ¹³ ) cells, and there are about 400 types of mature cells alone. For these cells, not only analyzing the functions of individual cells but also techniques for discriminating and identifying cell types in the preparation of cells for medical applications.

In order to identify a cell, a technique for detecting a marker factor on a cell surface specifically expressed with an antibody is generally known. However, in order to detect intracellular information with an antibody, it is necessary to fix the target cell and permeate the membrane, and there is a problem that it cannot be applied to sorting of living cells. In addition, a receptor capable of identifying a cell does not necessarily exist on the cell surface. In addition, a method of classifying one factor into two, positive / negative (negative or positive), such as detection of a marker factor of a cell by an antibody, is a method of qualitatively classifying a cell, and it is said that precise classification is difficult. There's a problem.

As a method of classifying cells more precisely, that is, a method of classifying cells quantitatively, for example, cell profiling based on multivariate measurement using a microarray or next-generation sequencing is known. With these methods, it is possible to classify cells quantitatively by simultaneously quantitatively measuring many types of intracellular molecules such as proteins and RNA, and using statistical analysis such as multivariate analysis. is there. However, since the measured cell is destroyed, there is a problem that it cannot be measured in a state where the cell is alive.

Therefore, focusing on the point of suppressing the gene expression of the marker using microRNA (hereinafter referred to as miRNA) specifically expressed in a desired cell, a cell separation method using the system is proposed. (Patent Document 1).

International Publication WO2015 / 105172

In the method of Patent Document 1, depending on the desired target cell, an appropriate miRNA for classifying the cell may not be selected, and an improvement of the system is desired.

The present inventors are able to detect a plurality of miRNAs in a countable manner with respect to mRNA containing two or more miRNA response sequences and a marker gene sequence operably linked thereto, and the detection sensitivity to each miRNA on the mRNA. It was found that the miRNA can be regulated by the position of the target sequence. That is, by summarizing information on multiple factors first, and then detecting the resulting synthesized parameters, even a limited number of signals that can be detected at the same time are derived from quantitative information on many factors in living cells. The present inventors have found that the synthesis parameter from which the essence is extracted can be directly detected, and have completed the present invention.

Therefore, the problems of the present invention can be solved by the following means.
[1] A method for designing mRNA containing two or more miRNA response elements, comprising the following steps, wherein mRNA containing two or more miRNA response elements is functionally linked to two or more miRNA response elements A mRNA comprising a marker gene sequence obtained;
(1) a step of measuring the translation inhibitory effect of mRNA having one miRNA response element in two or more target cells;
(2) A step of calculating the translational suppression effect of mRNA containing two or more miRNA response sequences in each target cell based on the measurement result of the step (1),
(3) A step of selecting mRNA containing two or more miRNA response elements that maximizes the difference in translational suppression effect between the two or more target cells based on the value calculated in the step (2).
[2] Based on the measurement result of the step (1) between the steps (1) and (2), the type of miRNA response element used for the calculation of the step (2) is determined using multivariate analysis. The method according to [1], further comprising a step of limiting.
[3] The translation inhibitory effect of the step (2) is obtained by adding the translation inhibitory effect -log (ρ) of each of the two or more miRNA response elements by the number of miRNAs. Translation suppression effect -log (ρ)

(In the formula, ρ represents the translation suppression effect by miRNA,
d [nt] represents the distance from the start codon to the miRNA target sequence,
ξ represents -0.576,
ρ ⁰ represents the hypothetical translational suppression effect at a distance of 0 [nt] for each miRNA)
The method according to [1] or [2], which is calculated based on the above.
[4] The step (3) includes the step of selecting mRNA containing two or more miRNA response elements that maximizes the distribution of the translation amount of the marker gene in each target cell. The method according to any one of the above.
[5] Designing mRNA by the method according to any one of [1] to [4];
A method for producing mRNA containing two or more miRNA response elements, comprising a step of synthesizing the designed mRNA by a genetic engineering technique.
[6] A method of separating two or more target cells using the mRNA designed by the method according to any one of [1] to [4], using the translation amount of the marker gene as an index.
[7] The mRNAs are four types of mRNAs, which are designed by the method according to any one of [1] to [4] and have different marker gene sequences and 5 ′ UTR sequences, respectively [6] The method described in 1.

According to the present invention, there is provided a method for designing an mRNA containing two or more miRNA response elements, which can separate desired cells with higher accuracy. The mRNA functions as a probe that responds to a large number of miRNAs per molecule and can arbitrarily adjust the degree of response to individual miRNAs (detection sensitivity). That is, according to the present invention, intracellular multifactor information was successfully extracted with a linear model. In the prior art, the number of signals that can be detected simultaneously in a non-invasive manner (without killing cells) is limited, and cannot be handled by a detection probe corresponding to one-to-one. Even with a limited number of signals that can be detected at the same time, it is now possible to directly detect the synthetic parameters from which the essences are derived, derived from quantitative information on a large number of living intracellular factors.

FIG. 1 shows a scheme for performing multivariate calculations in living cells. FIG. 1a shows that in the conventional method (left), individual intracellular information (in this case, miRNA information) is individually detected, and then information is extracted by multivariate analysis. (Right) shows that multivariate calculation is performed in advance in the cell and the result is detected directly. FIG. 1b shows a linear multivariate calculation using mRNAs that respond to multiple miRNAs. It suffices that the mRNA responds independently to a plurality of miRNA activities and can regulate the response performance to each miRNA. If the ratio is calculated using two mRNAs, a linear calculation with an arbitrary specific gravity can be realized for many miRNAs. FIG. 2 shows that synthetic mRNA can achieve a regulatable response independent of multiple miRNAs. FIG. 2a shows the 5'UTR design of mRNA with 5 slots. Each slot inserts a sequence complementary to the miRNA. AUG is the start codon of marker protein (hmAG1). The lower part is an mRNA series that responds to four types of miRNAs. Response sequences to gray, miR-34a-5p; bitumen, miR-17-5p; green, miR-21-5p; red, miR-92a-3p. FIG. 2b shows examples of mRNAs that respond to miR-17-5p and miR-92a-3p. The design of 5 ’UTR is shown at the top. Relative expression is expressed as a ratio to the average expression level of the marker in the presence of both. The product of Relative expression in the presence of only one miRNA inhibitor was designated as Estimated expression. FIG. 2c shows a comparison of experimentally measured Relative expression values (Observed relative expression) and Estimated Expression. The results of three independent experiments were plotted for 12 randomly designed types. FIG. 2d shows a series of mRNAs containing multiple response elements to the same miRNA. Each mRNA contains a miRNA response element at 1 to 5 sites. The expression level of mRNA is predicted to be the integrated value of the suppression effect (ρ, rho) in each slot. FIG. 2e shows a comparison of experimentally measured Relative expression values (Observed relative expression) and estimated values. Four types of miRNA were compared. FIG. 2f shows a plot of the inhibitory effect in each slot and the distance of the slot from the start codon. The suppression effect in each slot was calculated by fitting by the least square method from the data of e for each miRNA. Error bars indicate the mean ± standard deviation of three analysis results. FIG. 3 shows the classification of living cells based on a linear calculation of multiple miRNA activities. FIG. 3a shows the screening scheme. Three types of mRNA (1-slot s mRNA) 発 現 that expresses different fluorescent proteins (hmAG1, tagBFP, hdKRed) 応 答 in response to different miRNA (a, b, c), and hmKO2 1-slot control mRNA that does not respond to miRNA At the same time introduced. 270 miRNAs are 96 transfections including controls. Analyzed by flow cytometry, we obtained miRNA activity profiles in various cells. FIG. 3b shows a comparison of two independent screening results using human iPS cells (hiPS). The expression level of mRNA using green, hmAG1; bitumen, tagBFP; purple, hdKRed as marker proteins is shown. FIG. 3c shows a cell-to-cell comparison of the normalized data set. A comparison between normal human dermal fibroblast (NHDF) and hiPSC is shown as an example. FIG. 3d shows cell classification by principal component analysis. The results of performing principal component analysis by standardizing the results of screening for 8 types of cells and conditions are shown. The plot with the first component (Component 1) and the second component (Component 2) is shown. FIG. 3e shows a set of mRNAs that maximizes intercellular dispersion. Based on the standardized data set, the mRNA set that maximizes the dispersion between cells was calculated and designed using the ratio of hmAG1 and hmKO2 and the ratio of tagBFP and hdKRed as parameters. FIG. 3f shows the computational classification of the five types of cells according to the mRNA set of FIG. FIG. 3g shows the classification of living cells by mRNA set. Flow cytometry results are shown as a two-dimensional density plot of fluorescence ratio. Cells shown in red are red, and the other four types of cells are shown as black density. FIG. 4 shows the tracking of changes in hiPS cells based on linear calculations of multiple miRNA activity. FIG. 4a shows the secondary screening scheme. From the screening results shown in FIG. 3, 54 miRNA having a large difference between hiPSC and hiPSC (14d) was selected and used as a secondary screening for 24 transfections. FIG. 4b shows a tracking scheme over time. hiPSCs were naturally (randomly) differentiated in the absence of bFGF, losing pluripotency. After the indicated number of days of culture, the mRNA set was transfected and analyzed 24 hours later by flow cytometry. FIG. 4c shows a comparison of changes over time in miRNA activity. For the measured fluorescence ratio, a comparison between day 1 and day 3 is shown as an example. FIG. 4d shows the result of the principal component analysis of the screening result. The plot with the first component (Component 1) and the second component (Component 2) is shown. In addition, cells cultured for 1-3 days in the presence of bFGFb are shown in blue. FIG. 4e shows a set of mRNA that maximizes cell-to-cell dispersion when hiPSCs are differentiated by losing pluripotency in the absence of bFGF. An mRNA set that maximizes cell-to-cell dispersion was calculated and designed. FIG. 4f shows the computational classification of cells according to the mRNA set of FIG. 4e. FIG. 4g is a classification of living cells by mRNA set. Flow cytometry results are shown as a two-dimensional density plot of fluorescence ratio. Cells shown in red are red, and the other four types of cells are shown as black density. FIG. 5 shows the effect of miRNA inhibitor on miRNA activity measurement. FIG. 5a shows the results of confirming the independence of miRNA inhibitors used in this example. For each mRNA that responds to miRNA, five miRNA inhibitors were introduced into the cells. The expression level in each condition is shown as a ratio based on the expression level in the presence of the miRNA inhibitor to which mRNA responds. Error bars indicate the mean ± standard deviation of 3 experiments. It was confirmed that each miRNA inhibitor did not crosstalk. FIG. 5b shows the results of confirming the effect of the total amount of miRNA inhibitor introduced. The same experiment as in FIG. 5a was performed with the addition of miR-1 inhibitor 2 pmol or 4 pmol 実 験. The condition “/ w” corresponds to “miR-1 (n.c.)” in FIG. Under such conditions, the activity of miRNA detected does not vary depending on the presence or absence of an irrelevant miRNA inhibitor. FIG. 6 shows the difference in miRNA detectability depending on the position of the target sequence. FIG. 6a shows the structure of s5 ′ UTR of 5-slot mRNA and 1-slot mRNA. FIG. 6b shows the suppression effect in each slot. Each miRNA was calculated from the data in FIG. Error bars indicate the mean ± standard deviation of three analysis results. The right panel shows 1-slot mRNA results. 1-slot mRNA behaves like slot-5. FIG. 7 shows the standardization of the screening results. The upper figure shows the results of HeLa cells and normal human lung fibroblast cells (NHLF), and the lower diagram shows the comparison results of HeLa cells and human iPS cell cells (hiPSC). The green plot shows hmAG1, the blue plot shows tagBFP, and the purple plot shows the expression level of mRNA when hdKRed is used as a marker protein. FIG. 7a shows a comparison of the observed expression values. There is a cell-by-cell bias. FIG. 7b shows the normalization of the bias between cells. The difference in expression fluorescence between cells for each fluorescent protein was normalized based on the results of HeLa cells. However, the distribution position is different for each fluorescent protein. FIG. 7c shows the normalization of bias between fluorescent proteins. Distribution was normalized for each fluorescent protein. FIG. 8 shows the results of cell classification by different mRNA sets. Experiments similar to FIG. 3g were performed with different mRNA sets. Flow cytometry results are shown as a two-dimensional density plot of fluorescence ratio. Cells shown in red are red, and the other four types of cells are shown as black density. FIG. 8a shows the results using a set of control mRNAs that do not respond to miRNA. FIG. 8b shows the results using a set of 1-slot mRNAs that responded to miRNAs with large variations in activity. FIG. 8 c shows the results using the mRNA set (FIG. 4 e) when tracking changes in hiPSC. FIG. 9 shows the results of tracking hiPSC with different mRNA sets. The same experiment as in Fig. 4g was performed, and the results of flow cytometry were shown as a two-dimensional density plot of the fluorescence ratio. Cells shown in red are red, and the other eight types of cells are shown as black density. FIG. 9a shows the results of other culture conditions. Cells of different culture conditions were isolated using the mRNA set of FIGS. FIG. 9b shows the results of experiments using other mRNA sets. The mRNA set 時 (Fig. 3e) 時 when the cell types were classified was used.

Hereinafter, the present invention will be described in detail with reference to embodiments. The following embodiments do not limit the present invention.

According to the first embodiment, the present invention is a method for designing mRNA containing two or more miRNA response elements, wherein the mRNA containing two or more miRNA response elements has two or more miRNA response elements and a function thereof. The present invention relates to a method, wherein the mRNA comprises an operably linked marker gene sequence. The method according to the present embodiment includes the following steps.
(1) a step of measuring the inhibitory effect of mRNA having one miRNA response element in two or more target cells;
(2) a step of calculating the translational inhibitory effect of mRNA containing two or more miRNA response elements in each cell from the measurement result of the step (1),
(3) A step of selecting mRNA containing two or more miRNA response elements that maximizes the difference in translational suppression effect between two or more target cells from the value calculated in the step (2).

In the present invention, two or more miRNA response sequences (hereinafter also referred to as miRNA response sequences or miRNA target sequences) and a messenger RNA (mRNA) that includes a marker gene operably linked thereto are provided for each miRNA response sequence. The integrated value of the translational suppression effect has a translational suppression effect, and the translational suppression effect of each miRNA response element is based on the discovery that it is inversely proportional to the distance from the start codon of each miRNA response element.

In the present invention, two or more miRNA response elements and a marker gene are operably linked to each other in the 5′UTR of the open reading frame (including the start codon) encoding the marker gene. Means comprising a response sequence. Such mRNA can be used to separate cell types using the expression of the corresponding miRNA in the cell as an index. More specifically, when the corresponding miRNA is expressed in the cell, the translation of the marker gene is suppressed depending on the expression level. As used herein, “miRNA expression” refers to the presence of miRNA in a state in which a mature miRNA interacts with a predetermined plurality of proteins to form an RNA-induced silencing complex (RISC). Shall. “Mature miRNA” is a single-stranded RNA (20-25 bases), and is generated from pre-miRNA by cleavage by Dicer outside the nucleus, and “pre-miRNA” is obtained by partial cleavage by a nuclear enzyme called Drosha, It originates from pri-mRNA, a single-stranded RNA transcribed from DNA. The miRNA in the present invention can be selected from at least 10,000 miRNAs. Specifically, miRNA registered in database information (for example, http://www.mirbase.org/ or http://www.microrna.org/) and / or literature information described in the database Or can be selected from commercially available miRNAs from libraries. That is, in the present invention, the miRNA serving as an index is not limited to a specific miRNA.

An mRNA containing two or more miRNA response elements as described above used in the design method according to this embodiment is an miRNA-responsive mRNA, an n-slot mRNA (n is an integer of 2 or more), or a reporter mRNA. Also referred to. The basic structure of such miRNA-responsive mRNA will be described.

The schematic structure of mRNA designed in this embodiment is illustrated in the upper diagram of FIG. The mRNA shown in FIG. 2a encodes a marker gene from the 5 ′ end in a 5 ′ to 3 ′ direction, a Cap structure (7 methylguanosine 5 ′ phosphate), 5 slots into which miRNA response elements can be inserted, and a marker gene. With open reading frame and poly A tail.

Here, “slot” conceptually indicates a portion into which a miRNA target sequence can be inserted, and each slot is composed of a miRNA target sequence or a sequence that is not a miRNA target (also referred to as an empty slot). Is done. However, the same or different miRNA target sequences are inserted into two or more of the five slots. The number of bases of the miRNA target sequence inserted into each slot or the non-miRNA target sequence may be the same or different, but is generally 20 to 25 bases. Any number of bases may or may not exist between the slots. In addition, in an mRNA containing three or more slots, the number of bases and the sequence between the slots may be the same or different.

The miRNA target sequence refers to a sequence that can specifically bind to an indicator miRNA. The miRNA target sequence is preferably, for example, a sequence that is completely complementary to the indicator miRNA. Alternatively, the miRNA target sequence may have a mismatch (mismatch) with a completely complementary sequence as long as it can be recognized in the miRNA. The mismatch from the sequence that is completely complementary to the miRNA may be any mismatch that can be normally recognized by the miRNA in the desired cell, and the mismatch of about 40 to 50% in the original function in the cell in vivo. There is no problem. Such mismatch is not particularly limited, but 1 base, 2 bases, 3 bases, 4 bases, 5 bases, 6 bases, 7 bases, 8 bases, 9 bases, or 10 bases or 1% of the total recognition sequence, 5% %, 10%, 20%, 30%, or 40% discrepancy. In particular, like the miRNA target sequence on the mRNA provided by the cell, in particular, the part other than the seed region, that is, the 5 'side of the target sequence corresponding to about 3' side 16 base of the miRNA A region may contain a number of mismatches, and portions of the seed region may contain no mismatches, or may contain 1 base, 2 bases, or 3 bases mismatches. In addition, it is preferable to select the miRNA target sequence so that an AUG sequence that can serve as an initiation codon is not included.

The sequence that is not the target of miRNA that can be inserted into the slot (hereinafter also referred to as non-target sequence) is not particularly limited, but has low similarity to the miRNA target sequence and does not contain AUG as a sequence It is preferable that The low similarity may be, for example, 60% or more, 70% or more, or 80% or more of mismatched sequences, but is not limited thereto. When two or more empty slots exist on the same miRNA-responsive mRNA, the non-target sequences into which the empty slots are inserted may be the same or different. Further, in a set of a plurality of types of miRNA-responsive mRNAs designed with the separation of specific target cells as an index, non-target sequences into which empty slots of different miRNA-responsive mRNAs are inserted may be the same or different.

In miRNA-responsive mRNA, the number of bases and the type of base between the Cap structure and the slot located at the most 5 ′ side (slot-1 in FIG. 2) do not include AUG as the start codon, and the stem structure As long as a three-dimensional structure is not formed, it may be arbitrary. For example, the number of bases between the Cap structure and the miRNA target sequence is not particularly limited, and can be designed to be the number of bases according to the purpose and application, for example, 1000 bases or less, preferably 500 bases In the following, it can be designed more preferably in the range of 250 bases or less. The number of bases and the type of base between the slot closest to the start codon (slot-5 in FIG. 2) and the start codon may be arbitrary as long as they do not constitute a stem structure or a three-dimensional structure. Accordingly, there is no particular upper limit to the number of bases between the slot closest to the start codon and the start codon, for example, 1000 bases or less, preferably 500 bases or less, more preferably 250 bases or less, for example, 2 to It can be designed to be 20 bases, preferably 3 to 20 bases.

A marker gene is a gene that is translated in a cell, functions as a marker, and encodes an arbitrary protein that enables discrimination of a cell type. Examples of proteins that can be translated into cells and function as markers include, for example, proteins that can be visualized and quantified by assisting fluorescence, luminescence, coloration, or fluorescence, luminescence, or coloration. It may be. As fluorescent proteins, blue fluorescent proteins such as Sirius and EBFP; cyan fluorescent proteins such as mTurquoise, TagCFP, AmCyan, mTFP1, MidoriishiCyan, and CFP; TurboGFP, AcGFP, TagGFP, Azami-Green (for example, hmAG1), ZsGreen, EmGFP, Green fluorescent proteins such as EGFP, GFP2, and HyPer; Yellow fluorescent proteins such as TagYFP, EYFP, Venus, YFP, PhiYFP, PhiYFP-m, TurboYFP, ZsYellow, and mBanana; Orange fluorescent proteins such as KusabiraOrange (for example, hmKO2) and mOrange Red fluorescent proteins such as TurboRFP, DsRed-Express, DsRed2, TagRFP, DsRed-Monomer, AsRed2, mStrawberry, etc .; TurboFP602, mRFP1, JRed, KillerRed, mCherry, HcRed, KeimaRed (eg, hdKeimaRed), mRasberry, mPlum Examples include but are not limited to infrared fluorescent proteins.

An example of a photoprotein is aequorin, but is not limited thereto. Examples of proteins that assist fluorescence, luminescence, or coloration include, but are not limited to, enzymes that decompose fluorescence, luminescence, or color precursors such as luciferase, phosphatase, peroxidase, and β-lactamase. Here, in the present invention, when a protein that assists fluorescence, luminescence, or color is used as a marker gene, in the discrimination of a desired cell, the corresponding precursor is brought into contact with the cell, or the precursor corresponding to the inside of the cell. Can be done by introducing.

Further, another example of a protein that can function as a marker in a cell is a protein that directly affects the function of the cell. Cell growth protein, cell death protein, cell signal factor, drug resistance gene, transcription control factor, translation control factor, differentiation control factor, reprogramming induction factor, RNA binding protein factor, chromatin control factor, membrane protein can be exemplified However, it is not limited to these. For example, a cell growth protein functions as a marker by proliferating only cells that express it and specifying the proliferated cells. The cell killing protein causes cell death of the cell expressing it, thereby killing the cell itself containing or not containing a specific miRNA, and functions as a marker indicating cell viability. The cell signal factor functions as a marker by the cell that expresses it emits a specific biological signal and specifies this signal.

In the present invention, the marker gene may include a gene encoding a localization signal. Examples of the localization signal include a nuclear localization signal, a cell membrane localization signal, a mitochondrial localization signal, a protein secretion signal, and the like. Specifically, a classical nuclear translocation sequence (NLS), M9 Examples include, but are not limited to, sequences, mitochondrial target sequences (MTS), and endoplasmic reticulum translocation sequences. Such a localization signal is particularly advantageous when the discrimination step in the method of the present invention is performed on an image by imaging cytometry or the like described later.

The miRNA-responsive mRNA preferably contains a modified base such as pseudouridine or 5-methylcytidine instead of ordinary uridine and cytidine. This is to reduce cytotoxicity. The positions of the modified bases can be all or part of the uridine and cytidine independently, and if they are part of the base, they can be random positions at an arbitrary ratio.

In the illustrated conceptual diagram, the total number of slots is 5, but the total number of slots may be 2 or more, for example, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more. It may be. In the present embodiment, for simplification of description, miRNA-responsive mRNA having a total number of slots of 5 will be described as an example, but the present invention is not limited to this.

The mRNA having the basic structure shown in the upper diagram of FIG. 2a has the following characteristics (a) and (b).
(A) The translational suppression effect by mRNA containing two or more miRNA response elements is an integrated value of the translational suppression effect by the miRNA target sequence alone in each slot.

(B) The translational suppression effect of the miRNA target sequence alone in each slot is generally proportional to the constant power of the distance from the start codon to the miRNA target sequence, specifically to the power of -0.576, regardless of the type of miRNA.

In the formula of (a), miRNA response elements inserted into each slot are miRNA (1), miRNA (2), miRNA (3), miRNA (4), miRNA (5) in order from slot-1 respectively. Think about the case. The translation suppression effect (ρ ^slot-1 ) by miRNA (1) alone inserted in ^slot-1 is the measured translational suppression effect in the system to which inhibitors for all response elements including miRNA (1) were added. In the case of 1, it is represented by an actually measured translation suppression effect in a system to which inhibitors for all response elements other than miRNA (1) are added. The translation suppression effect (ρ ^slot- 2) for miRNA (2) inserted into slot-2 and the translation suppression effect (ρ ^slot- 3) for miRNA (3) inserted into ^slot-3 should be obtained in the same way. These integrated values are the effect of suppressing the predicted translation of mRNA having a plurality of slots. In addition, when considering with the logarithm of (rho) like the said numerical formula, a total value becomes a logarithm of the prediction translation suppression effect.

In the formula (b), the distance d from the start codon (AUG) of each slot is the base from the A of the start codon AUG to the 5 ′ side to the base at the 3 ′ end of the miRNA response element constituting each slot. It shall be a number (nt). Also, ρ ⁰ is a virtual translation suppression effect value when the distance d is 0. Derivation of equations (a) and (b) and experimental support are shown in the examples.

The design method according to the present embodiment relates to a method of selecting the type and number (but 2 or more) of miRNA response sequences to be inserted into each slot with respect to the basic structure of miRNA responsive mRNA having the above characteristics. Prior to the above step (1), as an optional pre-step, a step of screening a candidate miRNA target sequence that can be inserted into the slot can be performed. A miRNA response element that can be a candidate may be a response element for any miRNA as described above. Therefore, it is not limited to miRNA specified at the time of the present application, and includes any miRNA whose presence and function are specified in the future. Among these, it is preferable to use a miRNA response sequence that does not have a similar sequence among a plurality of candidate miRNA target sequences and does not contain an AUG that can correspond to the start codon. Here, similar sequences refer to sequences having sequence homology of, for example, about 80% or more, about 85% or more, about 90% or more, or about 95% or more. In the step of screening a candidate miRNA target sequence, as an example, miRNA as an index can be appropriately selected according to the characteristics of the cell type to be separated using the separation method described later. In particular, among a plurality of cells to be separated, a combination of miRNAs that are highly expressed (high activity) in one cell and low expressed (low activity) in another cell can be selected. . This is to increase the accuracy of separation.

Step (1) is a step of measuring the inhibitory effect of mRNA having one miRNA response element in two or more target cells. The mRNA having one miRNA response element used in step (1) is also referred to as 1-slot mRNA. In this step, 1-slot mRNA corresponding to each of a plurality of types of candidate miRNA target sequences optionally selected in the previous step is synthesized. The number of types of 1-slot mRNA that is measured for the translational suppression effect can be appropriately determined by those skilled in the art according to the purpose and application, and there is no theoretical upper and lower limit. For example, the effect of suppressing the translation of 20 or more, 50 or more, 70 or more, 100 or more 1-slot mRNAs can be measured.

Each 1-slot mRNA has a Cap structure (7-methyl guanosine 5 ′ phosphate), a response element of one miRNA, an open reading frame encoding a marker gene, in the 5 ′ to 3 ′ direction from the 5 ′ end, and , With poly A tail. That is, it can be said that the miRNA-responsive mRNA shown in FIG. 2a has only one slot, and if the translational inhibitory effect of 1-slot mRNA can be measured, the 5′UTR 5 ′ side and 3 ′ The number of bases and types of bases on the side may be arbitrary. This is because 1-slot mRNA is experimentally used to derive ρ ⁰ in the above equation (b).

In this embodiment, the target cell is a cell that serves as an index for designing miRNA-responsive mRNA, and two or more types, for example, 3, 4, 5, 6, 7, or 8 There may be more than species, and theoretically, the types of target cells are not limited.

The target cell can refer to a plurality of cell types to be separated in the assumed use situation in the cell separation method described later, but is not limited thereto. An example of cells to be separated may be a plurality of types of cells that are differentiated from the same pluripotent stem cell and have different degrees of differentiation. Alternatively, as another example, cells collected from living organisms, organs, and tissues, or cultured cells thereof, including a group of cells containing a plurality of types of cells, or unwanted cell contamination is suspected. Examples include cells included in the cell group. However, the target cell only needs to be an index for designing miRNA-responsive mRNA that reflects its intracellular characteristics (miRNA activity). It is not limited. Also, in any case, what indicators are effective depending on the purpose of separation, such as whether you want to roughly classify different cell types or whether you want to classify closely among similar cell types? It is different and can be appropriately determined by those skilled in the art.

The cell to be separated may be a cell contained in a cell group collected from a multicellular species, or may be a cell contained in a cell group obtained by culturing an isolated cell. The cells are particularly cells collected from mammals (eg, humans, mice, monkeys, pigs, rats, etc.) or cells obtained by culturing cells isolated from mammals or mammalian cell lines. It may be. Examples of somatic cells include keratinized epithelial cells (eg, keratinized epidermal cells), mucosal epithelial cells (eg, epithelial cells of the tongue surface), exocrine glandular epithelial cells (eg, mammary cells), hormone-secreting cells (eg, , Adrenal medullary cells), metabolism / storage cells (eg, hepatocytes), luminal epithelial cells that make up the interface (eg, type I alveolar cells), luminal epithelial cells of the inner chain (eg, blood vessels) Endothelial cells), ciliated cells with transport ability (eg, airway epithelial cells), cells for extracellular matrix secretion (eg, fibroblasts), contractile cells (eg, smooth muscle cells), blood and immune system Cells (eg, T lymphocytes), sensory cells (eg, sputum cells), autonomic nervous system neurons (eg, cholinergic neurons), sensory organs and peripheral neuron support cells (eg, associated cells), central nervous system Neurons and glial cells (eg, astrocytes), pigment cells (eg Retinal pigment epithelial cells), and progenitor cells (tissue progenitor cells), etc. There is no particular limitation on the degree of cell differentiation and the age of the animal from which the cells are collected, and this is the same for both undifferentiated progenitor cells (including somatic stem cells) and terminally differentiated mature cells. It can be used as the source of somatic cells in the invention. Examples of undifferentiated progenitor cells include tissue stem cells (somatic stem cells) such as neural stem cells, hematopoietic stem cells, mesenchymal stem cells, and dental pulp stem cells. In the present invention, the mammal individual from which somatic cells are collected is not particularly limited, but is preferably a human. Further, preferred cells are a group of cells that have been subjected to an artificial manipulation after collection of progenitor cells, and may include unwanted cells and / or may be heterogeneous. For example, a cell group comprising iPS cells prepared from the somatic cells, or a cell group obtained after differentiating pluripotent stem cells exemplified by ES sputum cells and iPS cells, and the desired cells It is a cell group which can contain the differentiated cell besides. In the present embodiment, it is preferable that the cell group to be discriminated is in a living state. In the present invention, the cell being in a viable state means a cell in a state where metabolic capacity is maintained. The present invention can be used for subsequent applications in which cells are subjected to the method of the present invention and remain in a viable state, particularly while maintaining their mitogenic potential, without losing their natural properties even after the separation method is completed. This is advantageous.

The measurement of the translation inhibitory effect in the target cell can be carried out by introducing 1-slot mRNA into the cell and obtaining the translation inhibitory effect of 1-slot mRNA based on the translation amount of the marker gene.

In the present invention, the step of introducing 1-slot mRNA into cells (hereinafter referred to as the introduction step) is performed by lipofection method, liposome method, electroporation method, calcium phosphate coprecipitation method, DEAE dextran method, microinjection method, gene Using a gun method or the like, one or more 1-slot mRNAs are directly introduced into cells included in a cell group. When measuring the inhibitory effect of 1-slot mRNA for 50 or more, 100 or more, or 200 or more different miRNA response elements, there are two, three or four or more miRNAs with different miRNA response elements and marker genes. Responsive mRNA and control mRNA can be co-introduced into the target cell. Control mRNA refers to mRNA that does not have a miRNA target site and encodes a marker gene different from the marker gene encoded by 1-slot mRNA. The activity ratio of marker proteins expressed from two or more co-introduced mRNAs is constant within the cell population. In addition, the introduction amount at this time varies depending on the cell group to be introduced, the mRNA to be introduced, the introduction method and the kind of the introduction reagent, and those skilled in the art can appropriately select these in order to obtain a desired translation amount. The amount of control mRNA introduced can also be appropriately selected by those skilled in the art to obtain a desired translation amount.

When miRNA-responsive mRNA is introduced into a cell, the translation amount of the marker gene encoded by the miRNA-responsive mRNA is controlled, for example, the amount of translation is suppressed, if a given miRNA exists as RISC in the cell. . The translation amount is quantitatively controlled according to the miRNA activity. On the other hand, when the predetermined miRNA does not exist in the cell, or when the predetermined miRNA does not exist as RISC, the translation amount of the marker gene encoded by the miRNA-responsive mRNA is not suppressed. Therefore, the amount of translation of the marker gene differs between cells in which a given miRNA is present as RISC and cells that are not present. In the present specification, the case where a predetermined miRNA is present as RISC is also referred to as “when miRNA activity is present”. On the other hand, the control mRNA expresses the marker protein regardless of the miRNA activity. This is because even when introduced, the miRNA target sequence does not exist, and therefore translational control is not performed according to the miRNA expression level.

In this embodiment, the translation suppression effect of 1-slot mRNA is measured in all of two or more target cells.

The translation amount of the marker gene can be obtained by detecting a signal from the marker protein using a predetermined detection device. Examples of the detection device include, but are not limited to, a flow cytometer, an imaging cytometer, a fluorescence microscope, a light emission microscope, and a CCD camera. As such a detection apparatus, those suitable for those skilled in the art can be used depending on the marker protein and the mode of discrimination. For example, when the marker protein is a fluorescent protein or a luminescent protein, the marker protein can be quantified using a detection device such as a flow cytometer, an imaging cytometer, a fluorescence microscope, or a CCD camera. In the case of a protein that assists in luminescence or coloration, a marker protein quantification method using a detection device such as a luminescence microscope, a CCD camera, or a luminometer is possible. When the marker protein is a membrane-localized protein, A cell surface protein-specific detection reagent such as an antibody and a marker protein quantification method using the above-described detection apparatus are possible. When the marker protein is a fluorescent protein, the intensity of light emitted from the fluorescent protein and the luminescent enzyme, which are marker proteins translated in individual cells, can be quantitatively obtained by using flow cytometry.

The method for measuring the effect of suppressing translation of 1-slot mRNA is not limited to the specific method described above, and can be performed by any other method. For example, the translational inhibitory effect of 1-slot mRNA can be measured by quantifying mRNA that interacts with miRNA using a method using microarrays or a method based on next-generation sequencing. The present invention also constitutes a measurement using the method.
In step (1), the principle of the above-described formula (b), that is, the translational suppression effect of the miRNA target sequence alone in each slot is generally a constant power of the distance from the start codon to the miRNA target sequence regardless of the miRNA type. In particular, it can be said that the invention relates to a method for adjusting the translation sensitivity of 1-slot mRNA based on the discovery that it is proportional to the power of -0.576. That is, by changing the distance from the start codon of an miRNA response element in an mRNA having one miRNA response element, the translation suppression efficiency is changed, that is, even if the mRNA has the same miRNA response element. Can do. By using such mRNA, it can be said that various cells can be separated.

Step (2) is a step of calculating the translational inhibitory effect of mRNA containing two or more miRNA response elements in each target cell based on the measurement result of step (1). In this step, the translational inhibitory effect of mRNA containing two or more miRNA response elements in each cell is obtained by calculation based on the aforementioned characteristics (a) and (b). More specifically, a hypothetical translational inhibition rate obtained when d is 0, obtained from the measurement value in the step (1), raised to the power of d ^−0.576 (d is the miRNA from the start codon of the marker gene Response element distance) is calculated by integrating the number of miRNA response elements.

From the measurement result of the step (1), an actual measurement value of the translational suppression effect ρ ^(d) at the distance d (nt) from the AUG can be obtained for each 1-slot mRNA. Based on the equation (b), ρ ⁽⁰⁾ that is a virtual mitigation effect of translation of a specific miRNA can be obtained from the actually measured value of ρ ^{(d) and} the tuning factor k value at the position d. Note that d may be the same or different between the 1-slot mRNAs designed in step (1), and the value of ρ ⁽⁰⁾ can be derived in the same way. is there. In the present invention, it is only necessary to be able to measure a desired translational suppression effect based on the principle shown by the formulas (a) and (b). For example, the test conditions (design conditions for 1-slot mRNA, etc.) of step (1) Depending on the case, it may not be necessary to directly calculate the value of ρ ⁽⁰⁾ .

Then, the translational suppression effect of a certain miRNA response element alone at a different distance d from AUG is expressed by ρ ⁽⁰⁾ ^ d ^−0.576 . Then, based on the formula (a), the translation inhibition effect prediction value (calculated value) of mRNA is obtained by accumulating this value by the number of miRNA response elements. In the formula (a), the translation suppression effect of empty slots is 1 (translation is not suppressed). According to this method, the translation suppression effect of miRNA-responsive mRNA having 5 slots can be comprehensively calculated, for example, for all selected miRNA response sequences.

However, since the miRNA response sequence has a length of about 20 bases, the slot cannot be designed at a position where the miRNA target sequence overlaps. Moreover, in the calculation by the formulas (a) and (b), mRNAs containing 1 or less miRNA target sequences are not excluded in principle.

After the above step (1) and before (2), based on the measurement result of the step (1), the type of miRNA response element used for the calculation of the step (2) is changed to a multivariate. A step of limiting using analysis, for example, principal component analysis or cluster analysis may be further included. This process is also referred to herein as a “limiting process”. In the step (1), for example, 1-slot mRNA to be measured may be 100 or more and 200 or more. For all miRNAs searched in step (1), all combinations of mRNAs are designed and searched. Then, the combination may become enormous. Therefore, the miRNA activity profile for each cell obtained in step (1) can be subjected to multivariate analysis such as principal component analysis and cluster analysis to limit miRNA effective for cell separation. In the principal component analysis, miRNAs each having a high absolute value of the principal component loading amount with respect to the principal components having a high contribution rate can be selected as miRNAs useful for cell separation. In cluster analysis, by selecting a representative miRNA from a group of miRNAs classified into the same cluster, other miRNAs that are expected to be less useful for cell separation can be excluded. In the example of the present invention, the step (2) is performed after narrowing down from 270 miRNA response elements to 26 miRNA response elements at this stage. By these multivariate analyses, the mRNA composition calculated in step (2) can be limited to one or more.

Also, at this stage, it is possible to design an mRNA configuration in which not only the type of miRNA response element but also the number and position of slots are fixed to some extent. In FIG. 2a and the examples described later, the step (2) is carried out by fixing the structure of 5-slot mRNA, the distance between slots of 2 nt, and the distance between the slot closest to the start codon and the start codon of 2 nt. Those skilled in the art can appropriately determine the position and number structure of such slots. In such a design, when two or more miRNA-responsive mRNAs are designed for cell separation, the positions and number of slots in the set of two or more miRNA-responsive mRNAs are the same. The total length of the 5′UTR of the miRNA-responsive mRNA may be the same as or different from each other. Both can be designed appropriately at this stage.

This limiting process is not essential in principle when the comprehensive calculation can be performed in the process (2), and may be optionally performed.

Step (3) is a step of selecting mRNA containing two or more miRNA response elements that maximize the difference in translational suppression effect between two or more target cells, based on the value calculated in the step (2). In this step, an miRNA response element and a non-miRNA response element to be inserted into each slot are selected so that the difference in translation suppression effect is maximized.

In step (2), it is possible to design an mRNA having an arbitrary number of slots at a position away from the AUG by an arbitrary number of bases, and each cell when the designed mRNA is introduced into a target cell. The expression level in can be obtained by calculation. That is, the result of introducing arbitrarily designed mRNA into cells can be estimated. If you design many types of mRNA in step (2) and calculate the estimated value in each cell when it is introduced into the cell, you can set an arbitrary condition to obtain an mRNA that has an effect according to your purpose. be able to. For example, if you want to obtain mRNA with high ability to separate cells when introduced into multiple types of target cells, using the dispersion of expression level in each cell for each designed mRNA as an index, It can be selected as an mRNA expected to have the highest cell separation ability. Even when two or more types of designed mRNAs are simultaneously introduced into a cell, the expression level of each mRNA can be obtained by calculation. A combination can be obtained. For example, when four types of mRNA are introduced into a cell and the cells are classified using the ratio of the expression level of each of the two types of mRNA as an index, the expression level in each cell for each of the four types of mRNA is determined by the process (2 ), The ratios can be calculated and two fluorescence ratios in each cell can be obtained as parameters for each set of four types of mRNA. Search for the mRNA set that maximizes the variation of the two parameters obtained, the difference between any two cells, or the integrated value of the difference for any mRNA set selected from four types of any mRNA. Can do.

The number of mRNAs to be searched in step (3) varies depending on the number of miRNA candidates to be inserted into slots and the number of slots to be narrowed in the limited step. For example, in a limited process, when the calculation is limited to a single mRNA or a set of 4 types of mRNA, the process (3) is substantially the same as the single that calculated the translational suppression effect in step (2). In some cases, it may be necessary to simply select one mRNA or a set of four mRNAs.

In the step (3), the final single mRNA or 2 to 2 based on the calculation result of the dispersion, the variation of multiple parameters, the difference between any two cells, or the integrated value of the difference. You can select a set of 4 mRNAs, and measure the effect of translational suppression using an experimental method, and calculate the variance, the variation of multiple parameters, the difference between any two cells, or the integrated value of the difference. Combined with the results, the final single mRNA or a set of 2 to 4 mRNAs can be selected. That is, for all mRNAs for which the translational suppression effect was calculated in step (2), or among the mRNAs for which the translational suppression effect was calculated in step (2), dispersion in step (3), variation of multiple parameters, any two About mRNA which was narrowed down to some extent from the calculation result of the difference between cells or the integrated value of the difference, these mRNAs are actually prepared by genetic engineering techniques, introduced into target cells, and the separation method is tested. As a result, the design of the present invention can be completed by actually selecting the mRNA that maximizes the dispersion of the target cells, the variation of multiple parameters, the difference between any two cells, or the integrated value of the difference. .

Next, steps (1) to (3) will be described based on examples.
Process (1)
Using 1-slot mRNA, the miRNA activity in each cell is specifically measured. Here, since the expression level of miRNA-responsive mRNA is considered, the smaller the expression level, the higher the activity. In addition, logarithm is considered for subsequent calculations. Therefore, 0 is the case where there is no activity, -1 is the case where the expression level is suppressed to 1/10, and -2 is the case where the expression level is suppressed to 1/100. Assume that cells A, B, C, and D are analyzed and the following results are obtained.

 

Process (2)
If the miRNA activity is known for each cell in the above step, the expression level in each cell can be predicted when a certain mRNA containing a plurality of miRNA target sequences is produced and introduced into the cell. At this time, the expected value is given by k x the sum of the logarithms. The value of k is determined by the distance (d [nt]) from the AUG of the slot, and is obtained by k = d ^-0.576 . However, in the following example, k is given directly for the sake of simplicity (assuming an mRNA with a slot at such d).

a = 0.1 x 0 + 0.4 x -1 = -0.4 (The expression level of this mRNA is calculated as 10 ^-0.4 = 0.398)
b = 0.1 x -0.5 + 0.4 x -0.1 = -0.09
c = 0.1 x -1 + 0.4 x -0.8 = -0.42
d = 0.1 x -2 + 0.4 x -0.2 = -0.28

For example, mRNA containing 100 slots of miRNA and including 5 slots with fixed distance from AUG in advance (for example, the distance of 5 slots is 105 nt, 80 nt, 65 nt, 40 nt, 15 nt) ) Each slot can contain 100 miRNA target sequences or 101 empty sequences. The 5-slot mRNA that can be created is 101 x 101 x 101 x 101 x 101 = 10,510,100,501 (however, one of them is all empty), and expression when that mRNA is introduced into each cell The amount can be calculated. With this calculation, 10,510,100,501 types of mRNAs that are fixed at different slot distances can be designed. For example, if the distance between 5 slots is 106 nt, ス ロ ッ ト 80 nt, 65 nt, 40 nt, 15nt, there are 10,510,100,501 types, and 104 nt, 80 nt, 65 nt, 40 nt, 15nt is also 106 nt, 83 nt , 62 nt, 41 nt, 13nt can also be 10,510,100,501 types. The same applies when the number of slots is other than 5, each of which can design an mRNA with an arbitrary number of slots at an arbitrary number of bases away from the AUG, and when the mRNA is introduced into these cells, The expression level in the cell can be obtained by calculation.

Step (3)
The degree of cell separation when a certain mRNA is used is calculated using the logarithmic variance of the expression level of that mRNA when that mRNA is used as an index.
When the slot configuration is mRNA of “empty / miR-1 / empty / miR-2 / empty” in order from the 5 ′ side, the logarithmic value of the expression level is (A, B, C, D) = (−0.4, -0.09, -0.42, -0.28) Therefore, the sample variance of this mRNA (the target cell when using) can be calculated as 0.0172. For example, when using mRNAs of “Sky / miR-1 / Sky / miR-2 / Sky” and “miR-3 / miR-4 / miR-4 / Sky / miR-3”, “miR-3 / miR -4 / miR-4 / empty / miR-3 "is expressed as follows.
e = (0.05 + 0.8) x -2 + (0.1 + 0.2) x -0.5 = -1.85
f = (0.05 + 0.8) x -2 + (0.1 + 0.2) x -0 = -1.7
g = (0.05 + 0.8) x -0.4 + (0.1 + 0.2) x -2 = -0.94
h = (0.05 + 0.8) x -1.2 + (0.1 + 0.2) x -0.1 = -1.05
The sample dispersion of this mRNA can be calculated as 0.156. In addition, when trying to classify cells by these two values, {A, B, C, D} = {(-0.4, -1.85), (-0.09, -1.7), (-0.42, -0.94), (-0.28, -1.05)} Therefore, calculating the correlation coefficient of two variables yields -0.320. For example, the sample variance and the correlation coefficient can be used as an index of two-parameter variation.

The miRNA-responsive mRNA can be synthesized by a person skilled in the art by any method known in genetic engineering if the miRNA-responsive mRNA is designed according to the above steps (1) to (3) and the sequence thereof is determined. In particular, it can be obtained by an in vitro transcription synthesis method using a template DNA containing a promoter sequence as a template.

Therefore, the present invention can also be regarded as an mRNA production method including a design process including the steps (1) to (3) and a synthesis process. The produced mRNA can be suitably used in the separation method described in detail in the second embodiment.

The present invention relates to a method for separating two or more target cells according to the second embodiment. The method uses the miRNA-responsive mRNA designed and synthesized by the method of the first embodiment to separate two or more target cells using the translation amount of the marker gene as an index. Measurement with miRNA-responsive mRNA can be said to be an index that reflects the characteristics of the target cell used in the method of the first embodiment. It is also possible to introduce into the mixture. However, it is an index that attempts to classify the combination of target cells that was the basis when designing mRNA as much as possible.

In the separation method according to this embodiment, mRNA synthesized and synthesized according to the first embodiment is used. Therefore, one type of miRNA-responsive mRNA designed and synthesized according to the first embodiment can be used, and two types, three types, four types or more of miRNA-responsive mRNAs can also be used. When two or more types of miRNA-responsive mRNA are used, according to the first embodiment, two or more types of miRNA responsiveness having different 5′UTR sequences and marker genes designed to maximize cell separation are used. Use a set of mRNAs. In particular, it is more preferable to use a set composed of four types of miRNA-responsive mRNAs having different 5 ′ UTR sequences and marker genes.

As a specific method, four types of miRNA-responsive mRNA can be co-introduced into a cell group including the target cell, and the target cell can be isolated using the translation amount of the marker gene as an index. Detailed experimental procedures and measurements The method is substantially the same as step (1) of the first embodiment. Therefore, the measurement can be performed by various methods, but the case where the marker gene is a fluorescent protein gene and the measurement is performed by flow cytometry will be described as an example. When the fluorescence intensity measured from four types of miRNA-responsive mRNA is FL1, FL2, FL3, and FL4, and the fluorescence intensity ratio of FL1 / FL2 and FL3 / FL4 is the X-axis and Y-axis, respectively. Cells can be separated with a dot plot. Alternatively, the fluorescence intensity measured from 6 types of miRNA-responsive mRNAs is similarly divided into 2 types of fluorescence intensity ratios, and the cells are separated by dot plots with the X-axis, Y-axis, and Z-axis, respectively. There is no theoretical upper limit.

Such separation can also be performed by imaging cytometry using an image analyzer. The image analyzer can obtain information on changes in the amount of translation of marker genes in the cell over time, and is excellent in terms of imaging and visualization. In addition to improving the amount of analysis per unit time, It is advantageous in that it can be applied to analysis including morphological information and positional information, identification of cells adhered to a culture vessel, and cells targeted for planar or three-dimensionally organized cells. .

Hereinafter, the present invention will be described in more detail using examples. The following examples do not limit the invention.

Sequence of mRNA The 5 ′ UTR sequence of 5-slot mRNA used in this example is shown in SEQ ID NOs: 2 to 115 in Table 3, and the 5 ′ UTR sequence of 1-slot mRNA is shown in SEQ ID NOs: 122 to 391 in Table 5. Show.

 

 

 

 

 

 

 

 

 

 
 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

Construction fluorescent protein fluorescence protein coding region and 3'UTR fragment hmAG1, hmKO2, protein coding region of hdKRed and tagBFP the plasmid DNA: pFucci-S / G2 / M Green (Amargaam), pFucci-G1 Orange (Amargaam), pAM Appropriate primer sets (Fwd / Rev hmAG1 (SEQ ID NO: 392) shown in Table 6 from -tagBFP (references [Miki, K., et al, Cell Stem Cell, 2015]) and pNP-hdKeima-Red (Amargaam), respectively. / 393), Fwd / Rev hmKO2 (SEQ ID NO: 394/395), Fwd / Rev tagBFP (SEQ ID NO: 396/397), Fwd / Rev hdKeimaRed (SEQ ID NO: 398/399))). The plasmid DNA in the PCR product was digested with the restriction enzyme Dpn I (Toyobo) at 37 ° C. for 30 minutes and purified using the MinElute PCR purification kit (QIAGEN) according to the manufacturer's instructions. The 3′UTR sequence was amplified by PCR using oligonucleotide temp3UTR (SEQ ID NO: 405) as a template and Fwd3UTR (SEQ ID NO: 403) and Rev3UTR (SEQ ID NO: 404) as primers. The 5′UTR sequence of the control mRNA was PCR amplified using the oligonucleotide temp5UTR (SEQ ID NO: 402) as a template and T7Fwd5UTR (SEQ ID NO: 400) and Rev5UTR (SEQ ID NO: 401) as primers. These PCR products were purified using MinElute PCR purification kit (QIAGEN) according to the manufacturer's instructions.

 

Construction of mRNA synthesis template DNA To generate mRNA synthesis template, PCR amplification fragment of marker protein coding region (final concentration 0.2 ng / μL), PCR amplification fragment of 3'UTR (final concentration 10 nM) and 5 'UTR sequence Were mixed with each other (final concentration 10 nM), PCR amplified using the T7Fwd and Rev120A primer sets shown in Table 7, and ligated. One oligonucleotide was used as a template for synthesis of 1-slot mRNA, and two oligonucleotides were used as a template for synthesis of 5-slot mRNA. However, when preparing control mRNA, the purified PCR fragment was used at a final concentration of 10 nM instead of the oligonucleotide for the 5 ′ UTR sequence. The PCR product was purified using MinElute PCR purification kit (QIAGEN) according to the manufacturer's instructions. The 5 ′ UTR sequence of 5-slot mRNA finally synthesized is shown in Table 3, the 5 ′ UTR sequence of 1-slot mRNA is shown in Table 5, and the ORF sequences of fluorescent proteins (SEQ ID NOs: 117, 118, 119, 120, And the 3 ′ UTR sequence (SEQ ID NO: 121, common to all mRNAs) is shown in Table 4. The 5′UTR sequence of 5-slot control mRNA (SEQ ID NO: 1) is shown in Table 3 and 5 of 1-slot control mRNA. The 'UTR sequence (SEQ ID NO: 116) is shown in Table 4.

 

mRNA synthesis and purification miRNA-responsive mRNA is prepared using the MegaScript T7 kit (Ambion) in a modified protocol (see reference [Miki, K., et al, Cell Stem Cell, 2015] below). did. In this reaction, pseudouridine-5′-triphosphate and 5-methylcytidine-5′-triphosphate (TriLink BioTechnologies) were used in place of uridine triphosphate and cytidine triphosphate, respectively. Prior to the IVT (mRNA synthesis) reaction, guanosine-5′-triphosphate was diluted 5-fold with Anti Reverse Cap Analog (New England Biolabs). The reaction mixture was incubated at 37 degrees for 4 hours, TURBO DNase (Ambion) was added and then incubated at 37 degrees for an additional 30 minutes. The obtained mRNA was purified by FavorPrep Blood / Cultured Cells total RNA extraction column (Favorgen Biotech) and incubated at 37 degrees for 30 minutes using Antarctic Phosphatase (New England Biolabs). Then, it further refine | purified by RNeasy MiniElute Cleanup Kit (QIAGEN).

In accordance with the conditions described in Table 8, reverse transfection was performed in 24-well format culture plates using StemFect (Stemgent) according to the manufacturer's instructions. However, forward transfection was performed when the time course of human iPS cells was followed.

 

Cells were detached from culture dishes 24 hours after flow cytometry transfection and analyzed by flow cytometry using a FACSAria II (BD Biosciences) through mesh. hmAG1, hmKO2, tagBFP and hdKRed are blue laser (488 nm) and FITC filter (530/30 nm), green laser (561 nm) and PE filter (585/42 nm), purple laser (405 nm) and Pacific Blue Detection was performed with a filter (450 nm / 40 nm), and a violet laser (405 nm) and a Qdot 605 filter (610/20 nm), respectively. Dead cells and debris were excluded based on forward and side light scatter values.

Correction analysis of fluorescence values detected by flow cytometry When four types of fluorescent proteins hmAG1, hmKO2, tagBFP and hdKRed are detected at the same time, cells transfected with only hmAG1, hmKO2, tagBFP or hdKRed mRNA are simultaneously detected. Analysis was performed to correct the fluorescence value detected as a fluorescent protein different from the actual protein (reference: [Endo, K. and Saito, H. Methods in Molecular Biology, 2014]).

Calculation of Relative expression By flow cytometry analysis, the fluorescence intensity of hmAG1 expressed from the reporter mRNA is divided by the fluorescence intensity of tagBFP expressed from the co-introduced control mRNA, and the geometric mean value in the analyzed cell population is calculated as the reporter mRNA. It was set as the fluorescence ratio. For each mRNA sequence, the relative ratio of the fluorescence ratio in the presence of the miR-1 inhibitor (target miRNA active state) is defined as “Relative expression” with reference to the fluorescence ratio in the presence of the miRNA inhibitor to which the reporter mRNA responds. Defined. In the case of mRNA containing two types of miRNA response elements, the “Relative expression” value was measured in the presence of one miRNA inhibitor, and the sum of the two measured values was defined as “Estimated expression”. In the case of mRNA containing three miRNA response elements, the “Relative expression” value is measured in the presence of two of the three miRNA inhibitors, and the sum of the measured three values is called “Estimated expression”. did.

Results Developed a design method for a probe (= mRNA) that responds to a large number of miRNAs with a single molecule and arbitrarily adjusts the degree of response to individual miRNAs (detection sensitivity). Successfully extracted. Since RNA Synthetic Device is based on the control of mRNA and post-transcriptional / translational steps, miRNAs that act on mRNA are used as marker molecules inside cells. For high-throughput analysis such as microarray and next-generation sequencing, first, (1) comprehensive quantitative detection of miRNA, then (2) easy handling from a large number of variables such as multivariate analysis A large number of synthetic variables are extracted, and the state of the cells is distinguished based on the extracted variables (the left figure in FIG. 1a). On the other hand, the number of signals that can be detected simultaneously non-invasively (without killing cells) is limited, and cannot be handled by a detection probe corresponding to one-to-one. Therefore, in order to detect multi-factor information in living cells, the multi-factor information is first summarized, and then the resulting synthesized parameter is detected (right diagram in FIG. 1a). Then, even with a limited number of signals that can be detected at the same time, it is possible to directly detect a synthetic parameter derived from the quantitative information of a large number of living cell factors and extracted from the essence thereof. We have found that one mRNA can detect multiple miRNAs countably, and that the sensitivity to each miRNA can be adjusted by the position of the miRNA target sequence on the mRNA.

Since multiple miRNA responses create mRNAs that respond to multiple types of miRNAs that are accumulated, mRNA that previously contained miRNA target sequences at only one location (1-slot mRNA, Fig. 6a, Miki, K. et al , Cell Stem Cell, 2015, International Publication WO2015 / 105172), 5 miRNA target sequence insertion sites (slots) were designed in succession in the 5 ′ UTR (5-slot mRNA, FIG. 2a). Examples of intracellular factors include miR-34a-5p as an example of miRNA showing weak activity in HeLa cells, miR-17-5p and miR-92a-3p as examples of strong activity, and examples of very strong activity We selected four types of miR-21-5p. Two or three different miRNA target sequences were inserted into five slots, and 12 kinds of mRNAs encoding hmAG1 as a marker protein were randomly designed to synthesize mRNAs (FIG. 2a, Table 3). The synthesized 5-slot mRNA was transfected into HeLa cells together with an mRNA encoding a fluorescent protein tagBFP and an inhibitor against miRNA to which 5-slot mRNA responds (miRVana miRNA inhibitor, Invitrogen) as a control for mRNA introduction. 24 hours after transfection, the fluorescence intensity of the marker protein was quantified using a flow cytometer, and the Relative expression value was analyzed.

For example, in the case of mRNA having the target sequence of miR-17-5p in slot-2 and the target sequence of miR-92a-3p in slot-4 (Fig. 2b), 5 when inhibitors for both miRNAs are introduced The expression level of -slot mRNA is considered to indicate the expression level when not affected by the activity of either miRNA, and this was taken as the reference value (= 1) of the expression level for each 5-slot mRNA. When either miRNA inhibitor (miR-17-5p or miR-92a-3p) 導入 is introduced, the expression level is only the other miRNA activity mi (miR-92a-3p or miR-17-5p), respectively. Is considered to be reflected. In addition, the expression level of in the absence of an miRNA inhibitor (actually, in the presence of inhibitor with respect to miRNA-1 used as mock 考 え) is considered to reflect the activities of both miRNAs. The value of Relative expression reflecting both miRNA activities was close to the integrated value of Relative expression values in response to individual miRNA activities (this is the predicted expression level, estimated expression).

Therefore, for each of 12 randomly generated 5-slot s mRNAs, Relative expression value (observed relative expression) 下 in the absence of miRNA inhibitors and Relative responsive to miRNA activity of only one of the responding miRNAs When the integrated values of expression values (estimated expression) were compared, showed a very high correlation (Fig. 2c). This indicates that a plurality of miRNA activities can be detected in an integrated manner by using a synthetic mRNA having a plurality of miRNA target sequences.

Synthesize 1-slot mRNA that responds to miR34a-5p, miR-92a-3p, miR17-5p or miR-21-5p and introduce it into HeLa cells along with tagBFP mRNA and 4 miRNA inhibitors that control mRNA transfer When Relative expression was calculated, the value of 導入 Relative expression was not affected even when miRNA inhibitors other than miRNA to which mRNA responded were introduced. Therefore, it was confirmed that under these experimental conditions, these miRNA inhibitors do not inhibit other three types of miRNA activities that are not intended (FIG. 5a). On the other hand, when a plurality of miRNA inhibitors are inserted, the total amount of miRNA inhibitors introduced into cells varies. As the amount of miRNA inhibitor against miR-1 used as a negative control was increased, the measured Relative expression value was not affected, so the effect of the total amount of miRNA inhibitor introduced under the current experimental conditions Was confirmed to be negligible.

Of the five slots whose response sensitivity can be adjusted depending on the slot position, a series of 5-slot mRNAs in which one, two or three are occupied by the same miRNA target sequence are miR34a-5p, miR-92a-3p. , miR17-5p or miR-21-5p were designed and mRNA was synthesized (Fig 2d, Table 3). The synthesized 5-slot mRNA was transfected into HeLa cells together with an inhibitor against miRNA to which the fluorescent protein tagBFP and 5-slot mRNA respond, respectively, as a control for mRNA introduction. 24 hours after transfection, the fluorescence intensity of the marker protein was quantified using a flow cytometer, and the Relative expression value was analyzed. However, in miR-21-5p analysis, the expression of the marker gene is strongly suppressed and the influence of the autofluorescence of the cells increases, so cells with tagBFP fluorescence values of 10,000 or less are removed and the amount of mRNA introduced is high. The cell group was evaluated.

At this time, the expression level (Relative expression) of each 5-slot mRNA is considered to be an integrated value of miRNA responsiveness (ρ) in each slot (Fig. 2d, robot). Therefore, miRNA responsiveness (ρ) in each slot was calculated for each of the four types of miRNAs from the experimentally measured data set of Relative expression value (observed relative expression). The integrated value of miRNA responsiveness in each slot determined by this analysis was defined as the estimated expression level (estimated expression) 5- for each 5-slot mRNA.

For all four miRNAs, the predicted expression level was in good agreement with the experimentally measured Relative expression value (observed relative expression) (Fig. 2e). This suggests that the behavior of 5-slot mRNA can be explained by miRNA responsiveness (Local relative expression, ρ) in each slot. In any of the four types of miRNAs, the 5′-side slot ’with a smaller slot number had a lower response sensitivity and the 3′-side slot had a higher response sensitivity (FIGS. 2f and 6).

On the other hand, when analyzing Relative expression values for 1-slot mRNA responding to miR34a-5p, miR-92a-3p, miR17-5p or miR-21-5p, 1-slot mRNA becomes slot-5 in any miRNA. Close values were shown (Fig. 6b). The miRNA target sequence of 1-slot mRNA is located ~ 20 nt from the 5 'end of mRNA and AUG 23 nt (Fig. 6a). Therefore, this result indicates that the miRNA target sequence is not the 5' end. This suggests that the closer to the start codon, the higher the response sensitivity of mRNA, and the lower the response sensitivity depending on the distance from the start codon.

It was obtained by analysis of 4 types of miRNAs with different activities in order to determine the relationship between the miRNA target sequence position and miRNA responsiveness as a parameter common to these miRNAs, not parameters unique to the four types of miRNAs analyzed The response sensitivity of each slot was simultaneously fitted to the following exponential function model (FIG. 2f). Fitting was performed using the method of least squares on logarithmic values.

At this time, d is the distance ([nt]) from the start codon, and ρ ⁰ is the virtual response sensitivity when the distance is 0 [nt] for each miRNA. ξ represents a common variable regardless of the type of miRNA. (That is, ρ (d) = {ρ ⁰ } ^{k (d)} = ρ ⁰ ^ d ^ξ .)

As a result of the fitting analysis, the miRNA translation suppression effect (local repression, -log (ρ)) can be determined by using the distance 開始 (d [nt]) from the start codon to the miRNA target sequence and the variable ξ regardless of the miRNA type. I was able to explain. In this analysis, 共通 -0.576 was obtained as the value of the common variable ξ. In other words, when designing an mRNA containing a miRNA target sequence in the 5'UTR mi, the translation suppression effect by miRNA is generally considered to be proportional to the -0.576 power of the distance from the start codon to the miRNA target sequence, regardless of the type of miRNA. .

Classification of many types of cells
In order to demonstrate this model, we attempted to classify and isolate multiple types of normal human cells according to the miRNA activity profile while remaining alive (FIG. 3).

First, the miRNA activity for each cell type was quantified exploratoryly. In order to prevent the target sequence from becoming 5A UTR, the miRNA not containing CAU was extracted from 270 (Fig. 3a, Table 4). At this time, we selected cells with relatively high expression levels from the past literature (references [Neveu, P., et al, Cell Stem Cell, 2010]) and miRNAs with high sequence similarity Excluded. Of the 270 miRNAs selected, 90 were synthesized as hmAG1, another 90 as tagBFP, and the remaining 90 as 1-slot mRNA that encodes hdKRed as a marker fluorescent protein. The 5 'UTR sequence of the prepared mRNA is shown in Table 5 IV. Analysis target cells are mixed with 1 type of 1-slot mRNA encoding hmAG1, tagBFP, and hdKRed with different miRNA target sequences inserted, and a total of 4 types of hmKO2 mRNA used as mRNA transfer controls. Transfected. That is, 1-slot mRNA was transfected with a total of 90 combinations. In addition, as an analytical control, 6 kinds of water containing no mRNA, a mixture of four kinds of mRNAs, hmAG1, tagBFP, hdKRed, and hmKO2 that do not respond to miRNA, or only one of each of these four kinds of mRNAs. Different types of transfection were performed. That is, 96 types of transfection were performed for each cell, and one set of search analysis was performed. In addition to HeLa cells, primary cells of human normal skin fibroblasts (NHDF), human normal lung fibroblasts (NHLF), human normal epidermal keratinocytes (NHEK), and human kidney epithelial cells (HRE) are analyzed. Four types of cells, mouse fibroblast cell (MEF) cell, iPS cell derived from human epithelial fibroblasts (hiPSC), and cell cell (hiPSC®14d) cell that partially differentiated hiPSC in the absence of bFGF were used.

After flow cytometry, using the corrected fluorescence intensity, divide the fluorescence intensity of hmAG1, tagBFP and hdKRed by the fluorescence intensity of hmKO2 for each analyzed cell to obtain the fluorescence ratio, and calculate the geometric mean of the analyzed cells. The fluorescence ratio of 1-slot mRNA was expressed as Fluorescence ratio. The search analysis was performed twice on the same cells and the resulting fluorescence ratio comparison is shown in FIG. 3b. These analysis results showed a high correlation, suggesting that this analysis system is stable.

In general, it is known that only about 1/3 of miRNA detected quantitatively has a translational inhibitory activity (reference document [Mullokandov, G. et al, Nature Methods, 2013]). Since about 2/3 of 270 miRNAs used in this analysis are considered not to be affected by miRNA activity, the fluorescence ratio of 1-slot mRNA is considered to be distributed diagonally as a whole. However, comparing the results of cell-to-cell search analysis, the observed fluorescence ratio is plotted off the diagonal line in the comparison between HeLa cells and NHLF (FIG. 7a). This is because there is a difference in the synthesis efficiency and stability of the fluorescent protein from cell to cell, and as a result, the distribution of the expression level is considered to be biased. Therefore, the expression level of fluorescent protein between cells was standardized using HeLa cells as a reference. Specifically, a scatter diagram of the fluorescence ratio of each cell and the fluorescence ratio of HeLa cells was created, linear regression was performed for each fluorescent protein, and linear correction was performed so that the slope became 1 (FIG. 7b). . As a result, the bias between cells was corrected, and each 1-slot mRNA was distributed diagonally when creating a scatter plot, but the distribution of fluorescence ratios of hmAG1, tagBFP and hdKRed did not match. This indicates that there is a bias due to the difference in the fluorescent protein used as a marker. Therefore, the distribution of each fluorescence ratio was standardized to an average of 0.5 and a standard deviation of 0.15 (FIG. 7c). A comparison of NHDF and hiPSC for the normalized fluorescence ratio is shown in FIG. 3c. Although it is expected that there is a large difference in miRNA activity between these cells, in fact, each point in the figure showing miRNA activity spreads in a direction perpendicular to the diagonal line, and this distribution Are evenly distributed regardless of the fluorescent protein used. This correction value was used as a miRNA activity profile for each cell, and analyzed by principal component analysis as an example of a linear model. For the principal component analysis, the prcomp function of the statistical package R was used. When the first principal component is plotted on the horizontal axis and the second principal component is plotted on the vertical axis, the eight cell conditions were classified (FIG. 3d). This indicates that these cells can be classified for statistical analysis according to the miRNA activity profile obtained by exploratory analysis using 1-slot mRNA.

Components obtained by principal component analysis are expressed as a linear combination of 270 miRNA activities. On the other hand, multiple miRNA activities in cells could be measured quantitatively by one mRNA with multiple target sequences, and the detection sensitivity of miRNA activity could be controlled by the distance between the target sequence and the start codon . Therefore, taking the logarithm of the expression level of the marker protein, a plurality of miRNA activities are aggregated in a linear model (FIG. 1b). However, the coefficient for miRNA activity (k = d ^-0.576 ) is always positive. Thus, by using two types of miRNA-responsive mRNAs that express different fluorescent proteins and taking the ratio of the two types of fluorescence values, any weight can be realized for each miRNA (FIG. 1b). That is, the multivariate miRNA activity profile in the cell can be aggregated as a ratio of one fluorescence value by designing two miRNA response mRNAs. Thus, as in the case of the principal component analysis, it is considered that the target cells can be classified alive by using mRNA that maximizes the variance of the values measured in the target cells.

Here, for simplicity, four types of 5-slot mRNA were used. In order to classify different cell types, miRNAs that were considered to have high cell separation ability were selected from 270 miRNA activity profiles. Principal component analysis and cluster analysis (Ward method) are performed on miRNA activity profile obtained by screening analysis, and 270 270 miRNA is separated into 49 clusters, and each cluster has the highest contribution to the first or second component. I chose. The dist function and hclust function of the statistical package R were used for cluster analysis. Principal component analysis was performed again using only the selected 49-miRNA, and 26 miRNAs that contributed to the first component or the second component were selected. The predicted value of the expression level of 5-slot mRNA containing the target sequence of 26 miRNA was calculated in 8 types of cells, and mRNA (Estimated expression <0.05) with extremely small expression level in any cell was excluded. For the remaining mRNA, the variance of the predicted expression level in 8 cells was calculated, and the mRNA that maximized the variance was defined as 5-slot mRNA # 1 (hmAG1). Next, the ratio of 5-slot mRNA # 1 to the predicted expression level was calculated, and the mRNA whose value maximizes the dispersion of 8 cells was defined as 5-slot mRNA # 2 (hmKO2). Furthermore, the correlation coefficient between the ratio of 5-slot mRNA # 1 and 5-slot mRNA # 2 と and the predicted expression level of the new mRNA is less than 0.3 か つ, and the dispersion of 8 cells is maximized in these two-dimensional data. MRNA to be obtained was determined and designated as 5-slot mRNA # 3 (tagBFP). Finally, 5-slot mRNA # 4 (hdKRed) す る that maximizes the dispersion of 8 cells in the two-dimensional data of 5-slot mRNA # 1 and # 2 比率 ratio and 5-slot mRNA # 3 and # 4 比率 ratio was obtained. . A set of 5-slot mRNA obtained by a series of computational searches is shown in Fig. 3e. In addition, FIG. 3f shows the estimation result of the distribution of 8 cells when this mRNA set was transfected.

The 4 designed 5-slot mRNAs were mixed and transfected into HeLa cells, NHEK, NHDF, hiPSC and hiPSC-14d, and flow cytometry was performed 24 hours later. After fluorescence correction, two fluorescence ratios of hmAG1 / hmKO2 and tagBFP / hdKRed were determined for each analyzed cell, and the cell distribution was shown in a density plot (FIG. 3g). These five types of cells were distributed in different locations using the same 5-slot mRNA set. Based on these results, two parameters extracted from 6 types of miRNA activity with a linear model were directly measured using 4 types of 5-slot mRNA, and 5 types of cells were isolated based on this parameter. It has become possible. In addition, similar experiments were performed with a set of mRNA that does not respond to miRNA and a set of 1-slot miRNA-responsive mRNA 、, but these cell types could not be separated (Fig. 8).

Tracking changes in the same cell (species) over time In addition, the process of losing differentiation ability of human iPS cells is tracked over time, and the difference in the resolution of the resulting mRNA set depends on the difference in the cells searched in step (1). Was verified (FIG. 4). First, from the results of exploratory quantification of 270 miRNA activities (Fig. 3c), 54 miRNAs with a large difference between hiPSCs and hiPSCs (iPSC 14d) partially differentiated in the absence of bFGF were selected ( FIG. 4a). For these miRNAs, a secondary exploratory analysis was performed to track changes in miRNA activity over time during the process of partial differentiation of hiPSC in the absence of bFGF (FIG. 4b). hiPSC was cultured in the absence of bFGF, and on the day of culture start (day 0) and 1 (day 1), 3 (day 3), 6 (day 6), 9 (day 9), or 14 days later (day 14), One kind of 1-slot mRNA encoding hmAG1, tagBFP, and hdKRed each having a different miRNA target sequence inserted, and a total of four kinds of mRNAs, hmKO2 mRNA used as a control for mRNA introduction, were mixed and transfected. At this time, the same transfection as the six types of controls used in the primary search analysis (FIG. 3a) was performed. That is, 24 types of transfection were performed in each culture condition (FIG. 4a).
Cells under each culture condition were analyzed by flow cytometry 24 hours after transfection. Unlike the results of the temporary search analysis (FIG. 7a), there was no significant bias in the fluorescence ratio between these culture conditions (FIG. 4c), so the value of the fluorescence ratio was directly used for the subsequent analysis. . The miRNA activity profile was subjected to principal component analysis, and the cells under these culture conditions were classified with the first principal component as the horizontal axis and the second principal component as the vertical axis (FIG. 4d).
Principal component analysis and cluster analysis (Ward method) were performed on the measured fluorescence ratio, and 54 miRNAs were separated into 23 clusters, and miRNAs with the highest contribution to the first or second component were selected in each cluster. . In the cluster analysis, the dist function and hclust function of the statistical package R were used as before. Principal component analysis was performed again with only the selected 23 miRNAs, and 11 miRNAs with high contribution to the first component or the second component were selected. Using these 11 miRNAs, the 5-slot mRNA set obtained by performing the same calculation as the primary search analysis (Fig. 3e, f) is shown in Fig. 4e, and when this mRNA set is transfected. FIG. 4f shows the estimation result of the cell distribution under each culture condition. In addition, FIG. 4g shows the results of actually transfecting these four 5-slot mRNAs into cells of each culture condition and performing flow cytometry. On the electronic tea plot, the cell population moves with time, and the changes over time in human iPS cell information over time are tracked alive using two parameters extracted from eight miRNA activities using a linear model. We were able to. Compared with the probe set (Fig. 3e) in which many types of cells were classified, changes in hiPSC could be separated into a wider range (Fig. 9). This indicates that the two synthesis parameters that classify cells can be changed arbitrarily depending on the mRNA design.

(References)
MicroRNA profiling reveals two distinct p53-related human pluripotent stemcell states.
Neveu P., Kye MJ., Qi S., Buchholz DE., Clegg DO., Sahin M., Park IH., Kim KS.,
Daley GQ., Kornblum HI., Shraiman BI., Kosik KS.
Cell Stem Cell, 7 (6): 671-81, 2010

Efficient Detection and Purification of Cell Populations Using Synthetic MicroRNA Switches.
Miki K., Endo K., Takahashi S., Funakoshi S., Takei I., Katayama S., Toyoda T., Kotaka M., Takaki T., Umeda M., Okubo C., Nishikawa M., Oishi A ., Narita M., Miyashita I., Asano K., Hayashi K., Osafune K., Yamanaka S., Saito H., Yoshida Y.
Cell Stem Cell, 16 (6): 699-711, 2015

Engineering protein-responsive mRNA switch in Mammalian cells.
Endo K., Saito H.
Methods Mol Biol, 1111: 183-96, 2014

High-throughput assessment of microRNA activity and function using microRNA sensor and decoy libraries.
Mullokandov G., Baccarini A., Ruzo A., Jayaprakash AD., Tung N., Israelow B., Evans MJ., Sachidanandam R., Brown BD.
Nat Methods, 9 (8): 840-6, 2012

Claims (7)

1. A method for designing an mRNA containing two or more miRNA response elements, comprising the following steps, wherein an mRNA containing two or more miRNA response elements is operably linked to two or more miRNA response elements: A method comprising mRNA comprising a sequence;
   (1) a step of measuring the translation inhibitory effect of mRNA having one miRNA response element in two or more target cells;
   (2) A step of calculating the translational suppression effect of mRNA containing two or more miRNA response sequences in each target cell based on the measurement result of the step (1),
   (3) A step of selecting mRNA containing two or more miRNA response elements that maximizes the difference in translational suppression effect between the two or more target cells based on the value calculated in the step (2).
2. Between the step (1) and the step (2), based on the measurement result of the step (1), the types of miRNA response elements used for the calculation of the step (2) are limited using multivariate analysis. The method of claim 1, further comprising a step.
3. The translation suppression effect of the step (2) is obtained by adding the translation suppression effect -log (ρ) of each of the two or more miRNA response elements by the number of miRNAs. -log (ρ) is
   

   

   
   (In the formula, ρ represents the translation suppression effect by miRNA,
   d [nt] represents the distance from the start codon to the miRNA target sequence,
   ξ represents -0.576,
   The method according to claim 1 or 2, wherein ρ ⁰ is calculated based on a hypothetical translational suppression effect at a distance 0 [nt] for each miRNA.
4. The step (3) maximizes the distribution of the translation amount of the marker gene in each target cell;
   The method according to any one of claims 1 to 3, comprising a step of selecting mRNA containing two or more miRNA response elements.
5. Designing mRNA by the method according to any one of claims 1 to 4,
   A method for producing mRNA containing two or more miRNA response elements, comprising a step of synthesizing the designed mRNA by a genetic engineering technique.
6. A method for separating two or more target cells using the mRNA designed by the method according to any one of claims 1 to 4 and using the translation amount of the marker gene as an index.
7. The method according to claim 6, wherein the mRNAs are 4 types of mRNAs designed by the method according to any one of claims 1 to 4 and having different marker gene sequences and 5 'UTR sequences.

Images