Single cell transcriptome sequencing technology based on Raman spectrum sorting

Abstract

The invention relates to the technical field of microorganism single-cell sequencing, in particular to a single-cell transcriptome sequencing method based on Raman spectrum sorting. Sorting, database building, gene sequencing and data analysis are carried out on single cells based on Raman spectrum, and it is found that the same sample has different cell morphologies, and different cell morphologies have different gene expressions; according to the single cell transcriptome sequencing method based on the Raman spectrum, the expression states of different single cells in the biological sample can be rapidly and efficiently achieved, and therefore analysis and identification of gene expression heterogeneity in one sample are achieved. According to the method, single cell sorting equipment depending on an expensive microfluidic technology is not needed, and visual single cell sorting can be carried out according to experimental requirements; the method does not need to depend on a combined label or a probe to mark the cells, and is a lossless cell sorting method; the method has the advantages of large number of detected genes, wide coverage, simple operation and low cost.

Description

Technical Field

[0001] The present invention relates to the technical field of microbial single-cell sequencing, and in particular to a single-cell transcriptome sequencing technology based on Raman spectroscopy sorting. Background Art

[0002] With the maturity and large-scale application of eukaryotic single-cell transcriptome sequencing technology, microbial single-cell transcriptome sequencing technology has also attracted more and more attention. Even in the same clonal population grown under the same laboratory conditions, gene expression in bacteria is highly heterogeneous (A.Raj, A.van Oudenaarden, Nature, Nurture, or chance: stochastic gene expression and its consequences. Cell 135, 216-226 (2008).). Bacteria can randomly differentiate into subpopulations and play different roles in the survival of the community, which is called a bet-hedging strategy (A.Eldar, MBElowitz, Functional roles for noise in genetic circuits. Nature 467, 167-173 (2010).). Single-cell transcriptome sequencing is a powerful tool for revealing single-cell heterogeneity (AM Klein et al., Droplet barcoding for single-cell transcriptomics applied to embryonic stem cells. Cell 161, 1187-1201 (2015).). However, eukaryotic single-cell transcriptome sequencing technology is not applicable to microorganisms. There are three main reasons: 1) The mRNA content of bacteria is very low, two orders of magnitude lower than that of human cells; 2) Bacterial mRNA is not adenylated, so it is difficult to separate bacterial mRNA and rRNA; 3) Compared with eukaryotic cells, bacteria have different cell walls and cell membranes, so it is difficult to lyse; 4) Finally, the small size of bacteria makes it difficult to separate microbial single cells using microfluidics.

[0003] With the development of technology, researchers have mainly developed the following microbial single-cell transcriptome sequencing methods:

[0004] (1) MATQ-seq (multiple annealing and deoxycytidine (dC) tailing-based quantitative scRNA-seq) sequencing method: This is a single-cell sequencing quantitative method based on multiple annealing and C (deoxycytidine). Single microbial cells are separated by flow cytometry and the cells are lysed; the RNA is then reverse transcribed to obtain cDNA; the obtained cDNA is subjected to C reaction so that the reads have polyC tails; the cDNA is then subjected to PCR amplification reaction; the PCR amplification product is connected to the sequencing tag; because rRNA reads account for a high proportion, this method adds a cas9 protein-mediated rRNA degradation method (depletion of abundant sequences by hybridization, DASH)) to remove rRNA reads. Finally, the remaining reads are PCR amplified and sequenced (C.Homberger, RJHayward, L.Barquist, J.Vogel, Improved Bacterial Single-CellRNA-Seq through Automated MATQ-Seq and Cas9-Based Removal of rRNA Reads.mBio14, e0355722(2023).). Therefore, this sequencing method requires the customization of expensive microfluidic bacterial single-cell sorting equipment to overcome the sorting difficulties caused by the small size of bacteria, and cannot guarantee that each droplet has only one microbial cell.

[0005] (2) SPLiT-seq (split-pool ligation transcriptomics sequencing) sequencing method: This sequencing method is a single-cell quantitative sequencing method based on combinatorial barcodes. Currently, two main sequencing methods have been derived: microSPLiT (microbial SPLiT-seq) and PETRI-seq (prokaryotic expression profiling by tagging RNA in situ and sequencing). The main steps of the microSPLiT sequencing method are: using soil mild lysozyme to penetrate and digest the formaldehyde-fixed bacteria; adding A (deoxyadenosine) to the bacteria; randomly dividing the cells into 96-well plates; reverse transcription reaction and two rounds of ligation reaction of RNA to obtain cDNA fragments with well-specific tags and sequencing tags (A. Kuchina et al., Microbial single-cell RNA sequencing by split-pool barcoding. Science 371, (2021).). The overall steps of PETRI-seq sequencing method and microSPLiT are similar. PETRI-seq does not require the addition of A reaction to the cells (SB Blattman, W. Jiang, P. Oikonomou, S. Tavazoie, Prokaryotic single-cell RNA sequencing by in situ combinatorial indexing. Nat Microbiol 5, 1192-1201 (2020).). These two methods do not rely on equipment for cell sorting, microfluidics or microplates to separate individual cells, but only require combined tags, so the cost of these two sequencing methods is relatively low. However, this sequencing method has many operating procedures, including enzyme-dependent multiple rounds of combined tag linking reactions, resulting in low cell retention rate. In addition, the 3' end bias of this sequencing method is relatively strong, and it cannot cover the full length of the gene, and the sequencing coverage is relatively poor.

[0006] (3) par-seqFISH (parallel sequential fluorescence in situ hybridization) sequencing method: This is a transcriptome sequencing method based on probes and microscope imaging. This method mainly achieves the identification of different bacteria and the same bacteria under different culture conditions by designing 16S RNA probes; by designing primary probes that specifically bind to gene sequences to achieve gene identification; by designing probe 2 that binds to primary probes to perform color development reaction; by observing the color development under a microscope, gene expression quantification can be further achieved. After measuring the expression of a group of genes, the color development of the previous round of probes is turned off and the next round of probes is performed, thereby obtaining the expression level of the next group of genes (D. Dar, N. Dar, L. Cai, DK Newman, Spatial transcriptomics of planktonic and sessile bacterial populations at single-cell resolution. Science 373, (2021).). Since this method relies on probes to identify genes, it is necessary to design many probe sequences for bacterial genes. Considering factors such as the specificity of the probes themselves, this is a very complex and costly process. According to reports, 1763 mRNA probes target 105 mRNAs, which shows that the number of genes detected by this method is relatively small. In addition, this method also relies on microfluidic devices and automatic photography to perform multiple rounds of fluorescent staining, photography, and elution, which is complicated and time-consuming.

[0007] Raman spectroscopy is a scattering spectrum, which is the phenomenon that the frequency of incident light changes after being scattered when the molecular bonds in the compound are excited to the virtual energy state but have not yet recovered to the original state. Each single-cell Raman spectrum consists of more than 1,500 Raman peaks corresponding to a type of chemical bond. The peaks can be used as a "molecular fingerprint" unique to a single cell, and then reflect the multi-dimensional information of the composition and content of chemical substances in a specific cell. The collection of single-cell Raman spectra of a cell population under a specific spatiotemporal state is called a "Raman group". The Raman group can quickly and inexpensively measure and monitor the "metabolome" with single-cell accuracy, and its changes can reflect and characterize the panoramic and almost infinite "state" and "function" of the cell system. In addition, water molecules do not have strong signal interference in the key fingerprint area, which enables it to have the ability to detect in vivo. Therefore, Raman group technology has almost all the characteristics required for ideal single-cell phenotyping or sorting technology, and is an ideal tool for single-cell phenotype group (phenome) identification. The spontaneous Raman signals of most cell phenotypes in nature are weak. In order to stably improve the sensitivity of Raman detection, a surface enhanced Raman probe technology based on a labeling strategy has been developed. By detecting the strong Raman signals of SERS probes, targeted detection of living tumors, detection of polysaccharides in living cells, specific detection of glycosylated proteins, telomere length assessment, micro RNA detection, dynamic tracking of cell membrane repair, and O-GIc NAcvlation imaging research in single living cells have been achieved.

[0008] Raman spectroscopy is a non-destructive detection technology that can quickly identify intracellular molecular components. As a label-free, non-invasive, real-time, and rapid phenotypic detection method, Raman spectroscopy can provide a large amount of information such as molecular conformation, electron distribution, and the interaction between molecular forces. The "single-cell Raman map" composed of the Raman signals of all compounds in the cell can be used as its "chemical fingerprint" to distinguish and identify cell species. It has the advantages of being fast, sensitive, and non-destructive, opening up a new field of microbial identification. Therefore, when there are differences in gene expression of the same bacteria, there are also differences in its Raman spectra. Researchers can further infer differences in gene expression through differences in Raman spectra. In addition, the ejection sorting of microbial single cells based on Raman spectroscopy can well overcome the defect of microfluidics technology that microbial single cells are too small to be sorted, and better realize the precise sorting of microbial single cells.

[0009] The basic principle of Raman ejection sorting (RACE) is to sputter a thin film (usually an aluminum film) on a Raman measurement substrate, spot a cell suspension of appropriate concentration on the substrate and air-dry at room temperature; after identifying the target cells by Raman spectroscopy, a high-frequency pulsed laser can be applied to eject the target cells into the collection hole. This method can flexibly realize the collection of single-hole single cells or multiple cells. Based on this platform, Wang et al. realized the screening and identification of human intestinal flora microorganisms based on metabolic activity (Wang Yi; Raman-activated sorting of antibiotic-resistant bacteria in human gut microbiota [J]. Environmental Microbiology, 2020, 22 (7): 2613-2614). This method is relatively simple to operate and has a high sorting accuracy, but to achieve cell ejection separation, it is necessary to flip the substrate after Raman detection. This process increases the complexity of system operation, and operation in an open environment is prone to contamination risks. Jing et al. used transparent ITO quartz as the substrate, which eliminated the need to flip the substrate after detection. They also developed an all-in-one receiving chip to overcome the contamination problem and ultimately achieved the separation and identification of marine carbon-fixing bacteria. However, the biggest drawback of this method is that dry-film conditional Raman detection and laser ejection sorting will cause certain damage to the cells, and the single-cell sequencing coverage generally does not exceed 20% (Song Yizhi et al; Single-cell genomics based on Raman sorting reveals novel carotenoid-containing bacteria in the Red Sea[J].Microbiology, 2017, 10(1):125-137). In addition, to reduce damage, Liang et al. added an agarose film to the ejection substrate to achieve the water phase ejection effect. The survival rate of eukaryotic cells (Saccharomyces cerevisiae) after ejection was 63%, Gram-negative bacteria (Escherichia coli) was 22%, and Gram-positive bacteria (Lactobacillus rhamnosus) was 74% (LIANG P, LIU B, et al. Isolation and culture of single microbial cells by laser ejection sorting technology [J]. Applied and Environmental Microbiology, 2022, 88 (3): e0116521).To further improve the coverage of single-cell sequencing, Su et al. optimized the multiple displacement amplification (MDA) process, thereby increasing the gene coverage from <20% to >50% (SU XL, et al. Rational optimization of Raman-activated cell ejection and secquencing for bacteria [J], Analytical Chemistry, 2020, 92 (12): 8081-8089). Currently, in order to obtain the target single cell, the technology of Raman ejection sorting of single cells still needs to be improved.

[0010] In view of the shortcomings of the current single-cell precision sorting and single-cell transcriptome sequencing methods, such as complex equipment, high cost, cumbersome operation process, low cell retention rate, poor sequencing coverage and a small number of genes detected, this application has developed a bacterial single-cell transcription sequencing method based on Raman spectroscopy sorting. This method does not require a complicated operating procedure. By combining Raman spectroscopy and optical microscopy, the required microbial single cells can be quickly and accurately identified, and the cells can be separated for subsequent library construction and sequencing. In addition, the process of library construction and sequencing does not require the addition of labels, and there is no difference from the library construction process of ordinary transcriptome sequencing. Sequencing data analysis does not need to rely on complex algorithms and personalized software. Therefore, the bacterial single-cell transcription sequencing method based on Raman sorting developed by this application has simple experimental operation and analysis methods, good sequencing coverage, and a relatively large number of genes tested, achieving "what you see is what you get" precise sorting and sequencing. Summary of the invention

[0011] The present invention sorts single cells, builds libraries, performs gene sequencing and data analysis based on Raman spectroscopy, and finds that different cell morphologies exist in the same sample, and different cell morphologies have different gene expressions, that is, the single-cell transcriptome sequencing method based on Raman spectroscopy in the present application can quickly and efficiently realize the analysis and identification of the expression states and gene expression heterogeneity of different single cells in biological samples. On this basis, the present invention is completed.

[0012] In a first aspect, the present invention provides a single-cell transcriptome sequencing method based on Raman spectroscopy sorting, the method comprising the following steps:

[0013] S01. Obtain the sample to be tested, perform pretreatment to remove excess cells, residues and other impurities, and obtain a spare sample;

[0014] S02. Place the pre-treated spare sample in step S01 on the Raman chip, air-dry, collect and sort the Raman spectrum, locate the laser point to the cells to be tested under the Raman microscope, eject the cells into the receiver for standby, and complete the sorting;

[0015] S03. After transferring the single cells in the receiver in the above step S02, the cells are lysed and digested, and sequenced by reverse transcription, amplification, library construction, and sequencing.

[0016] Furthermore, in the step S02 of collecting the Raman spectrum, the laser wavelength is 525-550 nm, preferably 532-535 nm.

[0017] Furthermore, the Raman spectrum acquisition time in step S02 is 2-8s, preferably 6-7s.

[0018] Furthermore, the laser intensity in the laser point positioning is 1-5 mW, preferably 3-4 mW.

[0019] Furthermore, the laser energy is 55-70 nj, preferably 60-65 nj.

[0020] Further, the cleavage and digestion as described in step S03 includes the following steps:

[0021] S03-1 Transfer the single cell ejected into the receiver to a container and add lysis solution, incubate and centrifuge to obtain a cell pellet;

[0022] S03-2: Add lysis buffer to the above cell pellet again for further digestion;

[0023] When the cell is a single cell of bacteria, the lysis buffer used contains 3%-10% SDS, 0.3-1M Tris-HCl buffer, 0.2%-2% β-mercaptoethanol 2-mercaptoethanol, 50-200mM EDTA, 1-10M NaCl, 50-200mM sodium pyrophosphate phosphatase inhibitor, 50-200mM KCl, 20-60μLRNase free H 2 O;

[0024] S03-3 The above cell mixture solution was repeatedly rapidly frozen and thawed, and then PBS and RNase inhibitor were added for digestion, and then inactivated and stored for later use.

[0025] Furthermore, the samples to be tested include those from sources such as food, medicine, cosmetics or the environment.

[0026] Furthermore, the Raman spectroscopy sorting is Raman spectroscopy sorting by ejection (RACE).

[0027] In a second aspect, the present invention provides a single-cell transcriptome sequencing system, the system comprising a sample preprocessing module, a sample single-cell sorting module, a single-cell lysis module, and a single-cell sequencing module, wherein:

[0028] The sample preprocessing module is to obtain the sample to be tested, remove excess cells, residues, etc. for preprocessing, and obtain a spare sample;

[0029] The sample single cell sorting module is to place the pre-treated sample on the Raman chip, air-dry it, collect and sort the Raman spectrum, locate the laser point to the cell to be tested under the Raman microscope, and eject the cell into the receiver to complete the sorting;

[0030] The single-cell transcriptome sequencing module is a method of lysing the sorted single cells in a lysis solution to obtain cell lysis products, and then performing reverse transcription, amplification and sequencing analysis under temperature-controlled conditions using primers and polymerase.

[0031] Furthermore, the laser wavelength in the Raman spectrum collection is 525-550nm, preferably 532-535nm.

[0032] Furthermore, the Raman spectrum acquisition time is 2-8 s, preferably 6-7 s.

[0033] Furthermore, the laser intensity in the laser point positioning is 1-5 mW, preferably 3-4 mW.

[0034] Furthermore, the laser energy is 55-70 nj, preferably 60-65 nj.

[0035] Furthermore, the lysis comprises the following steps:

[0036] S03-1 Transfer the single cell ejected into the receiver to a container and add lysis solution, incubate and centrifuge to obtain a cell pellet;

[0037] S03-2: Add lysis buffer to the above cell pellet again for further digestion;

[0038] When the cell is a single cell of bacteria, the lysis buffer used contains 3%-10% SDS, 0.3-1M Tris-HCl buffer, 0.2%-2% β-mercaptoethanol 2-mercaptoethanol, 50-200mM EDTA, 1-10M NaCl, 50-200mM sodium pyrophosphate phosphatase inhibitor, 50-200mM KCl, 20-60μLRNase free H 2 O;

[0039] S03-3 The above cell mixture solution was repeatedly rapidly frozen and thawed using liquid nitrogen, and then PBS and RNase inhibitor were added for digestion, and then inactivated and stored for later use.

[0040] Furthermore, the samples to be tested include those from sources such as food, medicine, cosmetics or the environment.

[0041] Furthermore, the Raman spectroscopy sorting is Raman spectroscopy sorting by ejection (RACE).

[0042] In a third aspect, the present invention provides an application of a single-cell transcriptome sequencing analysis method based on Raman spectroscopy in biological species identification, wherein the method is as described in the first aspect of the present invention.

[0043] Beneficial Effects

[0044] 1. By comparing the current microbial single-cell transcriptome sequencing methods, the present invention does not need to rely on expensive single-cell sorting equipment using microfluidics technology, and can perform visual single-cell sorting according to experimental needs. Therefore, the present invention can achieve visual and accurate sorting of microbial single cells, and the cost is relatively low.

[0045] 2. The present invention does not need to rely on combined tags or probes to mark cells, and is a non-destructive cell sorting method, thus overcoming the problem of low cell retention rate in existing sequencing methods.

[0046] 3. The present invention does not rely on multiple rounds of enzyme-based tag linking reactions and amplification reactions, so the 3' end bias of sequencing data is not obvious, the gene coverage is good, and the number of genes detected is relatively large.

[0047] 4. The present invention does not require probe design, and does not require multiple rounds of complex steps such as fluorescent staining, photography, and elution. It only needs to perform subsequent amplification and library construction according to the conventional transcriptome sequencing process after cell sorting. Therefore, the operation is simple and time-saving, and the sequencing cost is low. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 .Single-cell sample transcriptome library construction process.

[0049] Figure 2 The main cell morphologies of Bacillus licheniformis on days 0, 3, and 9 after citric acid stimulation (A) and the corresponding spectral clustering of the cell morphologies (B) and the corresponding Raman average spectra of each cell morphology (C).

[0050] Figure 3 .Single-cell transcriptome sequencing kraken2 analysis of read number and gene number statistics (A) Citrate stimulation of Bacillus licheniformis.

[0051] Figure 4 .Sequencing quality assessment.

[0052] Figure 5 .Raman spectroscopy and single-cell transcriptomics reveal bacterial expression heterogeneity.

[0053] Figure 6. Single-cell transcriptome analysis results of Bacteroides bacteria; (A) Species annotation of reads classification using kraken2 software; (B) Classification annotation of detected genes.

[0054] Figure 7 .Comparison of two library construction methods. DETAILED DESCRIPTION

[0055] The specific embodiments of the present invention are further described below. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention. In addition, the technical features involved in the embodiments described below can be combined with each other as long as they do not conflict with each other.

[0056] The experimental methods in the following examples are conventional methods unless otherwise specified, and the experimental materials used in the following examples are commercially available unless otherwise specified.

[0057] Example

[0058] Example 1

[0059] 1.1 Sample

[0060] Bacterial samples of Bacillus licheniformis were collected at different days after treatment with citric acid.

[0061] (1) Take out the strain at -80°C and add Columbia medium to activate it for 24 h;

[0062] (2) The activated bacteria were inoculated into 10% Columbia medium (control group) and 10% Columbia medium plus citric acid (1.125 mg / ml) (experimental group), and the pH of the medium was adjusted to 4.0 using citric acid;

[0063] (3) The two groups were cultured simultaneously in a 37°C constant temperature shaking incubator for 0, 3, and 9 days.

[0064] 1.2 Sample pretreatment:

[0065] Take citric acid-treated Bacillus licheniformis bacteria for different days, centrifuge at 300 rpm at room temperature for 10 min to remove excess cells, residues and other impurities; take 1 ml of the supernatant, centrifuge at 5000 rpm for 5 min; discard the supernatant, keep the bacterial pellet, add 200 μl of clean and sterile ultrapure water to resuspend, gently blow and set aside the sample.

[0066] 1.3 Tissue single cell Raman spectroscopy acquisition and sorting

[0067] (1) Take 2 μl of bacterial solution, spot it on the Raman chip, and air-dry;

[0068] (2) After the sample is air-dried, observe the sample under a Raman microscope and locate the laser spot on the cell to be tested;

[0069] (3) Raman acquisition conditions: laser wavelength is 532 nm, sample Raman spectrum acquisition laser intensity is 3 mW, Raman spectrum acquisition time is 2 s, and sample Raman sorting laser energy is 60 nJ.

[0070] 1.4 Transcriptome sequencing of sorted single cells

[0071] (1) Focus the laser on the selected microbial single cell and eject the selected cell into the receiver at 60 nJ;

[0072] (2) Prepare lysis buffer according to the following components and adjust the pH to 9. Take the total volume as 100 μl:

[0073]

[0074]

[0075] (3) After Raman sorting of single cells, lysates were performed according to the following conditions, with five samples under each condition:

[0076] Condition A: Place 1.5 μl PBS in a receiver and eject a single cell into the receiver. Then, turn the receiver upside down and place the PCR tube on top. Centrifuge the cells at 2000 rpm for 30 s into a PCR tube containing 4.5 μl PBS. Add 0.5 μl MetaPolyzyme enzyme and digest at 37°C for 1 h. Then add 0.5 μl RNase inhibitor and 4 μl lysis buffer and digest at 24°C for 10 min. Inactivate at 95°C for 3 min and store at 4°C.

[0077] Condition B: Place 1.5 μl of lysis buffer in a receiver and eject a single cell into the receiver. Then, invert the receiver into a PCR tube and centrifuge the cells at 2000 rpm for 30 seconds into a PCR tube containing 6.5 μl PBS. Add 0.5 μl of RNase inhibitor and 2.5 μl of lysis buffer and digest at 24°C for 10 minutes. Inactivate at 95°C for 3 minutes and store at 4°C.

[0078] Condition C: Place 1.5 μl of lysis buffer in a receiver and eject a single cell into the receiver. Then, turn the receiver upside down and place the cell in a PCR tube. Centrifuge the cells at 2000 rpm for 30 seconds to a tube containing 2.5 μl of lysis buffer and digest at 24°C for 10 minutes. Repeat rapid freeze-thaw for 3 times with liquid nitrogen and add the cells to a 6.5 μl PBS PCR tube. Add 0.5 μl of RNase inhibitor and digest at 24°C for 10 minutes. Inactivate at 95°C for 3 minutes and store at 4°C.

[0079] (4) Add 2 μl gDNA Wipeout Buffer, mix well and incubate at 42°C for 10 min to remove DNA;

[0080] (5) Prepare reverse transcription master mix according to the following composition, 42°C ~ 60 min, 95°C ~ 3 min, and store at 4°C;

[0081] Component Volume(μl) RT / PolymeraseBuffer 4 RandomPrimer 1 OligodTPrimer 1 QuantiscriptRTEnzymeMix 1 Total volume 7

[0082] (6) Prepare the master mix for the ligation reaction according to the following composition, 24℃~60min, 95℃~3min, and store at 4℃

[0083]

[0084]

[0085] (7) Prepare the cDNA amplification master mix according to the following composition, 30℃~2.5h, 65℃~5min, store at 4℃

[0086] Component Volume(μl) REPLI-gscReactionBuffer 29 REPLI-gSensiPhiDNAPolymerase 1 Total volume 30

[0087] (8) The amplified products are sequenced.

[0088] 1.5 Experimental Results

[0089] (1) Raman spectroscopy analysis results of Bacillus licheniformis

[0090] Using Raman microscopy, we observed the licheniformis treated with citric acid for 0 days, 3 days and 9 days. We found that the licheniformis treated with citric acid under three different conditions showed three different forms: I, II, and III ( Figure 2 A), suggesting that different treatment conditions lead to different gene expressions in Bacillus licheniformis.

[0091] Raman spectra were collected for cells of each morphology (more than 50 cells were selected for each morphology). Through dimensionality reduction and cluster analysis of the Raman spectra, it was found that although they were the same bacteria, the Raman spectra of the three morphologies of bacteria could be well separated, and each morphology had a unique characteristic spectrum ( Figure 2 B and C). The results show that Raman spectroscopy can well characterize the gene expression of bacteria.

[0092] (2) Single-cell transcriptome analysis results of Bacillus licheniformis

[0093] In order to explore the effect of experimental conditions on the construction of microbial single-cell transcriptome sequencing libraries, citric acid was used to stimulate Bacillus licheniformis for the same number of days, and transcriptome libraries and sequencing of microbial single cells were performed under different experimental conditions. The transcriptome data obtained by sequencing under each condition were analyzed using kraken software, and the reads in the sequencing data were classified according to species using kraken software. The number of reads belonging to Bacillus licheniformis and the corresponding number of genes obtained under conditions A, B, and C were counted ( Figure 3 ).

[0094] The single-cell transcriptome sequencing samples obtained under condition A showed that the number of reads of Bacillus licheniformis was 0, and the number of genes (excluding rRNA and tRNA related genes) was 0.

[0095] The single-cell transcriptome sequencing samples obtained under condition B showed that the number of reads of Bacillus licheniformis increased, and the genes belonging to the bacteria (excluding rRNA and tRNA related genes) could also be detected. However, neither the number of reads nor the number of genes could well describe the changes in the gene expression profile of the bacteria.

[0096] The method was improved under C conditions, and it was found that the number of reads and genes detected was significantly increased. Although different cells may express different genes, it was found that the number of genes detected under C conditions could reach 2189, while the total number of genes of the bacterium Bacillus licheniformis was 4474. Therefore, single-cell transcriptome sequencing under C conditions after Raman sorting can well depict changes in gene expression.

[0097] (3) Quality assessment of Bacillus licheniformis transcriptome sequencing

[0098] The quality of the sequencing data was further evaluated and it was found that the base composition in the sequencing data conformed to A=T, G=C, and the percentage of base quantity was around 25% ( Figure 4A). GC content analysis showed that the GC content in the sequencing data did not deviate from the normal range ( Figure 4 B). By analyzing the gene coverage of the sequencing data, it was found that the data obtained by this sequencing method would not show a 3' end bias (Figure C). The genes were sorted according to the expression level (0-25%, 25%-50%, 50%-75%, 75%-100%), and the gene expression was quantitatively evaluated, and it was found that most of the genes were accurately quantified (Figure D).

[0099] (4) Demonstration of sequencing results of Bacillus licheniformis

[0100] Using the established Raman sorting single-cell transcriptome sequencing method, the heterogeneity of bacterial gene expression profiles was explored. After 3 days of citric acid stimulation of Bacillus licheniformis, it can be clearly observed under a Raman microscope that the citric acid-stimulated Bacillus licheniformis presents five main morphologies ( Figure 5 A).

[0101] Four single cells were selected from each morphology for sequencing, and expression clusters of different cell types were found ( Figure 5 B), the number of genes expressed in different cell types is also inconsistent ( Figure 5 C).

[0102] Gene expression analysis showed that different cell types had different gene expressions. Cell type I mainly expressed FdhF / YdeP family oxidoreductases, manA (mannose-6-phosphate isomerase, class I), and PTS mannose / fructose / sorbose transporter subunit IIC, which help the cells respond to the low pH caused by citrate.

[0103] Cell type II is in a cell division state. This type of cell highly expresses DNA polymerase polX, DNA helicase (DEAD / DEAHbox helicase), etc., which participate in DNA synthesis and unwinding during cell division.

[0104] The gene expression of cell type III is similar to that of cell type II, but the gene expression level is relatively low.

[0105] When cells transform into cell type IV, GTP-related genes MnmE, elongation factor Tu and GTP 3′, 8-cyclase MoaA, 50S protein and GntR family transcriptional regulators are highly expressed, which regulate biological processes such as gene translation and cell motility.

[0106] When the cell type changes to V, the genes expressed are mainly transporters, Na+ / H+ antiporter NhaC, permeases and bacteriocins, which have functions such as stress resistance, material exchange and inhibition of spore germination ( Figure 5 D).

[0107] By analyzing the Raman spectra of different cell types, it was found that the average Raman spectra of different cell morphologies were not consistent, indicating that different cell types have different gene expressions ( Figure 5 E).

[0108] In addition, the experimental results showed that different types of cells express different sporulation response genes ( Figure 5 F).

[0109] Therefore, the Raman sorting-based single-cell transcriptome sequencing method established in this application can reveal the heterogeneity of cellular gene expression in complex samples.

[0110] Example 2 Results of single cell transcriptome analysis of Bacteroides bacteria

[0111] The sorting and sequencing methods established above were verified again in bacteria of the genus Bacteroides (Bacteroides xylanisolvens). Based on the same library construction and sequencing methods mentioned above, the obtained sequencing data were analyzed. Analysis of reads using kraken2 software showed that 5472 of the 10154 bacterial reads obtained by the method of the present application belonged to the genus Bacteroides, and 3017 of the 5472 Bacteroides reads belonged to Bacteroides xylanisolvens. Figure 6 A).

[0112] The sequenced genes were analyzed and it was found that 3616 genes were detected out of a total of 5307 genes of Bacteroides xylanisolans. Through the annotation of the 3616 genes, it was found that these genes included rRNA, tRNA and some coding genes ( Figure 6 B). The results show that the single-cell sequencing method established in this application can accurately measure gene expression.

[0113] Example 3 Lysis mode during the process of single-cell transcriptome library construction of microorganisms

[0114] 3.1 Test steps

[0115] The specificity of the lysis method used in the process of building a library of microbial single-cell transcriptomes of the present invention was further verified.

[0116] The lysis solution provided by the QIAGEN (#150065) kit was used for lysis of microbial single cell sorting. The experiments were carried out according to the experimental steps provided in the instructions and the improved method of this application, and the results were analyzed. After Raman sorting of single cells, lysis was carried out according to the following conditions, with five samples under each condition:

[0117] QIAGEN (#150065) method (A): Place 1.5 μl PBS in a receiver and eject a single cell into the receiver. Then, invert the receiver into a PCR tube and centrifuge the cells at 2000 rpm for 30 seconds into a PCR tube containing 5.5 μl PBS. Then, immediately add 4 μl lysis buffer and digest at 24°C for 5 minutes. Inactivate at 95°C for 3 minutes and store at 4°C.

[0118] The improved method (B) of the present application: 1.5 μl of the lysis solution in the kit is placed in a receiver, a single cell is ejected into the receiver, and then the receiver is inverted into a PCR tube, and the cells are centrifuged at 2000 rpm for 30 s to a lysis solution containing 2.5 μl of the kit, and digested at 24°C to 10 min, and then rapidly frozen and thawed three times using liquid nitrogen, added to a 6.5 μl PBS PCR tube, and then 0.5 μl of RNase inhibitor is added to digest at 24°C to 10 min, inactivated at 95°C to 3 min, and stored at 4°C.

[0119] The remaining experimental steps were the same as those provided above, and the sequencing quality was analyzed using kraken2 software.

[0120] 3.2 Test results

[0121] Kraken2 analysis results showed that the lysate provided by the QIAGEN (#150065) kit could not be used to successfully construct a library for microbial single cells according to the method provided in the kit manual, which is consistent with the suggestion in the manual that this method is not suitable for constructing a library for microbial single cell transcriptomes ( Figure 7 ).

[0122] Using the lysate provided by the QIAGEN (#150065) kit, the method of the present application can successfully detect the expression of single microbial cells. The method of the present application is universal and highly sensitive.

Claims

1. A single-cell transcriptome sequencing method based on Raman spectroscopy sorting, the method comprising the following steps: S01. Obtain the sample to be tested, perform pretreatment to remove excess cells, residues and other impurities, and obtain a spare sample; S02. Place the pre-treated spare sample in step S01 on the Raman chip, air-dry, collect and sort the Raman spectrum, locate the laser point to the cells to be tested under the Raman microscope, eject the cells into the receiver for standby, and complete the sorting; S03. After transferring the single cells in the receiver in the above step S02, the cells are lysed and digested, and sequenced by reverse transcription, amplification, library construction, and sequencing.

2. As claimed in claim 1, in step S02, the laser wavelength in the Raman spectrum acquisition is 525-550nm, preferably 532-535nm; the Raman spectrum acquisition time is 2-8s, preferably 6-7s; the laser intensity in the laser point positioning is 1-5mW, preferably 3-4mW; the laser energy is 55-70nj; preferably 60-65nj.

3. Step S03 as claimed in claim 1, the lysis and digestion comprises the following steps: S03-1: transferring the single cell ejected into the receiver to a container, adding a lysis solution, incubating, and centrifuging to obtain a cell pellet; S03-2: adding a lysis solution to the cell pellet again for further digestion; When the cell is a single bacterial cell, the lysis buffer used contains 3%-10% SDS, 0.3-1M Tris-HCl buffer, 0.2%-2% β-mercaptoethanol 2-mercaptoethanol, 50-200mM EDTA, 1-10M NaCl sodium chloride, 50-200mM sodium pyrophosphate phosphatase inhibitor, 50-200mM KCl potassium chloride, 20-60μLRNase freeH2O; S03-3 The above cell mixture solution was repeatedly rapidly frozen and thawed, and then PBS and RNase inhibitor were added for digestion, and then inactivated and stored for later use.

4. A single-cell transcriptome sequencing system, the system consisting of a sample preprocessing module, a sample single-cell sorting module, a single-cell lysis module, and a single-cell transcriptome sequencing module, wherein: The sample preprocessing module is to obtain the sample to be tested, remove excess cells, residues, etc. for preprocessing, and obtain a spare sample; The sample single cell sorting module is to place the pre-treated sample on the Raman chip, air-dry it, collect and sort the Raman spectrum, locate the laser point to the cell to be tested under the Raman microscope, and eject the cell into the receiver to complete the sorting; The single-cell transcriptome sequencing module is a method of lysing the sorted single cells in a lysis solution to obtain cell lysis products, and then performing reverse transcription, amplification and sequencing analysis under temperature-controlled conditions using primers and polymerase.

5. The lysis step as claimed in claim 5 comprises the following steps: S03-1: transferring the single cell ejected into the receiver to a container, adding a lysis solution, incubating, and centrifuging to obtain a cell pellet; S03-2: adding a lysis solution to the cell pellet again for further digestion; When the cell is a single bacterial cell, the lysis buffer used contains 3%-10% SDS, 0.3-1M Tris-HCl buffer, 0.2%-2% β-mercaptoethanol 2-mercaptoethanol, 50-200mM EDTA, 1-10M NaCl sodium chloride, 50-200mM sodium pyrophosphate phosphatase inhibitor, 50-200mM KCl potassium chloride, 20-60μLRNase freeH2O; S03-3 The above cell mixture solution was repeatedly rapidly frozen and thawed using liquid nitrogen, and then PBS and RNase inhibitor were added for digestion, and then inactivated and stored for later use.

6. The Raman spectrum acquisition as claimed in claim 5, wherein the laser wavelength is 525-550 nm, preferably 532-535 nm; the Raman spectrum acquisition time is 2-8 s, preferably 6-7 s.

7. The laser intensity in the laser point positioning as described in claim 5 is 1-5 mW; preferably 3-4 mW; the laser energy is 55-70 nj; preferably 60-65 nj.

8. An application of the single-cell transcriptome sequencing analysis method based on Raman spectroscopy as claimed in claim 1 in the identification of biological species.

9. The sample to be tested as described in any one of claims 1, 5 or 8 comprises a sample from a source such as food, medicine, cosmetics or the environment.

10. The Raman spectroscopy sorting according to any one of claims 1, 5 or 8 is Raman spectroscopy sorting by ejection (RACE).

Images