application single-cell sequencing technology is more and more common to research scientific problems. The most popular application is the Chromium solution of 10x Genomics (hereinafter referred to as 10X). But based on 2017 Christoph Ziegenhain et al [1] compared six single-cell transcriptome technologies and the team of the 2019 Broad Institute compared seven single-cell RNA sequencing methods [2], some In the single-cell transcriptome research of special or small cell samples, Smart-seq2 technology is still a research tool. What are the differences between Smart-seq2 and 10x technology and the applied scientific research scenarios? Zhang Zemin's team from Peking University gave the answer in a preprinted article "Direct Comparative Analysis of 10x Genomics Chromium and Smart-seq2" published in 2019. So today the editor once again clarified the technical principles and applications of Smart-seq 2 and 10x Genomics Chromium.
Smart-Seq Introduction
Smart-Seq (Switching mechanism at 5'end of the RNA transcript) was published in 2012[3], and in 2013, it published its improved technology application Smart-Seq2[4], In 2014, the Smart-Seq2 protocol was published [5]. Smart-Seq2 has made a number of improvements and optimizations to the original Smart-Seq experimental process. It no longer requires a purification step, which can greatly increase the yield. The most important improvements are the following two:
- TSO 3'the last guanosine The acid is replaced with locked nucleic acid (LNA).The thermal stability of LNA monomer is enhanced, and its annealing temperature enhances the 3'extension ability of non-template cDNA
- betaine (a methyl donor with two important roles: it increases the thermal stability of proteins, and By destroying the DNA helix to reduce or even eliminate the dependence of DNA thermal fusion on the base pair composition) combined with a higher MgCl2 concentration. Solve the problem that some RNAs form secondary structures (such as hairpins or loops) due to steric hindrance, which may lead to the extension of the enzyme termination chain
Smart technology is based on high-fidelity reverse transcriptase, template conversion and preamplification To increase the cDNA yield, the experimental procedure was 2 days, and the full-length transcript was obtained. This method has good coverage, can detect rare transcripts, and does not require additional professional equipment, so it has a wide range of applications.
Smart-Seq2 library construction principle
- Single cell sorting: Smart-seq2 uses flow cytometer or micromanipulation for cell sorting, the volume does not exceed 0.5 ul.
- Cell lysis: Transfer the separated cells directly to the cell lysis buffer for cell lysis.
- Reverse transcription (one-strand synthesis): Use Oligo(dT) primer to reverse transcription of RNA (main mRNA) with polyA tail. Since a special active reverse transcriptase (Moloney Murine Leukemia Virus) is used for reverse transcription, three Cs are added to the 3'end of the cDNA strand.
- template replacement (two-strand synthesis): This step uses TSO (template-switching oligo) primers to synthesize the two-strand cDNA, thereby replacing the RNA complementary to the one-strand cDNA. It should be noted that there are three Gs at the 3'end of TSO that can be complementary to the three Cs at the 3'end of a strand, and the +G at the end is a modified G, which can increase the thermal stability of TSO.And its ability to complement the free 3'end of a strand of cDNA.
- PCR amplification: In this step, light cDNA enrichment is performed, and the cDNA can be amplified to ng level.
- labeling: The modified highly active Tn5 transposase is used to break the DNA while adding linkers to both ends of the cDNA. The DNA fragment after labeling is usually 200-600bp.
- PCR enrichment and computer-based sequencing: After the last PCR amplification, the computer-based sequencing can be performed.
10x Genomics single-cell technology
ChromiumTM Single-cell technology
ChromiumTM Single-span 1span Cell _span 2span _pannomics 10span 3', and is based on one-time separation of Xpan 1span 5000 A single cell, and a technology that can detect at the single cell level. This method is based on Microfluidics-based approaches and has similar molecular biology principles to Smart. It uses template switching technology, but is different from Smart’s cell capture and flux. The droplet-based method is to wrap a single cell in a small oil drop (containing barcode and RT primer) to reverse transcription into cDNA, and then the oil drop breaks to release the cDNA. The library construction is performed uniformly, which increases the experimental throughput, but requires specialization. Of experimental equipment.
10x Genomics library building principle
- Prepared cell suspension, 10x barcode gel magnetic beads and oil droplets are added to different chambers of Chromium Chip B.The GEM is formed via a microfluidic "double cross" cross system. In order to obtain a single-cell reaction system, the cell suspension concentration is recommended to be controlled at 700-1200 cells/ul, so 90-99% of GEM produced does not contain cells, and most of the remaining GEM contains one cell.
- A single GEM is formed sequentially and then all are mixed, the cells are lysed, and the gel beads automatically dissolve to release a large number of primer sequences.
- The released primer contains a 30nt poly dT reverse transcription primer, and the RNA with polyA is reverse transcribed into a cDNA strand with 10x Barcode and UMI information, and then completed by SMART. Chain synthesis.
- The oil droplets were broken, the first strand of cDNA was purified by magnetic beads, and then the cDNA was amplified by PCR.
- After the cDNA amplification is completed, the cDNA is digested and fragmented, and the most suitable fragments are screened by magnetic beads. The read2 sequencing primers are connected by end repair, A and adapters, and then PCR is used to construct P5 and P7 adapters cDNA library is enough.
Smart-Seq2 Advantages and limitations
Advantages _ul ART-span _ul ART-span _ul ART-span _ul ART-span ul16 ,The cost is cheap ~12%, which provides the possibility to analyze a large number of cells.
restriction
li17strong li17strong- polyadenylation of RNA is selective, so non-poly(A) RNA cannot be analyzed
- sequencing reads do not have mRNA strand specificity
10x Genomics advantages and limitations _strong12 strong Disadvantages Data differences Zhang Zemin’s team directly compared scRNA-seq data from the same CD45 cell sample on the two platforms, and extensively and systematically analyzed and evaluated the characteristics of each data. Smart-seq2 detects more genes and more complex data sets in cells, especially lower abundance transcripts and alternatively spliced transcripts, but captures a higher proportion of mitochondrial genes. The combination of Smart-seq2 data is also more similar to bulk RNA-seq data.
Although all based on the poly(A) enrichment strategy, about 10-30% of all transcripts detected by the two platforms are derived from non-coding genes, and the proportion of lncRNA in 10x is even greater. high.
Smart-seq2 has higher sensitivity, the number of detected expressed genes is much higher than the number of detected expressed genes of 10x.
For the detected genes, Smart-seq2 data showed a unimodal distribution, and there were few low-expressed genes detected in all cells. In contrast, the 10x data shows a clear bimodal distribution due to a large number of genes close to zero expression, indicating that the mRNA in the 10x data has higher noise or random capture at very low expression levels.
10x-based data shows a more serious dropout problem (dropout is a major feature of scRNA-Seq data, that is, many genes cannot be detected at all in some cells, but in other cells. It was detected as high expression. It is generally believed that this dropout is because some genes were not successfully reverse transcribed during the library construction process.), especially for genes with low expression levels.
10x technology sequencing data has a strong bias at the 3'end of mRNA due to technical characteristics, while Smart-seq2 technology has a more uniform distribution on the gene. At the same time, 10x technical data is not applicable at the level of splicing analysis.
However, because 10x data can cover a large number of cells, it can better detect rare cell types. In addition, the different sets of differentially expressed genes between cell clusters detected by each platform indicate that these technologies are complementary.
Smart-seq2 technical case:
Björklund used Smart-seq2 single-cell transcriptome sequencing in 2016 to reveal the heterogeneity of human CD127+ innate lymphocytes [6]. Innate lymphocytes (ILCs) are a type of immune cells , which regulate the early immune response to viruses, bacteria and parasites by releasing soluble factors. These cells can be divided into three subtypes according to the transcription factors and cytokines on the cell surface, and each subtype has different functions. ILCs (Lin−CD127+) and NK cells (CD45+Lin−CD127−NKG2A+CD56+CD16−) in tonsils of patients with obstructive sleep apnea syndrome were sorted by flow cytometry.Use Smart-seq2 for single-cell transcriptome amplification to build a library; Illumina HiSeq 2000 3M reads/sample for sequencing analysis. PCA (principal component analysis) and t-SNE were used to classify the 847 expressed genes in ILC, and the cells were divided into 4 types: ILC1, ILC2, ILC3, and NK cells. The SCDE software package was used to analyze the differential genes of the four types of cells to find the common and unique differential genes. The results found that the expression of the common differential genes in CD127+ILCs was significantly higher than that in NK cells. ILC1, ILC2, and ILC3 differential gene expression analysis showed that there are 79 up-regulated genes in ILC1 cells, which are involved in the regulation of interferon gamma, and 58 up-regulated genes in ILC2 cells are involved in prostaglandin and Notch signaling pathways and environmental induction To play a role, ILC3 cells have a total of 371 up-regulated genes. According to the GO annotation, there are 85 immune-related genes with unknown functions.
PCA and t-SNE were used to analyze 1,958 annotated immune genes in ILC3, and ILC3 was divided into 3 subtypes: Cluster A, Cluster B, and Cluster C. Through the analysis of these three subtypes of cells Analysis revealed a new immune cell CD62L+ILC3. In this study, the single-cell transcriptome analysis of 648 single cells, using t-SNE cell typing and SCDE gene differential expression analysis, found a new immune cell CD62L+ILC3 in ILC3.
10x Genomics technical case:
The Belgian research team created the first complete lung cancer cell map in history by studying thousands of healthy and cancerous lung cells. The results of the study show that lung cancer is much more complicated than the previous understanding. It contains 52 different types of stromal cells. This new information can be used to develop new lung cancer treatment avenues [7].
Smart-seq2 and 10x Genomics joint technology case:
October 31, 2019, Peking University Biomedical Frontier Innovation Center (BIOPIC), School of Life Sciences, Beijing Future Genetic Diagnosis Professor Zhang Zemin and Associate Professor Ren Xianwen from the Innovation Center (ICG), together with Professor Peng Jirun from Beijing Shijitan Hospital Affiliated to Capital Medical University, and Dr. Liu Kang from Boehringer Ingelheim, published a titled Landscape and Dynamics of Single Immune Cells in Hepatocellular Carcinoma on Cell Research paper [8]. The study combines two single-cell RNA sequencing technologies, 10x Genomics and SMART-seq2, to systematically characterize the immune cells of multiple tissues of liver cancer patients, analyze the characteristics of dynamic migration and state transition of immune cells, and explore their The potential value of liver cancer treatment.
Overall design of the project
Main findings
- There are big differences in the immune composition of different tissues. Macrophages in tumors are the main source of myeloid cells
- Macrophages in tumors show two different states (TAM-like and MDSC-like) passes and in other data 3. LAMP3+ DCs in tumors are mature DCs with hepatic lymph nodes Migration and potential ability to interact with multiple lymphocytes
References
1. Ziegenhain, C., et al., Comparative Analysis of Single-Cell RNA Sequencing Methods. Mol Cell, 2017. 65 (4): p. 631-643 e4.
2. Vieth, B., et al., A systematic evaluation of single cell RNA-seq analysis pipelines. Nat Commun, 2019. 10 (1): p. 4667.
3. Ramskold, D., et al., Full-length mRNA-Seq from single-cell levels of RNA and individual circulating tumor cells. Nat Biotechnol, 2012. strong 11strong 30 (8): p. 777-82.
4. Picelli, S., et al., Smart-seq2 for sensitive full-length transcriptome profiling in single cells. Nat Methods, 2013. 10 (11): p. 1096-8.
5. Picelli, S., et al., Full-length RNA-seq from single cells using Smart-seq2. Nat Protoc, 2014. 9 (1) : p. 171-81.
6. Bjorklund, AK, et al., The heterogeneity of human CD127(+) innate lymphoid cells revealed by single-cell RNA sequencing. Nat Immunol, 2016. 17 (4 ): p. 451-60.
7. Lambrechts, D., et al., Phenotype molding of stromal cells in the lung tumor microenvironment. Nat Med, 2018. 24 (8): p. 1277- 1289.
8. Zhang, Q., et al., Landscape and Dynamics of Single Immune Cells in Hepatocellular Carcinoma. Cell, 2019. 179 (4 ): p. 829-845 e20.
.
