There are about 37.2 trillion cells in an adult's body, and for these basic units of life, human beings have not yet mastered the genetic "codebook" and have a full "map" from the cells. "The total number of cells is a huge number, but the classification of human cells, as well as the study of the respective functions of cells, is very important for the treatment of human diseases."
Recently, Liu Longqi of Shenzhen Huada Life Science Research Institute said in an exclusive interview with the surging news (www.thepaper.cn) reporter that at this stage, there are still many challenges in the comprehensive analysis of human cell maps, and they have focused on crab-eating macaca fascicularis (Macaca fascicularis) with a genetic similarity of up to 93% in recent years. For the first time, 35 research teams from six countries demonstrated a large-scale cellular transcriptome atlas of this non-human primate (NHP).
The study was published online on the evening of April 13 in the top academic journal Nature, titled "Cell transcriptomic atlas of the non-human primate Macaca fascicularis." The research was jointly completed by research teams from 6 countries, including Shenzhen Huada Life Science Research Institute, Beijing Huada Life Science Research Institute, Shenzhen National Gene Bank, Jilin University, Guangzhou Institute of Biomedicine and Health, Chinese Academy of Sciences, Karolinska Institute of Medicine and Health in Sweden, University of Cambridge in the United Kingdom, ICREA Research Institute in Spain, and ASTAR in Singapore.
"This map is like a 'map', and with it it is equivalent to having a high-precision instrument that explores the resolution of living cells, can 'see' which cells are in each organ, and can also be fine-grained to the specific molecular characteristics of each cell and the interaction with other cells." Dr. Han Lei, the first author of the paper and a graduate student of The Life Sciences Institute of Huada University in Shenzhen, said, "This lays a foundation for us to better understand the basic structure of life, explore the relationship between diseases and cells, and provide a new direction for the precise treatment of diseases." ”
Academician Pu Muming, director of the Center for Excellence in Brain Science and Intelligent Technology of the Chinese Academy of Sciences, also analyzed the significance of this research in biology and application. He believes that as an important model animal for clinical drug research and development, in the next five to ten years, various disease models of cynobial monkeys will continue to appear. "The data obtained today is the best foundation, the best database for drug development, and a very important milestone."
It is worth noting that the research team also set up a non-human primate million single cell resource website NHPCA (https://db.cngb.org/nhpca/). The database is open and interactive, Liu Longqi told the surging news reporter, "can achieve user self-exploration, but also can achieve user data and our data integration." ”
He stressed that the above concept is very important and is also a major trend in the field of cell mapping in the future. "In the future, when the cell map volume is getting larger and larger, how to integrate data from different platforms and different laboratories is very important. How to build a better database is also a very important science in itself. ”
Dr. Longqi Liu, Shenzhen Huada Life Science Research Institute.
Human Cell Atlas Challenge
How do we peer into the mysteries of cells that carry genetic material? By placing the cell under a microscope, its morphology and structure can be observed, but science has evolved so far that scientists want to resolve the nuances between cells at the molecular level.
The advent of single-cell sequencing technology has opened up an unprecedented opportunity to "read" cells in depth. In 2009, Tang Fuhui (currently a researcher at the Beijing Future Genetic Diagnostic High-precision Innovation Center and a professor at the BIOPIC Center of the School of Life Sciences of Peking University) published the world's first single-cell mRNA sequencing article during his postdoctoral period, achieving a leap from 0 to 1. This important attempt ushered in the era of single-cell transcriptome sequencing.
In recent years, the development of high-throughput single-cell sequencing technology has also spawned the "Human Cell Atlas". In October 2017, Aviv Regev, a computational biologist at the MIT-Harvard Broad Institute in the United States, and Sarah A. Teichmann, head of cytogenetics at the Sanger Institute of the Wellcome Trust in the United Kingdom, published an article in the paper preprint platform bioRxiv, the title of which is "The Human Cell Atlas", which systematically discussed the significance, goals, and objectives of HCA. Tasks and implementation paths. Nearly a year before the article was published, Regev and others had already held a meeting on the plan in London, England.
Guo Guoji, a professor at the Center for Stem Cell and Regenerative Medicine of Zhejiang University School of Medicine, said in an interview with the surging news (www.thepaper.cn) reporter that as an international big science program that has emerged in the field of life sciences in recent years, "the human cell mapping program aims to describe the detailed characteristics of each cell in the human body, present the 3D structure of different types of cells in human tissue, outline the interconnection of all human systems, and reveal the relationship between map changes and health and disease." The Human Cell Atlas program will radically improve people's understanding, diagnosis and treatment of disease. ”
Previously, in March 2020, Guo Guoji et al. published a research result in Nature, in which they performed Microwell-seq high-throughput single-cell sequencing analysis on 60 human tissue samples and 7 cell culture samples, systematically mapping human cells spanning embryonic and adult stages and covering eight major systems. Guo Guoji said at that time that this research work has constructed the basic framework of human cell atlas, which is of guiding significance for the further improvement of the atlas in the future.
However, as of now, the construction of the human cell map is still challenging. Liu Longqi told the surging news reporter, "The biggest challenge lies in the flux and cost of the current tool, which is also the most direct difficulty." "If you calculate according to commercial tools, the cost of measuring a cell is a few dollars," the cost of sequencing 37 trillion cells is very high and impossible to support." ”
The cost constraint also lies in the fact that compared to the Human Genome Project (HGP), which was officially launched in 1990, the Human Cell Atlas Project does not currently have a fixed funding support, "The Human Genome Project is relatively fixed, and the Human Cell Atlas Project is now funded by various scientists, which brings great challenges." ”
The second challenge is that the project objectives are not particularly clear. "The organization is relatively free, everyone has their own research plan, and the division of labor of different people may be difficult to be very clear, and at present, everyone is doing it according to their own interests." Liu Longqi believes that overall, the human cell atlas program has not yet achieved clear goals and task division.
The third challenge lies in the "barriers" between laboratories. "At this stage, the tools used by different laboratories may vary, and the end result is that it brings different challenges to data sharing and analysis mining between people." This also stems from the fact that the tools in the field as a whole are not yet very mature.
Autonomous and controllable tools come first
The latest study, led by Liu Longqi's Huada Institute of Life Sciences in Shenzhen, first set its sights on crab-eating macaques, which are genetically similar to humans by up to 93%.
The research team writes in the paper that non-human primates (NHPs) represent the closest human choice given evolutionary proximity. Constructing a non-human primate cell atlas (NHPCA) will produce a catalogue of features that can be used to study human physiology, disease, and aging. It will also provide insight into the evolutionary mechanisms of different bodily functions between humans and non-human primates.
Of course, non-human primates also include a large and diverse group of species that vary greatly in ecology, diet, movement, and behavior. In view of the more frequent reproductive cycle and wider distribution, especially the crab-eating macaque, it is often used in research for scientific research purposes around the world.
Liu Longqi also said, "From the perspective of genomic sequence, the difference between macaques and humans is much smaller than that of model animals such as mice, which is the first reason for choosing monkeys; second, if you want to really understand human diseases, such as Parkinson's, Alzheimer's disease, etc., you can't do research directly on humans, we must build animal disease models, non-human primates especially in cognitive and nervous system disease research has significant advantages." ”
Liu Longqi said that at present, there is less such data, and macaques are still a blank, "so the release of these data is very important for a series of studies in the future, including understanding the disease mechanism and drug screening." ”
Single-cell library building system DNBelab C4.
All experiments in this study used the single-cell library building system DNBelab C4, as well as the DNBSEQ family of sequencing platforms. Liu Longqi believes that in the case of too high a cost to support, "tool breakthrough is the first problem that must be solved." ”
DNBelab C4 was independently developed by BGI Intelligent Manufacturing and officially released to the public in October 2019. In addition, BGI has its own DNBSEQ series sequencing platform, "We have a very significant advantage in building a cell map, from upstream library construction to downstream sequencing, we have full control over the tools." ”
DNBSEQ sequencing platform.
And how to isolate from tissues and organs to cell profile production? Liu Longqi introduced that for samples analyzed with single-cell nuclear RNA sequencing (snRNA-seq), they are first stored in liquid nitrogen (-190 °C) after collection until nuclei extraction is carried out. The purer nucleus was obtained by density gradient centrifugation, and after the single cell (nucleus) suspension was prepared, the nucleus was wrapped into droplets by DNBelab C4, the cell lysis and bead capture mRNA were completed in the droplets, and after the microbeads were effectively recovered, the reverse transcription and cDNA synthesis were completed in a single tube, and the library was prepared according to the high-throughput sequencing process and sequencing and letter analysis were carried out.
"In fact, the single-cell tool does one thing, to each single cell nucleus with a DNA tag, just like a two-dimensional code, after the labeled cells are put together, we can carry out the next step of library building and sequencing." After sequencing, we differentiate each cell according to the label, and through this process we can generate a cell map. Liu Longqi further explained graphically.
One of the notable details of the study was that a small number of tissue samples used the usual single-cell RNA sequencing (scRNA-seq) technique, while most tissues used single-cell nuclear RNA sequencing (snRNA-seq) for profiling.
Liu Longqi explained, "First of all, although most of the mRNA is in the cytoplasm, the mRNA is transcribed from the nucleus, and there are still many mRNAs in the nucleus, so this is theoretically feasible." ”
Another key reason is that single-cell nuclear RNA sequencing is "very friendly" to sample processing. Liu Longqi mentioned that the study involved 45 organs and tissues, most of the organ sample dissociation strategies are different, for each cell needs to be optimized methods, and when the researchers analyze the whole cell, there are differences in the size of cells in an organ, "such as the liver, the liver parenchymal cells are larger, and the immune cells are very small, so in the process of tissue dissociation, it is easy to lose some cells according to the traditional method." In addition, single-cell tools based on droplet wrapping technology are more sensitive to cell size, "large cells are difficult to wrap, small cells are easier to wrap, these will bring very obvious preference." ”
Overall, the nucleus size is almost fixed, around 1 micron, which eliminates the preference for data. In addition, regardless of the number of organs, the research team can use a unified method to isolate and extract, thus eliminating the impact of sample processing. "We've minimized these impacts." Of course, for tissues such as the pancreas, it is difficult to isolate high-quality cells, and a more stable nucleus is the only choice.
For free cells like bone marrow, the team still employed single-cell RNA sequencing. "The cells in the bone marrow are very small overall, and the nucleus is relatively more difficult than other tissues, but it is very easy to isolate the cells."
According to Liu Longqi, for single-cell sequencing and single-cell nucleus sequencing, the research team conducted a systematic analysis, "Some cells we have done both cells and nuclei, from a global perspective, their consistency is about 90%-95%." But when you look at individual genes, there may be differences in the expression of certain genes between the cell and the nucleus. ”
The research team believes that the results of this study also confirm the utility of single-cell nuclear RNA sequencing for the generation of large-scale cell maps.
The world's first whole-cell atlas of non-human primates
Based on BGI's self-developed single-cell library and sequencing platform, the research team finally performed single-cell sequencing analysis on about 1.14 million cells in 45 organs of adult macaque monkeys. It is divided into 113 major cell types and 463 cell subclasses.
It is also the world's first non-human primate whole cell atlas. "In fact, it can continue to be subdivided, depending on how large the particle size the researchers hope to get, theoretically it can be divided into thousands of species, as long as the function of the cell can be understood." Liu Longqi believes that the fine classification of cells is the basis for many subsequent research work.
Crab-eating macaque cell atlas.
What information does this atlas of macaques provide us at the moment? The research team selected some important research topics for preliminary exploration. Among them, based on this atlas, they constructed a virus database containing 126 virus-susceptible cell types such as COVID-19, hepatitis B, and rabies viruses. It's like a "virus dictionary" that allows you to quickly query the cell type most likely to infect the virus and see the organs that are likely to be distributed by that cell type.
Of particular concern is that, given that the COVID-19 pandemic is still spreading globally, this study focused on the receptors ACE2 (angiotensin-converting enzyme 2) and TMPRSS2 (transmembrane serine protease 2) for the new coronavirus (SARS-CoV-2). Multiple previous studies have demonstrated that the entry of SARS-CoV-2 into host cells is initiated by binding spike protein (S protein) to ACE2 on the cell surface, and TMPRSS2 lysing the S protein. ACE2 is likened to a "door lock" for virus intrusion, while TMPRSS2 is considered a "lubricant".
The study evaluated their expression in macaque tissues. Their analysis found that TMPRSS2 was widely expressed in multiple tissues of macaque monkeys, and the expression of ACE2 was relatively limited. ACE2 is most expressed in the gallbladder (mucus cells, endothelial cells, glandular cells, and smooth muscle cells), testicular support cells, renal epithelial cells (mainly proximal tubular cells), lungs (ciliary cells, rod cells, especially alveolar type 2 cells), and liver (liver cells, especially bile duct cells).
Previous studies have suggested that ACE2+TMPRSS2+ double positive cells have a higher risk of SARS-CoV-2 infection. The research team noted that these cells were the most numerous among macaque gallbladder cells, which is also consistent with reports of acute cholecystitis in patients with COVID-19. Considerable co-expression was also observed in lung and kidney cells, and less in other cell types such as bladder epithelial cells, pancreatic ducts, and islet cells.
Next, the research team also conducted a comparative analysis of the expression of ACE2 and TMPRSS2 in monkeys and humans. The results showed that in these two different species, the liver pattern was similar, while the patterns of the gallbladder, kidneys and lungs were even more different.
They analyzed this during the discussion session, and the study showed that a comparison between macaques and humans found that ACE2 and TMPRSS2 were not expressed differently in different cell types. "This could influence the pathogenesis study of SARS-CoV-2 and may explain why drugs such as hydroxychloroquine, despite providing promising results in monkey cell lines in vitro, are ineffective in humans."
Of course, such examples also remind that there are differences in target cell susceptibility between humans and macaques, and although macaques are better mode animals than mice in neurological diseases, etc., they still need to be cautious about the results of the experiment.
The study also mentioned that in addition to ACE2 and TMPRSS2, many other molecules are also involved in promoting the pathogenesis of SARS-CoV-2 to the cell surface or COVID-19. "Their expression or co-expression in macaque tissues, other associations and interactions between viruses and hosts, as well as differences between species, can all be studied through the NHPCA website."
Liu Longqi said that the new crown is just a typical example, "We have built a database of the entire virus susceptibility, you can get the corresponding receptors such as hepatitis B virus and rabies virus according to this database, which points out the direction for the clinical diagnosis and treatment of the disease." ”
It is worth mentioning that this research was led by the BGI team and was jointly completed by 35 research teams from 6 countries. "The study covered 45 organs, each of which is very complex, and it is difficult to say which independent team will be very familiar with each organ, so we work with many very top teams in the world to jointly interpret it." For example, Liu Longqi said that for the kidney organ, the team collaborated with Professor Patrick Maxwell, a well-known expert in the field of nephrology and dean of the Cambridge University Clinical Medical School, to "ensure that we fully understand the information obtained in this organ." ”
Liu Longqi believes that this way of global cooperation, especially through the way of big science plan organization to carry out joint interpretation, "this should be an important strategy for the development of cell atlas in the future." ”
The research team also created the database. "As the basic data, I believe that many people in the world will use this map for data mining, and we have now received a lot of demand in this regard, and the current data is also public." Liu Longqi said, "First of all, the volume of data is very large, and the way ordinary scientific research users contact your data may be to rely on a very clear database. ”
In addition, the database is interactive. "Whether you understand data analysis or not, you can explore through the click of the mouse, we are now building a very complete database, on the one hand, it involves the display of the entire data, on the other hand, the user can explore and choose through this data for the genes he is interested in, the phenotype of diseases, etc." Liu Longqi said.
Liu Longqi also said that in the field of single-cell sequencing, it is still looking forward to the emergence of technologies and tools with "dimensionality reduction" effects in terms of cost and performance. Better collaboration between global research teams and the realization of the human cell map may be looking forward to the next qualitative leap.