Publication types Research Support, N. Gov't Review. Substances Endonucleases. One of the first applications of multimodal omics technologies arose from the desire to connect cell surface phenotypes with gene expression profiles. Several well-characterized biological systems, particularly immune cell subtypes and hematopoiesis, have benefited from in-depth characterization of cell surface markers for a variety of functionally distinct cellular populations As a result, quantitative phenotypic information of selected cell surface markers can permit inference of cellular function.
Fluorescence-activated cell sorting FACS in combination with index sorting allows simultaneous recording of cell surface protein levels prior to deposition in lysis buffer for downstream destructive molecular assay, such as the Smart-seq2 protocol for gene expression profiling The application of such approaches has allowed the linkage of stem cell function with global molecular profile for the first time and provided numerous insights into our understanding of transcriptional heterogeneity throughout hematopoiesis 44 , — Strategies involving index sorting and downstream scRNA-seq are particularly powerful when combined with functional outcome analyses.
Wilson et al. These methods would be particularly useful in linking T cell function to distinct gene expression profiles, allowing for the identification of subpopulations of cells that are associated with specific clinical outcomes.
Nevertheless, isolation strategies of functional cell types frequently do not achieve homogeneity and contaminating cells cannot be fully excluded from destructive molecular assays. This is in contrast to selective single-cell functional assays that can distinguish truly functional cells from contaminants, meaning that cellular heterogeneity is often the first to be identified i.
Furthermore, cell isolation by FACS requires prior knowledge of distinct cell types, thereby precluding the discovery of novel cell types.
In addition, index-sorting FACS-based approaches are not compatible with droplet-based high-throughput sequencing platforms. To overcome these limitations, Stoeckius et al.
Here, antibodies against cell surface proteins of interest are labelled using unique oligonucleotide barcodes. Antibody-labelled cells are subjected to the Drop-seq protocol, encapsulating single cells in droplets containing beads to introduce unique cellular barcodes to mRNA and the antibody-derived tags ADTs. Subsequently, ADT counts are used to quantify antibody-bound cell surface proteins and provide a link to the corresponding single-cell gene expression profiles.
Consistent surface proteome quantification and resolution were achieved compared to traditional flow cytometry approaches, while providing a theoretically unlimited scope for antibody multiplexing The application of CITE-seq in tumor microenvironment biology has been noted previously , Praktiknjo et al. By performing CITE-seq, the authors were able to construct a comprehensive gene expression atlas and simultaneously recorded a comprehensive set of 63 immune-specific cell surface proteins.
Most notably, they derived a comprehensive cell atlas of the tumor microenvironment, using gene expression profiles and quantification of cell surface proteins, underscoring the utility of CITE-seq in the discovery of novel tumor-specific cell surface antigens for cell therapy.
By linking surface protein quantification with gene expression profiling at single cell resolution, CITE-seq can identify novel antigens associated with specific clones within heterogeneous cancer tissues, ultimately raising the prospect of a broader spectrum of effective cell therapies. Such studies provide a prominent example how single-cell multiomics can provide rapid insight into previously unknown diseases and help inform the development of effective therapeutics. Large-scale perturbation screens have previously provided unprecedented insights into gene functions and their role in complex biological mechanisms In Perturb-seq Figure 3 and Table 2 , a pool of barcoded single-guide RNAs sgRNAs is constructed against a set of 24 transcription factors and transduced cells are subjected to high-throughput droplet-based sequencing, whereby unique cell barcodes are also introduced.
The dual barcoding approach allows connection of single-cell gene expression profiles with a respective perturbation. Such single-cell CRISPR screens and their ability to interrogate transcriptional consequences of perturbations provided a novel method to assess the functional effectors of complex biological mechanism and tissues , Of note, Jin et al.
To interrogate the underlying molecular mechanisms driving autism, the authors introduced a guide RNA pool against risk genes to the forebrain of a developing embryo in utero. The progeny of perturbed cells was then collected at P7 for downstream scRNA-seq analysis, providing key insights into the molecular mechanisms of neocortical cell types. Perturb-seq can be very useful in trying to understand larger pathways that integrate multiple signals.
For example, Adamson et al. This type of data has the potential to disentangle larger signaling networks, all of which is important for understanding complex processes such as immune responses. Despite the demonstrated efficacy, application of Perturb-seq is limited by the sequencing depth of high-throughput approaches.
Acquired data is subject to significant background noise and low-abundant transcripts are frequently missed 47 , Furthermore, the multiplicity problem of combining multiplexed perturbations with single-cell gene expression profiles poses a computational challenge. Schraivogel proposed an intriguing adaptation, termed targeted Perturb-seq TAP-seq By performing targeted amplification of a selected set of genes prior to sequencing, the cost and analytical complexity could be significantly reduced.
This approach provides a powerful tool for screening cellular pathways with defined genetic biomarkers. In the context of cell therapy, TAP-seq could thus provide a cost-effective tool for identifying underlying molecular mechanisms of immune cell evasion of CAR T therapy. There have been a wide variety of additional approaches to integrate single-cell perturbation screens with the surface proteome of the same cell.
Most notably, Mimitou et al. In brief, Mimitou et al. Thus, sgRNA, transcripts, antibody-oligonucleotides and up to 2 other parameters can be recorded for individual cells More recently, Frangieh et al. Here, the authors demonstrated the benefits of Perturb-CITE-seq by identifying molecular pathways driving immune evasion of a melanoma cell line against primary tumor infiltrating lymphocytes Overall, the ability to connect gene expression profiles and the cell surface proteome from single cells under perturbation provides a comprehensive characterisation of complex molecular systems.
As demonstrated by Frangieh et al. Recent work by Lee-Six et al. The authors isolated single HSPCs from a healthy donor and were able to retrospectively reconstruct the phylogenetic tree of single cell-derived colonies, based on a broad set of shared or unique acquired somatic mutations.
By simultaneously screening mature cells isolated from peripheral blood samples of the same individual, Lee-Six et al. Using this approach in a 59 year old human, the authors could map all the way back to the most recent common ancestor for blood and buccal epithelium, observed an early expansion of the stem cell compartment and confirmed hematopoietic activity of a large number of diverse HSC clones estimated to be between 50, and , actively contributing HSCs , This technique could be powerfully applied to gain insight into the clonal dynamics of HSCs used in gene therapy.
Careful patient monitoring must be undertaken to ensure therapeutic efficacy and restoration of normal tissue function. As multipotent cells provide the most common target for gene therapies, gene corrections can significantly impact the clonal dynamics of the target tissue. Intriguingly, previous efforts to track therapeutic efficacy of corrective therapies large depended on monitoring progeny cells, their homeostatic function and particularly the proportion of target cells expressing the desired gene edit , However, such approaches do not provide sufficient resolution to fully characterize clonal dynamics of corrected cell types and their impact on homeostatic tissue function.
WGS of single cell-derived colonies allows to monitor naturally occurring somatic mutations in multipotent cells and their progeny to establish their relationship and infer clonal dynamics of single cells When applied to a pool of edited cell and mature cell progeny post-gene therapy, such approaches can provide a direct insight into therapeutic efficacy and long-term tissue health.
In contrast, upfront labelling of target cells followed by temporal tracking of their progeny can reveal patterns of clonal evolution. Here, the advent of routine and cost-effective sequencing also revolutionised lineage tracing, providing a compelling alternative to traditional imaging-based approaches.
In the context of diabetes, lineage tracing has been used to track the various cell types which originate from pancreatic progenitor cell populations — and identify cell types that are able to transdifferentiate into insulin-secreting cells , , High-throughput screening at single cell resolution and integration into multimodal approaches greatly expand the scope of lineage tracing While fluorescent tags limit the capacity of parallel barcoding, DNA sequence complexity provides a scalable barcoding approach.
In principle, unique DNA barcodes are first introduced into a large population of target cells. Subsequently, amplification of the unique set of DNA barcodes in cell progeny can be used to compute lineage phylogenies , The resulting insertions and deletions indels create unique cellular barcodes, which evolve over time. By sequencing such regions, the mutational patterns can be used to establish phylogeny and clonal evolution. Raj et al. The use of droplet-based high-throughput gene expression thus provides cell type information, otherwise lost in previous lineage tracing protocols.
Thus, providing a crucial quality control mechanism prior to performing computational- and capital-intense sequencing , While prospective lineage tracing is not possible in humans, the use of these techniques in preclinical studies has the potential to unlock cellular relationships that are relevant to understanding cell origins in normal and diseased tissues. Furthermore, lineage tracing may also be used to link immature immune cell types to their immunologically active terminally differentiated counterparts.
This could feed into refinements of CAR T cell production protocols for example, allowing for the selection of specific populations with maximal effector function Nevertheless, these multimodal lineage tracing technologies are currently in their infancy and a variety of experimental and computational limitations require attention.
Furthermore, Spanjaard et al. Thus, if not excluded, high-frequency scars can result in false inference of lineage relationship. To address the issue of barcode duplications and noise, Zafar et al. While these advances are promising, further computational innovation will be of paramount importance for the adoption of single-cell lineage tracing in gene and cell therapy developments.
Single-cell sequencing technologies and their multimodal integration continue to push the boundaries of understanding the mechanisms governing complex tissue organization. However, such single-cell screening protocols are largely based on removing the cells and destroying them, typically discarding any spatial information of the underlying tissue from which they were extracted.
The crucial role of cellular location and spatial gene expression throughout early embryogenesis has been widely recognized Similarly, cellular location in heterogeneous tumors and the surrounding tumor microenvironment are vital to cell function Therefore, resolving spatial dimensions and linking these with gene expression profiles to infer gene function and cell identity can help us understand disease pathology and complex tissue function.
Here, we discuss selected technological developments in spatial transcriptomics and their prospect in the development of novel cell and gene therapies [spatial omics protocols are comprehensively described elsewhere: , ]. The development of fluorescence in situ hybridisation FISH techniques first enabled the detection of DNA and RNA molecules in structurally preserved, fixed tissue sections 43 , , Oligonucleotides, complementary to a target nucleotide sequence, are labelled with single or multiple fluorophores.
In turn, fluorescently labelled oligos bound to a target region can be observed using optical microscopy. Here, the authors constructed a library, consisting of short single fluorophore-labelled oligos, against a single mRNA target to estimate the number of mRNA molecules in a single cell, screening up to 3 mRNA sequences in parallel.
To enable high-throughput spatial transcriptomic screening, Lubeck et al. In brief, multiple single fluorophore-labelled probes are used for mRNA labelling during a single hybridization round. By stripping probes and performing multiple rounds of hybridisation, the number of unique barcoding increases exponentially.
Shah et al. This strategy avoids optical crowding by effectively diluting mRNA molecules into separate images. The result was a robust protocol for screening 10, genes in spatially resolved tissues, spanning thousands of cells A recent study by Lohoff et al.
In parallel, Chen et al. While the methods outlined above drove innovation in spatial transcriptomics, their relative infancy is accompanied by experimental and computational complexity, which currently provides a barrier to wide-spread adoption. Several commercially available platforms have been established to provide a standardised experimental framework. The Visium platform utilised NGS for deriving spatially resolved gene expression profiles , Here, a tissue section of interest is deposited onto a slide, coated with uniquely barcoded arrays barcode spacing permits 55um resolution.
However, the current barcode spacing prevents interrogation of neighbouring cells. Here, in situ analysis can provide a complementary approach, allowing interrogation of a defined set of mRNA targets at spatial singe cell resolution — Collectively, spatial transcriptomics technologies are currently in the developmental and early adaption phase.
As a result, several key limitations persist. For instance, the tissue-dependent optimisation and sequential hybridisation rounds require significant experimental time, while the use of customised equipment also impacts implementation.
However, increasing throughput and the desire to reach whole-transcriptome coverage will greatly increase imaging time and data complexity, making the most prominent limiting factor the development of robust analytical tools. To overcome the computational barrier, recent advances aim to address key unmet needs in data analysis and its scalability , Despite these challenges, several major advances have already been made using spatial transcriptomics, including studies in tumor heterogeneity and transcriptional changes in the microenvironment.
In one study, Berglund et al. The authors uncovered significant transcriptional differences between the tumor core and its periphery. Intriguingly, thorough interrogation of stromal and immune cell types, surrounding the primary tumor, facilitated the identification of heterogeneous gene expression networks in the tumor microenvironment Spatial transcriptomics has also been applied for mapping the localisation of Cxclabundant reticular cells in the bone marrow niche and for the characterisation of stromal cell heterogeneity in tumor microenvironments , These and other studies demonstrate that the potential of spatial transcriptomics in deciphering tumor architecture, heterogeneity and microenvironments has been widely recognised.
Beside its role in therapeutic discovery and disease pathology, spatially resolved gene expression profiles can become of paramount importance for monitoring therapeutic outcomes of cell therapies and identify evasion mechanisms in response to cell therapies. In addition, spatial characterisation post CAR T cell therapy could provide an insight into the impact of off-target effects on the function of proximal tissues.
Similarly, spatial transcriptomics could aid in long-term monitoring of patients undergoing corrective gene therapies. The past decade has produced an abundance of novel single-cell molecular tools, facilitating the unbiased screening of a wide array of molecular dimensions at unprecedented resolution.
Unimodal sequencing technologies have proved particularly impactful in the first wave of wide-scale adoption, but more approaches have been focused on combining such techniques into multimodal screens to allow simultaneous capture of multiple molecular dimensions from the same cell.
These technologies have allowed researchers to unpick the molecular mechanisms driving disease pathology at a scale not previously considered possible. Tissue and disease heterogeneity, previously masked in bulk sequencing approaches, are now routinely being explored at single cell resolution.
Techniques such as scRNA-seq have been widely adopted due to the production of robust experimental protocols and increasing consensus surrounding the computational approaches for quality control and data analysis. On the other hand, multimodal screens have not yet enjoyed similar uptake due to their reliance on high sequencing costs, advanced integrative computational tools and technical expertise.
However, just as moving to single cells was a technical hurdle of 10 years ago, the research benefits derived from novel multimodal screening platforms will push the limits of discovery and accelerate technical development and method standardization. The next few years should see these technical and computational approaches streamlined to create reproducible protocols and standardised analytical pipelines to facilitate rapid adoption rates, as has occurred for scRNA-seq historically.
Concomitant with the technical challenges and need for standardization, the increased accessibility of single-cell technologies has exponentially increased the amount of data generated during these studies.
This provides a unique opportunity to leverage the power of these studies by integrating datasets but also makes for substantial computing and processing challenges.
Batch correction and data integration across experiments and different sequencing platforms are areas that will require particular attention and novel computational approaches for handling and analysing increasing amounts of data will be of paramount importance.
Ultimately, the continuous technical improvements and aggregation of data could provide the foundation for a fully characterized reference atlas of the human body at single cell resolution. The drive towards such a resource is evident in the recently announced efforts to establish a common coordinate framework CCF for data collection and integration In line with that, initiatives such as the Human Biomolecular Atlas Program and the CCF aim to provide a publicly available tool to help researchers map data from diseased states onto healthy single-cell datasets and provide a reference for the entire scientific community , A number of recent studies have clearly demonstrated the utility of these approaches in 1 understanding complex biological processes such as cell fate determination and immune response, 2 dissecting tissue and disease heterogeneity, and 3 stimulating innovative research aimed at developing novel therapeutics — Within the next decade, it is anticipated that an increasing number of patients across many disease types will be treated with gene and cell therapy.
Using samples obtained from these growing patient cohorts, single-cell technologies will undoubtedly be used to answer essential questions related to the relationships between disease-causing cells, normal or corrected cell types, tumor-targeting lymphocytes such as CAR T cells, and endogenous immune populations. For autoimmune diseases such as type 1 diabetes where the risk of relapse is relatively high due to immunogenicity, this level of detail will be essential to finding new ways to increase treatment efficacy.
Additionally, due to the relatively recent wider application of these therapeutics, only a limited number of gene or cell therapy clinical trial patients have been monitored for more than 10 years following treatment initiation 65 , 84 , — Depending on the stability of edited cells and the influence of other comorbidities, detailed studies using single-cell approaches may also become relevant during long-term follow up.
As patients enter the later decades of life, the intersection of age-related and treatment-related abnormalities may present unique clinical challenges.
Further refinements and innovations to single-cell profiling technologies have the potential to unlock and disentangle relationships between key drivers of disease phenotypes, leading to wider delivery of authentic personalised medicine. DB and AC wrote and compiled the review. JR-L designed and created the figures. DK supervised the work and edited the manuscript. All authors contributed to the article and approved the submitted version. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors would like to thank Tim Lohoff for the valuable discussions and advice on current development in multi-omics and spatial transcriptomics. National Center for Biotechnology Information , U.
Front Immunol. Published online Jul Daniel Bode , 1 , 2 Alyssa H. Cull , 3 Juan A. Rubio-Lara , 3 and David G. Alyssa H. Juan A. David G. Author information Article notes Copyright and License information Disclaimer. Kent, ku. This article was submitted to Immunological Tolerance and Regulation, a section of the journal Frontiers in Immunology.
Received Apr 29; Accepted Jun The use, distribution or reproduction in other forums is permitted, provided the original author s and the copyright owner s are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Abstract Single-cell molecular tools have been developed at an incredible pace over the last five years as sequencing costs continue to drop and numerous molecular assays have been coupled to sequencing readouts.
Keywords: cell therapy, gene therapy, single-cell sequencing, scRNA-seq, multimodal omics, multiomics, CAR T cell therapy, disease heterogeneity. Introduction The crucial role that single-cell approaches play in understanding cell function has been recognised for decades.
Open in a separate window. Figure 1. Cell Therapy as a Promising Treatment for More Complex Diseases While gene therapy has revolutionized the treatment of primary immunodeficiencies and monogenic disorders, other strategies may be required to treat more complex diseases. Figure 2. Using Single-Cell Approaches to Refine Treatment And Inform the Development of Novel Therapeutics Although great strides have been made in gene and cell therapy, applications to a wider range of diseases requires more information.
Table 1 Unmet needs and addressable questions in gene and cell therapy. Prior to therapy What is the underlying clonal diversity for complex diseases such as cancer or diabetes? Can understanding the heterogeneity of diseases refined diagnosis? Are T cells obtained from different individuals inherently different? What contributes to CAR T cell product variability? Can we adjust this to improve treatment efficacy? Do HSPCs acquire mutations or epigenetic changes during ex vivo expansion and transduction steps?
What makes a successful T cell product? Post-treatment follow-up Gene therapy ex vivo and in vivo Cell therapy What are the clonal dynamics of edited cells over time and how does that change in relation to unedited cells? Can low level leukemic clones be detected prior to overt leukemias for patients?
When using in vivo approaches, what are the consequences of gene correction or transgene expression in cells that do not usually express the gene of interest? Can in vivo gene therapy approaches be designed to specifically target disease-causing cells? How can on-target, off-tumor toxicities be minimized?
Which CAR T cells survive over time and are some better at targeting tumor cells than others? Are there differences between CAR T cell populations in the blood versus those present in tumor tissue? How do cancer cells especially in solid tumors adapt to evade targeting by CAR T cells? Single-Cell Multi-Omics Platforms and Their Prospect in Gene and Cell Therapy A wide array of screening platforms have been developed to interrogate molecular states at the single cell level to give insight into tumor heterogeneity and clonal evolution of complex tissues.
Genome The first protocol for DNA sequencing at the single cell level, termed single nucleus sequencing SNS , was described by Navin and colleagues In fact, in spite of advancements in knowledge of the CNS function, the treatment of neurological disorders with modern medical and surgical approaches remains difficult for many reasons, such as the complexity of the CNS, the limited regenerative capacity of the tissue, and the difficulty in conveying conventional drugs to the organ due to the blood-brain barrier.
Gene therapy, allowing the delivery of genetic materials that encodes potential therapeutic molecules, represents an attractive option.
Moreover, she says, the technology can be used to evaluate heterogeneous cell samples for the integration patterns and integrity of the transgene. And with TLA, unlike a lot of technologies, you only need to know very limited sequence information on your locus of interest. TLA is hypothesis neutral. A similar point is made by Baghbaderani.
He adds that gene therapy developers can benefit from working with a contract development and manufacturing organization such as Lonza.
According to Baghbaderani, Lonza can offer extensive analytical expertise—ranging from early-stage development to downstream manufacturing—under a single roof. Looking to the future, Baghbaderani sees gene therapy analytics as crucial to the development of streamlined, robust GMP-compliant manufacturing processes. Using process control analytics to ensure a consistency of product is also important, adds Chadwick.
Companies are also looking to provide a wider range of services. Sciex, for example, is repurposing its work on AAV vectors to meet a growing demand for assays designed for lentiviruses, Darling points out. The company is also looking into gene therapies that use liposomes instead of viral vectors, and it is developing techniques for RNA analysis.
Cergentis is also developing a broader range of services. The company, Bergboer notes, is making its TLA analysis service available as a kit for in-house use. Log in to leave a comment. Sign in. Forgot your password? Get help.
0コメント