Sample multiplexing methods through lipid nanoparticle labels and applications in single-cell RNA sequencing

In recent years, high dimensional analytical technologies have significantly revolutionized our capacity to characterize biological systems, where single-cell RNA sequencing (scRNA-seq) standing out as a particularly transformative technique. With the development of advanced sequencers and cell capturing devices, powered by bioinformatic tools, the per-test capacity of scRNA-seq has reached whole-transcriptome single-cell profiling for thousands of cells, from which the global transcriptomic patterns and high heterogeneity of cell populations are captured and provide deep biological insights. Sample multiplexing, a method enabling the analysis of multiple samples within a single experiment via individual sample pre-labelling, has been realized and used in multiple analytical techniques, such as mass spectrometry and sequencing. This reduces batch-to-batch variation that is difficult to control during multi-step experiments, and significantly reduces cost. Sample multiplexing is also highly desirable for scRNA-seq which has even more steps and expense than the other methods. Furthermore, challenges are significant in pre-labeling living cells compared with chemically stable peptides or DNA molecules. Recently, there are emerging strategies to provide multiplexing for scRNA-seq of living cell samples. Cells are labeled with oligonucleotide barcodes, through barcode conjugated binding motifs that target cell surface receptors or the plasma membrane. After labeling and sample mixing, the demultiplexing of samples depends on a bimodal distribution in barcode counts that will set thresholds for labeled and unlabeled populations. These strategies have shown robust multiplexing and demultiplexing results in human peripheral bone marrow cells (PBMC) samples. However, recent application of these strategies reported incomplete labeling or failed bimodal distribution that occurred frequently in other sample types that indicated limited labeling sensitivity of certain cells, especially when deriving from digests of solid tissues. In my project, I developed a method (‘Nanocoding’) that utilizes cell surface amine groups as anchoring sites, and applied lipid nanoparticle formulations to label barcodes efficiently and stably onto living cells. Nanocoding labels cells through a combination of biophysical adsorption and irreversible labeling through surface amine-targeted NHS linkers followed by streptavidin and biotin-functionalized barcode-encapsulated LNPs. In contrast to the currently used tools that are 1:1 conjugates of oligonucleotide and binding motif, this method utilizes LNP formulations to densely encapsulate oligonucleotides in colloidal nanoparticles with surface functionalized groups to target cells. In this project, with application to multiple biological models and characterizations through various analytical methods, I showed that Nanoncoding could achieve both strong bimodal distributions of labels and efficient labeling of cells independent of cell surface epitope features, making it robust and efficient to label cells from a breadth of sample types. And when compared with current labeling using direct conjugate targeting cell surfaces, only Nanocoding showed significant stability with minimal crosstalk after several hours of pooling multiple samples. The development and application of Nanocoding methods are described in Chapters 1 to 4. In Chapter 1, I discuss the background of sample multiplexing development and its importance in modern analytical methods. In Chapter 2, I discuss the design rationale of Nanocoding, development and optimization of its workflow, and evaluation of its advantages compared with other cell barcoding methods using fluorescent analogs. In Chapter 3, I show the application of Nanocoding method to sample multiplexing of mouse spleen cells, stromal vascular fraction (SVF) cells and cultured cells. I discuss the performance of barcoding efficiency and accuracy of Nanocoding in challenging sample types, and its strength in labeling such samples that could not be achieved by current methods. Additionally, in Chapter 4, Nanocoding is used to explore the biology of spleen and adipose tissues in the context of obesity and age related changes at the single-cell level using bioinformatic analysis. In earlier years of my PhD study, I worked on metabolic glycoengineering probe design and optimization for a two-step strategy of targeted drug delivery for anti-cancer treatment. In Chapter 5, I discuss my development of an enhanced cancer cell targeting metabolic glycoengineering probe, which could modify cell surfaces with azides groups with improved abundance compared with previously reported molecular designs. In this study, I demonstrate the development of this enhanced molecule through screening a panel of candidates, and characterize its cell labeling activity both in vitro and in vivo. I further evaluate its anti-cancer activity when coupled with an azide-targeting dibenzocyclooctyne(DBCO)-conjugated camptothecin prodrug on a mouse xenograft colon cancer model.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Materials Science & Engr

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2024

Author dc:creator

• Feng, Yujun

Contributors dc:contributor

• Smith, Andrew
• Schroeder, Charles
• Wang, Hua
• Cheng, Jianjun

• Nanomaterials
• Oligonucleotide Functionalization
• Single-cell Rna Sequencing

Statement dc:rights

• Copyright 2024 Yujun Feng

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/127394

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F127394