Применение вероятностей событий VDJ рекомбинации для изучения T-клеточных рецепторов (Application of VDJ recombination events probability for TCR study) тема диссертации и автореферата по ВАК РФ 00.00.00, кандидат наук Смирнова Анастасия Олеговна

• Смирнова Анастасия Олеговна
• кандидат наук
• 2025

Оглавление диссертации кандидат наук Смирнова Анастасия Олеговна

Table of Contents

Chapter 1. Introduction

Chapter 2. Literature review

2.1. T cells receptors: structure and functions in adaptive immunity

2.1.1. Role of T cell receptors in immunity

2.1.2. Structure of TCRs loci and V(D)J rearrangement

2.1.3. The study of TCR repertoire by high-throughput sequencing

2.1.4. V(D)J rearrangement models

2.2. y5 T cells: biology and their TCRs application in clinical cases

2.2.1. y5 T cells as leukocyte population

2.2.2. y5 T cells in inflammatory bowel diseases

2.2.3. Biology of colorectal cancer and tumor-specific T cells

2.2.4. TCRy and TCRS rearrangements as genomic markers and their application in leukemia screening

Chapter 3. Materials and methods

3.1. Sample collection

3.2. DNA and RNA isolation

3.3. Libraries preparation

3.4. Sequencing data analysis

3.4.1. TCR and BCR repertoires reconstruction

3.4.2. TCR repertoires and sequences analysis

3.4.3. Modeling of TCR generation probability

3.4.4. Statistical analysis

Chapter 4. D-D rearrangements in TCRp locus

4.1. Introduction

4.2. Results

4.2.1. The pipeline development for the detection of partial TCR and IGH rearrangements

4.2.2. The test of the pipeline efficacy

4.2.3. TCRp D-D rearrangement detection at the DNA level

4.2.4. Bimodal distribution of TCRp D-D rearrangements length and relation to RS sites

4.2.5. The analysis of D-D signal joint

4.2.6. D1-D2 coding joint contribute to productive V-D1-D2-J2 chains

4.3. Discussion

Chapter 5. Amplification bias correction in MPlex libraries

5.1. Introduction

5.2. Results

5.2.1. The Over Amplification Rate measure definition and versatility test

5.2.2. The versatility and accuracy test

5.2.3. The practical implementation of OAR

5.2.4. Comparison of iROAR and spike-in-based approach for amplification bias detection

5.3. Discussion

Chapter 6. Computational models for TCRy and TCR8 rearrangements

6.1. Introduction

6.2 Results

6.2.1. The development of TCR generation models for complete TCRy and TCRS rearrangements

6.2.2. Evaluation of V, D, and J choice skewness

6.2.3. Evaluation of training set size bias

6.2.4. The impact of rearrangement events on recombinational entropy

6.2.5. TCR generation model accuracy

6.2.6. Analysis of V-D and D-J insertions scenarios in TCRS

6.2.7. Generational models for partial TCRS rearrangements

6.3. Discussion

Chapter 7. TCRy and TCR8 rearrangements as MRD markers

7.1. Introduction

7.2.1. Application of TCRy and TCRS models for QC in MRD monitoring

7.2.2. The filtration of false-positive TCR rearrangements as MRD markers

7.3. Discussion

Chapter 8. TCRy and TCR8 rearrangements in IBD

8.1. Introduction

8.2. Results

8.2.1. The general characteristics of the repertoires

8.2.2. The detection of IBD-specific TCRy clonotypes

8.3. Discussion

Chapter 9. Cancer-specific TCR clonotypes in CRC with different L1 activity

9.1. Introduction

9.2. Results

9.2.1. The general characteristics of the repertoires

9.2.2. The detection of IBD-specific TCR clonotypes

9.2.3. Clonal cluster analysis

9.2.4. Analysis of clonal enrichment

9.3. Discussion

Chapter 10. Conclusions

Acknowledgements

List of Abbreviations

Bibliography

List of Figures

List of Tables

Appendix A. Information on datasets

Appendix B. Supplementary figures and tables

Введение диссертации (часть автореферата) на тему «Применение вероятностей событий VDJ рекомбинации для изучения T-клеточных рецепторов (Application of VDJ recombination events probability for TCR study)»

Chapter 1. Introduction

Research topic relevance. T cell receptors (TCRs) are the key molecules in adaptive immunity. After antigen recognition, they activate lymphocyte proliferation, resulting in the production of multiple identical cells (clone) with the same receptor sequence (clonotype) for an effective immune response [7].

The variable region of TCR subunits, binding to an antigen, is formed by a semi-stochastic mechanism called V(D)J recombination. In this process, genomically encoded fragments (V, D and J genes) are randomly combined, the ends of the coding regions are truncated, and random nucleotide insertions are added at the segment junctions. This forms a hypervariable region (complementarity determining region 3, or CDR3) capable of recognizing any potential antigen [114].

Since CDR3 formation is a combinatorial process, recombination events can be described by probabilistic models, which are based on the extracted statistics from each rearrangement event [56]. The probability of generating different receptor sequences is not uniform, and this information can be used for several purposes. It can be applied in identification of T cell receptors responsible for recognizing the same immune challenge in different human individuals. Such immune challenges include bacterial and viral infectious agents, tumor antigens, vaccines, self-antigens, etc. Distributions of rearrangement events were calculated for TCRa and TCRp, but not for TCRy and TCRS.

yS T cells are less common among circulating T cells compared to ap T cells. However, they are abundant in epithelial tissues and constitute a significant proportion of intraepithelial lymphocytes (IELs) in the intestinal epithelium. yS T cells are not restricted to MHC-dependent activation and are capable of recognizing some non-protein antigens [54]. These factors provide a potential for yS T cells to be used in targeted therapies of disorders localized in the gastrointestinal tract, including colorectal cancer (CRC).

CRC is one of the most common malignancies with a high mortality rate [142]. The etiology of CRC is still not fully understood. However, chronic inflammatory processes associated with inflammatory bowel diseases (IBDs), such as Crohn's disease (CD) or ulcerative colitis (UC), have been shown to correlate with CRC tumorigenesis. IBDs lead to dysregulation of the mucosal immune system, which affects tumor cell proliferation, growth, and invasion [46]. T cells can infiltrate tumors such as sarcomas and carcinomas. They are attracted to tumor-associated antigens, some of which are products of genes not normally expressed in normal cells. Abnormal DNA repair mechanisms in tumor cells can lead to the creation of new proteins known as neoantigens [57]. Molecular alterations such as microsatellite instability (MSI) can also cause disruptions in the genome. MSI can also be caused by retrotransposition of mobile elements, including LINE-1. Although most of its copies are normally inactive due to accumulated mutations in their promoters, the uncontrolled mutations in cancer cells can potentially induce their retrotransposition activity [16, 111].

TCRy and TCRS recombination products are also abundant in T and B cell leukemic repertoires [45, 67]. For this reason, they are often used as molecular markers for minimal residual disease (MRD) monitoring in leukemia therapy to predict disease relapse [65]. Although immune receptors allow tracking of different malignant clones and subclones during treatment [120], their use is limited by the possibility of independent appearances of leukemic clone-specific rearrangements in normal cells. This is particularly relevant for TCRy, where the potential diversity is severely limited due to the absence of a D segment. The use of TCR generation probability models can solve this problem and help to distinguish leukemic clonotypes from normal T-cell rearrangements, thus estimating their reliability as MRD [94].

Research objectives. The main aim of this research was to develop new tools for immune receptor repertoires analysis and demonstrate their implication for clinical cases. The goals of this research are subdivided into the following groups:

1) Create an algorithm for the identification of partial TCR and BCR rearrangements in Rep-seq data;

2) Build a pipeline for amplification bias correction to improve immune repertoire data quality;

3) Develop a model for the calculation of TCRy and TCRS rearrangement generation probability;

4) Demonstrate the implication of the developed model for clinical immune repertoire data, including:

a) Evaluation of TCRy locus rearrangements as molecular markers for minimal residual disease monitoring;

b) Identification of TCRy and TCRS rearrangements associated with inflammatory bowel disease

c) Detection of specific TCR rearrangements specific to colorectal cancer and associated with tumors having different levels of retrotransposon activity. Scientific novelty. This study demonstrated the existence of tandem TCRp D-D,

which was previously considered impossible according to current immunology textbooks. In addition, we proposed TCRy clonotype and TCRa clusters that could potentially be associated with IBD and CRC progression.

As a secondary outcome of this study, we introduced several pipelines for TCR and BCR repertoire data processing that have not been previously published or have demonstrated higher efficacy than existing approaches: the fully in silico amplification bias correction for BCR and TCR repertoires, the pipeline for the detection of partial BCR and BCR rearrangements, and the first generation models for TCRy and TCRS rearrangements.

Theoretical and practical significance. Apart from advancing the theoretical knowledge of the V(D)J rearrangement process, partial TCRp D-D could be used in clinical practice (e.g. malignant clonotype tracking). The information on IBD-specific

TCRy and CRC-specific TCRa clonotypes could potentially be used in the development of targeted therapies.

The developed approach for amplification bias correction can be applied to other studies of BCR and TCR repertoires. Furthermore, the information about the generation probability of TCRy and TCR5 and the presence of partial BCR and TCR rearrangements within Rep-seq libraries can be used for MRD monitoring during leukemia therapy. Main results submitted for the defense are the following:

• TCRp locus can form D1-D2 rearrangements, contributing to TCRp diversity;

• Amplification bias in DNA Rep-seq can be corrected using the ratio of read count and clonotype count-based frequencies for non-functional clonotypes in MPlex PCR-based immune repertoires;

• The uneven distribution of V(D)J rearrangement events leads to reproduction of some TCRy and TCR5 rearrangements detected in malignant cells in the repertoire of healthy individuals. The information about their generation probability allows us to eliminate false positive results in marker selection for MRD monitoring.

• Some y5 and ap T cell clones could potentially be associated with IBD and CRC progression, respectively.

Validity of the obtained results. The results reliability was confirmed by statistical analysis using various criteria and the use of randomization procedures. In case of novel approaches, their reliance and effectiveness was validated by comparison to existing ones on the same dataset.

Based on the dissertation materials, three articles have been published in peer-reviewed scientific journals. The results were reported at the following conferences: the European Human Genetics Virtual Conference in 2021 and 2022 (E-conference), 20th EAACI Immunology Winter School "Basic Immunology Research in Allergy and Clinical Immunology" in 2022 (E-conference), Вычислительная биология и искусственный интеллект для персонализированной медицины in 2022 (E-conference), The

Kind-Philipp-Meeting in 2023 (Wilsede, Germany), and 18th International Congress of Immunology in 2023 (Cape Town, South Africa).

Personal contribution of the author. The role of the dissertation author was to perform the computational part of this research. Details of the personal contribution to each study are described in the Introduction section of the corresponding chapters.

Publications. The main results of this thesis have been published in 3 scientific journals indexed in Web of Science and Scopus systems:

1. Smirnova A.O., Miroshnichenkova A. M., Belyaeva L.D., Kelmanson I.V., Lebedev Y.B., Mamedov I.Z., Chudakov D.M., Komkov A.Y. Novel bimodal TRBD1-TRBD2 rearrangements with dual or absent D-region contribute to TRB V-(D)-J combinatorial diversity //Frontiers in Immunology. - 2023. - V. 14. - P. 1245175. (Main author)

2. Smirnova A.O., Miroshnichenkova A. M., Olshanskaya Y.V., Maschan M.A., Lebedev Y.B., Chudakov D.M., Mamedov I.Z., Komkov A.Y. The use of non-functional clonotypes as a natural calibrator for quantitative bias correction in adaptive immune receptor repertoire profiling //Elife. - 2023. - V. 12. - P. e69157. (Main author)

3. Urazbakhtin S., Smirnova A., Volakhava A., Zerkalenkova E., Salyutina M., Doubek M., Jelinkova H., Khudainazarova N., Volchkov E.,Belyaeva L., Komech E., Pavlova S., Lebedev Y, Plevova K., Olshanskaya Y., Komkov A., Mamedov I. The Absence of Retroelement Activity Is Characteristic for Childhood Acute Leukemias and Adult Acute Lymphoblastic Leukemia //International Journal of Molecular Sciences. -2022. - V. 23. - №. 3. - P. 1756. (Co-author)

Beside, the results of this dissertation were also presented on 6 international conferences:

1. Smirnova A.O., Lebedev Y.B., Mamedov I.Z., Komkov Y.A. iROAR: a computational algorithm for multiplex PCR bias detection and correction in TCR repertoires datasets // European Human Genetics Virtual Conference ESHG 2021 (E-conference, 2021)

2. Smirnova A., Grigoryeva T., Komkov Y., Lebedev Y., Mora T., Walczak A., Chudakov D., Mamedov I. Estimation of recombination events probabilities and construction of rearrangement model for TCRy // 20th EAACI Immunology Winter School "Basic Immunology Research in Allergy and Clinical Immunology". (E-conference, 2022)

3. Smirnova A., Mamedov I., Mora T., Walczak A., Grigoryeva T., Lebedev Y., Chudakov D. Evaluation of TCRy generation probability using Bayesian statistics // European Human Genetics Virtual Conference ESHG 2022 (E-conference, 2022)

4. Smirnova A.O., Komkov A.Y., Salutina M.V., Tkachenko V.K., Tsukanov A.S., Snezhkina A.V., Kudryavtseva A.V., Shubin V.P., Lebedev Y.Effect of LINE-1 activity on the infiltration of specific T cells in colorectal cancer tumors. Computational biology and artificial intelligence (in Russian, E-conference, 2022)

5. Smirnova A., Grigoryeva T., Komkov Y., Lebedev Y., Mora T., Walczak A., Mamedov I., Chudakov D. Application of VDJ recombination events probability for y5 T-cells in MRD monitoring // The Kind-Philipp-Meeting (Wilsede, 2023) (https://doi.org/10.1055/s-0043-1768549)

6. Smirnova A., Grigoryeva T., Komkov Y., Mora T., Walczak A., Lebedev Y., Mamedov I., Chudakov D. Modeling of y5 T-cells VDJ recombination in MRD monitoring Gamma Delta T Cells // 18th International Congress of Immunology (Cape Town, 2023)

Thesis structure. The dissertation has 10 chapters, including Introduction, Literature review, Materials and methods, and Conclusions. The full length of the dissertation is 159 pages, including 30 figures, 5 tables, and 21 objects in Appendix. Bibliography consists of 142 references.

The current introduction (Chapter 1) summarizes the main blind spots in the covered topics and sets the objectives of this thesis by formulating the research questions. Chapter 2 provides a comprehensive literature review and explains the background

necessary for understanding the following sections. Chapter 3 describes the materials and methods. The results obtained during the work on the thesis are divided into Chapters 4-9, depending on the covered topic. Each Chapter consists of a short introduction to the topic, revision of results, and its discussion. Chapter 4 is about the study of TCRp D-D rearrangements and partial BCR and TCR rearrangements in general. Chapter 5 explains the developed algorithm for in silico correction of amplification bias in TCR and BCR repertoires. Chapter 6, which is the main part of this thesis, is dedicated to generation models of TCRy and TCRS rearrangements. In Chapters 7-9, these models were used in MRD monitoring, and in the search for clonotypes specific to IBD and CRC, respectively. Chapter 10 summarizes the obtained results, their relevance to the set goals, and the perspectives for further research.