Immune system learning and memory quantified by graphical analysis of B-lymphocyte phylogenetic trees

Detection of Human Bladder Epithelial Cancerous Cells with Atomic Force Microscopy and Machine Learning

The development of noninvasive methods for bladder cancer identification remains a critical clinical need. Recent studies have shown that atomic force microscopy (AFM), combined with pattern recognition machine learning, can detect bladder cancer by analyzing cells extracted from urine. However, these promising findings were limited by a relatively small patient cohort, resulting in modest statistical significance. In this study, we corroborated the AFM technique's capability to identify bladder cancer cells with high accuracy using a controlled model system of genetically purified human bladder epithelial cell lines, comparing cancerous cells with nonmalignant controls. By processing AFM adhesion maps through machine learning algorithms, following previously established methods, we achieved an area under the ROC curve (AUC) of 0.97, with 91% accuracy in cancer cell identification. Furthermore, we enhanced cancer detection by incorporating multiple imaging channels recorded with AFM operating in Ringing mode, achieving an AUC of 0.99 and 93% accuracy. These results demonstrated strong statistical significance (p < 0.0001) in this well-defined model system. While this controlled study does not capture the biological variation present in clinical settings, it provides independent support for AFM-based detection methods and establishes a rigorous technical foundation for further clinical development of AFM imaging-based methods for bladder cancer detection.

Mutational patterns along different evolution paths of follicular lymphoma

Follicular lymphoma (FL) is an indolent disease, characterized by a median life expectancy of 18-20 years and by intermittent periods of relapse and remission. FL frequently transforms into the more aggressive diffuse large B cell lymphoma (t-FL). In previous studies, the analysis of immunoglobulin heavy chain variable region (IgHV) genes in sequential biopsies from the same patient revealed two different patterns of tumor clonal evolution: direct evolution, through acquisition of additional IgHV mutations over time, or divergent evolution, in which lymphoma clones from serial biopsies independently develop from a less-mutated common progenitor cell (CPC). Our goal in this study was to characterize the somatic hypermutation (SHM) patterns of IgHV genes in sequential FL samples from the same patients, and address the question of whether the mutation mechanisms (SHM targeting, DNA repair or both), or selection forces acting on the tumor clones, were different in FL samples compared to healthy control samples, or in late relapsed/transformed FL samples compared to earlier ones. Our analysis revealed differences in the distribution of mutations from each of the nucleotides when tumor and non-tumor clones were compared, while FL and transformed FL (t-FL) tumor clones displayed similar mutation distributions. Lineage tree measurements suggested that either initial clone affinity or selection thresholds were lower in FL samples compared to controls, but similar between FL and t-FL samples. Finally, we observed that both FL and t-FL tumor clones tend to accumulate larger numbers of potential N-glycosylation sites due to the introduction of new SHM. Taken together, these results suggest that transformation into t-FL, in contrast to initial FL development, is not associated with any major changes in DNA targeting or repair, or the selection threshold of the tumor clone.

GLaMST: grow lineages along minimum spanning tree for b cell receptor sequencing data

Background

B cell affinity maturation enables B cells to generate high-affinity antibodies. This process involves somatic hypermutation of B cell immunoglobulin receptor (BCR) genes and selection by their ability to bind antigens. Lineage trees are used to describe this microevolution of B cell immunoglobulin genes. In a lineage tree, each node is one BCR sequence that mutated from the germinal center and each directed edge represents a single base mutation, insertion or deletion. In BCR sequencing data, the observed data only contains a subset of BCR sequences in this microevolution process. Therefore, reconstructing the lineage tree from experimental data requires algorithms to build the tree based on partially observed tree nodes.

Results

We developed a new algorithm named Grow Lineages along Minimum Spanning Tree (GLaMST), which efficiently reconstruct the lineage tree given observed BCR sequences that correspond to a subset of the tree nodes. Through comparison using simulated and real data, GLaMST outperforms existing algorithms in simulations with high rates of mutation, insertion and deletion, and generates lineage trees with smaller size and closer to ground truth according to tree features that highly correlated with selection pressure.

Conclusions

GLaMST outperforms state-of-art in reconstruction of the BCR lineage tree in both efficiency and accuracy. Integrating it into existing BCR sequencing analysis frameworks can significant improve lineage tree reconstruction aspect of the analysis.

Immunoglobulin gene analysis as a tool for investigating human immune responses

The human immunoglobulin repertoire is a hugely diverse set of sequences that are formed by processes of gene rearrangement, heavy and light chain gene assortment, class switching and somatic hypermutation. Early B cell development produces diverse IgM and IgD B cell receptors on the B cell surface, resulting in a repertoire that can bind many foreign antigens but which has had self-reactive B cells removed. Later antigen-dependent development processes adjust the antigen affinity of the receptor by somatic hypermutation. The effector mechanism of the antibody is also adjusted, by switching the class of the antibody from IgM to one of seven other classes depending on the required function. There are many instances in human biology where positive and negative selection forces can act to shape the immunoglobulin repertoire and therefore repertoire analysis can provide useful information on infection control, vaccination efficacy, autoimmune diseases, and cancer. It can also be used to identify antigen-specific sequences that may be of use in therapeutics. The juxtaposition of lymphocyte development and numerical evaluation of immune repertoires has resulted in the growth of a new sub-speciality in immunology where immunologists and computer scientists/physicists collaborate to assess immune repertoires and develop models of immune action.

B cell and/or autoantibody deficiency do not prevent neuropsychiatric disease in murine systemic lupus erythematosus

Background

Neuropsychiatric lupus (NPSLE) can be one of the earliest clinical manifestations in human lupus. However, its mechanisms are not fully understood. In lupus, a compromised blood-brain barrier may allow for the passage of circulating autoantibodies into the brain, where they can induce neuropsychiatric abnormalities including depression-like behavior and cognitive abnormalities. The purpose of this study was to determine the role of B cells and/or autoantibodies in the pathogenesis of murine NPSLE.

Methods

We evaluated neuropsychiatric manifestations, brain pathology, and cytokine expression in constitutively (JhD/MRL/lpr) and conditionally (hCD20-DTA/MRL/lpr, inducible by tamoxifen) B cell-depleted mice as compared to MRL/lpr lupus mice.

Results

We found that autoantibody levels were negligible (JhD/MRL/lpr) or significantly reduced (hCD20-DTA/MRL/lpr) in the serum and cerebrospinal fluid, respectively. Nevertheless, both JhD/MRL/lpr and hCD20-DTA/MRL/lpr mice showed profound depression-like behavior, which was no different from MRL/lpr mice. Cognitive deficits were also observed in both JhD/MRL/lpr and hCD20-DTA/MRL/lpr mice, similar to those exhibited by MRL/lpr mice. Furthermore, although some differences were dependent on the timing of depletion, central features of NPSLE in the MRL/lpr strain including increased blood-brain barrier permeability, brain cell apoptosis, and upregulated cytokine expression persisted in B cell-deficient and B cell-depleted mice.

Conclusions

Our study surprisingly found that B cells and/or autoantibodies are not required for key features of neuropsychiatric disease in murine NPSLE.

Immunoglobulin gene repertoire diversification and selection in the stomach - from gastritis to gastric lymphomas

Chronic gastritis is characterized by gastric mucosal inflammation due to autoimmune responses or infection, frequently with Helicobacter pylori. Gastritis with H. pylori background can cause gastric mucosa-associated lymphoid tissue lymphoma (MALT-L), which sometimes further transforms into diffuse large B-cell lymphoma (DLBCL). However, gastric DLBCL can also be initiated de novo. The mechanisms underlying transformation into DLBCL are not completely understood. We analyzed immunoglobulin repertoires and clonal trees to investigate whether and how immunoglobulin gene repertoires, clonal diversification, and selection in gastritis, gastric MALT-L, and DLBCL differ from each other and from normal responses. The two gastritis types (positive or negative for H. pylori) had similarly diverse repertoires. MALT-L dominant clones (defined as the largest clones in each sample) presented higher diversification and longer mutational histories compared with all other conditions. DLBCL dominant clones displayed lower clonal diversification, suggesting the transforming events are triggered by similar responses in different patients. These results are surprising, as we expected to find similarities between the dominant clones of gastritis and MALT-L and between those of MALT-L and DLBCL.

Integrating B cell lineage information into statistical tests for detecting selection in Ig sequences

Detecting selection in B cell Ig sequences is critical to understanding affinity maturation and can provide insights into Ag-driven selection in normal and pathologic immune responses. The most common sequence-based methods for detecting selection analyze the ratio of replacement and silent mutations using a binomial statistical analysis. However, these approaches have been criticized for low sensitivity. An alternative method is based on the analysis of lineage trees constructed from sets of clonally related Ig sequences. Several tree shape measures have been proposed as indicators of selection that can be statistically compared across cohorts. However, we show that tree shape analysis is confounded by underlying experimental factors that are difficult to control for in practice, including the sequencing depth and number of generations in each clone. Thus, although lineage tree shapes may reflect selection, their analysis alone is an unreliable measure of in vivo selection. To usefully capture the information provided by lineage trees, we propose a new method that applies the binomial statistical framework to mutations identified based on lineage tree structure. This hybrid method is able to detect selection with increased sensitivity in both simulated and experimental data sets. We anticipate that this approach will be especially useful in the analysis of large-scale Ig sequencing data sets generated by high-throughput sequencing technologies.

Ig gene diversification and selection in follicular lymphoma, diffuse large B cell lymphoma and primary central nervous system lymphoma revealed by lineage tree and mutation analyses

Follicular lymphoma (FL), diffuse large B cell lymphoma (DLBCL) and primary central nervous system lymphoma are B cell malignancies. FL and DLBCL have a germinal center origin. We have applied mutational analyses and a novel algorithm for quantifying shape properties of mutational lineage trees to investigate the nature of the diversification, somatic hypermutation and selection processes that affect B cell clones in these malignancies and reveal whether they differ from normal responses. Lineage tree analysis demonstrated higher diversification and mutations per cell in the lymphoma clones. This was caused solely by the longer diversification times of the malignant clones, as their recent diversification processes were similar to those of normal responses, implying similar mutation frequencies. Since previous analyses of antigen-driven selection were shown to yield false positives, we performed a corrected analysis of replacement and silent mutation patterns, which revealed selection against replacement mutations in the framework regions, responsible for the structural integrity of the B cell receptor, but not for positive selection for replacements in the complementary determining regions. Most replacements, however, were neutral or conservative, suggesting that if at all selection operates in these malignancies it is for structural B cell receptor integrity but not for antigen binding.

Adaptive force produced by stress-induced regulation of random variation intensity

The Darwinian theory of life evolution is capable of explaining the majority of related phenomena. At the same time, the mechanisms of optimizing traits beneficial to a population as a whole but not directly to an individual remain largely unclear. There are also significant problems with explaining the phenomenon of punctuated equilibrium. From another perspective, multiple mechanisms for the regulation of the rate of genetic mutations according to the environmental stress have been discovered, but their precise functional role is not well understood yet. Here a novel mathematical paradigm called a Kinetic-Force Principle (KFP), which can serve as a general basis for biologically plausible optimization methods, is introduced and its rigorous derivation is provided. Based on this principle, it is shown that, if the rate of random changes in a biological system is proportional, even only roughly, to the amount of environmental stress, a virtual force is created, acting in the direction of stress relief. It is demonstrated that KFP can provide important insights into solving the above problems. Evidence is presented in support of a hypothesis that the nature employs KFP for accelerating adaptation in biological systems. A detailed comparison between KFP and the principle of variation and natural selection is presented and their complementarity is revealed. It is concluded that KFP is not a competing alternative, but a powerful addition to the principle of variation and natural selection. It is also shown KFP can be used in multiple ways for adaptation of individual biological organisms.

Generation of Immunoglobulin diversity in human gut-associated lymphoid tissue

The organised gut associated lymphoid tissue (GALT) exists adjacent to an extensive and diverse luminal flora. The follicle associated epithelium and associated dendritic cells and lymphocytes form a tightly fortified gateway between the flora and the host that permits connectivity between them and chronic activation of the lymphoid compartment. As a consequence, plasma cell precursors are generated continuously, and in abundance, in GALT by clonal proliferation. Clonal proliferation alone on this scale would reduce the spectrum of B cell specificity. To compensate, GALT also houses molecular machinery that diversifies the receptor repertoire by somatic hypermutation, class switch recombination and receptor revision. These three processes of enhancing the diversity of mature B cells ensure that although clonally related plasma cells may secrete immunoglobulin side by side in the mucosa they rarely have identical antigen binding sites.

IgTree: creating Immunoglobulin variable region gene lineage trees

Lineage trees describe the microevolution of cells within an organism. They have been useful in the study of B cell affinity maturation, which is based on somatic hypermutation of immunoglobulin genes in germinal centers and selection of the resulting mutants. Our aim was to create and implement an algorithm that can generate lineage trees from immunoglobulin variable region gene sequences. The IgTree program implements the algorithm we developed, and generates lineage trees. Original sequences found in experiments are assigned to either leaves or internal nodes of the tree. Each tree node represents a single mutation separating the sequences. The mutations that separate the sequences from each other can be point mutations, deletions or insertions. The program can deal with gaps and find potential reversion mutations. The program also enumerates mutation frequencies and sequence motifs around each mutation, on a per-tree basis. The algorithm has proven useful in several studies of immunoglobulin variable region gene mutations.

Antigen-driven selection in germinal centers as reflected by the shape characteristics of immunoglobulin gene lineage trees: a large-scale simulation study

During the immune response, the generation of memory B lymphocytes in germinal centers involves affinity maturation of the cells' antigen receptors, based on somatic hypermutation of receptor genes and antigen-driven selection of the resulting mutants. Affinity maturation is vital for immune protection, and is the basis of humoral immune learning and memory. Lineage trees of somatically hypermutated immunoglobulin genes often serve to qualitatively illustrate claims concerning the dynamics of affinity maturation in germinal centers. Here, we derive the quantitative relationships between parameters characterizing affinity maturation dynamics (proliferation, differentiation and mutation rates, initial affinity of the Ig to the antigen, and selection thresholds) and the mathematical properties of lineage trees, using a computer simulation which combines mathematical models for all mature B cell populations, stochastic models of hypermutation and selection, lineage tree generation and measurement of graphical tree characteristics. We identified seven key lineage tree properties, and found correlations of these with initial clone affinity and with the selection threshold. These two parameters were found to be the main factors affecting lineage tree shapes in both primary and secondary response trees. The results also confirm that recycling from centrocytes back to centroblasts is highly likely.

Models for antigen receptor gene rearrangement: CDR3 length

Despite the various processing steps involved in V(D)J recombination, which could potentially introduce many biases in the length distribution of complementarity determining region 3 (CDR3) segments, the observed CDR3 length distributions for complete repertoires are very close to a normal-like distribution. This raises the question of whether this distribution is simply a result of the random steps included in the process of gene rearrangement, or has been optimized during evolution. We have addressed this issue by constructing a simulation of gene rearrangement, which takes into account the DNA modification steps included in the process, namely hairpin opening, nucleotide additions, and nucleotide deletions. We found that the near-Gaussian- shape of CDR3 length distribution can only be obtained under a relatively narrow set of parameter values, and thus our model suggests that specific biases govern the rearrangement process. In both B-cell receptor (BCR) heavy chain and T-cell receptor beta chain, we obtained a Gaussian distribution using identical parameters, despite the difference in the number and the lengths of the D segments. Hence our results suggest that these parameters most likely reflect the optimal conditions under which the rearrangement process occurs. We have subsequently used the insights gained in this study to estimate the probability of occurrence of two exactly identical BCRs over the course of a human lifetime. Whereas identical rearrangements of the heavy chain are highly unlikely to occur within one human lifetime, for the light chain we found that this probability is not negligible, and hence the light chain CDR3 alone cannot serve as an indicator of B-cell clonality.

Novel Analysis of Clonal Diversification in Blood B Cell and Bone Marrow Plasma Cell Clones in Immunoglobulin Light Chain Amyloidosis

Immunoglobulin light chain amyloidosis (AL) is characterized by a limited clonal expansion of plasma cells and amyloid formation. Here, we report restriction in the diversity of VL gene usage with a dominance of clonally related B cells in the peripheral blood (PB) isotype-specific repertoire of AL patients. A rigorous quantification of lineage trees reveals presence of intraclonal variations in the PB clones compared to the bone marrow (BM) clones, which suggests a common precursor that is still subject to somatic mutation. When compared to normal BM and PB B cells, AL clones showed significant but incomplete impairment of antigenic selection, which could not be detected by conventional R and S mutation analysis. Therefore, graphical analysis of B cell lineage trees and mathematical quantification of tree properties provide novel insights into the process of B cell clonal evolution in AL.

Lineage tree analysis of immunoglobulin variable-region gene mutations in autoimmune diseases: chronic activation, normal selection

Autoimmune diseases show high diversity in the affected organs, clinical manifestations and disease dynamics. Yet they all share common features, such as the ectopic germinal centers found in many affected tissues. Lineage trees depict the diversification, via somatic hypermutation (SHM), of immunoglobulin variable-region (IGV) genes. We previously developed an algorithm for quantifying the graphical properties of IGV gene lineage trees, allowing evaluation of the dynamical interplay between SHM and antigen-driven selection in different lymphoid tissues, species, and disease situations. Here, we apply this method to ectopic GC B cell clones from patients with Myasthenia Gravis, Rheumatoid Arthritis, and Sjögren's Syndrome, using data scaling to minimize the effects of the large variability due to methodological differences between groups. Autoimmune trees were found to be significantly larger relative to normal controls. In contrast, comparison of the measurements for tree branching indicated that similar selection pressure operates on autoimmune and normal control clones.

Designing an A* algorithm for calculating edit distance between rooted-unordered trees

Tree structures are useful for describing and analyzing biological objects and processes. Consequently, there is a need to design metrics and algorithms to compare trees. A natural comparison metric is the "Tree Edit Distance," the number of simple edit (insert/delete) operations needed to transform one tree into the other. Rooted-ordered trees, where the order between the siblings is significant, can be compared in polynomial time. Rooted-unordered trees are used to describe processes or objects where the topology, rather than the order or the identity of each node, is important. For example, in immunology, rooted-unordered trees describe the process of immunoglobulin (antibody) gene diversification in the germinal center over time. Comparing such trees has been proven to be a difficult computational problem that belongs to the set of NP-Complete problems. Comparing two trees can be viewed as a search problem in graphs. A* is a search algorithm that explores the search space in an efficient order. Using a good lower bound estimation of the degree of difference between the two trees, A* can reduce search time dramatically. We have designed and implemented a variant of the A* search algorithm suitable for calculating tree edit distance. We show here that A* is able to perform an edit distance measurement in reasonable time for trees with dozens of nodes.

Feedback Loops, Reversals and Nonlinearities in Lymphocyte Development

Systems of differentiating cells are often regarded by experimental biologists as unidirectional processes, in which cells spend a fixed time at each successive developmental stage. However, mathematical modeling has in several cases revealed that differentiating cell systems are more complex than previously believed. For example, non-linear transitions, feedback effects, and even apparent reversals have been suggested by our studies on models for the development of lymphocytes and their receptor repertoires, and are reviewed in this paper. These studies have shown that cell population growth in developing lymphocyte subsets is usually nonlinear, as it depends on the density of cells in each compartment. Additionally, T cell development has been shown to be subject to feedback regulation by mature T cell subsets, and B cell development has been shown to include a phenotypic reflux from an advanced to an earlier developmental stage. The challenges we face in our efforts to understand how the repertoires of these cells are generated and regulated are also discussed here.

Immunoglobulin variable-region gene mutational lineage tree analysis: application to autoimmune diseases

Lineage trees have frequently been drawn to illustrate diversification, via somatic hypermutation (SHM), of immunoglobulin variable-region (IGV) genes. In order to extract more information from IGV sequences, we developed a novel mathematical method for analyzing the graphical properties of IgV gene lineage trees, allowing quantification of the differences between the dynamics of SHM and antigen-driven selection in different lymphoid tissues, species, and disease situations. Here, we investigated trees generated from published IGV sequence data from B cell clones participating in autoimmune responses in patients with Myasthenia Gravis (MG), Rheumatoid Arthritis (RA), and Sjögren's Syndrome (SS). At present, as no standards exist for cell sampling and sequence extraction methods, data obtained by different research groups from two studies of the same disease often vary considerably. Nevertheless, based on comparisons of data groups within individual studies, we show here that lineage trees from different individual patients are often similar and can be grouped together, as can trees from two different tissues in the same patient, and even from IgG- and IgA-expressing B cell clones. Additionally, lineage trees from most studies reflect the chronic character of autoimmune diseases.

Quantitative analysis of clonal bone marrow CD19+ B cells: use of B cell lineage trees to delineate their role in the pathogenesis of light chain amyloidosis

Light chain amyloidosis (AL) is a bone marrow (BM) plasma cell neoplasia with systemic deposition of Ig light chain amyloid fibrils. Here, we report the identification of clonal CD19 B cells in the BM and the use of a novel mathematical algorithm to generate B cell lineage trees of the clonal CD19 B cells and CD138 plasma cells from the BM of AL patients to delineate the relationship between these two clonal populations. The CD19+ clonal B cells in the BM of AL patients related to the clonal plasma cells represent a pre-plasma cell precursor population. The B cell lineage trees from AL patients also show significant differences in clonal diversification and antigenic selection compared to clones from normal, healthy controls. These data provide a robust example of the use of graphical quantification methods in delineating the role of neoplastic precursors in the pathogenesis of hematopoietic malignancies.