Clonify: unseeded antibody lineage assignment from next-generation sequencing data

Longitudinal analysis of drift in the circulating human antibody repertoire over four years

The human immune system is a sophisticated network of cells and molecules that plays a vital role in safeguarding the body against a broad range of foreign agents. One of the key components of the immune system is the circulating antibody repertoire, a vast collection of different antibodies that recognize and eliminate foreign agents specifically. The diversity and specificity of the circulating antibody repertoire are essential for effective immunity. In a study of ten healthy donors, we previously demonstrated the accessible human antibody repertoire to be on the order of 1015-1018 distinct members, although an individual will only sample a fraction of that repertoire at a given time point. The mode of generation of the naïve antibody repertoire through random recombination before antigen contact suggests that the composition of the circulating antibody repertoire may drift over time, but the magnitude of this drift is poorly understood. Here, we used high-throughput sequencing to analyze two donors from our original study, approximately four years later. Our results reveal conservation of the size, overall diversity, and gross features of the circulating antibody repertoire, such as V- and J-gene usage and CDRH3 length distribution, over time but suggest substantial changes to fine features of the repertoire, such as combinations of V, J, D gene segments and P and N sequences, which may influence primary responses to newly encountered pathogens as well as to immunization procedures.

Combining mutation and recombination statistics to infer clonal families in antibody repertoires

B-cell repertoires are characterized by a diverse set of receptors of distinct specificities generated through two processes of somatic diversification: V(D)J recombination and somatic hypermutations. B-cell clonal families stem from the same V(D)J recombination event, but differ in their hypermutations. Clonal families identification is key to understanding B-cell repertoire function, evolution, and dynamics. We present HILARy (high-precision inference of lineages in antibody repertoires), an efficient, fast, and precise method to identify clonal families from single- or paired-chain repertoire sequencing datasets. HILARy combines probabilistic models that capture the receptor generation and selection statistics with adapted clustering methods to achieve consistently high inference accuracy. It automatically leverages the phylogenetic signal of shared mutations in difficult repertoire subsets. Exploiting the high sensitivity of the method, we find the statistics of evolutionary properties such as the site frequency spectrum and dN/dS ratio do not depend on the junction length. We also identify a broad range of selection pressures spanning two orders of magnitude.

Benchmarking antibody clustering methods using sequence, structural, and machine learning similarity measures for antibody discovery applications

Antibodies are proteins produced by our immune system that have been harnessed as biotherapeutics. The discovery of antibody-based therapeutics relies on analyzing large volumes of diverse sequences coming from phage display or animal immunizations. Identification of suitable therapeutic candidates is achieved by grouping the sequences by their similarity and subsequent selection of a diverse set of antibodies for further tests. Such groupings are typically created using sequence-similarity measures alone. Maximizing diversity in selected candidates is crucial to reducing the number of tests of molecules with near-identical properties. With the advances in structural modeling and machine learning, antibodies can now be grouped across other diversity dimensions, such as predicted paratopes or three-dimensional structures. Here we benchmarked antibody grouping methods using clonotype, sequence, paratope prediction, structure prediction, and embedding information. The results were benchmarked on two tasks: binder detection and epitope mapping. We demonstrate that on binder detection no method appears to outperform the others, while on epitope mapping, clonotype, paratope, and embedding clusterings are top performers. Most importantly, all the methods propose orthogonal groupings, offering more diverse pools of candidates when using multiple methods than any single method alone. To facilitate exploring the diversity of antibodies using different methods, we have created an online tool-CLAP-available at (clap.naturalantibody.com) that allows users to group, contrast, and visualize antibodies using the different grouping methods.

Anchor Clustering for million-scale immune repertoire sequencing data

BACKGROUND

The clustering of immune repertoire data is challenging due to the computational cost associated with a very large number of pairwise sequence comparisons. To overcome this limitation, we developed Anchor Clustering, an unsupervised clustering method designed to identify similar sequences from millions of antigen receptor gene sequences. First, a Point Packing algorithm is used to identify a set of maximally spaced anchor sequences. Then, the genetic distance of the remaining sequences to all anchor sequences is calculated and transformed into distance vectors. Finally, distance vectors are clustered using unsupervised clustering. This process is repeated iteratively until the resulting clusters are small enough so that pairwise distance comparisons can be performed.

RESULTS

Our results demonstrate that Anchor Clustering is faster than existing pairwise comparison clustering methods while providing similar clustering quality. With its flexible, memory-saving strategy, Anchor Clustering is capable of clustering millions of antigen receptor gene sequences in just a few minutes.

CONCLUSIONS

This method enables the meta-analysis of immune-repertoire data from different studies and could contribute to a more comprehensive understanding of the immune repertoire data space.

Antigen pressure from two founder viruses induces multiple insertions at a single antibody position to generate broadly neutralizing HIV antibodies

Vaccination strategies aimed at maturing broadly neutralizing antibodies (bnAbs) from naïve precursors are hindered by unusual features that characterize these Abs, including insertions and deletions (indels). Longitudinal studies of natural HIV infection cases shed light on the complex processes underlying bnAb development and have suggested a role for superinfection as a potential enhancer of neutralization breadth. Here we describe the development of a potent bnAb lineage that was elicited by two founder viruses to inform vaccine design. The V3-glycan targeting bnAb lineage (PC39-1) was isolated from subtype C-infected IAVI Protocol C elite neutralizer, donor PC39, and is defined by the presence of multiple independent insertions in CDRH1 that range from 1-11 amino acids in length. Memory B cell members of this lineage are predominantly atypical in phenotype yet also span the class-switched and antibody-secreting cell compartments. Development of neutralization breadth occurred concomitantly with extensive recombination between founder viruses before each virus separated into two distinct population "arms" that evolved independently to escape the PC39-1 lineage. Ab crystal structures show an extended CDRH1 that can help stabilize the CDRH3. Overall, these findings suggest that early exposure of the humoral system to multiple related Env molecules could promote the induction of bnAbs by focusing Ab responses to conserved epitopes.

Efficient isolation of rare B cells using next-generation antigen barcoding

The ability to efficiently isolate antigen-specific B cells in high throughput will greatly accelerate the discovery of therapeutic monoclonal antibodies (mAbs) and catalyze rational vaccine development. Traditional mAb discovery is a costly and labor-intensive process, although recent advances in single-cell genomics using emulsion microfluidics allow simultaneous processing of thousands of individual cells. Here we present a streamlined method for isolation and analysis of large numbers of antigen-specific B cells, including next generation antigen barcoding and an integrated computational framework for B cell multi-omics. We demonstrate the power of this approach by recovering thousands of antigen-specific mAbs, including the efficient isolation of extremely rare precursors of VRC01-class and IOMA-class broadly neutralizing HIV mAbs.

Vaccination induces HIV broadly neutralizing antibody precursors in humans

Broadly neutralizing antibodies (bnAbs) can protect against HIV infection but have not been induced by human vaccination. A key barrier to bnAb induction is vaccine priming of rare bnAb-precursor B cells. In a randomized, double-blind, placebo-controlled phase 1 clinical trial, the HIV vaccine-priming candidate eOD-GT8 60mer adjuvanted with AS01B had a favorable safety profile and induced VRC01-class bnAb precursors in 97% of vaccine recipients with median frequencies reaching 0.1% among immunoglobulin G B cells in blood. bnAb precursors shared properties with bnAbs and gained somatic hypermutation and affinity with the boost. The results establish clinical proof of concept for germline-targeting vaccine priming, support development of boosting regimens to induce bnAbs, and encourage application of the germline-targeting strategy to other targets in HIV and other pathogens.

Dependence on a variable residue limits the breadth of an HIV MPER neutralizing antibody, despite convergent evolution with broadly neutralizing antibodies

Broadly neutralizing antibodies (bNAbs) that target the membrane-proximal external region (MPER) of HIV gp41 envelope, such as 4E10, VRC42.01 and PGZL1, can neutralize >80% of viruses. These three MPER-directed monoclonal antibodies share germline antibody genes (IGHV1-69 and IGKV3-20) and form a bNAb epitope class. Furthermore, convergent evolution within these two lineages towards a ^111.2GW^111.3 motif in the CDRH3 is known to enhance neutralization potency. We have previously isolated an MPER neutralizing antibody, CAP206-CH12, that uses these same germline heavy and light chain genes but lacks breadth (neutralizing only 6% of heterologous viruses). Longitudinal sequencing of the CAP206-CH12 lineage over three years revealed similar convergent evolution towards ^111.2GW^111.3 among some lineage members. Mutagenesis of CAP206-CH12 from ^111.2GL^111.3 to ^111.2GW^111.3 and the introduction of the double GWGW motif into CAP206-CH12 modestly improved neutralization potency (2.5–3-fold) but did not reach the levels of potency of VRC42.01, 4E10 or PGZL1. To explore the lack of potency/breadth, viral mutagenesis was performed to map the CAP206-CH12 epitope. This indicated that CAP206-CH12 is dependent on D⁶⁷⁴, a highly variable residue at the solvent-exposed elbow of MPER. In contrast, VRC42.01, PGZL1 and 4E10 were dependent on highly conserved residues (W⁶⁷², F⁶⁷³, T⁶⁷⁶, and W⁶⁸⁰) facing the hydrophobic patch of the MPER. Therefore, while CAP206-CH12, VRC42.01, PGZL1 and 4E10 share germline genes and show some evidence of convergent evolution, their dependence on different amino acids, which impacts orientation of binding to the MPER, result in differences in breadth and potency. These data have implications for the design of HIV vaccines directed at the MPER epitope.

Germline-targeting immunogens are a promising HIV vaccine design strategy. This approach is reliant on the identification of broadly neutralizing antibody (bNAb) classes, which use the same germline antibody genes to target the same viral epitopes. Here, we compare four HIV Envelope MPER-directed antibodies (4E10, VRC42.01, PGZL1 and CAP206-CH12) that despite having shared antibody genes, show distinct neutralization profiles. We show that CAP206-CH12 is dependent on a highly variable residue in the MPER, which results in low neutralization breadth. In contrast, the 4E10, PGZL1 and VRC42.01 mAbs are dependent on highly conserved residues in the MPER, resulting in exceptional neutralization breadth. Our data suggest that while shared germline genes within bNAb epitope classes are required, in some cases these are not sufficient to produce neutralization breadth, and MPER immunogens will need to trigger responses to conserved sites.

Similarity measurements of B cell receptor repertoire in baseline mice showed spectrum convergence of IgM

BACKGROUND

The B cell receptor (BCR) repertoire is highly diverse among individuals. Poor similarity of the spectrum among inbred baseline mice may limit the ability to discriminate true signals from those involving specific experimental factors. The repertoire similarity of the baseline status lacks intensive measurements.

RESULTS

We measured the repertoire similarity of IgH in blood and spleen samples from untreated BALB/c and C57BL/6J mice to investigate the baseline status of the two inbred strains. The antibody pool was stratified by the isotype of IgA, IgG and IgM. Between individuals, the results showed better convergence of CDR3 and clonal lineage profiles in IgM than in IgA and IgG, and better robustness of somatic mutation networks in IgM than in IgA and IgG. It also showed that the CDR3 clonotypes and clonal lineages shared better in the spleen samples than in the blood samples. The animal batch differences were detected in CDR3 evenness, mutated clonotype proportions, and maximal network degrees. A cut-off of 95% identity in the CDR3 nucleotide sequences was suitable for clonal lineage establishment.

CONCLUSIONS

Our findings reveal a natural landscape of BCR repertoire similarities between baseline mice and provide a solid reference for designing studies of mouse BCR repertoires.

Adaptive Immune Receptor Repertoire (AIRR) Community Guide to Repertoire Analysis

Adaptive immune receptor repertoires (AIRRs) are rich with information that can be mined for insights into the workings of the immune system. Gene usage, CDR3 properties, clonal lineage structure, and sequence diversity are all capable of revealing the dynamic immune response to perturbation by disease, vaccination, or other interventions. Here we focus on a conceptual introduction to the many aspects of repertoire analysis and orient the reader toward the uses and advantages of each. Along the way, we note some of the many software tools that have been developed for these investigations and link the ideas discussed to chapters on methods provided elsewhere in this volume.

IgV somatic mutation of human anti-SARS-CoV-2 monoclonal antibodies governs neutralization and breadth of reactivity

Abs that neutralize SARS-CoV-2 are thought to provide the most immediate and effective treatment for those severely afflicted by this virus. Because coronavirus potentially diversifies by mutation, broadly neutralizing Abs are especially sought. Here, we report a possibly novel approach to rapid generation of potent broadly neutralizing human anti-SARS-CoV-2 Abs. We isolated SARS-CoV-2 spike protein-specific memory B cells by panning from the blood of convalescent subjects after infection with SARS-CoV-2 and sequenced and expressed Ig genes from individual B cells as human mAbs. All of 43 human mAbs generated in this way neutralized SARS-CoV-2. Eighteen of the forty-three human mAbs exhibited half-maximal inhibitory concentrations (IC50) of 6.7 × 10-12 M to 6.7 × 10-15 M for spike-pseudotyped virus. Seven of the human mAbs also neutralized (with IC50 < 6.7 × 10-12 M) viruses pseudotyped with mutant spike proteins (including receptor-binding domain mutants and the S1 C-terminal D614G mutant). Neutralization of the Wuhan Hu-1 founder strain and of some variants decreased when coding sequences were reverted to germline, suggesting that potency of neutralization was acquired by somatic hypermutation and selection of B cells. These results indicate that infection with SARS-CoV-2 evokes high-affinity B cell responses, some products of which are broadly neutralizing and others highly strain specific. We also identify variants that would potentially resist immunity evoked by infection with the Wuhan Hu-1 founder strain or by vaccines developed with products of that strain, suggesting evolutionary courses that SARS-CoV-2 could take.

Vaccination with prefusion-stabilized respiratory syncytial virus fusion protein induces genetically and antigenically diverse antibody responses

An effective vaccine for respiratory syncytial virus (RSV) is an unrealized public health goal. A single dose of the prefusion-stabilized fusion (F) glycoprotein subunit vaccine (DS-Cav1) substantially increases serum-neutralizing activity in healthy adults. We sought to determine whether DS-Cav1 vaccination induces a repertoire mirroring the pre-existing diversity from natural infection or whether antibody lineages targeting specific epitopes predominate. We evaluated RSV F-specific B cell responses before and after vaccination in six participants using complementary B cell sequencing methodologies and identified 555 clonal lineages. DS-Cav1-induced lineages recognized the prefusion conformation of F (pre-F) and were genetically diverse. Expressed antibodies recognized all six antigenic sites on the pre-F trimer. We identified 34 public clonotypes, and structural analysis of two antibodies from a predominant clonotype revealed a common mode of recognition. Thus, vaccination with DS-Cav1 generates a diverse polyclonal response targeting the antigenic sites on pre-F, supporting the development and advanced testing of pre-F-based vaccines against RSV.

Alignment free identification of clones in B cell receptor repertoires

Following antigenic challenge, activated B cells rapidly expand and undergo somatic hypermutation, yielding groups of clonally related B cells with diversified immunoglobulin receptors. Inference of clonal relationships based on the receptor sequence is an essential step in many adaptive immune receptor repertoire sequencing studies. These relationships are typically identified by a multi-step process that involves: (i) grouping sequences based on shared V and J gene assignments, and junction lengths and (ii) clustering these sequences using a junction-based distance. However, this approach is sensitive to the initial gene assignments, which are error-prone, and fails to identify clonal relatives whose junction length has changed through accumulation of indels. Through defining a translation-invariant feature space in which we cluster the sequences, we develop an alignment free clonal identification method that does not require gene assignments and is not restricted to a fixed junction length. This alignment free approach has higher sensitivity compared to a typical junction-based distance method without loss of specificity and PPV. While the alignment free procedure identifies clones that are broadly consistent with the junction-based distance method, it also identifies clones with characteristics (multiple V or J gene assignments or junction lengths) that are not detectable with the junction-based distance method.

B cell engagement with HIV-1 founder virus envelope predicts development of broadly neutralizing antibodies

Determining which immunological mechanisms contribute to the development of broad neutralizing antibodies (bNAbs) during HIV-1 infection is a major goal to inform vaccine design. Using samples from a longitudinal HIV-1 acute infection cohort, we found key B cell determinants within the first 14-43 days of viremia that predict the development of bNAbs years later. Individuals who develop neutralization breadth had significantly higher B cell engagement with the autologous founder HIV envelope (Env) within 1 month of initial viremia. A higher frequency of founder-Env-specific naive B cells was associated with increased B cell activation and differentiation and predictive of bNAb development. These data demonstrate that the initial B cell interaction with the founder HIV Env is important for the development of broadly neutralizing antibodies and provide evidence that events within HIV acute infection lead to downstream functional outcomes.

Using the antibody-antigen binding interface to train image-based deep neural networks for antibody-epitope classification

High-throughput B-cell sequencing has opened up new avenues for investigating complex mechanisms underlying our adaptive immune response. These technological advances drive data generation and the need to mine and analyze the information contained in these large datasets, in particular the identification of therapeutic antibodies (Abs) or those associated with disease exposure and protection. Here, we describe our efforts to use artificial intelligence (AI)-based image-analyses for prospective classification of Abs based solely on sequence information. We hypothesized that Abs recognizing the same part of an antigen share a limited set of features at the binding interface, and that the binding site regions of these Abs share share common structure and physicochemical property patterns that can serve as a "fingerprint" to recognize uncharacterized Abs. We combined large-scale sequence-based protein-structure predictions to generate ensembles of 3-D Ab models, reduced the Ab binding interface to a 2-D image (fingerprint), used pre-trained convolutional neural networks to extract features, and trained deep neural networks (DNNs) to classify Abs. We evaluated this approach using Ab sequences derived from human HIV and Ebola viral infections to differentiate between two Abs, Abs belonging to specific B-cell family lineages, and Abs with different epitope preferences. In addition, we explored a different type of DNN method to detect one class of Abs from a larger pool of Abs. Testing on Ab sets that had been kept aside during model training, we achieved average prediction accuracies ranging from 71-96% depending on the complexity of the classification task. The high level of accuracies reached during these classification tests suggests that the DNN models were able to learn a series of structural patterns shared by Abs belonging to the same class. The developed methodology provides a means to apply AI-based image recognition techniques to analyze high-throughput B-cell sequencing datasets (repertoires) for Ab classification.

Systems Biology Methods Applied to Blood and Tissue for a Comprehensive Analysis of Immune Response to Hepatitis B Vaccine in Adults

Conventional vaccine design has been based on trial-and-error approaches, which have been generally successful. However, there have been some major failures in vaccine development and we still do not have highly effective licensed vaccines for tuberculosis, HIV, respiratory syncytial virus, and other major infections of global significance. Approaches at rational vaccine design have been limited by our understanding of the immune response to vaccination at the molecular level. Tools now exist to undertake in-depth analysis using systems biology approaches, but to be fully realized, studies are required in humans with intensive blood and tissue sampling. Methods that support this intensive sampling need to be developed and validated as feasible. To this end, we describe here a detailed approach that was applied in a study of 15 healthy adults, who were immunized with hepatitis B vaccine. Sampling included ~350 mL of blood, 12 microbiome samples, and lymph node fine needle aspirates obtained over a ~7-month period, enabling comprehensive analysis of the immune response at the molecular level, including single cell and tissue sample analysis. Samples were collected for analysis of immune phenotyping, whole blood and single cell gene expression, proteomics, lipidomics, epigenetics, whole blood response to key immune stimuli, cytokine responses, in vitro T cell responses, antibody repertoire analysis and the microbiome. Data integration was undertaken using different approaches-NetworkAnalyst and DIABLO. Our results demonstrate that such intensive sampling studies are feasible in healthy adults, and data integration tools exist to analyze the vast amount of data generated from a multi-omics systems biology approach. This will provide the basis for a better understanding of vaccine-induced immunity and accelerate future rational vaccine design.

Emerging Strategies for Therapeutic Antibody Discovery from Human B Cells

Monoclonal antibodies from human sources are being increasingly recognized as valuable options in many therapeutic areas. These antibodies can show exquisite specificity and high potency while maintaining a desirable safety profile, having been matured and tolerized within human patients. However, the discovery of these antibodies presents important challenges, since the B cells encoding therapeutic antibodies can be rare in a typical blood draw and are short-lived ex vivo. Furthermore, the unique pairing of V_H and V_L domains in each B cell contributes to specificity and function; therefore, maintaining antibody chain pairing presents a throughput limitation. This work will review the various approaches aimed at addressing these challenges with an eye to next-generation methods for high-throughput discovery from the human B-cell repertoire.

PyIR: a scalable wrapper for processing billions of immunoglobulin and T cell receptor sequences using IgBLAST

BACKGROUND

Recent advances in DNA sequencing technologies have enabled significant leaps in capacity to generate large volumes of DNA sequence data, which has spurred a rapid growth in the use of bioinformatics as a means of interrogating antibody variable gene repertoires. Common tools used for annotation of antibody sequences are often limited in functionality, modularity and usability.

RESULTS

We have developed PyIR, a Python wrapper and library for IgBLAST, which offers a minimal setup CLI and API, FASTQ support, file chunking for large sequence files, JSON and Python dictionary output, and built-in sequence filtering.

CONCLUSIONS

PyIR offers improved processing speed over multithreaded IgBLAST (version 1.14) when spawning more than 16 processes on a single computer system. Its customizable filtering and data encapsulation allow it to be adapted to a wide range of computing environments. The API allows for IgBLAST to be used in customized bioinformatics workflows.

AncesTree: An interactive immunoglobulin lineage tree visualizer

High-throughput sequencing of human immunoglobulin genes allows analysis of antibody repertoires and the reconstruction of clonal lineage evolution. The study of antibodies (Abs) affinity maturation is of specific interest to understand the generation of Abs with high affinity or broadly neutralizing activities. Moreover, phylogenic analysis enables the identification of the key somatic mutations required to achieve optimal antigen binding. The Immcantation framework provides a start-to-finish set of analytical methods for high-throughput adaptive immune receptor repertoire sequencing (AIRR-Seq; Rep-Seq) data. Furthermore, Immcantation's Change-O package has developed IgPhyML, an algorithm designed to build specifically immunoglobulin (Ig) phylogenic trees. Meanwhile Phylip, an algorithm that has been originally developed for applications in ecology and macroevolution, can also be used for the phylogenic reconstruction of antibodies maturation pathway. To complement Ig lineages made by IgPhyML or Dnaml (Phylip), we developed AncesTree, a graphic user interface (GUI) that aims to give researchers the opportunity to interactively explore antibodies clonal evolution. AncesTree displays interactive immunoglobulins phylogenic tree, Ig related mutations and sequence alignments using additional information coming from specialized antibody tools. The GUI is a Java standalone application allowing interaction with Ig tree that can run under Windows, Linux and Mac OS.

Comparisons of the antibody repertoires of a humanized rodent and humans by high throughput sequencing

The humanization of animal model immune systems by genetic engineering has shown great promise for antibody discovery, tolerance studies and for the evaluation of vaccines. Assessment of the baseline antibody repertoires of unimmunized model animals will be useful as a benchmark for future immunization experiments. We characterized the heavy chain and kappa light chain antibody repertoires of a model animal, the OmniRat, by high throughput antibody sequencing and made use of two novel datasets for comparison to human repertoires. Intra-animal and inter-animal repertoire comparisons reveal a high level of conservation in antibody diversity between the lymph node and spleen and between members of the species. Multiple differences were found in both the heavy and kappa chain repertoires between OmniRats and humans including gene segment usage, CDR3 length distributions, class switch recombination, somatic hypermutation levels and in features of V(D)J recombination. The Inference and Generation of Repertoires (IGoR) software tool was used to model recombination in VH regions which allowed for the quantification of some of these differences. Diversity estimates of the OmniRat heavy chain repertoires almost reached that of humans, around two orders of magnitude less. Despite variation between the species repertoires, a high frequency of OmniRat clonotypes were also found in the human repertoire. These data give insights into the development and selection of humanized animal antibodies and provide actionable information for use in vaccine studies.

Rapid and Focused Maturation of a VRC01-Class HIV Broadly Neutralizing Antibody Lineage Involves Both Binding and Accommodation of the N276-Glycan

The VH1-2 restricted VRC01-class of antibodies targeting the HIV envelope CD4 binding site are a major focus of HIV vaccine strategies. However, a detailed analysis of VRC01-class antibody development has been limited by the rare nature of these responses during natural infection and the lack of longitudinal sampling of such responses. To inform vaccine strategies, we mapped the development of a VRC01-class antibody lineage (PCIN63) in the subtype C infected IAVI Protocol C neutralizer PC063. PCIN63 monoclonal antibodies had the hallmark VRC01-class features and demonstrated neutralization breadth similar to the prototype VRC01 antibody, but were 2- to 3-fold less mutated. Maturation occurred rapidly within ∼24 months of emergence of the lineage and somatic hypermutations accumulated at key contact residues. This longitudinal study of broadly neutralizing VRC01-class antibody lineage reveals early binding to the N276-glycan during affinity maturation, which may have implications for vaccine design.

A comparison of unamplified and massively multiplexed PCR amplification for murine antibody repertoire sequencing

Sequencing antibody repertoires has steadily become cheaper and easier. Sequencing methods usually rely on some form of amplification, often a massively multiplexed PCR prior to sequencing. To eliminate potential biases and create a data set that could be used for other studies, our lab compared unamplified sequencing results from the splenic heavy-chain repertoire in the mouse to those processed through two commercial applications. We also compared the use of mRNA vs total RNA, reverse transcriptase, and primer usage for cDNA synthesis and submission. The use of mRNA for cDNA synthesis resulted in higher read counts but reverse transcriptase and primer usage had no statistical effects on read count. Although most of the amplified data sets contained more antibody reads than the unamplified data set, we detected more unique V-gene segments in the unamplified data set. Although unique CDR3 detection was much lower in the unamplified data set, RNASeq detected 98% of the high frequency CDR3s. We have shown that unamplified profiling of the antibody repertoire is possible, detects more V-gene segments, and detects high frequency clones in the repertoire.

A comparison of unamplified and massively multiplexed PCR amplification for murine antibody repertoire sequencing

Sequencing antibody repertoires has steadily become cheaper and easier. Sequencing methods usually rely on some form of amplification, often a massively multiplexed PCR prior to sequencing. To eliminate potential biases and create a data set that could be used for other studies, our laboratory compared unamplified sequencing results from the splenic heavy-chain repertoire in the mouse to those processed through two commercial applications. We also compared the use of mRNA vs total RNA, reverse transcriptase, and primer usage for cDNA synthesis and submission. The use of mRNA for cDNA synthesis resulted in higher read counts but reverse transcriptase and primer usage had no statistical effects on read count. Although most of the amplified data sets contained more antibody reads than the unamplified data set, we detected more unique variable (V)-gene segments in the unamplified data set. Although unique CDR3 detection was much lower in the unamplified data set, RNASeq detected 98% of the high-frequency CDR3s. We have shown that unamplified profiling of the antibody repertoire is possible, detects more V-gene segments, and detects high-frequency clones in the repertoire.

Tracing Antibody Repertoire Evolution by Systems Phylogeny

Antibody evolution studies have been traditionally limited to either tracing a single clonal lineage (B cells derived from a single V-(D)-J recombination) over time or examining bulk functionality changes (e.g., tracing serum polyclonal antibody proteins). Studying a single B cell disregards the majority of the humoral immune response, whereas bulk functional studies lack the necessary resolution to analyze the co-existing clonal diversity. Recent advances in high-throughput sequencing (HTS) technologies and bioinformatics have made it possible to examine multiple co-evolving antibody monoclonal lineages within the context of a single repertoire. A plethora of accompanying methods and tools have been introduced in hopes of better understanding how pathogen presence dictates the global evolution of the antibody repertoire. Here, we provide a comprehensive summary of the tremendous progress of this newly emerging field of systems phylogeny of antibody responses. We present an overview encompassing the historical developments of repertoire phylogenetics, state-of-the-art tools, and an outlook on the future directions of this fast-advancing and promising field.

5' Rapid Amplification of cDNA Ends and Illumina MiSeq Reveals B Cell Receptor Features in Healthy Adults, Adults With Chronic HIV-1 Infection, Cord Blood, and Humanized Mice

Using 5′ rapid amplification of cDNA ends, Illumina MiSeq, and basic flow cytometry, we systematically analyzed the expressed B cell receptor (BCR) repertoire in 14 healthy adult PBMCs, 5 HIV-1+ adult PBMCs, 5 cord blood samples, and 3 HIS-CD4/B mice, examining the full-length variable region of μ, γ, α, κ, and λ chains for V-gene usage, somatic hypermutation (SHM), and CDR3 length. Adding to the known repertoire of healthy adults, Illumina MiSeq consistently detected small fractions of reads with high mutation frequencies including hypermutated μ reads, and reads with long CDR3s. Additionally, the less studied IgA repertoire displayed similar characteristics to that of IgG. Compared to healthy adults, the five HIV-1 chronically infected adults displayed elevated mutation frequencies for all μ, γ, α, κ, and λ chains examined and slightly longer CDR3 lengths for γ, α, and λ. To evaluate the reconstituted human BCR sequences in a humanized mouse model, we analyzed cord blood and HIS-CD4/B mice, which all lacked the typical SHM seen in the adult reference. Furthermore, MiSeq revealed identical unmutated IgM sequences derived from separate cell aliquots, thus for the first time demonstrating rare clonal members of unmutated IgM B cells by sequencing.

Analyzing Immunoglobulin Repertoires

Somatic assembly of T cell receptor and B cell receptor (BCR) genes produces a vast diversity of lymphocyte antigen recognition capacity. The advent of efficient high-throughput sequencing of lymphocyte antigen receptor genes has recently generated unprecedented opportunities for exploration of adaptive immune responses. With these opportunities have come significant challenges in understanding the analysis techniques that most accurately reflect underlying biological phenomena. In this regard, sample preparation and sequence analysis techniques, which have largely been borrowed and adapted from other fields, continue to evolve. Here, we review current methods and challenges of library preparation, sequencing and statistical analysis of lymphocyte receptor repertoire studies. We discuss the general steps in the process of immune repertoire generation including sample preparation, platforms available for sequencing, processing of sequencing data, measurable features of the immune repertoire, and the statistical tools that can be used for analysis and interpretation of the data. Because BCR analysis harbors additional complexities, such as immunoglobulin (Ig) (i.e., antibody) gene somatic hypermutation and class switch recombination, the emphasis of this review is on Ig/BCR sequence analysis.

Immune Repertoire Sequencing Using Molecular Identifiers Enables Accurate Clonality Discovery and Clone Size Quantification

Unique molecular identifiers (MIDs) have been demonstrated to effectively improve immune repertoire sequencing (IR-seq) accuracy, especially to identify somatic hypermutations in antibody repertoire sequencing. However, evaluating the sensitivity to detect rare T cells and the degree of clonal expansion in IR-seq has been difficult due to the lack of knowledge of T cell receptor (TCR) RNA molecule copy number and a generalized approach to estimate T cell clone size from TCR RNA molecule quantification. This limited the application of TCR repertoire sequencing (TCR-seq) in clinical settings, such as detecting minimal residual disease in lymphoid malignancies after treatment, evaluating effectiveness of vaccination and assessing degree of infection. Here, we describe using an MID Clustering-based IR-Seq (MIDCIRS) method to quantitatively study TCR RNA molecule copy number and clonality in T cells. First, we demonstrated the necessity of performing MID sub-clustering to eliminate erroneous sequences. Further, we showed that MIDCIRS enables a sensitive detection of a single cell in as many as one million naïve T cells and an accurate estimation of the degree of T cell clonal expression. The demonstrated accuracy, sensitivity, and wide dynamic range of MIDCIRS TCR-seq provide foundations for future applications in both basic research and clinical settings.

HIV Envelope Glycoform Heterogeneity and Localized Diversity Govern the Initiation and Maturation of a V2 Apex Broadly Neutralizing Antibody Lineage

Understanding how broadly neutralizing antibodies (bnAbs) to HIV envelope (Env) develop during natural infection can help guide the rational design of an HIV vaccine. Here, we described a bnAb lineage targeting the Env V2 apex and the Ab-Env co-evolution that led to development of neutralization breadth. The lineage Abs bore an anionic heavy chain complementarity-determining region 3 (CDRH3) of 25 amino acids, among the shortest known for this class of Abs, and achieved breadth with only 10% nucleotide somatic hypermutation and no insertions or deletions. The data suggested a role for Env glycoform heterogeneity in the activation of the lineage germline B cell. Finally, we showed that localized diversity at key V2 epitope residues drove bnAb maturation toward breadth, mirroring the Env evolution pattern described for another donor who developed V2-apex targeting bnAbs. Overall, these findings suggest potential strategies for vaccine approaches based on germline-targeting and serial immunogen design.

Reconstructing Antibody Repertoires from Error-Prone Immunosequencing Reads

Transforming error-prone immunosequencing datasets into Ab repertoires is a fundamental problem in immunogenomics, and a prerequisite for studies of immune responses. Although various repertoire reconstruction algorithms were released in the last 3 y, it remains unclear how to benchmark them and how to assess the accuracy of the reconstructed repertoires. We describe an accurate IgReC algorithm for constructing Ab repertoires from high-throughput immunosequencing datasets and a new framework for assessing the quality of reconstructed repertoires. Surprisingly, Ab repertoires constructed by IgReC from barcoded immunosequencing datasets in the blind mode (without using information about unique molecular identifiers) improved upon the repertoires constructed by the state-of-the-art tools that use barcoding. This finding suggests that IgReC may alleviate the need to generate repertoires using the barcoding technology (the workhorse of current immunogenomics efforts) because our computational approach to error correction of immunosequencing data is nearly as powerful as the experimental approach based on barcoding.

Glycans Function as Anchors for Antibodies and Help Drive HIV Broadly Neutralizing Antibody Development

Apex broadly neutralizing HIV antibodies (bnAbs) recognize glycans and protein surface close to the 3-fold axis of the envelope (Env) trimer and are among the most potent and broad Abs described. The evolution of apex bnAbs from one donor (CAP256) has been studied in detail and many Abs at different stages of maturation have been described. Using diverse engineering tools, we investigated the involvement of glycan recognition in the development of the CAP256.VRC26 Ab lineage. We found that sialic acid-bearing glycans were recognized by germline-encoded and somatically mutated residues on the Ab heavy chain. This recognition provided an "anchor" for the Abs as the core protein epitope varies, prevented complete neutralization escape, and eventually led to broadening of the response. These findings illustrate how glycan-specific maturation enables a human Ab to cope with pathogen escape mechanisms and will aid in optimization of immunization strategies to induce V2 apex bnAb responses.

Rare, high-affinity mouse anti-PD-1 antibodies that function in checkpoint blockade, discovered using microfluidics and molecular genomics

Conventionally, mouse hybridomas or well-plate screening are used to identify therapeutic monoclonal antibody candidates. In this study, we present an alternative to hybridoma-based discovery that combines microfluidics, yeast single-chain variable fragment (scFv) display, and deep sequencing to rapidly interrogate and screen mouse antibody repertoires. We used our approach on six wild-type mice to identify 269 molecules that bind to programmed cell death protein 1 (PD-1), which were present at an average of 1 in 2,000 in the pre-sort scFv libraries. Two rounds of fluorescence-activated cell sorting (FACS) produced populations of PD-1-binding scFv with a mean enrichment of 800-fold, whereas most scFv present in the pre-sort mouse repertoires were de-enriched. Therefore, our work suggests that most of the antibodies present in the repertoires of immunized mice are not strong binders to PD-1. We observed clusters of related antibody sequences in each mouse following FACS, suggesting evolution of clonal lineages. In the pre-sort repertoires, these putative clonal lineages varied in both the complementary-determining region (CDR)3K and CDR3H, while the FACS-selected PD-1-binding subsets varied primarily in the CDR3H. PD-1 binders were generally not highly diverged from germline, showing 98% identity on average with germline V-genes. Some CDR3 sequences were discovered in more than one animal, even across different mouse strains, suggesting convergent evolution. We synthesized 17 of the anti-PD-1 binders as full-length monoclonal antibodies. All 17 full-length antibodies bound recombinant PD-1 with K_D < 500 nM (average = 62 nM). Fifteen of the 17 full-length antibodies specifically bound surface-expressed PD-1 in a FACS assay, and nine of the antibodies functioned as checkpoint inhibitors in a cellular assay. We conclude that our method is a viable alternative to hybridomas, with key advantages in comprehensiveness and turnaround time.

Zika virus activates de novo and cross-reactive memory B cell responses in dengue-experienced donors

Zika virus (ZIKV) shares a high degree of homology with dengue virus (DENV), suggesting that preexisting immunity to DENV could affect immune responses to ZIKV. We have tracked the evolution of ZIKV-induced B cell responses in three DENV-experienced donors. The acute antibody (plasmablast) responses were characterized by relatively high somatic hypermutation and a bias toward DENV binding and neutralization, implying the early activation of DENV clones. A DENV-naïve donor in contrast showed a classical primary plasmablast response. Five months after infection, the DENV-experienced donors developed potent type-specific ZIKV neutralizing antibody responses in addition to DENV cross-reactive responses. Because cross-reactive responses were poorly neutralizing and associated with enhanced ZIKV infection in vitro, preexisting DENV immunity could negatively affect protective antibody responses to ZIKV. The observed effects are epitope-dependent, suggesting that a ZIKV vaccine should be carefully designed for DENV-seropositive populations.

Identification and specificity of broadly neutralizing antibodies against HIV

Beginning in 2009, studies of the humoral responses of HIV‐positive individuals have led to the identification of scores, if not hundreds, of antibodies that are both broadly reactive and potently neutralizing. This development has provided renewed impetus toward an HIV vaccine and led directly to the development of novel immunogens. Advances in identification of donors with the most potent and broad anti‐HIV serum neutralizing responses were crucial in this effort. Equally, development of methods for the rapid generation of human antibodies from these donors was pivotal. Primarily these methods comprise single B‐cell culture coupled to high‐throughput neutralization screening and flow cytometry‐based sorting of single B cells using HIV envelope protein baits. In this review, the advantages and disadvantages of these methodologies are discussed in the context of the specificities targeted by individual antibodies and the need for further improvements to evaluate HIV vaccine candidates.

Novel Approaches to Analyze Immunoglobulin Repertoires

Analysis of immunoglobulin (Ig) repertoires aims to comprehend Ig diversity with the goal of predicting humoral immune responses in the context of infection, vaccination, autoimmunity, and malignancies. The first next-generation sequencing (NGS) analyses of bulk B cell populations dramatically advanced sampling depth over previous low-throughput single-cell-based protocols, albeit at the expense of accuracy and loss of chain-pairing information. In recent years the field has substantially differentiated, with bulk analyses becoming more accurate while single-cell approaches have gained in throughput. Additionally, new platforms striving to combine high throughput and chain pairing have been developed as well as various computational tools for analysis. Here we review the developments of the past 4-5 years and discuss the open challenges.

Hierarchical Clustering Can Identify B Cell Clones with High Confidence in Ig Repertoire Sequencing Data

Adaptive immunity is driven by the expansion, somatic hypermutation, and selection of B cell clones. Each clone is the progeny of a single B cell responding to Ag, with diversified Ig receptors. These receptors can now be profiled on a large scale by next-generation sequencing. Such data provide a window into the microevolutionary dynamics that drive successful immune responses and the dysregulation that occurs with aging or disease. Clonal relationships are not directly measured, but they must be computationally inferred from these sequencing data. Although several hierarchical clustering-based methods have been proposed, they vary in distance and linkage methods and have not yet been rigorously compared. In this study, we use a combination of human experimental and simulated data to characterize the performance of hierarchical clustering-based methods for partitioning sequences into clones. We find that single linkage clustering has high performance, with specificity, sensitivity, and positive predictive value all >99%, whereas other linkages result in a significant loss of sensitivity. Surprisingly, distance metrics that incorporate the biases of somatic hypermutation do not outperform simple Hamming distance. Although errors were more likely in sequences with short junctions, using the entire dataset to choose a single distance threshold for clustering is near optimal. Our results suggest that hierarchical clustering using single linkage with Hamming distance identifies clones with high confidence and provides a fully automated method for clonal grouping. The performance estimates we develop provide important context to interpret clonal analysis of repertoire sequencing data and allow for rigorous testing of other clonal grouping algorithms.

SONAR: A High-Throughput Pipeline for Inferring Antibody Ontogenies from Longitudinal Sequencing of B Cell Transcripts

The rapid advance of massively parallel or next-generation sequencing technologies has made possible the characterization of B cell receptor repertoires in ever greater detail, and these developments have triggered a proliferation of software tools for processing and annotating these data. Of especial interest, however, is the capability to track the development of specific antibody lineages across time, which remains beyond the scope of most current programs. We have previously reported on the use of techniques such as inter- and intradonor analysis and CDR3 tracing to identify transcripts related to an antibody of interest. Here, we present Software for the Ontogenic aNalysis of Antibody Repertoires (SONAR), capable of automating both general repertoire analysis and specialized techniques for investigating specific lineages. SONAR annotates next-generation sequencing data, identifies transcripts in a lineage of interest, and tracks lineage development across multiple time points. SONAR also generates figures, such as identity-divergence plots and longitudinal phylogenetic "birthday" trees, and provides interfaces to other programs such as DNAML and BEAST. SONAR can be downloaded as a ready-to-run Docker image or manually installed on a local machine. In the latter case, it can also be configured to take advantage of a high-performance computing cluster for the most computationally intensive steps, if available. In summary, this software provides a useful new tool for the processing of large next-generation sequencing datasets and the ontogenic analysis of neutralizing antibody lineages. SONAR can be found at https://github.com/scharch/SONAR, and the Docker image can be obtained from https://hub.docker.com/r/scharch/sonar/.

Tailored Immunogens Direct Affinity Maturation toward HIV Neutralizing Antibodies

Induction of broadly neutralizing antibodies (bnAbs) is a primary goal of HIV vaccine development. VRC01-class bnAbs are important vaccine leads because their precursor B cells targeted by an engineered priming immunogen are relatively common among humans. This priming immunogen has demonstrated the ability to initiate a bnAb response in animal models, but recall and maturation toward bnAb development has not been shown. Here, we report the development of boosting immunogens designed to guide the genetic and functional maturation of previously primed VRC01-class precursors. Boosting a transgenic mouse model expressing germline VRC01 heavy chains produced broad neutralization of near-native isolates (N276A) and weak neutralization of fully native HIV. Functional and genetic characteristics indicate that the boosted mAbs are consistent with partially mature VRC01-class antibodies and place them on a maturation trajectory that leads toward mature VRC01-class bnAbs. The results show how reductionist sequential immunization can guide maturation of HIV bnAb responses.

Priming HIV-1 broadly neutralizing antibody precursors in human Ig loci transgenic mice

A major obstacle to a broadly neutralizing antibody (bnAb)-based HIV vaccine is the activation of appropriate B cell precursors. Germline-targeting immunogens must be capable of priming rare bnAb precursors in the physiological setting. We tested the ability of the VRC01-class bnAb germline-targeting immunogen eOD-GT8 60mer (60-subunit self-assembling nanoparticle) to activate appropriate precursors in mice transgenic for human immunoglobulin (Ig) loci. Despite an average frequency of, at most, about one VRC01-class precursor per mouse, we found that at least 29% of singly immunized mice produced a VRC01-class memory response, suggesting that priming generally succeeded when at least one precursor was present. The results demonstrate the feasibility of using germline targeting to prime specific and exceedingly rare bnAb-precursor B cells within a humanlike repertoire.

Early Antibody Lineage Diversification and Independent Limb Maturation Lead to Broad HIV-1 Neutralization Targeting the Env High-Mannose Patch

The high-mannose patch on HIV Env is a preferred target for broadly neutralizing antibodies (bnAbs), but to date, no vaccination regimen has elicited bnAbs against this region. Here, we present the development of a bnAb lineage targeting the high-mannose patch in an HIV-1 subtype-C-infected donor from sub-Saharan Africa. The Abs first acquired autologous neutralization, then gradually matured to achieve breadth. One Ab neutralized >47% of HIV-1 strains with only ∼11% somatic hypermutation and no insertions or deletions. By sequencing autologous env, we determined key residues that triggered the lineage and participated in Ab-Env coevolution. Next-generation sequencing of the Ab repertoire showed an early expansive diversification of the lineage followed by independent maturation of individual limbs, several of them developing notable breadth and potency. Overall, the findings are encouraging from a vaccine standpoint and suggest immunization strategies mimicking the evolution of the entire high-mannose patch and promoting maturation of multiple diverse Ab pathways.