Comprehensive Guide to Antibody Sequencing Methods

Antibody sequencing refers to the analysis of antibody gene sequences by different technical means in order to deeply understand the diversity, specificity and function of immune system. The following are several common antibody sequencing methods:

Traditional Sanger sequencing

The basic principle of Sanger sequencing is to stop the extension of the chain by introducing a terminal nucleotide (i.e. dideoxynucleotide) in the process of DNA synthesis, so as to obtain a series of DNA fragments with different lengths, and then separate these fragments by electrophoresis and read the sequence by autoradiography, and finally reconstruct the DNA sequence.

Technology:

Usually, DNA is extracted from B cells or hybridoma cells or RNA is inverted into cDNA, and specific short primers are designed to amplify the variable region genes (including heavy chain and light chain) of antibodies by PCR. Four common deoxynucleotides (dATP, dTTP, dCTP and dGTP) and a small amount of dideoxynucleotide (ddATP, ddTTP, ddCTP, ddGTP) need to be added for extension reaction. At each extension, the pre-added dideoxynucleotide will pair with the corresponding base in the template DNA, resulting in the termination of chain synthesis at a specific position. Because of the phenomenon of multiple chain termination, the PCR reaction will generate multiple DNA fragments with different lengths, and the terminal position of each fragment corresponds to the terminal base (ddNTP) at the corresponding position. The DNA fragments produced by the reaction were separated by Capillary electrophores. Because Capillary Electrophoresis separate DNA fragments according to their lengths, shorter fragments move faster and longer fragments move slower, so DNA fragments of various lengths can be separated. After separation, the terminal ddNTP of each fragment is different, and the fluorescence signal emitted by each terminated nucleotide (A, T, C or G) is different. The fluorescence signal is read by a detector and converted into a digital signal to generate an electrophoresis map. Each peak of the spectrum represents a termination position. By analyzing these data, the complete sequence of DNA fragments can be inferred.

Sanger sequencing and analysis (Singh L et al., 2022).

Advantages and disadvantages:

The advantage is high accuracy, and the general error rate is lower than 0.001%, which used to be the gold standard; Suitable for sequence determination of single antibody; Disadvantages are low throughput (the sequencing length of each reaction is short (usually 500-900 bases), and only one DNA fragment can be sequenced at a time), high sample starting amount, complicated operation, and difficulty in handling large-scale or complex samples (for longer sequences, it may be necessary to splice multiple fragments, which will become complicated when dealing with larger antibody genes).

• Monocytes/macrophages can express TCRαβ/γδ, and eosinophils can also express TCRγδ. Studies have shown that there are macrophage subsets that can express TCRαβ in chronic inflammatory diseases (such as tuberculosis and atherosclerosis) and tumor microenvironment. In order to verify its accuracy, the author used the peripheral blood mononuclear cell subsets of tumor patients and healthy people and the tumor microenvironment of five randomly selected human tumor entities to confirm that there is a second subgroup of macrophages that can express immunoglobulin (Ig) in vitro. The Ig(IgM, IgG and IgA) of macrophages and tumor macrophages were sequenced by Sanger sequencing and deep sequencing techniques, and it was confirmed that these Ig genes had V(D)J rearrangement, and their rearrangement and expression were individual-specific (for example, Ig heavy chains (such as IgG1, IgG3, IgM) and Igκ, Igλ light chains were found in TAM derived from melanoma). The use of Ig V chain and J chain in macrophages is biased and individual-specific. Sanger sequencing technology was used to study Ig CDR3 variants (IgM, IgG, Igκ and Igλ). It was found that the diversity of Ig in TAM was significantly lower than that in B cells of patients with normal or inflammatory diseases. Compared with the Ig library of B cells, the heavy chain and light chain variants of Ig in TAM showed some limitations, and specific Ig variants were shared among multiple TAM populations (Fuchs T et al., 2018).
• Mycobacterium tuberculosis is the pathogen of tuberculosis, and it has synergistic effect with other infectious factors such as HIV. The prevention and treatment of tuberculosis faces multiple challenges, including the increase of Mycobacterium in drug-resistant tb and the complicated immune escape mechanism. The researchers pointed out that the virulence of Mycobacterium tuberculosis is related to its ability to escape the host immune system, especially to inhibit the fusion of phagocytes and lysosomes in macrophages, thus avoiding degradation. Some virulence factors, such as KatG, SodA, PhoS1, GroES, etc., may become targets for diagnosis or treatment of tuberculosis, but they lack the sequence characteristics of IgV.In this study, the 5'-terminal rapid amplification (RACE) PCR technique was used to amplify the gene sequences of IgV IgVH and IgVL, and these sequences were analyzed in detail by Sanger sequencing and NGS, and the IGVs of homologous antibodies from hybridomas PhoS1/PstS1NRC-2410, SodANRC-13810, KatGNRC-49680 and GroESNRC-2894 were determined. By cloning each putative IgV sequence and transferring it to a common IgG2A κ–FC vector, the isotype transformation was carried out. IgVH and IgVL constructs were co-expressed in 293F cells and recognized glycosylation. Western blot and ELISA showed that both rAbNRC-2894 and hyAbNRC-2894 could recognize 10 kDa monomer and 25 kDa polymer complex related to GroES, and the recognition specificity and affinity of rAb and hyAb antigens were similar, which proved that rAb antigen recognition specificity and affinity, and recombinant IgV could recognize and bind Mycobacterium tuberculosis-related antigens (Foreman HC et al., 2021).

NGS

NGS refers to modern high-throughput sequencing technology, including Illumina, Ion Torrent, PacBio, Oxford Nanopore and other platforms, which can simultaneously sequence a large number of DNA or RNA samples in a short time, with high throughput, low cost and fast speed.

Technology:

RNA is extracted from B cells of target organisms (such as mice, humans or other animals). The extracted RNA is mainly mRNA, because it contains the coding information of antibody. Transcribe mRNA into cDNA by Reverse Transcription. The variable regions of antibody (including V, D and J regions of heavy chain and light chain) were amplified by specific primers. PCR and other techniques are used for amplification to ensure that enough antibody gene sequences are obtained for subsequent sequencing. The amplified antibody gene fragment needs to be transformed into a sequencing library suitable for NGS platform. The process of library construction includes connecting aptamers and adding sequencing primers. After the construction of the library, thousands of antibody genes were simultaneously sequenced using NGS platforms (such as Illumina, Ion Torrent, PacBio or Oxford Nanopore). The final sequencing data is usually composed of short sequences, and the complete antibody gene sequence needs to be restored by splicing, redundancy removal and other steps.

Hybridoma Ig sequencing using MiSeq Illumina NGS and de novo transcript assembly (Foreman HC et al., 2021).

Advantages and disadvantages:

The advantage is that the Qualcomm quantity can be used to sequence thousands of antibody genes in parallel in a short time; NGS can fully capture all antibody sequences produced by b cells; Be able to analyze the diversity of antibody library, including antibody affinity, affinity site, clone amplification and other information; Low-frequency antibody clones can be detected, and even rare antibody variations can be captured; Can reveal the dynamic changes of the overall immune response; The antibody gene sequencing at single cell level can be realized. The disadvantage is that the rearrangement and diversity of antibody genes are high, and the analysis and interpretation of data are very complicated; The cost is higher.

• Multiple myeloma (MM) is a malignant tumor that occurs in bone marrow and affects the function of plasma cells. The clinical features of MM include bone injury, anemia, high blood calcium level and renal failure, which seriously affect the health and quality of life of patients. Through targeted therapy, especially monoclonal antibody therapy (such as daratumumab and elotuzumab), the prognosis of MM patients has improved. In order to explore new therapeutic targets, the author combined with Phage Display and NGS to carry out experiments on antibody libraries produced by mice (PCL, MNC, PBMC) immunized with malignant plasma cells from patients. Through the screening of mouse immune library, it was found that after two rounds of panning, scFv phage antibodies mainly came from two subgroups: IGHV1(45.5%) and IGHV5(28.8%), and a small amount of antibodies came from IGHV14(19.7%). Most (80%) of antibodies from subgroup IGHV1 have CDRH3 sequences of 8 to 16 aa in length. In IGHV5 subgroup, the CDRH3 length of antibody is usually 10 and 16 aa, and one specific VH clone (CDRH3 length is 10 aa) is the most abundant in this group. Antibodies of subgroups IGHV1 and IGHV5 usually bind to corresponding VL chains (such as IGKV1 or IGKV6) and are paired based on specific VH-VL. However, some antibodies (such as #16 and #32) show great matching flexibility and may be paired with different VL chains (such as VL from IGKV1 or IGKV6). And all 9 antibodies showed the potential to bind to myeloma plasma cell antigen. The new antibody #5-CD38-IgG1 was found to have similar binding epitopes to those of known antibodies daratumumab and isatuximab, and #5-CD38-IgG1 could recognize CD38 on malignant plasma cells in bone marrow mononuclear cells (BM MNC) from three myeloma patients (Krohn S et al., 2022).
• NGS technology can quickly collect millions of antibody sequences from individuals, which increases the opportunity to provide alternatives for therapeutic antibody discovery. At present, the degree of similarity between antibodies from natural sources and antibodies developed for therapeutic purposes is not clear. In order to find the similarity between therapeutic antibodies and natural antibodies, the author established the Observed Antibody Space (OAS) database, which contains about 100 million heavy chain and 60 million light chain sequences from 60 independent studies (including different immune states) from various species (such as humans, mice, rhesus monkeys, rabbits, camels and rats). Comparing the matching degree between natural antibody sequence and CST antibody sequence in OAS database, about 37.1% of CST heavy chain and OAS sequence are more than or equal to 90%, while 65.2% of therapeutic light chain and OAS sequence are more than or equal to 90%. Among them, the identity of some CST heavy chains and light chains reached 95% or 100%. The CDR regions of CST heavy chain and light chain have high matching degree, especially in the CDR region of light chain. About 19% of CST heavy chain CDR and OAS have a matching degree of ≥90%, while 64.4% of CST light chain CDR and OAS have a matching degree of ≥90%. CDR-H3 is the part with the most diversity in sequence and structure in antibody, and it plays a key role in binding antigen. There are still some CDR-H3 sequences of CST that are completely matched in OAS. There are important similarities between CST and NGS sequences from natural sources, and paired NGS technology can provide a more comprehensive comparative perspective for natural sources and CST (Krawczyk K et al., 2019).
• NGS can be used for antibody screening. Based on phage display technology, a semisynthetic library of scFv Gen3 was constructed, including three different novel coronavirus target proteins: spinous trimer, its monomer S1 and receptor binding domain (RBD). NGS was used to analyze the antibody diversity and affinity of different populations. The affinity range and folding enrichment of antibody were analyzed by stacking bar graph. At the concentration of 1 nM antigen sorting, the percentage of conjugate is directly proportional to the relative NGS abundance, and it shows an increasing enrichment trend from 10 nM to 1 nM. At very low abundance (such as < 0.001%) or multiple enrichment (< 0.01 times), antibodies with sub-nano molar affinity are still found in these populations. In RBD (receptor binding domain) population, 30 kinds of antibodies with affinity ≤100 pM were found, and there was no significant correlation between the abundance and affinity of these antibodies. The proportion of antibodies with abundance less than 0.001% is 10%, 37%, 57% and 90% respectively, and most of the antibodies with low abundance also have sub-nano molar affinity. It is revealed that antibodies in different HCDR3 clusters usually show similar binding patterns, while antibodies in different LCDR3 clusters may show more dispersed binding kinetics.By measuring different CDR regions in antibody population (such as HCDR3, HCDR3+LCDR3, tandem CDR, etc.), the relationship between relative abundance and enrichment and antibody affinity was obtained (Erasmus MF et al., 2023).

Phip-Seq

Phip-Seq (PHIGE Immunoprecisions Sequencing) is a technique for large-scale screening and recognition of antibodies binding to specific antigens. This method combines phage display technology and NGS, and is widely used in antibody identification, antibody-antigen interaction research and the discovery of new antibodies.

Basic process:

(1) Construct a phage display library containing thousands of different peptide sequences (different peptide segments of target antigen or random peptides with specific binding ability to antigen). (2) The solution containing phage library is mixed with immune samples (such as antibodies from serum). Antibodies will combine with peptides on the surface of phage to form antibody-phage complexes. Then, the antibody-phage complex is separated from the solution by immunoprecipitation technique (such as magnetic particle binding antibody).(3) By washing for many times, the phage that is not specifically bound is removed, and only the phage that is bound to the target antibody is retained. (4) Finally, NGS was performed on the enriched phage to identify the peptide sequence bound to the antibody. By analyzing these peptide sequences, we can identify the antigenic peptide segments that bind specifically to antibodies.

Advantages:

Thousands of different peptides can be screened at the same time, and a small amount of specific antibodies can be recognized. Without purifying antibodies, immune samples are directly used for analysis, and different types of antibodies, including IgM, IgG and other subtypes, can be recognized.

• The screening and identification of antibodies against COVID-19 caused by SARS-CoV-2 is complicated. The serum of convalescent patients is an important source of antibodies, but the number of antigen-specific B cells in circulating blood is very small, so it is difficult to isolate and identify these antibodies. In order to identify the effective antibody against Covid-19, the author studied the plasma samples of volunteers by FACS, scRNA-seq and PhIP-seq. Using Phage1 (including full-size RBD and its immunodominant fragment), the enriched immune library was analyzed, and the dominant VH-VL combinations were screened out. Some antibodies have high affinity in different SARS-CoV-2 variants (such as Wuhan, α, γ, ω), but low affinity in the variants. According to IC50 value and neutralization effect, the antibody can be divided into four types: weak (IC50 > 100nM), mild (20 < IC50 < 100nM), moderate (1 < IC50 < 20nM) and strong (IC50 < 1nM). Sixteen functional VH-VL combinations were found by NGS analysis, and these combinations came from different heavy chains and light chains (Lomakin YA et al., 2024).

TGS

Third-generation sequencing (TGS) is a kind of technology based on single-molecule DNA sequencing, and its representative platforms are PacBio and Oxford Nanopore. Compared with the traditional second-generation sequencing (such as Illumina), TGS is characterized by its ability to directly read single-molecule DNA without PCR amplification, longer reading length and higher flexibility, and is suitable for complex genomes, structural variations, long fragments and RNA sequencing. TGS mainly relies on the principle of single molecule sequencing, and reads bases in real time through the interaction between single molecule DNA template and polymerase.

• Single molecule real-time sequencing (SMRT): This technology was proposed by PacBio. Its principle is based on the amplification of single-molecule DNA template, which uses polymerase to distinguish different bases by the fluorescence signal released when each fluorescently labeled dNTP is added in the synthesis process. Each dNTP(A, T, C, G) has a different fluorescent label, and the released signal is monitored in real time by the sensor and the corresponding base is identified.

• Nanopore sequencing: Oxford nanopore technology uses a nanopore sensor. When a single molecule of DNA passes through a nanopore, the interaction between DNA and the pore wall will affect the current passing through the pore, and this current change can be monitored in real time and decoded into a base sequence. The core of this technology is to identify DNA sequences through the current change of nano-holes.

Technology:

Obtain the cDNA sequence of antibody gene (such as monoclonal antibody or B cell antibody gene). It is the core of TGS to melt the double-stranded cDNA by chemical method or enzymatic method to obtain single-stranded DNA, and combine the single-stranded DNA with polymerase, which will catalyze the addition of dNTP and complete the extension of DNA. Put the processed single-stranded DNA into sequencing chip or nanopore. For PacBio, single-molecule DNA and polymerase are immobilized in micropores, and real-time sequence reading is carried out under the drive of fluorescence signal; For Oxford Nanopore technology, single-molecule DNA is decoded through nanopores accompanied by current changes.

In the process of sequencing, the fluorescence signal or current change is monitored in real time, and the signal is transmitted to the computer system, and the DNA sequence is analyzed and resolved in real time through a specific algorithm. PacBio platform generally uses special software (such as SMRT Analysis) for data processing, while Nanopore uses MinION's analysis software for signal decoding. After sequencing, all the signal data will be converted into DNA sequences. Because TGS can directly read long fragments of DNA, it can provide higher quality data for downstream genome assembly and structural variation analysis.

Advantages and disadvantages:

The advantage is Long Reads; Direct sequencing with single-molecule DNA does not require PCR amplification process, thus avoiding the problems of amplification deviation, amplification product pollution and amplification failure introduced by PCR. Detection of complex structural variations, such as large-scale insertion, deletion, inversion, etc. Real-time sequencing and fast feedback. The disadvantage is that the error rate is high (for example, on PacBio platform, the error rate of base recognition may sometimes be as high as 15%); The cost is high and the technology is not mature enough.

• T1DM (caused by pancreatic β-cell injury) and T2DM (caused by insulin resistance) are very common endocrine diseases. Early diagnosis and classification of type I diabetes are very important for treatment and prognosis, and its diagnosis depends on detecting a series of autoimmune antibodies against pancreatic β-cell secretory proteins, especially GADAb. At present, the detection of GADAb depends on traditional immunological methods, such as ELISA, RIA and immunofluorescence, but its limitations are too great. The translocation behavior of GAD65, GADAb and their immune complex (GAD65-GADAb complex) was studied by Shi Ying glass nanopore technique. Preparation of 30-50 nm glass capillary solid-state nanopores, through the current signal analysis of the mixed solution of GAD65 and GADAb, it was found that there were characteristic signal peaks in the nanopores under different voltages (300 mV to 500 mV and below 300 mV), which could identify different signal patterns of monoclonal antibodies and polyclonal antibodies. In the low voltage (200 mV), the types of antibodies in GADAb-poly were preliminarily identified, which proved that there were at least two. It was also found that by mixing GAD65 with GADAb-mono and using nanopore for detection, three different protein binding states can be directly identified without modifying or labeling the sample (Tao C et al., 2024).
• The genome similarity between macaques and humans makes them an important object of HIV infection research, especially in the study of VH and VL domains of antibodies. It is found that macaque antibodies target the envelope glycoprotein (Env) of HIV in many cases, but these antibodies are not always effective in neutralizing the virus. In order to explore the potential of broadly neutralizing antibodies (bnAbs) and non-neutralizing antibodies (nnAbs) produced by vaccination stimulation in preventing HIV infection, the phage display library extracted from rhesus monkeys infected with SIV was determined by NGS and PacBio. 32 unique single chain antibody (scFv) sequences were screened by SIV Env gp140 antigen. The NGS and SMART of Pan0 and Pan4 libraries show that there are many representative genes of IGHV gene subgroups in Pan0 data set (such as IGHV1, IGHV4, IGHV7, etc.), and the diversity of IGHD genes is high (such as IGHD1, IGHD2, igd3, etc.). The data set of Pan4 shows different compositions of IGHV genes, especially IGHV3 is dominant, and the sequence number of IGHD2-2 gene is overwhelming. In the Pan0 library, the length of VH CDR3 ranged from 5 to 31 AA, and most of them (78.29%) were 17 AA or less. In Pan4 database, the length of VH CDR3 ranges from 9 to 25 AA, of which the most common length is 20 AA, accounting for 82.04% of the total sequence. There are 69 different IGKV genes in Pan0, and the most abundant ones are IGKV3-8 and IGKV3-9.In Pan4, the use of IGKV gene is not as diverse as that of Pan0. The length of V-κ CDR3 in Pan0 library is mostly 9 and 8 AA (Han SY et al., 2018).
• Phage display technology combined with NGS has limitations in antibody screening and discovery. By combining SMRT sequencing and phage display technology, functional scFv antibodies against CD160 and CD123 were quickly isolated from the library within a few days. In CD160, a total of 9434, 9336, 22660 and 14368 function readings were obtained from the sequence library of pan 0, 1, 2 and 3, respectively. The cloning of the library was gradually enhanced, and after the third round of screening, the clones in the library began to show high aggregation. In the initial stage, CDR length was widely distributed, but in pan 3, the proportion of 12 AA HCDR3 sequence clones increased greatly. The analysis of HCDR3 region shows that some amino acid sequences are biased in the process of library screening. PacBio sequencing technology was used to analyze the antibody library produced by rats immunized with targeted antigen CD123, and the clones with specific binding to CD123 positive cells were identified. IGHV1 and IGHV5 are the most abundant v- genes, but in IGK, the use of V genes is more diversified, and some specific genes, such as IGKV4 and IGKV22, occupy the main part of the library. Thirty-seven heavy chain clones were selected, and three different light chain candidates were selected for each heavy chain clone. The clones were tested by transient expression and binding to cells expressing CD123, and it was found that 30 clones had specific binding. Among 24 clones selected by NGS, about 12 clones have functional light chains, and in most clones, the most common light chain produces functional scFv (single chain antibody) (Nannini F et al., 2021).

Summary Table of Antibody Sequencing Methods

References

1. Fuchs T, Hahn M, Ries L, Giesler S, Busch S, Wang C, Han J, Schulze TJ, Puellmann K, Beham AW, Kaminski WE, Neumaier M. "Expression of combinatorial immunoglobulins in macrophages in the tumor microenvironment." PLoS One. 2018;13(9):e0204108.doi: 10.1371/journal.pone.0204108
2. Foreman HC, Frank A, Stedman TT. "Determination of variable region sequences from hybridoma immunoglobulins that target Mycobacterium tuberculosis virulence factors." PLoS One. 2021 ;16(8):e0256079. doi: 10.1371/journal.pone.0256079
3. Krohn S, Boje AS, Gehlert CL, Lutz S, Darzentas N, Knecht H, Herrmann D, Brüggemann M, Scheidig AJ, Weisel K, Gramatzki M, Peipp M, Klausz K. "Identification of New Antibodies Targeting Malignant Plasma Cells for Immunotherapy by Next-Generation Sequencing-Assisted Phage Display." Front Immunol. 2022;13:908093. doi: 10.3389/fimmu.2022.908093
4. Krawczyk K, Raybould MIJ, Kovaltsuk A, Deane CM. "Looking for therapeutic antibodies in next-generation sequencing repositories." MAbs. 2019 ;11(7):1197-1205. doi: 10.1080/19420862.2019.1633884
5. Lomakin YA, Ovchinnikova LA, Terekhov SS, Dzhelad SS, Yaroshevich I, Mamedov I, Smirnova A, Grigoreva T, Eliseev IE, Filimonova IN, Mokrushina YA, Abrikosova V, Rubtsova MP, Kostin NN, Simonova MA, Bobik TV, Aleshenko NL, Alekhin AI, Boitsov VM, Zhang H, Smirnov IV, Rubtsov YP, Gabibov AG. "Two-dimensional high-throughput on-cell screening of immunoglobulins against broad antigen repertoires." Commun Biol. 2024 ;7(1):842. doi: 10.1038/s42003-024-06500-2
6. Erasmus MF, Ferrara F, D'Angelo S, Spector L, Leal-Lopes C, Teixeira AA, Sørensen J, Nagpal S, Perea-Schmittle K, Choudhary A, Honnen W, Calianese D, Antonio Rodriguez Carnero L, Cocklin S, Greiff V, Pinter A, Bradbury ARM. "Insights into next generation sequencing guided antibody selection strategies." Sci Rep. 2023;13(1):18370. doi: 10.1038/s41598-023-45538-w
7. Tao C, Bai Y, Chen J, Lu J, Bi Y, Li J. "Detection of Glutamate Decarboxylase Antibodies and Simultaneous Multi-Molecular Translocation Exploration by Glass Nanopores."Biosensors (Basel). 2024;14(5):255. doi: 10.3390/bios14050255
8. Han SY, Antoine A, Howard D, Chang B, Chang WS, Slein M, Deikus G, Kossida S, Duroux P, Lefranc MP, Sebra RP, Smith ML, Fofana IBF. "Coupling of Single Molecule, Long Read Sequencing with IMGT/HighV-QUEST Analysis Expedites Identification of SIV gp140-Specific Antibodies from scFv Phage Display Libraries." Front Immunol. 2018 ;9:329. doi: 10.3389/fimmu.2018.00329
9. Nannini F, Senicar L, Parekh F, Kong KJ, Kinna A, Bughda R, Sillibourne J, Hu X, Ma B, Bai Y, Ferrari M, Pule MA, Onuoha SC. "Combining phage display with SMRTbell next-generation sequencing for the rapid discovery of functional scFv fragments." MAbs. 2021;13(1):1864084. doi: 10.1080/19420862.2020.1864084
10. Brochu HN, Tseng E, Smith E, Thomas MJ, Jones AM, Diveley KR, Law L, Hansen SG, Picker LJ, Gale M Jr, Peng X. "Systematic Profiling of Full-Length Ig and TCR Repertoire Diversity in Rhesus Macaque through Long Read Transcriptome Sequencing." J Immunol. 2020 ;204(12):3434-3444. doi: 10.4049/jimmunol.1901256