henhaobio
Sign in

bioRxivImmunologyPreprint

Novel human monoclonal antibodies with enhanced sensitivity for lipoarabinomannan antigens present in urines of TB patients

Alok Kumar ChoudharyiD, Deendayal PateliD, William J. HonneniD, Afsal KolloliiD, Charles ReichmaniD, Kamaljit KauriD, Ruixiang Blake ZhengiD, Elizabeth NakabugoiD, Emmanuel NasingheiD, Lydia NakiyingiiD, Todd L. LowaryiD, Abraham PinteriD

PDFDOI
In-depth full-text digest available · Jump to digest ↓

Abstract

Lipoarabinomannan (LAM) is a useful biomarker for detection of M. tuberculosis infection and disease. Related antigens can be detected in urine samples of TB patients by combinations of monoclonal antibodies (mAbs) directed against specific epitopes expressed in LAM. While sensitive for samples from patients with active TB disease who have HIV-1 co-infections, these assays are less effective for other populations, and there is therefore a need for more sensitive antibodies that can improve the sensitivity of these assays. Here we characterize the antigen and epitope specificities, sequence diversity and isotype dependencies of eight LAM-specific human mAbs that target five distinct arabinose- and mannose-dependent epitopes present in LAM and lipoarabinomannan (LM). Whereas all of the mAbs recognized ManLAM, only a few, including A194-01, consistently detected antigens in TB+ urine samples. Converting A194-01 from the IgG1 to the IgM isotype resulted in broader recognition of poly-Ara glycan epitopes, and increased sensitivity for clinical antigens when combined with several capture reagents, including RU95-C1, a novel antibody targeting the mannan domain of LAM. These results define novel epitopes that are differentially expressed in bacterial and urinary forms of LAM, and identify novel antibody combinations which possess enhanced diagnostic utility for clinical forms of LAM.

In-depth summary

Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (27 Jul 2026) · CC BY 4.0

Converting human antibody A194-01 from IgG1 to IgM broadens glycan recognition and boosts urinary lipoarabinomannan detection, and RU95-C1, a mannan-targeting antibody, works as a capture partner.

Urine LAM tests work for HIV-1 co-infected TB patients but are too insensitive for other populations, and recent efforts have not found antibodies that significantly improve them. This work characterizes eight human antibodies, six of them new, and tests their pairings against clinical urine samples.

Key results

  1. SuggestedIn capture assays with S4-20 or RU95-C1, the IgM form of A194-01 gave stronger signals than IgG for ManLAM and five of six TB-positive urines, while healthy control urines stayed near background.

    IgM stronger for ManLAM and five of six TB/HIV-1-positive Ugandan urines. Signal enhancement for 1:2 diluted urines ranged from 7.8 to 53.5, highest for samples with weak IgG signal. Seven negative control urines gave very low background, similar to blocking reagent. · 7. The A194-01 IgM isoform possesses enhanced…; Figure legends

  2. ShownAntibody valency drives A194-01 affinity. Competition IC50-type values fell steadily from monovalent to decavalent forms, and IgG conversion of the natural IgMs P30B9 and RU61-H5 abolished binding.

    Relative affinities against ManLAM: Fab 303 nM, scFv 82 nM, scFv dimer 3.1 nM, IgG 2.8 nM, tetrameric IgG-scFv 1.3 nM, IgM 0.6 nM. IgM, but not IgG, competed with FIND170 IgG and P30B9 IgM. IgG forms of P30B9 and RU61-H5 lost all binding. · 6. Effect of A194-01 isotype on binding…; 2c. Role of antibody isotype in antibody…

  3. SuggestedMost antibody pairs detected purified ManLAM, but few detected clinical urinary LAM. A194-01 IgM as detector gave the strongest urine signals with A194-01 or RU83-A8 capture; S4-20 and RU95-C1 capture also detected several urines.

    44/81 combinations gave S/N >100 for ManLAM, but only 6/81 gave S/N >10 for most of five urines. S4-20/A194-01 IgM: S/N 18,476 and 900 in two urines. RU95-C1/A194-01 IgM: S/N 61-484 in 4/5 urines. · 8. Immunoreactivity of human mAb panel against…; Discussion

  4. SuggestedGlycan array and blot data grouped the eight mAbs into four epitope classes. RU95-C1 appears to target an α-(1→6)-linked trimannose in the mannan core shared by LAM and LM, with no binding to mannose-cap structures.

    A194-01, RU83-A8, RU49-01 and RU49-02 recognized mannose-free polyarabinose; RU61-H5, RU28-01, P30B9 and RU95-C1 required mannose. RU95-C1 and RU61-H5 bound ManLAM, LM and the Endo-MA1 mannan fragment; an α-(1→2) Man branch on the middle Man blocked RU95-C1. · 1. Antigen specificity of mAbs; 2a. Epitope specificities of mAbs; 2b. Fine specificities of four mannose-dependent…; Discussion

  5. SuggestedUnusual HCDR2 insertions in the mannose-cap mAbs P30B9 and RU28-01 were needed for full function in germline reversion tests, and A194-01 and RU83-A8 share germline genes, which the authors read as a convergent response.

    P30B9 germline reversion lost ManLAM binding; GL + 6aa insert fully restored it. RU28-01 GL + 5aa insert partially restored activity. IGHV mutation 6.3% to 20%, IGVL 2.3% to 13.7%. A194-01 and RU83-A8 share IGHV3-20*04 and IGHJ6*03. · 3. Sequence analyses of anti-LAM mAbs; 4. Role of hypermutations and HCDR2 insertions…; 5. Functional relationship between A194-01 and…

Methods, data and reproducibility

Human anti-LAM mAbs from memory B cells were profiled by ELISA, Western blot and glycan microarray, then by germline reversion, chimera, isotype and valency variants. Pairs were tested in chemiluminescent sandwich immunoassays on mycobacterial LAM and clinical urines.

  • Eight human anti-LAM mAbs cloned from memory B cells of TB-infected individuals; variable region sequences deposited · human · n = 8 mAbs · GenBank MC402173–MC402178
  • Purified lipoglycans from BEI Resources: ManLAM (M. tuberculosis), PILAM (M. smegmatis), LepLAM (M. leprae), LM (M. tuberculosis) · M. tuberculosis, M. smegmatis, M. leprae
  • Microarray of synthetic mycobacterial glycans conjugated to BSA for epitope mapping · n = 61 glycans in Results; 63 in Methods
  • Urine from TB-positive, HIV-1 co-infected adults at Makerere University, Kampala, plus negative control urines from individuals without tuberculosis · human · n = six positive urines (Fig. 6a); five (checkerboard)
  • Urine from TB patients and healthy volunteers collected in New Jersey (Rutgers); plasma and PBMC used for B-cell antibody isolation · human
Methods & tools
Direct and chemiluminescent ELISA, Sandwich ELISA with ELISA Bright substrate, Western blotting, Synthetic glycoconjugate microarray, Endo-MA1 arabinofuranosidase digestion, IMGT germline assignment, Clustal Omega, MEGA X (Neighbor-Joining, 50 bootstraps), GenePix Pro 7.3.0.0, S4-20 capture mAb (FujiLAM-type pairing comparator)
Reagents & environment
GenePix Pro 7.3.0.0 for microarray image analysis; MEGA X for phylogenetic trees; Clustal Omega alignment; IMGT database for germline assignment.
Reproducibility gaps
No analysis code is described. The article says supporting data are in the paper and its Supplement, with variable region sequences in GenBank. A patent on some of the mAbs is noted in the competing interests.
Limitations (3)
  • Antibody pair specificity was not addressed; the authors say larger clinical panels are needed to screen the new combinations. · stated by the authors
  • The authors describe the experiments as exploratory and comparative, in line with early-stage diagnostic development, so the urine findings are preliminary. · noted in this summary
  • Clinical evidence rests on very small urine panels (six positive for IgM versus IgG, five for the checkerboard) with no formal statistics, and there is no comparison with a deployed format such as lateral flow with silver amplification. · noted in this summary

Read this if you develop immunoassays for urinary lipoarabinomannan and want epitope-level data on antibody isotype, valency and capture/detector pairing.

Citations
0

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISCantera '22Cantera '21Mukundan '20Singh '26Grace '25Panda '23
85% semantic matchMicrobiology
J. L. Cantera, L. M. Lillis, R. B. Peck+15 more·bioRxiv·Apr 12, 2022