Maturation of HIV-1 neutralizing antibodies in a germinal center conditional expression mouse model

1. Problem Statement

Developing an HIV-1 vaccine that elicits broadly neutralizing antibodies (bnAbs) is a major challenge. While bnAbs exist in some infected individuals, they typically require years of "affinity maturation" within germinal centers (GCs) to evolve from germline precursors into potent, broad neutralizers.

• The Bottleneck: This process often stalls at specific "rate-limiting" steps involving rare, functionally critical somatic hypermutations (SHMs), such as insertions/deletions (indels) or mutations in "cold spots" of the SHM process.
• The VRC01 Example: VRC01-class antibodies target the CD4 binding site (CD4bs) of the HIV-1 Envelope (Env). A major roadblock is the N276 glycan on the Env, which sterically hinders antibody access. Overcoming this requires specific mutations that are difficult to select for using standard prime-boost strategies starting from germline precursors.
• Limitations of Current Models: Existing mouse models that constitutively express bnAb precursors often fail because:
  1. Precursors may be deleted by central tolerance checkpoints due to autoreactivity.
  2. They express the antibody in naive B cells, which respond differently to immunogens than the GC/memory B cells that are the physiological targets for boost immunogens.
  3. They require multiple immunization steps to reach the intermediate stage, introducing variability and asynchrony.

2. Methodology

The authors developed a novel Germinal Center (GC) Conditional Expression Mouse Model to bypass these limitations and test boost immunogens directly on stalled intermediates.

• Genetic Engineering:
  • Conditional Cassette: They constructed heavy chain (HC) and light chain (LC) expression cassettes containing two tandem variable region exons: a proximal Germline (GL) precursor and a distal Intermediate (IA) antibody (specifically the 1538-79 VRC01 intermediate).
  • Switching Mechanism: The GL exon and polyadenylation signals were flanked by loxP sites. In the presence of Cre recombinase, the GL exon is deleted, allowing expression of the distal IA exon.
  • GC-Specific Cre: A streamlined mini-AID-cre transgene was integrated into the IgH locus. This transgene is expressed only in GC B cells (driven by Aicda regulatory elements), ensuring the switch from GL to IA occurs specifically within the GC.
  • Breeding: The HC cassette and mini-AID-cre were linked on one allele; the LC cassette was on a separate allele. Breeding these lines created the complete model.
• Experimental Strategy:
  • Adoptive Transfer: To mimic physiological precursor frequencies, splenic B cells from the GC model were transferred into wild-type F1 recipient mice.
  • Prime Immunization: Mice were immunized with CH505.core (a modified gp120 lacking N463 glycan but retaining N276). This antigen binds the GL precursor to activate B cells but has higher affinity for the IA intermediate, driving the selection of cells that have switched to IA expression.
  • Boost Immunization:
    • Boost 1: CH505.gp120.v2 (contains N276 glycan but lacks N463). This acts as a "bridge" to select for mutations that begin to accommodate the N276 glycan.
    • Boost 2: CH505.gp120 (wild-type with both N276 and N463 glycans). This applies full selective pressure to overcome the N276 barrier.
  • Analysis: Single-cell RT-PCR, sequencing, ELISA, and neutralization assays (TZM-bl assay against 119 HIV-1 pseudoviruses) were performed on sorted GC B cells and recombinant antibodies.

3. Key Contributions

1. Novel Mouse Model: The first model to conditionally express a specific bnAb intermediate only in GC B cells. This bypasses central tolerance deletion and ensures the antibody is tested in the correct physiological context (GC B cells competing with other GC B cells).
2. Bridge Immunogen Strategy: The development of a "bridge" immunogen (CH505.gp120.v2) that removes a non-obstructive glycan (N463) to allow initial engagement of the stalled intermediate, while retaining the obstructive glycan (N276) to drive the necessary maturation mutations.
3. Proof of Concept for Roadblock Bypass: Demonstrating that a stalled VRC01 intermediate can be matured beyond the N276 glycan barrier through targeted boost immunization, resulting in heterologous neutralization.

4. Key Results

• Model Validation: The model successfully expressed the GL-VRC01 precursor in naive B cells. Upon immunization and GC formation, Cre-mediated recombination occurred, and the B cells switched to express the IA-VRC01 intermediate.
• Successful Maturation:
  • After the boost regimen, the evolved antibodies (IA-VRC01.v2) acquired significant somatic mutations, including point mutations, CDR H3 insertions (e.g., "GT" or "GN"), and CDR L1 deletions.
  • A specific combination of mutations was identified as critical for broad binding: G56A (a canonical VRC01-class mutation), CDR H3 insertion, and CDR L1 deletion.
• Enhanced Binding and Neutralization:
  • Binding: The top evolved antibodies (e.g., 8g, 26g, 41g) showed significantly improved binding to diverse Env trimers compared to the original IA-VRC01, approaching the breadth of the mature VRC01 bnAb.
  • Neutralization Breadth: The original IA-VRC01 neutralized only 1.7% of a 119-strain panel. The evolved IA-VRC01.v2 antibodies neutralized 16.8% to 31.1% of the panel.
  • N276 Accommodation: Crucially, the evolved antibodies successfully neutralized viruses containing the N276 glycan, a feat the original intermediate could not achieve.
  • Potency: While the breadth improved dramatically, the potency (IC50) remained in the µg/ml range (lower than mature VRC01), indicating further maturation is needed.

5. Significance

• Vaccine Design Tool: This study provides a powerful platform for testing and refining boost immunogens. Instead of waiting for rare, stochastic mutations to occur in a germline-targeting scheme, researchers can start with a known "stalled" intermediate and empirically determine which immunogens can drive the specific mutations needed to overcome roadblocks.
• Overcoming Glycan Barriers: The success in maturing antibodies against the N276 glycan barrier suggests that similar "bridge" strategies (modifying non-critical glycans to allow engagement while retaining critical barriers) could be applied to other bnAb lineages.
• General Applicability: The GC conditional expression system is not limited to VRC01; it can be adapted for any bnAb lineage where development is stalled at a specific intermediate, offering a general solution to the "roadblock" problem in HIV vaccine design.
• Complementary Approach: The authors propose that these GC models (testing intermediates) should be used in tandem with germline-targeting models (testing precursors) to create a complete, multi-stage vaccine regimen.