Mouse models with human antibody repertoires for inducing multiple lineages of HIV-1 broadly neutralizing antibodies

The variable regions of antibody heavy chains (HCs) and light chains (LCs) are assembled by V(D)J recombination in progenitor B cells to generate an immense repertoire of primary B cell receptors (BCRs), the precursors of affinity-matured antibodies secreted in response to antigen stimulation. The complementarity determining region (CDR) 1, 2 and 3 of antibodies are the principal antigen contact sites, with CDR3 being highly diverse due to V(D)J junctional diversification by terminal deoxynucleotidyl transferase (TdT). The HC CDR3 (CDR H3) plays a prominent role in broadly neutralizing antibodies (bnAbs) against the human immunodeficiency virus-1 (HIV-1). BnAbs overcome the genetic heterogeneity of HIV-1 by recognizing conserved epitopes on the HIV-1 Envelope (Env) protein. Reaching these targets requires long CDR H3s that penetrate through the glycan shield or other structural hindrances on the Env protein. The shortage of human antibodies with such long CDR H3s poses a challenge for bnAb elicitation by vaccination. To aid immunogen design, we generated six mouse models for inducing bnAbs against particular HIV-1 Env epitopes. In each mouse model, we integrated the human HC VH, D, JH segments and LC VL, JL segments of a bnAb lineage into the mouse HC and LC loci, with each set engineered to undergo V(D)J recombination and to generate diverse human HC and LC variable regions. Combined action of V(D)J recombination and TdT- mediated junctional diversification in developing B cells generated a range of human variable region exons for a given bnAb lineage that contained highly diverse CDR3s in each mouse model. Moreover, these repertoires contained humanized antibodies that had bnAb-like long CDR H3s that could potentially serve as bnAb precursors. Therefore, these mouse models can be used to test whether immunogens can induce bnAbs from rare and diverse precursors in a complex antibody repertoire. Author summaryThe human immunodeficiency virus-1 (HIV-1) is the causative agent of acquired immunodeficiency syndrome (AIDS). An efficacious HIV-1 vaccine is needed to control the AIDS pandemic. However, in multiple clinical trials, vaccine candidates failed to confer protection against HIV-1 infection. The lack of efficacy is mainly due to the enormous heterogeneity of HIV-1 strains in human circulation. A breakthrough in the field has been the identification of broadly neutralizing antibodies (bnAbs) in a small fraction of HIV-1 infected patients. Because these antibodies recognize conserved targets on different HIV-1 strains, they can inhibit a wide spectrum of viruses. Eliciting HIV-1 bnAbs is a top priority for vaccine development. For this endeavor, a major difficulty is that most bnAbs have unusual properties. To induce bnAbs, vaccines must be highly selective for rare human antibodies that can develop into bnAbs. To facilitate this effort, we have generated a panel of mouse models that can produce potential precursors for major types of HIV-1 bnAbs. We engineered mouse models to produce diverse precursors in complex antibody repertoires, which mimic the challenging condition in human vaccination. These mouse models can be used to assess and optimize vaccine candidates at the preclinical stage.