In vivoaffinity maturation of murine B cells reprogrammed to express human antibodies

Abstract

CRISPR-edited murine B cells engineered to express human antibody variable chains proliferate, class switch, and secrete these antibodies in vaccinated mice. However, current strategies disrupt the heavy-chain locus, resulting in inefficient somatic hypermutation without functional affinity maturation. Here we show that recombined murine heavy- and kappa-variable genes can be directly and simultaneously overwritten, using Cas12a-mediated cuts at their 3'-most J segments and 5' homology arms complementary to distal V segments. Cells edited in this way to express the HIV-1 broadly neutralizing antibodies 10-1074 or VRC26.25-y robustly hypermutated and generated potent neutralizing plasma in vaccinated recipient mice. 10-1074 variants isolated from these mice bound and neutralized HIV-1 envelope glycoprotein more efficiently than wild-type 10-1074 while maintaining or improving its already low polyreactivity and long in vivo half-life. We further validated this approach by generating substantially broader and more potent variants of the anti-SARS-CoV-2 antibodies ZCB11 and S309. Thus, B cells edited at their native loci affinity mature, facilitating development of broad, potent, and bioavailable antibodies and expanding the potential applications of engineered B cells.

In vivoaffinity maturation of the HIV-1 Env-binding domain of CD4

“…We have recently shown that when heavy- and light-chain sequences of HIV-1 and SARS-CoV-2 neutralizing antibodies were introduced at the V(D)J-recombined regions of their respective mouse loci, the resulting BCR affinity matured after immunization 18, 32 . However, many protein-based therapeutics are not antibodies.…”

“…1c and d). This efficiency is higher than necessary for expansion and affinity maturation of engineered B cells 16,18,32 . Thus, primary murine B cells can be engineered ex vivo to efficiently express D1D2-OKT3-V H .…”

“…We have previously shown that human heavy- and light-chain variable regions can directly replace their murine counterparts in primary mature B cells, a technique we describe as ‘native-loci’ editing 18 . When these CRISPR/Cas12a-engineered cells were adoptively transferred to recipient mice, their B-cell receptors (BCRs) affinity matured in response to antigen, facilitating identification of broader, more potent, and more bioavailable forms of the original human antibody.…”

In vivo affinity maturation of the HIV-1 Env-binding domain of CD4

“…We have shown that introduction of human antibody heavy chain complementarity determining regions (HCDR3s) or heavy-and light-chain variable genes into their respective murine loci enabled in vivo affinity maturation of the resulting chimeric BCRs 18,32 . Extension of this approach to immunoadhesins such as CD4-Ig is complicated by the absence of the first heavy constant region 1 (CH1) that noncovalently associates with a light-chain constant domain.…”

“…1b). Accordingly, we introduced a double-stranded break by CRISPR/Mb2Cas12a ribonucleoproteins (RNPs) into the 3'-most JH segment, JH4 in mice 18,32,33 . Repair of this break was guided by a homology-directed repair template (HDRT) delivered by recombinant adeno-associated virus DJ (AAV-DJ).…”