# Deep Learning-based Protein Design of HIV-1 Env GP120 Core Immunogens for CD4 Binding Site Germline Targeting

> **NIH NIH R01** · FRED HUTCHINSON CANCER CENTER · 2024 · $832,117

## Abstract

PROJECT SUMMARY/ABSTRACT
This research proposal aims to develop new immunogens for an HIV-1 vaccine by utilizing advanced protein design
techniques and deep learning methods. Conventional structure-guided approaches have limitations in achieving
desired structural characteristics. Therefore, this study proposes using RFdiffusion, ProteinMPNN, and AlphaFold2,
to generate new germline-targeting gp120 cores based on the pre-fusion native-like structure of the HIV-1 strain
426c. The designed immunogens prioritize maintaining the structural integrity of the pre-fusion gp120 while
removing the bridging sheet and re-designing specific regions to maintain the pre-fusion native-like backbone
structure. In addition, particular attention is directed towards masking off-target epitopes. In vitro characterization
will be performed to evaluate the binding characteristics of these immunogens with germline, broadly neutralizing,
and non-neutralizing antibodies. Subsequently, transgenic mice expressing human germline VRC01-class BCRs
will be utilized to assess the immunogenicity of selected immunogens presented on nanoparticles and analyze their
impact on germinal center B cell responses and memory B cell repertoire. Monoclonal antibodies (mAbs) obtained
from immunized animals will be structurally characterized, shedding light on antibody maturation pathways
influenced by the immunogen structure. Additionally, this research plan aims to expand the immunogen repertoire
by designing gp120 cores based from diverse HIV-1 clades/strains. Furthermore, to enhance immunogenicity,
membrane-bound and nanoparticles delivery through self-amplifying mRNA will be explored. The ultimate objective
of this research is to gain valuable insights into novel antigen design techniques and their application in HIV-1
vaccine strategies for broadly neutralizing antibodies development. The findings from this study will contribute to
the development of immunogens that closely resemble the pre-fusion gp120 state, potentially leading to enhanced
B-cell responses capable of generating broadly neutralizing antibody lineages. By addressing the complex features
of the HIV-1 Env protein and advancing immunogen design strategies, this research aims to make significant
contributions towards the development of an effective HIV-1 vaccine.

## Key facts

- **NIH application ID:** 10897643
- **Project number:** 1R01AI183406-01
- **Recipient organization:** FRED HUTCHINSON CANCER CENTER
- **Principal Investigator:** Yoann Aldon
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $832,117
- **Award type:** 1
- **Project period:** 2024-02-16 → 2028-01-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10897643

## Citation

> US National Institutes of Health, RePORTER application 10897643, Deep Learning-based Protein Design of HIV-1 Env GP120 Core Immunogens for CD4 Binding Site Germline Targeting (1R01AI183406-01). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10897643. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*