# Analysis of Factors That Regulate HIV-1 Fusion in Reverse

## TL;DR

This study explores how HIV-1 envelope proteins can fuse with lentiviruses in a reverse process, revealing factors that influence this mechanism and its implications for infection.

## Contribution

The study introduces a novel method called 'fusion in reverse' to analyze HIV-1 envelope protein activity and its regulation.

## Key findings

- Infection via fusion in reverse depends on cell surface Env levels and is inhibited by an HIV-1-specific fusion inhibitor.
- Lentiviral pseudotyping with a GPI-anchored CD4 variant and a truncated CXCR4 protein is required for efficient infection.
- Membrane cholesterol levels do not affect Env activity, but sphingolipid deficiency and lipid scramblase incorporation alter infectivity.

## Abstract

Based on observations that HIV-1 envelope (Env) proteins on the surfaces of cells have the capacity to fuse with neighboring cells or enveloped viruses that express CD4 receptors and CXCR4 co-receptors, we tested factors that affect the capacities of lentiviral vectors pseudotyped with CD4 and CXCR4 variants to infect Env-expressing cells. The process, which we refer to as fusion in reverse, involves the binding and activation of cellular Env proteins to fuse membranes with lentiviruses carrying CD4 and CXCR4 proteins. We have found that infection via fusion in reverse depends on cell surface Env levels, is inhibitable by an HIV-1-specific fusion inhibitor, and preferentially requires lentiviral pseudotyping with a glycosylphosphatidylinositol (GPI)-anchored CD4 variant and a cytoplasmic tail-truncated CXCR4 protein. We have demonstrated that latently HIV-1-infected cells can be specifically infected using this mechanism, and that activation of latently infected cells increases infection efficiency. The fusion in reverse approach allowed us to characterize how alteration of CD4 plus CXCR4 lipid membranes affected Env protein activities. In particular, we found that perturbation of membrane cholesterol levels did not affect Env activity. In contrast, viruses assembled in cells deficient for long-chain sphingolipids showed increased infectivities, while viruses that incorporated a lipid scramblase were non-infectious. Our results yield new insights into factors that influence envelope protein functions.