MxB sensitivity of HIV-1 is determined by a highly variable and dynamic capsid surface.

Richard J Miles,Laura Hilditch,Greg J Towers,Claire Kerridge,David A Jacques,Christopher Monit

doi:10.7554/elife.56910

Richard J Miles, Laura Hilditch + Show 4 more

Open Access

https://doi.org/10.7554/elife.56910

Copy DOI

Abstract

The type one interferon induced restriction factor Myxovirus resistance B (MxB) restricts HIV-1 nuclear entry evidenced by inhibition of 2-LTR but not linear forms of viral DNA. The HIV-1 capsid is the key determinant of MxB sensitivity and cofactor binding defective HIV-1 capsid mutants P90A (defective for cyclophilin A and Nup358 recruitment) and N74D (defective for CPSF6 recruitment) have reduced dependency on nuclear transport associated cofactors, altered integration targeting preferences and are not restricted by MxB expression. This has suggested that nuclear import mechanism may determine MxB sensitivity. Here we have use genetics to separate HIV-1 nuclear import cofactor dependence from MxB sensitivity. We provide evidence that MxB sensitivity depends on HIV-1 capsid conformation, rather than cofactor recruitment. We show that depleting CPSF6 to change nuclear import pathway does not impact MxB sensitivity, but mutants that recapitulate the effect of Cyclophilin A binding on capsid conformation and dynamics strongly impact MxB sensitivity. We demonstrate that HIV-1 primary isolates have different MxB sensitivities due to cytotoxic T lymphocyte (CTL) selected differences in Gag sequence but similar cofactor dependencies. Overall our work demonstrates a complex relationship between cyclophilin dependence and MxB sensitivity likely driven by CTL escape. We propose that cyclophilin binding provides conformational flexibility to HIV-1 capsid facilitating simultaneous evasion of capsid-targeting restriction factors including TRIM5 as well as MxB.

Highlights

Mx proteins (MxA and Myxovirus resistance B (MxB) in humans [Aebi et al, 1989]) are interferon-induced, high molecular weight, GTPases with globular N-terminal domains forming an enzymatic active site and C-terminal leucine zipper domains connected by a hinge-like bundle-signalling element (BSE) (Fribourgh et al, 2014)
MxB sensitivity is variable amongst primary viral isolates that are dependent on the same cofactors for nuclear import
Previous studies of MxB have suggested that MxB sensitivity is determined by CA sequence and influenced by changes in CA sequence driven by adaptive immune pressure, from cytotoxic T lymphocyte (CTL) escape (Liu et al, 2015; Merindol et al, 2018; Wei et al, 2016)

Summary

Introduction

Mx proteins (MxA and MxB in humans [Aebi et al, 1989]) are interferon-induced, high molecular weight, GTPases with globular N-terminal domains forming an enzymatic active site and C-terminal leucine zipper domains connected by a hinge-like bundle-signalling element (BSE) (Fribourgh et al, 2014). MxA has well-characterised antiviral activity against influenza A virus (Pavlovic et al, 1990), bunyaviruses (Kochs et al, 2002) and rhabdoviruses (Pavlovic et al, 1990), whereas closely related MxB has been shown to inhibit HIV-1 infection (Goujon et al, 2013; Kane et al, 2013; Liu et al, 2013) and more recently hepatitis C virus (Yi et al, 2019) and herpesviruses (Crameri et al, 2018; Schilling et al, 2018). MxB dimerisation and oligomerisation appears to be important for anti-HIV activity, but larger helical assemblies are thought not to be required (Alvarez et al, 2017; Buffone et al, 2015; Fricke et al, 2014). MxA does not restrict HIV-1 but the transfer of the N-terminal nuclear

Objectives

Methods

Results

Conclusion