Cryo-Electron Tomography and Nucleocapsid Protein Labeling by Tomo-Bubblegram Imaging Reveal a Role for HIV-1 Integrase in Viral Maturation

Abstract

HIV-1 maturation involves proteolysis of the precursor polyproteins, Gag and Gag-Pol, and assembly of a conical core containing the viral ribonucleoprotein (vRNP), comprised of the vRNA genome and associated proteins, notably nucleocapsid (NC). Allosteric integrase inhibitors (ALLINIs) are antiviral drugs initially thought to impede the viral integrase from inserting the DNA version of the genome into a host chromosome. However, recent studies indicate that their primary effects are on maturation. To investigate these effects, we imaged virions produced in the presence of ALLINIs by cryo-electron tomography (cryo-ET) and “tomo-bubblegram” imaging, a novel labeling technique that exploits the fact that vitrified protein subjected to relatively high levels of electron irradiation generates bubbles of hydrogen gas, which are readily visible. First, a regular cryo-tomogram is calculated. Then, after priming the specimen with additional exposures, a second tilt series is recorded, visualizing the bubbles. With HIV, we exploited the serendipitously discovered susceptibility of NC to bubbling.Most wild-type virions have conical cores with internal density, putatively the vRNP. In contrast, most ALLINI-treated virions have non-conical or conical cores with sparse internal contents but contain “eccentric condensates”, outside the capsid. Tomo-bubblegrams showed that the bubbles specifically label the contents of filled cores and eccentric condensates. In immature virions, in which the domains of Gag and Gag-Pol are radially ordered in a spherical shell, the bubbles appear in the NC layer of that shell. Immature virions in which NC is replaced by a leucine zipper (another RNA binder) yielded no bubbles, confirming NC as the bubbling component. Taken together, these observations provide strong evidence that ALLINIs sabotage capsid assembly and vRNP encapsidation. They also illustrate the potential of bubblegram imaging for mapping components of large macromolecular complexes.