Allosteric loss-of-function mutations in HIV-1 Nef from a long-term non-progressor

Abstract

Activation of Src family kinases by human immunodeficiency virus type 1 (HIV-1) Nef may play an important role in the pathogenesis of HIV/AIDS. Here we investigated whether diverse Nef sequences universally activate Hck, a Src family member expressed in macrophages and other HIV-1 target cells. In general, we observed that Hck activation is a highly conserved Nef function. However, we identified an unusual Nef variant from an HIV-positive individual that did not develop AIDS which failed to activate Hck despite the presence of conserved residues linked to Hck SH3 domain binding and kinase activation. Amino acid sequence alignment with active Nef proteins revealed differences in regions not previously implicated in Hck activation, including a large internal flexible loop absent from available Nef structures. Substitution of these residues in active Nef compromised Hck activation without affecting SH3 domain binding. These findings show that residues at a distance from the SH3 domain binding site influence Nef interactions allosterically with a key effector protein linked to AIDS progression.

Figure 1

Yeast-based screen for HIV-1 Nef-induced Hck activation. A) Yeast cultures were co-transformed with expression constructs for Hck and Csk together with Consensus, ELI, LAI, NL4-3, SF2, or YU-2 Nef alleles. Top: Liquid cultures were normalized for cell density and spotted on galactose-agar plates at increasing dilutions, incubated for 3 days at 30 °C, and scanned. Yeast patches appear as dark circles against the translucent agar background. Bottom: Lysates from the same cultures were immunoblotted for Hck, Csk, or Nef expression as indicated. B) Hck-YEI replaces Hck plus Csk co-expression in the yeast assay. Yeast cultures were co-transformed with expression constructs for Hck-YEEI and Consensus, ELI, LAI, NL4-3, SF2, or YU-2 Nef alleles. Top: Yeast cultures were grown on galactose-agar plates as described in part A. Bottom: Lysates from the same cultures were immunoblotted for Hck or Nef as indicated.

Figure 2

Activation of Hck-YEEI by primary Nef alleles from LTNPs. A) Yeast assay. Yeast cultures were transformed with expression constructs for Hck-YEEI alone (-Nef) or together with Nef-SF2 as a positive control or the LTNP Nef alleles indicated. A Nef-SF2 PxxP to AxxA mutant (PA) which cannot bind or activate Hck-YEEI served as a negative control. Top: Liquid cultures were normalized for cell density and spotted on galactose-agar plates at increasing dilutions, incubated for 3 days at 30 °C, and scanned. Yeast patches appear as dark circles against the translucent agar background. Bottom: Lysates from the same cultures were separated via SDS-PAGE and immunoblotted for Hck or Nef proteins as indicated. B) Nef-LTNP4 fails to activate Hck-YEEI in vitro. Recombinant Hck-YEEI was assayed in the presence of increasing molar ratios of purified recombinant Nef-SF2 or Nef-LTNP4 using a peptide substrate and a FRET-based kinase assay. Each condition was repeated in quadruplicate and the extent of phosphorylation is expressed as the mean percent phosphorylation relative to a control phosphopeptide ± S.D. This experiment was repeated twice with comparable results; a representative example is shown.

Figure 3

LTNP-derived amino acid substitutions outside of the SH3-binding surface suppress Hck activation by Nef-SF2 in vitro. A) Model of Nef:SH3 complex. The Nef backbone is shown in grey, and the SH3 surface is colored violet. The conserved PxxP motif and hydrophobic pocket residues of Nef involved in SFK SH3 binding are highlighted in cyan. Residues substituted in Nef-LTNP4 associated with loss of SFK activation are highlighted in red (F191) and green (A156). The full-length Nef model was originally created by Geyer and Peterlin; note that the large unstructured loop and flexible N-terminal region are not present in existing Nef:SH3 structures., B) Replacement of Nef-SF2 residues F191 and A156 with corresponding residues of Nef-LTNP4 reduces its Hck-activating function. Wild-type Nef-SF2 (WT) as well as the Nef-SF2 point mutants A156I, F191L and the corresponding double mutant (AI-FL) were expressed in E. coli and purified to homogeneity. Recombinant Hck-YEEI activity was assayed either alone or in the presence of a 5- or 10-fold molar excess of each Nef protein using a peptide substrate and FRET-based kinase assay. Each condition was repeated in quadruplicate and the extent of phosphorylation is expressed as the mean percent phosphorylation relative to a control phosphopeptide ± S.D. This experiment was repeated twice with comparable results.

Figure 4

Determination of the relative binding affinity between Hck SH3 and Nef by HXMS. A–D) Recombinant purified Hck SH3 was labeled with deuterium for the periods indicated at the left of each panel. Changes in mass for the +5 charge state are shown for the Hck SH3 domain incubated alone (A) or in the presence of the non-binding control Nef-ELI (B), wild-type Nef-SF2 (C), or Nef-LTNP4 (D). Na+/K+ adducts are indicated with an asterisk. Labeling of SH3 in the presence of the Nef-SF2 mutants A156I, F191L and AI-FL gave identical results to those shown in panel C for wild-type Nef-SF2 (data not shown). Panel E shows the slowdown factor for the partial unfolding of the Hck SH3 domain either alone or in the presence of each Nef variant. The slowdown factor provides a relative measure of SH3 binding affinity for each Nef protein and is calculated from the rate constants for SH3 unfolding in the presence and absence of Nef as described elsewhere.