A Conserved Acidic Residue in the C-Terminal Flexible Loop of HIV-1 Nef Contributes to the Activity of SERINC5 and CD4 Downregulation

Abstract

The host transmembrane protein SERINC5 is incorporated into retrovirus particles and inhibits HIV-1 infectivity. The lentiviral Nef protein counteracts SERINC5 by downregulating it from the cell surface and preventing its incorporation into virions. The ability of Nef to antagonize the host factor varies in magnitude between different HIV-1 isolates. After having identified a subtype H nef allele unable to promote HIV-1 infectivity in the presence of SERINC5, we investigated the molecular determinants responsible for the defective counteraction of the host factor. Chimeric molecules with a subtype C Nef highly active against SERINC5 were constructed to locate Nef residues crucial for the activity against SERINC5. An Asn at the base of the C-terminal loop of the defective nef allele was found in place of a highly conserved acidic residue (D/E 150). The conversion of Asn to Asp restored the ability of the defective Nef to downregulate SERINC5 and promote HIV-1 infectivity. The substitution was also found to be crucial for the ability of Nef to downregulate CD4, but not for Nef activities that do not rely on the internalization of receptors from the cell surface, suggesting a general implication in promoting clathrin-mediated endocytosis. Accordingly, bimolecular fluorescence complementation revealed that the conserved acidic residue contributes to the recruitment of AP2 by Nef. Altogether, our results confirm that Nef downregulates SERINC5 and CD4 by engaging a similar machinery and indicates that, in addition to the di-leucine motif, other residues in the C-terminal flexible loop are important for the ability of the protein to sustain clathrin-mediated endocytosis.

Keywords: HIV-1; Nef; SERINC5.

Figure 1

The ability of Nef to counteract SERINC5 is variable across the different viral isolates. (A,B) Nef proteins from isolates derived from different genetic subtypes have a variable ability to promote HIV-1 infectivity in the presence of SERINC5. (A) Infectivity of HIV-1 limited to a single cycle of replication produced in the presence of SERINC5 and the indicated nef alleles was determined on TZM-bl-ZsGreen indicator cells after normalization by RT-activity. Bar graphs represent the mean infectivity relative to the infectivity in the absence of SERINC5. Error bars represent s.d. from three biological replicates, unpaired two-tailed t-test, * p < 0.05, *** p < 0.001 (B) Detection of HA-tagged Nef proteins expressed in trans in producer cells from (A) by western blot. The letters in parentheses indicate the genetic subtype of the viral isolate. (C,D) The ability of Nef proteins to counteract the inhibition of infectivity by SERINC5 correlates with their ability to downregulate SERINC5 from the cell surface. (C) Residual SERINC5-iFLAG cell surface levels in cells co-transfected with the indicated nef alleles expressed upstream of an IRES2-eGFP cassette to allow for the exclusive gating of Nef-expressing cells. Nef^Fs stands for Nef-frameshift, a control plasmid with the Nef ORF disrupted as explained in Materials and Methods. The residual SERINC5 surface expression was measured as median fluorescence intensity, and values are expressed as a percentage of the Nef^Fs control. Error bars represent s.d. from three biological replicates, unpaired two-tailed t-test, *** p < 0.001 (Graphpad). (D) Dot plots correlating the relative infectivity from (A) and the residual SERINC5 surface expression from (C), with a logarithmic regression curve fitted.

Figure 2

The defective SERINC5 counteracting activity of Nef^H maps in the C-terminus third of the molecule. (A): Pairwise alignment of Nef^H and Nef^C showing the divergent residues (in red), some notable functional motifs (green rectangles), and the portions of the molecules swapped for the generation of chimeric proteins (brackets). (B): The indicated HA-tagged chimeric Nef proteins are detected by western blot in the lysates of virus producing cells. (C): Infectivity of HIV-1 limited to a single cycle of replication produced in the presence of SERINC5 and the indicated Nef proteins. Bar graphs represent the mean infectivity relative to the infectivity of the virus produced with Nef^C. Mean ± s.d. (Biological quadruplicates), unpaired two-tailed t-test, *** p < 0.001, (GraphPad).

Figure 3

A 21 amino acid sequence from Nef^C rescues the SERINC5 counteracting activity of Nef^H. (A): Pairwise alignment of the C-terminus third of Nef^H and Nef^C showing the divergent residues (in red) and the portions of the molecules swapped for the generation of chimeric proteins within the C-terminal third of Nef (brackets). (B,D): Western blots of lysates from virus producing cells for the detection of HA-tagged chimeric Nef molecules, as indicated. (C,E): Infectivity of HIV-1 limited to a single cycle of replication produced in the presence of SERINC5 and the indicated Nef proteins. Bar graphs represent the mean infectivity relative to the infectivity of the virus produced with Nef^C. Mean ± s.d. (biological quadruplicates), unpaired two-tailed t-test, ** p <0.01, *** p <0.001, (GraphPad).

Figure 4

Conversion of Asn to Asp in position 150 fully restores the counteraction activity of Nef^H. (A) Pairwise alignment of the 21 aa minimal sequence which transfers the SERINC5 counteracting activity from Nef^H to Nef^C , indicating the presence of three divergent residues (in red). (B,C): Within the minimal sequence rescuing the activity of Nef^H, D150 is required for SERINC counteraction. (D,E): D150 is sufficient to restore the SERINC5 counteracting activity in Nef^H. (B,E) show western blots of lysates from virus producing cells for the detection of the indicated HA-tagged Nef. (C,E) show the infectivity of HIV-1 limited to a single cycle of replication produced in the presence of SERINC5 and the indicated Nef proteins, relative to the infectivity measured with Nef^C. Bar graphs represent the mean relative infectivity. Mean ± s.d. (biological quadruplicates), unpaired two-tailed t-test, ** p < 0.01, *** p < 0.001 (GraphPad). (F): Sequence logo representing the conservation of the amino acid residues of a sequence including residue 150, built from HIV-1 6487 sequences in the Los Alamos database (accessed on 28 November 2022).

Figure 5

The D150N substitution restores the ability of Nef^H to downregulate SERINC5 and CD4. (A): the effect of different Nef proteins on the surface expression level of SERINC5, CD4 and MHC-I. Relative cell surface levels in cells co-transfected with the indicated nef sequences expressed upstream of an IRES2-eGFP cassette to allow the exclusive gating of Nef-expressing cells. Residual surface expression was measured as median fluorescence intensity, and values are expressed as a percentage of the Nef^Fs control. The well-characterized mutation of the di-leucine sorting signal in the context of the Nef^LAI protein was also tested as a control. Error bars represent s.d. from three biological replicates, an unpaired two-tailed t-test, *** p > 0.0001 (Graphpad) –(B): representative confocal micrographs and quantification of Lck retargeting in Jurkat (TAg) T cells co-expressing the indicated Nef variants and mutants as a fusion protein with eGFP and Lck as a fusion protein with RFP. The quantification of total cells shows a dense accumulation of Lck at the trans-Golgi network when counting at least 100 eGFP-expressing cells per condition (plotted as %). Nef^SF2 and the corresponding loss-of-function mutant AxxA were used as controls. Scale bar: 10 µm. Bars represent the mean ± s.d. (biological triplicates), unpaired two-tailed t-test, *** p < 0.001, ** p < 0.01 (GraphPad).

Figure 6

An Asp at Nef residue 150 associates with the higher recruitment of AP2. (A): Mean fluorescence intensity (MFI) of BiFC signal from three independent experiments in cells expressing the indicated Nef proteins tagged with Venus N-HA and AP2α/σ2 hemicomplex tagged with V5-Venus C-terminal fragment. Mutation D150N reduced complementation fluorescence from Nef^C: AP2α/σ2, while substitution N150D in Nef^H increased the signal associated with AP2α/σ2 hemicomplex. As a control, the abrogation of AP2 binding motif in Nef^LAI (LL/AA) also reduced the association with the AP2α/σ2 hemicomplex, which produced no signal when expressed alone. MFI was measured in three independent experiments (left), and presented as average fold BiFC signal over control (WT) Nef (right). Mean ± s.d. Shown is the P-value derived from one-sample t-test (GraphPad) (B). A western blot analysis of AP2α/σ2-VN-V5 and Nef-HA-VC shows no difference in cellular expression between variants.