KHNYN is a manganese-dependent endoribonuclease required for ZAP-mediated antiviral restriction

Figures

• 

  Figure 1.. Structural Domains in ZAP and KHNYN. (A) Schematic representation of human KHNYN. Lines represent the primary sequence from N- to C-termini. The position of structural/functional domains are highlighted; green (di-KH domain: residues 11–200), cyan (ex-PIN domain: residues 410–594, extended region 410–435 is hashed), and magenta (CUElike domain: residues 627–678). (B) Multiple sequence alignment using CLUSTAL-W of Homo sapiens KHNYN ex-PIN domain with PIN domains from N4BP1, MCPIP1 (ZC3H12A), MCPIP2 (ZC3H12B), and MCPIP3 (ZC3H12C). Uniprot accession numbers are shown on the right. The alignment is colored according to the CLUSTAL scheme [64], sequence numbers correspond to the KHNYN sequence. The positions of α-helix (cylinders) and β-strand (arrows) secondary structure elements in the KHNYN ex-PIN domain are shown above. The tetra-Asp motif and other residues conserved at the active site are indicated by the asterisks.

• 

  Figure 2.. KHNYN stability. (A–G) Thermal denaturation of the KHNYN ex-PIN domain measured by DSF upon addition of different metal ions at 0.1, 1, 5, and 10 mM (A) buffer, (B) Mn2+, (C) Mg2+, (D) Ca2+, (E) Zn2+, (F) Co2+, and (G) Ni2+. Samples were heated at a rate of 1.5°. min−1 from 20 to 95°C whilst the fluorescence intensity was recorded at 330 and 350 nm (λex = 280 nm). Curves from one experiment set are (left) the F350/F330 ratio derived from raw thermograms plotted against temperature at each divalent ion concentration and (right) the first differential plot of the ratio data. In each panel, the buffer only melting profile (black) is shown for comparison. (H) The average Tm plotted against metal ion concentration for three independent repeat experiment sets. Bars are color coded corresponding to metal ion identity in (A) to (G). Bar heights represent the mean value for Tm and error bars represent the standard deviation of the measurements from the three independent experiments.

• 

  Figure 3.. KHNYN endonuclease activity. (A) Divalent metal ion dependency of ex-PIN RNA hydrolysis. 10 µM 33mer 5′-FAM-labeled U-rich RNA incubated with 4 µM ex-PIN and 2.5 mM of the indicated divalent metal ions for 0 to 180 min. Substrate and hydrolysis products were separated by urea denaturing gel electrophoresis and visualized by fluorescence imaging. Reaction incubation times are indicated, NP, no protein control. (B) ex-PIN catalytic residues. 10 µM 33mer 5′-FAM-labeled U-rich RNA was incubated with 4 µM ex-PIN or 40 µM ex-PIN (D524A/D525A) for 0 to 30 min. Products were separated and visualized as in (A). (C) Abolition of ex-PIN endonuclease activity. 10 µM 33mer 5′-FAM-labeled U-rich RNA was incubated for 0.5 to 30 min with 4 µM ex-PIN either prior heat-treated at 70°C for 10 min or with 10 mM EDTA. NP; no protein control; 24/25, negative control 30-min incubation with 40 µM ex-PIN (D524A/D525A). Ca2+, Mn2+, RNase-inhibitor, Mg2+, Me-; 30-min reaction containing 2.5 mM CaCl2, MnCl2, MnCl2 + RNase inhibitor (40 units), MgCl2, or no metal respectively. Products were separated and visualized as in (A). (D and E) ex-PIN and ex-PIN-CUElike concentration dependency of hydrolysis…

• 

  Figure 4.. KHNYN endonuclease specificity. (A) ex-PIN nucleic acid specificity. 50 nM 5′-FAM-labeled DNA substrates; blunt double-stranded 70mer (ds 70mer), double-stranded 25mer – 5′ overhang (ds 5′-25mer), double-stranded 70mer – 3′ overhang (ds 3′-70mer), single-stranded 70mer (ss 70mer), and single-stranded 25mer (ss 25mer) incubated with (+) and without (−) 500 nM ex-PIN at 37°C for 60 min. Products were separated by urea denaturing gel electrophoresis and visualized by fluorescence imaging. M, ss DNA size markers. Slower migrating bands in double-stranded samples lanes are residual nondenatured duplex DNA. (B) ex-PIN substrate RNA sequence specificity, CpG dependence. (Upper) 10 µM 33mer 5′-FAM-labeled CpG-rich RNA incubated with or without 4.0 µM ex-PIN for 10 min at 37°C. RNAs were separated by urea denaturing gel electrophoresis alongside FAM-labeled RNA size markers, lengths indicated, and visualized by fluorescence imaging. (Lower) Sequence of CpG rich RNA derived from the recoded region in HIV-CpG, the CpG di-nucleotides are highlighted in red. (C) ex-PIN substrate RNA sequence specificity. 10 µM 33mer 5′-FAM-labeled A-rich, U-rich AU-rich and UA-rich RNAs were incubated with…

• 

  Figure 5.. KHNYN endonuclease metal-ion inhibition. (A) ex-PIN inhibition by divalent metal ions. 10 µM 33mer 5′-FAM-labeled U-rich RNA incubated with 4 µM ex-PIN and 1.0 mM MnCl2 with addition 1.0, 5.0, and 10 mM MgCl2 or CaCl2. Samples were taken from reaction time courses at 5.0 min, substrate and products separated by urea denaturing gel electrophoresis and visualized by fluorescence imaging. NP, no protein control, Mn2+, 5-min reaction with no additional metal ions. (B) Quantitation of substrate (U-rich RNA) intensity derived from (A). The integrated band intensity determined for remaining substrate in the Mn2+ and reactions containing additional Mg2+ and Ca2+ ions after 5-min incubation is plotted against metal ion concentration. Bars are average integrated intensity and error bars represent the s.d. of data from three independent experiments. Dashed line represents the average value of U-rich RNA in the no protein control.

• 

  Figure 6.. Structure of the KHNYN ex-PIN domain. (A) The dimer asymmetric unit of the KHNYN ex-PIN domain crystal structure. The protein backbone of the two ex-PIN protomers, Monomer-A (wheat) and Monomer-B (cyan) are shown in cartoon representation (left) and in surface (right). N- and C-termini of each monomer are labeled in the left-hand panel. (B) Secondary structure of ex-PIN. The protein backbone of the ex-PIN Monomer-B is shown in cartoon, two views rotated by 180°. Secondary structure α-helices and β-strands are labeled sequentially from the N- to C-terminus. (C) 3D DALI structural alignment of ex-PIN Monomer-A and Monomer-B (RMSD 1.4 Å; 167 of 183 Cα atoms). The β3–α6 loop (residues 510–524) and α7–α8 (residues 540–562) of the α6–α8 helical wing that have alternative conformations are boxed in gray. (D) 3D DALI structural alignment of the PIN structure from (PDB: 8WZC) with ex-PIN Monomer-A (RMSD 1.7 Å; 146 of 183 Cα atoms) (left) and Monomer-B (RMSD 1.2 Å; 161 of 183 Cα atoms) (right). Structures are shown in cartoon representation, PIN (green), ex-PIN Monomer-A (wheat), and Monomer-B (cyan). The ex-PIN Nt-Arm (residues 410–436), absent in the PIN structure, is highlighted and…

• 

  Figure 7.. KHNYN PIN domain active site. DALI 3D structural alignment of the ex-PIN active site with (A) PIN (PDB: 8WZC). RMSD 1.2 Å over 161 Cα atoms and (B) the ZC3H12B PIN domain (PDB: 6SJD) RMSD (1.6 Å over 161 Cα atoms). Protein backbones, in cartoon, are colored cyan (ex-PIN; Monomer-B), green (PIN), and pink (ZC3H12B). The ex-PIN Nt-Arm is shown in gray. Conserved residues surrounding the active sites including tetra-Asp motifs (KHNYN; D443, D524, D525, and D543) and (ZC3H12B; D195, D279, D280, and D298) are shown in sticks. Mg ions bound in the PIN and ZC3H12B structures are shown as pale lemon-green spheres.

• 

  Figure 8.. Ex-PIN domain mutants and antiviral activity. Infectious virus production from KHNYN CRISPR KO HeLa cells co-transfected with pHIV-WT or pHIV-CpG and increasing amount of CRISPR-resistant (A) WT pKHNYN or pKHNYN(524/525A) (C) WT pKHNYN or pKHNYN(Δ410–434) and (E) WT pKHNYN or pKHNYN(N4BP1PIN). Each data point represents the value from an independent experiment normalized to the value obtained for pHIV-WT in the absence of KHNYN. Three of the four replicates for KHNYN-WT + HIV-WT and KHNYN-WT + HIV-CpG are the same datapoints in panels (C) and (E). (B, D, and F) Representative western blots of the protein levels in restriction assays, (B) WT KHNYN and KHNYN(524/525A), (D) WT KHNYN and KHNYN(Δ410–434) and (F) WT KHNYN and (KHNYN-N4BP1PIN). Membranes are probed with antibodies to detect HIV-1 p24Gag in virions, KHNYN-FLAG expression in the transfected cell lysate and Actin as the loading control. Membranes were also probed to detect the of level of HIV-1 Gag and Env in producer cell lysates, presented in Supplementary Fig. S2.