Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jan 3;12(1):e0169052.
doi: 10.1371/journal.pone.0169052. eCollection 2017.

Substrates and Inhibitors of SAMHD1

Affiliations
Free PMC article

Substrates and Inhibitors of SAMHD1

Joseph A Hollenbaugh et al. PLoS One. .
Free PMC article

Abstract

SAMHD1 hydrolyzes 2'-deoxynucleoside-5'-triphosphates (dNTPs) into 2'-deoxynucleosides and inorganic triphosphate products. In this paper, we evaluated the impact of 2' sugar moiety substitution for different nucleotides on being substrates for SAMHD1 and mechanisms of actions for the results. We found that dNTPs ((2'R)-2'-H) are only permissive in the catalytic site of SAMHD1 due to L150 exclusion of (2'R)-2'-F and (2'R)-2'-OH nucleotides. However, arabinose ((2'S)-2'-OH) nucleoside-5'-triphosphates analogs are permissive to bind in the catalytic site and be hydrolyzed by SAMHD1. Moreover, when the (2'S)-2' sugar moiety is increased to a (2'S)-2'-methyl as with the SMDU-TP analog, we detect inhibition of SAMHD1's dNTPase activity. Our computational modeling suggests that (2'S)-2'-methyl sugar moiety clashing with the Y374 of SAMHD1. We speculate that SMDU-TP mechanism of action requires that the analog first docks in the catalytic pocket of SAMHD1 but prevents the A351-V378 helix conformational change from being completed, which is needed before hydrolysis can occur. Collectively we have identified stereoselective 2' substitutions that reveal nucleotide substrate specificity for SAMHD1, and a novel inhibitory mechanism for the dNTPase activity of SAMHD1. Importantly, our data is beneficial for understanding if FDA-approved antiviral and anticancer nucleosides are hydrolyzed by SAMHD1 in vivo.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Cartoon for SAMHD1 homotetramerization.
SAMHD1 monomers bind GTP (or dGTP) at allosteric 1 (A1) site leading to the formation of homodimers. Next, dNTPs bind to allosteric 2 (A2) sites allowing for the formation of homotetramers. The catalytic (Cat) site then can accompany dNTPs for hydrolysis, leading to the generation of deoxynucleosides (dNs) and inorganic triphosphates (iPPP).
Fig 2
Fig 2. Examining the role of L150 for nucleotide specificity.
A) dCTP, C) (2'R) 2'-F-dCTP and E) CTP nucleotides (in green) are modeled in the catalytic site of SAMHD1. L150 clashes with (2'R) 2'-F-dCTP and CTP, but not dCTP (see arrows) within the catalytic pocket of SAMHD1. B, D and F) Determining if dCTP, (2'R) 2'-F-dCTP and CTP can be hydrolyzed for SAMHD1 in vitro. Structures of the compounds are above the HLPC graphs. Using semi-quantitative HLPC analysis method, compounds were incubated with and without 1.6 μM of SAMHD1 enzyme plus dGTP (A1 site activator) to determine if they are substrates of SAMHD1. Data are presented as the percent compound remaining (y-axis). dCTP and dGTP were significantly hydrolyzed (p < 0.001; T test). No significant (n.s.) differences were detected between samples with and without SAMHD1 protein for (2'R) 2'-F-dCTP and CTP analogs. HPLC analysis of each nucleoside was done twice in triplicate. Mean and SEM are plotted with significant or no significant (n.s.) differences determined by T test analysis.
Fig 3
Fig 3. Role of 3'-OH sugar moiety for SAMHD1’s substrate specificity.
A-D) Using semi-quantitative HLPC analysis method, 2',3'-ddATP, 2',3'-ddGTP, 2',3'-ddCTP and 2',3'-ddITP are incubated with and without 1.6 μM of SAMHD1 enzyme plus dGTP (A1 site activator) to determine if they are substrates of SAMHD1. Data are presented as the percent compound remaining (y-axis), showing that none of these nucleoside analogs were hydrolyzed by SAMHD1. dGTP is also used as an internal positive control and is significantly hydrolyzed (p < 0.001) by SAMHD1 in the presence of all the 2'3'-ddNTP. Mean and SEM are plotted with significant or no significant (n.s.) differences determined by T test analysis.
Fig 4
Fig 4. Role of Y374 and C2' sugar moiety substitution in acting as substrates of SAMHD1.
A) dCTP, C) ara-CTP and E) SMDU-TP nucleotides (in green) are modeled within the catalytic site of SAMHD1. Both dCTP and ara-CTP do not clash with Y374 (see arrow). However the model shows that the (2'S)-2'-methyl group of SMDU-TP clashes with Y374 in the catalytic pocket of SAMHD1. B, D and F) Determining if dCTP, ara-CTP and SMDU-TP can be hydrolyzed for SAMHD1 in vitro. Structures of the compounds are above the HLPC graphs with experimental conditions described in Fig 2. Data are presented as the percent compound remaining (y-axis). dCTP and ara-CTP are significantly hydrolyzed (p < 0.001). SMDU-TP and dGTP, in the same reaction tube, had no significant hydrolysis in the presence of SAMHD1. HPLC analysis of each nucleoside was done twice in triplicate. Mean and SEM are plotted with significant or no significant (n.s.) differences determined using T test analysis.
Fig 5
Fig 5. Ara-CTP does not fit into the A2 site of SAMHD1.
A) Evaluating ara-CTP hydrolysis in the presence of dGTP, using as A1site activator. When dGTP was present, ara-CTP and dCTP were significantly hydrolyzed (p < 0.001) by SAMHD1. Data are presented as the percent compound remaining (y-axis). B) Determining if ara-CTP is hydrolysis by SAMHD1 in the presence of GTP. GTP will only fit into the A1 site of SAMHD1, thus requiring ara-CTP to occupy the A2 and catalytic sites for ara-CTP hydrolysis to occur. The percentage of ara-CTP remained constant with and without SAMHD1, indicating that ara-CTP cannot occupy the A2 site. Reactions containing dCTP was conducted and led to hydrolysis of dCTP in the presence of SAMHD1. Mean and SEM are plotted with significant or no significant (n.s.) differences determined using T test analysis.
Fig 6
Fig 6. Biochemical assessing SMDU-TP.
(A) Representative TLC plate showing 32-PPP accumulations over 20–300 sec. time course when using 1.25 μCi/μL [γ-32P]-dTTP with 3000 (left side) and 1000 (right side) μM of dTTP (cold) as the substrate in the presence of 1 μM of SAMHD1 enzyme. (B) Graphing data from TLC analysis to generate slopes for the different dTTP concentrations tested. Data displayed as Product (μM) (y-axis) vs. Time (sec) (x-axis). Km of dTTP was calculated to be 845 ± 229 μM from the slopes generated using Prism software. (C) TLC analysis was done at various concentrations of dTTP (3000–30 μM) in the presence of various concentrations of SMDU-TP (1000–30 μM). Data are graphed as 1/V (μM/s) (y-axis) vs. 1/[dTTP] (μM) (x-axis). From the slopes generated, the Ki of SMDU-TP was calculated to be 256 ± 70 μM. (D) Biochemical HLPC analysis of reactions with and without SAMHD1, and in the presence of pppCH2dU or SMDU-TP analog is graphed. We observed that both pppCH2dU and SMDU-TP analogs inhibit SAMHD1’s activity ((p < 0.01) by one-way ANOVA analysis with Bonferroni’s multiple comparisons), leading to more dGTP substrate remaining after the 2 h incubation with enzyme. All data are representative of two independent studies with mean and SEM displayed.
Fig 7
Fig 7. Monitoring ara-CTP and gem-TP concentrations in MDMs.
A) Monocyte-derived macrophages (MDMs) were pretreated with virus-like particles (VLP) one day prior to replacing the medium with fresh medium plus compounds: 10 μM of ara-C (cytaribine-13C3) or 10 μM of gemcitabine. Whole cell lysates were collected at 0, 24 and 48 h post VLP addition. Lysates were analyzed by immunoblotting for SAMHD1 and GAPDH, loading control. SAMHD1 protein levels were reduced at 24 h after Vpx+ VLP exposure. Two human primary MDM donors are shown. Cellular nucleotide extracts were generated at 4, 12 and 24 h post drug addition from treated MDMs. The intracellular concentrations of B) gem-TP (2',2'-diF-dCTP) and (C) ara-CTP were quantified from the extracts using HPLC-MS/MS analysis. Data are plotted as pmol/million cells (y-axis) vs. time (h) (x-axis). Gem-TP is a significantly lower (**, p < 0.01; T test) in the Vpx+ VLP treated MDMs at 4 h after drug addition. However, the rate of gem-TP decay is comparable between the two groups, suggesting that gem-TP degradation is SAMHD1 independent. Ara-CTP concentrations are significantly higher (***, p < 0.001) at 4, 12 and 24 h for the Vpx+ VLP treated MDMs. Moreover, the rate of decay of ara-CTP is slower in Vpx+ VLP treated group, suggesting ara-CTP turnover is SAMHD1 dependent. Data are from two independent donors tested in duplicate.
Fig 8
Fig 8. Determining nucleotide analog specificity for SAMHD1.
A) Modification of the 2' sugar position of a nucleotide can lead to several different outcomes. First, (2'R)-2'-F and (2'R)-2'-OH sugar moieties have been shown not to be substrates for SAMHD1. Additional analogs with (2'R)-2'-F and (2'R)-2'-OH sugar moieties would be predicted not to be substrates for SAMHD1. Second, canonical dNTPs and the non-canonical dUTP are substrates for SAMHD1. Our data shows that ((2'S)-2'-OH) arabinose nucleoside-5'-triphosphates are also substrates for SAMHD1. Therefore, we also predict several other arabinose nucleoside analogs would be substrates for SAMHD1. Moreover, clofarabine-TP ((2'S)-2'-F) was reported hydrolyzed by SAMHD1 [43]. Finally, we found the SMDU-TP, (2'R)-2'-methyl sugar moiety, inhibited the triphosphohydrolase activity of SAMHD1. We postulate that the (2'R)-2'-methyl moiety may prevent the conformational change in the catalytic site of SAMHD1 due to the size of the methyl group clashing with Y374. Therefore, we predicted that nucleotides with a (2'S)-2'-cyano moiety may also inhibit dNTPase activity of SAMHD1. B) A SAMHD1 inhibitor has been reported [34]. The pppCH2-dU analog has a 5'-methylene modification, making the analog non-hydrolysable in the catalytic site, but also was shown to block homotetramerization when present in the A2 site [34]. C) Modification of the 3'-OH sugar moiety is not permissive. NRTIs and ddNTPs lack a 3'-OH moiety, making them chain terminators for DNA polymerases, are not substrates for SAMHD1. D) Base modifications for different nucleoside analogs are permissive substrates for SAMHD1.

Similar articles

See all similar articles

Cited by 15 articles

See all "Cited by" articles

References

    1. Goldstone DC, Ennis-Adeniran V, Hedden JJ, Groom HC, Rice GI, Christodoulou E, et al. HIV-1 restriction factor SAMHD1 is a deoxynucleoside triphosphate triphosphohydrolase. Nature. 2011;480(7377):379–82. Epub 2011/11/08. nature10623 [pii] 10.1038/nature10623 - DOI - PubMed
    1. Powell RD, Holland PJ, Hollis T, Perrino FW. The Aicardi-Goutieres syndrome gene and HIV-1 restriction factor SAMHD1 is a dGTP-regulated deoxynucleotide triphosphohydrolase. J Biol Chem. 2011;286(51):43596–600. Epub 2011/11/10. C111.317628 [pii] 10.1074/jbc.C111.317628 - DOI - PMC - PubMed
    1. Yan J, Kaur S, Delucia M, Hao C, Mehrens J, Wang C, et al. Tetramerization of SAMHD1 is required for biological activity and inhibition of HIV infection. The Journal of biological chemistry. 2013;288(15):10406–17. Epub 2013/02/22. 10.1074/jbc.M112.443796 - DOI - PMC - PubMed
    1. Brandariz-Nunez A, Valle-Casuso JC, White TE, Laguette N, Benkirane M, Brojatsch J, et al. Role of SAMHD1 nuclear localization in restriction of HIV-1 and SIVmac. Retrovirology. 2012;9:49 Epub 2012/06/14. PubMed Central PMCID: PMC3410799. 10.1186/1742-4690-9-49 - DOI - PMC - PubMed
    1. Zhu C, Gao W, Zhao K, Qin X, Zhang Y, Peng X, et al. Structural insight into dGTP-dependent activation of tetrameric SAMHD1 deoxynucleoside triphosphate triphosphohydrolase. Nature communications. 2013;4:2722 Epub 2013/11/13. 10.1038/ncomms3722 - DOI - PubMed

MeSH terms

Feedback