Restrictive influence of SAMHD1 on Hepatitis B Virus life cycle

Abstract

Deoxynucleotide triphosphates (dNTPs) are essential for efficient hepatitis B virus (HBV) replication. Here, we investigated the influence of the restriction factor SAMHD1, a dNTP hydrolase (dNTPase) and RNase, on HBV replication. We demonstrated that silencing of SAMHD1 in hepatic cells increased HBV replication, while overexpression had the opposite effect. SAMHD1 significantly affected the levels of extracellular viral DNA as well as intracellular reverse transcription products, without affecting HBV RNAs or cccDNA. SAMHD1 mutations that interfere with the dNTPase activity (D137N) or in the catalytic center of the histidine-aspartate (HD) domain (D311A), and a phospho-mimetic mutation (T592E), abrogated the inhibitory activity. In contrast, a mutation diminishing the potential RNase but not dNTPase activity (Q548A) and a mutation disabling phosphorylation (T592A) did not affect antiviral activity. Moreover, HBV restriction by SAMHD1 was rescued by addition of deoxynucleosides. Although HBV infection did not directly affect protein level or phosphorylation of SAMHD1, the virus upregulated intracellular dATPs. Interestingly, SAMHD1 was dephosphorylated, thus in a potentially antiviral-active state, in primary human hepatocytes. Furthermore, SAMHD1 was upregulated by type I and II interferons in hepatic cells. These results suggest that SAMHD1 is a relevant restriction factor for HBV and restricts reverse transcription through its dNTPase activity.

Figure 1. Silencing of SAMHD1 increases HBV replication in resting HepG2.2.15 cells.

(a) SAMHD1 or actin protein levels were detected in cell lysates from SAMHD1 knockdown or control HepG2.2.15 cells by Western blotting. (b) Phosphorylation on residue T592 of SAMHD1 and cyclin B1 expression were detected by Western blotting of lysates from control shRNA HepG2.2.15 cells cultured for 3 days in complete versus serum-free medium. (c-e) HepG2.2.15 cells stably expressing two different SAMHD1 shRNAs or a control shRNA were cultured in serum-free medium for 3 days. (c) Luciferase activity (relative luciferase units, RLU) was detected in cells 24 hours post infection with a single-round HIV-1-based lentiviral vector that encoded luciferase. (d) Equal cell growth and viability between different shRNA cell lines was assessed using ATPLite. (e) The amount of HBV DNA in the supernatant was determined by qPCR and normalized to the control shRNA. In (c,d), the mean ± standard error mean (SEM) of three biological replicates of one representative experiment is depicted. In (e), the fold-changes to negative control were calculated for each individual experiment based on the median of three biological replicates, each measured in three technical replicates. The mean ± SEM of the fold-changes of four independent experiments is depicted. Statistical significance was determined using a one-way ANOVA with multiple comparisons according to Dunnett (*p < 0.05; ****p < 0.00005).

Figure 2. Overexpression of SAMHD1 reduces extracellular and intracellular levels of HBV DNA in HepG2.2.15 cells without affecting viral RNA.

HepG2.2.15 cells were transfected with (a) increasing amounts or (b–d) 1 μg of a SAMHD1-coding plasmid. The total amount of plasmid DNA was adjusted to 1 μg by addition of empty vector. Twenty-four hours after transfection, the cells were washed with phosphate-buffered saline (PBS) and further cultured in serum-free medium without supplements. Forty-eight hours later, HBV DNA from the supernatant (a,d) and intracellular HBV DNA from cytoplasmic fractions (c) were quantified by qPCR. (b) The levels of intracellular HBV RNAs were determined by RT-qPCR using primer sets that amplify the indicated HBV RNAs (3.5 kb; 3.5 kb and 2.4 kb; 3.5 kb, 2.4 kb and 2.1 kb, or all four HBV RNAs). (a) The mean ± SEM of the technical triplicates is shown. Statistical significance was determined using a one-way ANOVA with multiple comparisons according to Dunnett (*p < 0.05; ***p < 0.0005; ****p < 0.00005). (b–d) For each independent experiment, fold-changes were calculated based on the median of three technical replicates divided by the median of three technical replicates of two negative controls. The mean ± SEM of the fold-changes of four independent experiments is represented. Statistical significance was determined using an unpaired t-test (**p < 0.005; ***p < 0.0005).

Figure 3. Restriction of HBV by SAMHD1 is dependent upon dNTPase activity, is inhibited by phosphorylation of T592 and is rescued by addition of deoxynucleosides.

(a,b) HepG2.2.15 cells were transfected with 1 μg of plasmids coding for wild-type or mutant FLAG-tagged SAMHD1 or with 1 μg of empty vector. Twenty-four hours after transfection, the cells were washed and cultured in serum-free medium without supplements for 48 hours. (a) Detection of overexpressed FLAG-SAMHD1 by Western blotting. Detection of actin was used as a loading control. (b) HBV DNA from the supernatant was quantified by qPCR. The mean ± SEM of technical triplicates from one representative experiment out of three independent experiments is represented. (c) HepG2 cells were transfected with 1 μg of plasmid coding for wild-type FLAG-tagged SAMHD1 or empty vector. Twenty-four hours after transfection, the cells were washed and cultured in serum-free medium without supplements to arrest the cell cycle. When indicated, 10 mM of hydroxyurea (HU) were added at the time of medium change. Quantification of dATP levels 3 days post medium change were determined by single-nucleotide incorporation assay. The mean ± SEM of technical duplicates is represented. (d) HepG2.2.15 cells were transfected with 1 μg of plasmid coding for wild-type FLAG-tagged SAMHD1 or with an empty vector. Twenty-four hours after transfection, the cells were washed and cultured in serum-free medium without supplements. When indicated, deoxynucleosides (dNs, 2 mM each) or solvent were added at the time of medium change. HBV DNA from the supernatant was quantified by qPCR 3 days post medium change. For each independent experiment, fold-changes to empty vector were calculated based on the median of three technical replicates. The mean ± SEM of the fold-changes of three independent experiments is represented. (b,d) Statistical significance was determined using a one-way ANOVA with multiple comparisons according to Dunnett (**p < 0.005; ***p < 0.0005; ****p < 0.00005; n.s.: non-significant).

Figure 4. SAMHD1 restricts HBV replication in infected HepG2-NTCP cells.

Detection of (a) SAMHD1 or (b) NTCP in SAMHD1 knockdown or control HepG2-NTCP cells by Western blotting. Actin or GAPDH were used as loading controls. (c) Equal cell growth and viability was assessed was assessed using ATPLite. Depicted are mean values (±SEM) of three biological replicates of one representative experiment out of three. (d) Control shRNA cells were cultured for 10 days in complete medium supplemented (+) or not (−) with 2.5% DMSO. When indicated cells were infected with HBV 48 hours after start of DMSO treatment. Phosphorylation at residue T592 in SAMHD1, as well as expression of total SAMHD1, cyclin B1 and actin, were detected by Western blotting. (e) Cells were cultured for 10 days in 2.5% DMSO-containing medium and infected with a single-round luciferase HIV-1 virus. Luciferase activity was detected 24 hours post infection. Depicted are mean values (±SEM) of three biological replicates of one representative experiment out of three. (f-i) Cells were infected with HBV inoculum, and where indicated, the entry inhibitor MyrcludexB (MyrB, 200 nM), or the reverse transcription inhibitor lamivudine (Lami, 0,5 μM) were added. Cells were cultured in medium containing 2.5% DMSO. Ten days post infection, HBV DNA from the supernatant (f), intracellular total HBV DNA (g) and cccDNA (h) were quantified by qPCR. (i) The intracellular HBV RNAs levels were determined 10 days post infection by RT-qPCR using primer sets that amplify the indicated HBV RNAs (3.5 kb; 3.5 kb and 2.4 kb; 3.5 kb, 2.4 kb and 2.1 kb, or all four HBV RNAs). (f) For each individual experiment, fold-changes to the untreated control shRNA were calculated based on the median of three biological replicates, each measured in three technical replicates. The mean ± SEM of the fold-changes of three independent experiments is depicted. (g–i) For each independent experiment, fold-changes to untreated control shRNA were calculated based on the median of three technical replicates. The mean ± SEM of the fold-changes of at least 3 independent experiments is represented. (f–i) Statistical significance was determined using a one-way ANOVA with multiple comparisons according to Dunnett (*p < 0.05, **p < 0.005; ****p < 0.00005).

Figure 5. dATP levels are affected by HBV infection. SAMHD1 is de-phosphorylated at T592 in primary human hepatocytes, and is not affected by HBV infection.

(a) HepG2-NTCP- control-shRNA cells were infected or not with HBV in presence of 2.5%DMSO and dATP content was measured 4 days post infection. The mean ± SEM of technical duplicates is represented. (b, c) Primary human hepatocytes (PHH) from one donnor were infected or not with HBV for 8 or 24 hours. (b) SAMHD1 phosphorylation on T592, total SAMHD1, cyclin B1 and GAPDH were detected by Western blotting. Cycling or serum starved HepG2.2.15-control-shRNA cells were used respectively as positive and negative controls for pT592-SAMHD1 and cyclin B1. A representative experiment out of two technical replicates is represented. (c) Levels of intracellular HBV RNAs from infected PHH used in (b) were determined by RT-qPCR using HBV primer set 3 (see Fig. 2b and Table 1; amplified HBV RNA species: 3.5 kb, 2.4 kb and 2.1 kb RNAs). B.d.: below detection. The mean ± SEM of technical triplicates is represented.

Figure 6. Interferon induces SAMHD1 expression in hepatic cell lines and in primary human hepatocytes.

(a) HepG2 were cultured with 20 or 2000 U/mL, while (b) HepaRG cells were cultured with 2000 U/mL of interferon α2a (IFNα2a), interferon β (IFNβ) or interferon γ (IFNγ) for 20 hours. Cells were collected, lysed and probed for SAMHD1, STAT1 and GAPDH by Western blotting. (c) Primary human hepatocytes (PHH) from one donor were treated with 500 U/ml of IFNα2a or IFNγ for 3 or 24 hours. The relative level of SAMHD1 mRNA was determined by RT-qPCR. Fold-changes to untreated PHHs at the respective time points were calculated based on the median of three technical replicates. The mean ± SEM of the fold-changes of three technical replicates is represented. Statistical significance was determined using a one-way ANOVA with multiple comparisons according to Dunnett (*p < 0.05, **p < 0.005, n.s.: non-significant).