A role for oxidized DNA precursors in Huntington's disease-like striatal neurodegeneration

Abstract

Several human neurodegenerative disorders are characterized by the accumulation of 8-oxo-7,8-dihydroguanine (8-oxodG) in the DNA of affected neurons. This can occur either through direct oxidation of DNA guanine or via incorporation of the oxidized nucleotide during replication. Hydrolases that degrade oxidized purine nucleoside triphosphates normally minimize this incorporation. hMTH1 is the major human hydrolase. It degrades both 8-oxodGTP and 8-oxoGTP to the corresponding monophosphates. To investigate whether the incorporation of oxidized nucleic acid precursors contributes to neurodegeneration, we constructed a transgenic mouse in which the human hMTH1 8-oxodGTPase is expressed. hMTH1 expression protected embryonic fibroblasts and mouse tissues against the effects of oxidants. Wild-type mice exposed to 3-nitropropionic acid develop neuropathological and behavioural symptoms that resemble those of Huntington's disease. hMTH1 transgene expression conferred a dramatic protection against these Huntington's disease-like symptoms, including weight loss, dystonia and gait abnormalities, striatal degeneration, and death. In a complementary approach, an in vitro genetic model for Huntington's disease was also used. hMTH1 expression protected progenitor striatal cells containing an expanded CAG repeat of the huntingtin gene from toxicity associated with expression of the mutant huntingtin. The findings implicate oxidized nucleic acid precursors in the neuropathological features of Huntington's disease and identify the utilization of oxidized nucleoside triphosphates by striatal cells as a significant contributor to the pathogenesis of this disorder.

Figure 1. Construction and characterization of a transgenic mouse expressing the hMTH1 cDNA.

(A) A BamH1-EcoRV fragment (509 bp) derived from pcDEBΔ encoding the hMTH1 cDNA was subcloned into the gWIZ vector under the control of the CMV promoter. This vector transfected into wild-type MEFs expressed the hMTH1 protein (data not shown). The MscI-KpnI fragment (2481 bp) was used in the construction of the transgenic mouse. (B) Determination of transgene copy number. Genomic DNA from mouse tails was analysed by Southern blot analysis (left panel). In the right panel 2, 5, 10, 20, 40 copies of the MscI-KpnI fragment were analysed. Both blots were probed with the same probe. Copy number was determined by comparison. The arrow indicates the DNA from the mouse that was used as a founder for the colony. (C) Analysis by FISH of the number of hMTH1 integrations in hemizygous (hMTH1-Tg^+/−) and homozygous (hMTH1-Tg^+/+) strains. (D) Expression of hMTH1 mRNA. RT-PCR was performed using total RNA from the indicated organs and human specific primers. RT-PCR for the GADPH gene is used as an internal control. hMTH1 and GAPDH fragments are respectively 200 bp and 330 bp. (E) Western blot analysis of transgene expression. Total proteins (20–40 µg) from a range of tissues were separated by SDS polyacrylamide electrophoresis, blotted and probed with an antibody against hMTH1. β-tubulin was used as a loading control.

Figure 2. hMTH1 protein in hMTH1-Tg^+/+ MEFs and protection against oxidative stress.

(A) Expression of hMTH1 in wild-type and homozygous hMTH1-Tg^+/+ MEFs. Total cell extracts were separated by 12%-SDS polyacrylamide electrophoresis, blotted and probed with an antibody against hMTH1 . The human SHSY5S neuroblastoma and SV40-transformed MRC5 cell lines are shown for comparison (left panel). Western blotting of subcellular fractions of hMTH1-Tg^+/+ MEFs with anti-hMTH1 (right panel). Proteins were separated on 18%-SDS polyacrylamide electrophoresis and cytocrome c was used to quantify mitocondrial cell extracts. (B) Steady-state levels of oxidized guanine. DNA and RNA from wild-type (filled bars) and hMTH1-Tg^+/+ (open bars) MEFs were digested to nucleosides and 8-oxodG and 8-oxoG were separated and quantified by HPLC-EC. Values are expressed as ratio to DNA and RNA guanine, respectively. Values are the mean±SE of 3 independent determinations. Asterisks indicate statistically significant differences (p-value = 0.005 and 0.015 for DNA and RNA, respectively; t-test). (C) Levels of oxidized guanine following oxidant treatment. 8-oxodG and 8-oxoG were measured by HPLC-EC in DNA and RNA extracted from MEFs exposed to 40 mM KBrO₃ for 30 min (time 0) and after the indicated times of repair incubation in drug-free medium (30, 60, 120 min). Wild type (filled bars) and hMTH1-Tg^+/+ (open bars). Values are the mean±SE of 3 independent determinations. Asterisks indicate a p-value≤0.05; t-test.

Figure 3. KBrO₃ sensitivity of hMTH1-Tg^+/+ and wild-type MEFs.

Wild-type (closed symbols) and hMTH1-Tg^+/+ (open symbols) MEFs were treated with KBrO₃ for 30 min at the indicated concentrations. Viability was measured by MTT assay 48 hr later. The graphs are the mean±SD of 3 independent experiments.

Figure 4. Paraquat-induced DNA 8-oxodG in wild-type and hMTH1-Tg^+/+ mice.

(A) Groups of mice (n = 13) were injected 5 times every other day with 10 mg/kg paraquat per injection. DNA from the indicated organs was digested and levels of 8-oxodG were determined by HPLC-EC. Wild-type (filled bars); hMTH1-Tg^+/+ (open bars). Significant differences (Student's t-test) between wild-type and hMTH1-Tg^+/+ organs are shown with an asterisk. The p values were 0.01, 0.03 and 0.04 for brain, heart and small intestine, respectively. SI, small intestine. BM, bone marrow. Values are the mean±standard errors. (B) Groups of control mice (n = 10) that had received saline injections on the same schedule of Paraquat-treated animals were sacrificed 2 days after the final injection. DNA 8-oxodG was determined in various organs as described above. Wild-type (filled bars) and hMTH1-Tg^+/+ (open bars). The p values (Student's t-test) were 0.02, 0.01 and 0.01 for brain, heart and small intestine, respectively. Values are the mean±standard errors.

Figure 5. 3-NP-induced toxicity in wild-type and hMTH1-Tg^+/− and hMTH1-Tg^+/+ transgenic mice.

Groups of wild-type (n = 20), hMTH1-Tg^+/− (n = 16) and hMTH1-Tg^+/+ (n = 16) mice were injected i.p. twice daily for 5 days with 60 mg/kg 3-NP. Wild-type (black bars), hMTH1-Tg^+/− (grey bars) and hMTH1-Tg^+/+ (white bars). (A) Weight loss. Body weight, measured immediately before the first injection on the indicated days, is expressed as a percentage of the pretreated body weight. (B) Motor impairment. Mice were monitored twice a day for dystonia and/or gait abnormalities. Neurological score was as follows: intermittent dystonia of one hindlimb: 1; intermittent dystonia of two hindlimbs: 2; permanent dystonia of hindimbs: 3; uncoordinated and wobbling gait or recumbency: 3; near death recumbency: 4. For each animal, the highest neurological score reached at any time of the observation period was considered. Values are mean±standard errors. (C) Cumulative mortality. The non-surviving fraction at the end of 5-day treatment is expressed as a percentage of starting total. (D) Striatal lesion formation. The percentage of animals with detectable post-mortem striatal lesions is shown. (E) Size of striatal lesions. Postmortem measurements of striatal lesions along the rostrocaudal axis. The asterisks indicate a P<0.05 vs wild-type according to One-way Anova and Tukey multiple comparison post-hoc test for panels A, B and E and to χ² test for panels C and D.

Figure 6. Oxidative DNA damage in the brain.

Brains from wild-type mice that had been treated twice per day for 5 days with 60 mg/kg 3-NP were examined for DNA 8-oxodG by immunohistochemistry. Nuclei of striatum, parietal and frontal cortex were counterstained by DAPI.

Figure 7. MTH1 expression and 3-NP-induced oxidative DNA damage in the brain.

(A) Immunofluorescence of MTH1 in the striatum of untreated wild-type mice (top panel, left) or 3-NP-treated (60 mg/kg twice daily for 5 days) wild-type, hMTH1-Tg^+/−, and hMTH1-Tg^+/+ animals. Nuclei of striatum counterstained by DAPI are shown in the right panels. (B) 8-oxodG immunoreactivity in the striatum of 3-NP-treated wild-type, hMTH1-Tg^+/−, and hMTH1-Tg^+/+ animals. Nuclei of striatum counterstained by DAPI are shown in the bottom panels.

Figure 8. Sensitivity to 3-NP of striatal cells expressing hMTH1 and wild-type or mutant murine htt.

Striatal cells derived from wild-type Hdh^Q7/Q7 and mutant Hdh^Q111/Q111 mice were transfected with hMTH1. (A) Proteins were separated and probed with an antibody against hMTH1. (B) Intracellular localization of hMTH1 (green fluorescence) in Hdh^Q111/Q111 and Hdh^Q111/Q111+hMTH1. Nuclei were counterstained by DAPI. (C) LDH release. LDH release from striatal cells into culture medium was measured 24 hr after continuous exposure to 20 mM 3-NP. Hdh^Q7/Q7 (grey bar) and Hdh^Q7/Q7+hMTH1 (dashed bar); Hdh^Q111/Q111 (black bars) and Hdh^Q111/Q111+hMTH1 (open bar). Mean±SE, n = 4. (D) Clonal survival. Cloning efficiency was determined at 33°C after 24 hr continuous exposure to the indicated 3-NP concentrations. Mean±SD, n = 2. (E) Coulter counter assay. Survival of non-proliferating cells measured in a Coulter Counter. Mean±SD, n = 3. The asterisks indicate significant differences by Student's t-test (p values = 0.02) between Hdh^Q111/Q111 and Hdh^Q111/Q111+hMTH1.