4-Hydroxynonenal oxidatively modifies histones: implications for Alzheimer’s disease
Abstract
There is increasing evidence of DNA oxidation and altered DNA repair mechanisms in Alzheimer’s disease (AD) brain. Histones, which interact with DNA, conceivably could provide a protective shield for DNA against oxidative stress. However, because of their abundant lysine residues, histones may be a target for 4-hydroxynonenal (HNE) modification. In this study, we have shown that HNE binds to histones and that this binding affects the conformation of the histone, measured by electron paramagnetic resonance in conjunction with a protein- specific spin label. The covalent modification to the histone by HNE affects the ability of the histone to bind DNA. Interestingly, acetylated histones appear to be more susceptible to HNE modifications than control histones. Conceivably, altered DNA-histone interactions, subsequent to oxidative modification of histones by the lipid peroxidation product HNE, may contribute to the vulnerability of DNA to oxidation in AD brain.
Keywords: Alzheimer’s disease; DNA oxidation; Histone-DNA interactions; Spin labeling; Protein conformation
The Alzheimer’s disease (AD) brain is under extensive oxidative stress, evidenced by significant protein oxidation, lipid peroxidation, and DNA oxidation [3 – 5]. 8-Hydroxy- 20-deoxyguanosine (8-OHdG) is elevated in mitochondrial and nuclear DNA in AD brain [8,14,16].
Lovell et al. [11] reported increased free 8-OHdG in intact DNA from ventricular AD CSF. Levels of free 8- OHdG in CSF of AD subjects were decreased significantly [11]. The activity of the enzyme 8-oxoguanine glycosylase is decreased in AD brain [12]. These findings indicate that, although oxidized bases are increased in AD brain, the system has been altered in such a way that base excision repair mechanisms cannot clear the oxidized bases sup- ported by many reports of increased levels of DNA strand breaks [17,20], DNA nicking, and fragmentation [6]. Additionally, helicase activity is increased in AD brain but may actually impede DNA repair by keeping DNA in the single-stranded state, thereby preventing enzymes that operate on double stranded DNA, like 8-oxoguanine glycosylase, from repairing DNA [12].
Nuclear DNA also can be protected by interaction with histones through electrostatic interactions. There are four classes of core histones (H2A, H2B, H3 and H4), and all contain a significant amount of lysine and arginine (30 – 40%) [9]. A reactive endproduct of lipid peroxidation, 4- hydroxy-2-nonenal (HNE) can form a covalent adduct with cysteine, lysine, or histidine through Michael addition [7]. HNE has been found to be increased in AD brain [10,15], and glutathione transferase, an enzyme that protects neurons from HNE damage [4], has decreased activity in AD brain [13]. The abundance of lysine residues on the histones may make histones a target for oxidative modification by HNE. We tested the hypothesis that HNE would bind histones and thereby decrease the ability of histones to bind DNA, potentially accounting for the increased oxidation of these oligonucleotides.
Crude histones (calf thymus IIA) were purchased from Sigma (St. Louis, MO). HNE was purchased from Calbiochem (San Diego, CA). Purified control histones (^sodium butyrate) were purchased from Upstate (Char- lottesville, VA). Anti-HNE was obtained from Alpha Diagnostics International (San Antonio, TX). All other chemicals were purchased from Sigma in their highest purity.DNA was isolated using standard chloroform/phenol procedure as described previously [19].Histones (calf thymus IIA, Sigma) were diluted to 4 mg/0.5 ml of 20 mM phosphate buffer. The aliquots were treated with 10 or 50 mM of HNE for 1 h at 378C. The protein specific spin label Mal-6 (2,2,6,6-tetramethyl-4- maleimidopiperidin-1oxyl) was used to monitor the confor- mation of histones [2]. Mal-6 (2.5 mg) was dissolved in 100 ml of acetonitrile and diluted to 25 ml in 20 mM phosphate buffer. Samples were spin-labeled by incubation with 0.5 ml of the Mal-6 solution at 48C for 16 h. Samples were dialyzed against 20 mM phosphate buffer for 72 h, changing the bulk buffer at 1, 6, 24, and 48 h. Electron paramagnetic resonance (EPR) spectra were obtained on a Bruker EMX spec- trometer with the following parameters: microwave fre- quency, 9.78 GHz; microwave power, 20 mW; modulation frequency, 100, modulation amplitude, 0.32 G; gain,1 105; time constant, 1.28 msec. Spectra were analyzed using WIN-EPR software. The rotational correlation time, t [2], was calculated using the following equation h at 37 8C. Five ml aliquots were then assayed for HNE- bound proteins.
Fig. 2. HNE immunoreactivity of histones before spin labeling. With 10 and 50 mM HNE treatment, HNE bound proteins were significantly increased compared to control (*P , 0:00006; **P , 0:002), respectively.
Samples (5 ml) were incubated with 10 ml of 2 £ modified Laemmli sample buffer (0.125 M Trizma base, pH 6.8, 4% SDS, 20% glycerol). A total of 250 ng of protein were loaded onto the slot blot apparatus and standard immunochemical methods followed as described above. The goat anti-rabbit IgG alkaline phosphatase secondary antibody was obtained from Sigma.
The histone-DNA binding assay was performed with slight modifications [21]. DNA (12 mg) was incubated with 6 mg of histone (calf thymus IIA, Sigma). Histones and DNA were incubated with 0, 10, or 50 mM HNE in 50 mM potassium phosphate buffer, 150 mM NaCl (pH 7.4) at 25 8C
where DHpp is the width of the center line (MI ¼ 0). A(0) and A(þ1) are the peak-to-peak amplitudes of resonance lines of the MI 0 and MI 1 peaks, respectively. A decrease in t indicates faster rotational motion of the spin label.
Aliquots (5 ml) of samples prepared for the spin labeling studies were taken prior to the addition of Mal-6 and assayed for HNE-bound proteins. In the case of the more purified histones (Upstate), 50 mg aliquots were incubated with 0, 10, 50 mM HNE in a final volume of 50 ml PBS for 1 increase the salt concentration to 400 mM NaCl. The percent association was reported by comparing each sample to the control in 150 mM NaCl.
Histones were incubated with 10 and 50 mM HNE for 1 h at 37 8C. Using EPR spin labeling, protein conformational changes induced in histones by HNE were monitored. The correlation time parameter (t), approximated by the amount of time that it takes for a spin to rotate through an angle of one radian, reflect changes in the conformation of the histone, which would alter the environment of the spin label. The less hinderance to motion, the shorter the correlation time. HNE, bound to the lysine-rich histones, altered their conformation, reflected by a decreased correlation time or faster motion of the spin label (Fig. 1). That HNE was bound to the histones was confirmed by immunochemical methods (Fig. 2).
DNA and histone interactions were monitored in the presence of 10 and 50 mM HNE (Fig. 3). At 150 mM NaCl concentration, 50 mM HNE incubation significantly decreased the association of DNA and the histones (P , 0:03). Histone-DNA interaction also was disturbed by increasing salt concentration to 400 mM. Again, even under the higher ionic strength conditions, 50 mM HNE treatment of the histone significantly affected the histone- DNA interaction (P , 0:005), suggesting that HNE affects the ability of histones to bind DNA properly.
As shown in the product literature (Upstate catalogue # 13 – 113) of histones isolated from HeLa cells treated with sodium butyrate, increased acetylated histones are found. After 50 mM HNE treatment, histones which had been purified with sodium butyrate, an inhibitor of histone deacetylase, had significantly increased HNE-bound pro- teins compared to histones with basal amounts of acety- lation (histones minus sodium butyrate) (Fig. 4).
In this study, we observed that HNE binds to histones, altering the conformation of the histones and affecting the ability of histones to bind DNA. The binding of HNE to the histones, at lysine residues, alters hydrophobicity, making the histones less positively charged, and thereby changing the electrostatic association of histones to DNA [18]. The decreased correlation time suggests that after HNE treat- ment histones may adopt a less globular tertiary structure, which may account for the decreased ability of DNA to associate with the histones in the histone-DNA binding assay.
The acetylated histone, also less positive, can, in its ‘open’ conformation, dissociate from DNA, but in the process may increase the likelihood of oxidative modifi- cations. Acetylation altered histone conformation in such a manner that unacetylated lysine residues were more accessible to HNE attack, evidenced by the significantly increased HNE-bound proteins in acetylated histones as compared to the control histones (Fig. 4). HNE bound to histones may affect transcription by preventing DNA association with the histone.
Altered histone-DNA interaction may make DNA more susceptible to oxidative damage because DNA is less tightly bound to the histone. This could potentially lead to the observed strand breaks and oxidized bases in AD [4]. The damaged DNA may affect transcription, altering gene expression, and perhaps contribute to aging and neurode- generation [1]. The role of oxidatively-modified histones.
Fig. 3. Histones treated with HNE had altered interaction with DNA. Fifty mM HNE treatment caused a statistically significant difference in DNA interaction compared to control (*P , 0:03). Upon dissociation because of the increased salt concentration, histones treated with 50 mM HNE had significantly less interaction with DNA (**P , 0:005).
Fig. 4. Histones which are hyperacetylated (sodium butyrate treated, black bars) are more susceptible to HNE-induced modification. Control histones (no sodium butyrate, hatched bars) had significantly increased HNE-bound proteins with 50 mM HNE treatment (*P , 0:003). Histones that were hyperacetylated had significantly increased HNE-bound proteins compared to hyperacetylated control (**P , 0:00002) and to the 50 mM HNE treated histone control (***P , 0:03). DNA oxidation needs to be further evaluated in AD brain, and such studies are underway in our laboratory.