DNA fragmentation (Fig. 2) induces necrosis and apoptosis (1,5). DNA oxidation and fragmentation are very early effects of oxidative stress in the brain, with measurable changes in DNA occurring within a few minutes of an insult (1,5-7). A high dose of an oxidative stress-inducing agent such as f-butylhydroperoxide (tBuOOH) induces immediate, large-scale DNA fragmentation and predominantly necrosis (1,5). A small dose of tBuOOH induces a small amount of DNA
fragmentation initially. However, apoptotic processes are triggered by this small-scale DNA fragmentation such that endonucleases are eventually activated, resulting in large amounts of DNA fragmentation and apoptosis that is maximal by 24 h or later (1,5). Necrosis is induced by massive insults that overcome the defense mechanisms of cells.
Apoptosis is induced by moderate insults that activate apoptotic processes and cause a cell to kill itself. Examples of moderate insults include mild ischemia and reperfusion such as in the penumbra area (8,9), low doses of tBuOOH (1,5), and low doses of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) (4). As much as 20-30% or more of the lesion produced in ischemia and reperfusion is due to apoptotic cell death (8). DNA damage causes the activation of the nuclear enzyme poly(ADP-ribose) polymerase (PARP) (Fig. 2), which binds to DNA nicks and fragments and is activated by the binding process (10). PARP contains two zinc fingers in the N-terminal portion of the protein, which bind to and stabilize DNA nicks (11). This allows the DNA to eventually be repaired. The C-terminal portion of PARP then catalyzes the synthesis of poly(ADP-ri-bose) from NAD and the conjugation of poly(ADP-ribose) onto proteins, including the middle region of PARP (10,11). PARP is also involved in base excision repair processes (12) and stimulates DNA transcription (13). Excessive PARP activity can result in depletion of brain NAD and another adenine containing compound, ATP (1,14, manuscript in preparation). This may compromise cellular energetics. Mice that express an inactive form of PARP can be resistant to ischemia- and reperfusion-induced brain damage (14,15). Susceptibility to brain damage may depend on the form of PARP expressed by the mice (16). Mice that have a mutation of the C-terminal region of PARP and express an inactive form of PARP that is still capable of binding to and stabilizing DNA nicks may be resistant to ischemia and reperfusion. However, mice expressing a form of PARP with a mutation in the N-terminal area, which cannot bind to DNA nicks, may not be resistant to ischemia and reperfusion.
We have used tBuOOH to induce oxidative stress and neurodegeneration in the brain. tBuOOH-treated mice are a model for the study of the effects of oxidative stress in the brain, such as the oxidative stress associated with stroke and other conditions. Administration of tBuOOH intracerebroventricularly allows the compound to rapidly penetrate into the brain and induce oxidative stress (17). When administered in high doses that result in 1 mM tBuOOH in the brain (18), tBuOOH causes DNA fragmentation, glutathione (GSH) oxidation, protein sulf-hydryl oxidation, edema, and lipid peroxidation (15,17,19). The brain mounts various defense mechanisms against tBuOOH toxicity such as increased GSH turnover, increased cellular availability of GSH, and increased activity of gluta-
thione disulfide (GSSG) reductase (20,21). tBuOOH affects a number of neurons in the brain including dopaminergic, serotonergic, GABAergic, cholinergic, and others (19). The compound also affects astrocytes, endothelial cells, and oligo-dendroctyes (19). Older and senescent mice are more susceptible to the toxic effects of tBuOOH than are younger mice (3,20,21). Perhaps the most age-sensitive effect of tBuOOH is the fragmentation of DNA, to which senescent mice are particularly susceptible (1,22).
We have also used MPTP, which is 1-methyl-4-phenyl-1,2,3,6-tetrahydropyri-dine, to induce oxidative stress and neurodegeneration in the brain, which is partly due to its ability to interfere with mitochondrial function (23). MPTP has been used to model Parkinson's disease in animals. MPTP is bioactivated by monoamine oxidase B to MPP+, which is 1-methyl-4-phenylpyridine, such that monoamine oxidase B knock-out mice, which do not express the enzyme, are resistant to MPTP toxicity (24). MPP+ is reduced by a number of different enzymes including tyrosine hydroxylase, monoamine oxidase, xanthine dehydrogenase, aldehyde dehydrogenase, lipoamide dehydrogenase, and NADH dehydrogenase (25-27). The result of the two-electron reduction is the induction of oxygen radical formation and redox cycling of MPP + . MPTP administration causes brain levels of GSH to change, GSSG levels to increase, and protein sulfhydryl levels to decrease in some brain regions (26,27). GSSG reductase is involved in protection against MPTP toxicity (28,29), since it reduces the GSSG produced during MPTP-induced oxidative stress. Vitamin E-deficient mice are more susceptible to MPTP toxicity than are normal mice (30). These facts demonstrate the importance of oxidative stress in the neurotoxicity of MPTP. MPTP induces necrosis and apoptosis in the brain and causes DNA fragmentation (4). Dopaminergic neurons in the midbrain have been observed to undergo apoptosis following MPTP administration (Fig. 1).
Focal cerebral ischemic stroke results in the neurodegeneration of cortical and other brain tissue and loss of function (31,32). Cessation of blood flow to the brain, whether permanent or transient, results in pannecrosis (infarction) to those areas that are truly ischemic. The penumbral areas surrounding the anoxic and aglycemic core are susceptible to a variety of toxicities created by ischemia that include overexposure to glutamate, oxygen radicals, and cytokines. Recent evidence indicates that, especially in the penumbra, delayed cellular death (apoptosis) by caspase activation (see Sec. IX) and DNA fragmentation can be a major contributor to the total infarcted tissue (33-39). In an intraluminal thread paradigm of transient focal cerebral ischemia, Li and Chopp and colleagues (8,33) studied in detail the involvement of apoptotic mechanisms in the progression of tissue infarction. They reported a temporal progression of cells including neurons that label histochemically with the TUNEL method for identifying double-stranded DNA breaks and, using northern analysis of DNA in ischemic tissue, found a characteristic laddering of fragments indicative of apoptotic processes. Similarly, Du et al., in a model of mild transient ischemia (9), found a substantial involvement of TUNEL-positive cells (with coincident DNA laddering) contributing to the progress of a reportedly delayed cortical infarction. Cortical infarction is significantly reduced when apoptosis is presumably inhibited in animals treated with caspase inhibitors (40-42). In the present work we report some of our findings from experiments with a model of focal ischemia and reperfusion.
Long-Evans male rats (body weight 300-350 g, Harlan Sprague-Dawley, Indianapolis, IN) were used in this study. Housing and anesthesia concurred with guidelines established by the institutional Animal Studies Committee and were in accordance with the PHS Guide for the Care and Use of Laboratory Animals, USDA Regulations, and the AVMA Panel on Euthanasia guidelines. Rats were allowed free access to water and rat chow (Wayne, Chicago, IL) before and after surgery. Under isoflurane anesthesia in a mixture of 30% O2 and 70% N2, animals were first implanted with an intraventricular cannula connected to an Alzet mini-osmotic pump (43,44), and pretreated with PBS vehicle or glial cell line-derived neurotrophic factor (GDNF, Amgen, Inc., Thousand Oaks, CA) infused at 10 |g/ day for 1 day prior to the middle cerebral artery (MCA) occlusion; other animals were left untreated without a cannula.
Mild transient focal cerebral ischemia was accomplished by occlusion of the MCA in a modification of the procedure described by Du et al. (9), as modified by Buchan et al. (45). The core body temperature of anesthetized animals was maintained at 37 ± 0.5°C throughout the surgical procedure using a homeother-mic water blanket under the animal and a heating lamp linked to an electronic temperature controller (YSI Model 73A, Yellow Springs, OH) regulated by a rectal thermometer. The right MCA was occluded using a Sundt #1 microaneu-rysm clip within 1 mm distal to its crossing of the rhinal fissure and the inferior cerebral vein. Both common carotid arteries were then occluded using nontraumatic aneurysm clips. The MCA was occluded for 45 min. At the end of the ischemic period, the aneurysm microclip and the carotid arterial clips were removed. Reperfusion of the arteries was confirmed by visual observation.
At 1 and 3 days after reperfusion, the animals were reanesthetized and perfused intracardially with 4% buffered paraformaldehyde. The brains were removed and sectioned into seven 2-mm coronal blocks used traditionally for analysis of infarct volume using TTC histochemistry. The blocks were embedded in paraffin, which allowed 5-|m sections to be collected from those blocks representing regions 2, 4, and 6 of the traditional seven-block set. The sections were processed with TUNNEL histochemistry (Apotag, Intergen, Purchase, NY) to label cells with double-stranded DNA breaks. The number of TUNEL-positive cells was counted throughout the affected cortex in each of three sections per brain from untreated, vehicle-treated, and GDNF-treated rats (n = 6 per group). In addition, tissue was collected from the ischemic cortex at 3 days after the mild transient ischemia and prepared for analysis of DNA fragmentation by northern blot.
3. DNA Extraction Method, Gel Electrophoresis, and Autoradiography
Brain tissue (50-100 mg) was gently disrupted in a hand-operated glass/glass Dounce homogenizer in 500 |l of lysis buffer (5 mM Tris-HCl, pH 7.4, containing 0.5% Triton X-100) and incubated on ice for 20 min. Samples were then spun for 20 min at 800g to remove nuclei and cellular debris. The supernatant was extracted with an equal volume of phenol chloroform isoamyl alcohol mixture (25:24:1). After centrifugation for 30 min at 2700g, DNA in the aqueous phase was precipitated twice with ethanol. The DNA pellet was resuspended in 20 |l of Tris-EDTA buffer (pH 8.0) and treated with 20 |g/ml RNase A for 45 min at 37°C to digest RNA. After quantitation, 2 |g of DNA was labeled with 32P-deoxycytosine triphosphate (dCTP) by a standard protocol involving the Klenow fragment of DNA polymerase 1. DNA in the samples was then reex-tracted as above to get rid of unincorporated nucleotide, resuspended in DNA loading buffer, and applied onto a 2% agarose submerged gel. After electrophore-sis, the gel was dried in a vacuum-driven gel drier and exposed for 4-24 h to autoradiographic film, which was then developed.
4. Results and Discussion—Focal Ischemia and Reperfusion
Figure 3 illustrates the profusion of TUNEL-positive cells in the ischemic cortex 3 days following a mild 45-min transient focal ischemia, in agreement with Du et al. (9). Such profiles were not frequent at 1 day of reperfusion (data not shown). Double labeling with glial fibrillary acidic protein (GFAP) immunohistochemis-try, a marker for reactive astrocytes, resulted in only 10% of the TUNEL-positive cells double labeling with GFAP, indicating that most of the TUNEL-positive cells are neurons. As seen in Fig. 3, most of the TUNEL-positive cells have neuronal morphology. In addition, the TUNEL-positive cells are all dark and shrunken with very dark or pyknotic nuclei, which is indicative of apoptosis. Figure 4 shows DNA laddering in the cerebral cortex that suffered ischemia and reperfusion. This is further evidence of apoptosis.
Was this article helpful?