Expression of Human Apolipoprotein E Downregulates Amyloid Precursor Protein–Induced Ischemic Susceptibility
Background and Purpose— Epidemiological findings and experimental data on transgenic mice show that Alzheimer’s disease–related changes render the brain more susceptible to ischemic damage. We studied whether the previously observed vulnerability in mice overexpressing the 751–amino-acid isoform of human amyloid precursor protein (APP751) is regulated by human apolipoprotein E (apoE) alleles, which determine the relative risk for Alzheimer’s disease and the susceptibility to various forms of acute brain damage.
Methods— Aged apoE knock out (KO) mice, mice overexpressing APP751 in the apoE KO background and mice expressing either human apoE3 or apoE4 and APP751 in the apoE KO background were exposed to permanent occlusion of the middle cerebral artery (MCA). Infarct volumes were quantified from T2-weighted magnetic resonance images 24 hours after the MCA occlusion. Local cortical blood flow was monitored by laser Doppler flowmetry. Ischemia-induced microgliosis was detected by immunohistochemistry.
Results— Overexpression of human APP751 significantly increased the infarct volumes in apoE KO mice. Furthermore, this APP751-induced ischemic vulnerability was attenuated by the coexpression of either human apoE isoform. MCA occlusion resulted in a similar relative reduction in cortical blood flow in all mouse groups. Vascular anatomy showed no variation in the MCA territory between the groups. Instead, the expression of human apoE isoforms reduced the ischemia-induced microgliosis.
Conclusions— Expression of either the human apoE3 or apoE4 isoform protects against the increased ischemic vulnerability observed in aged mice overexpressing human APP751, probably by modulating the inflammatory response induced by MCA occlusion.
Alzheimer’s disease (AD) is characterized clinically by dementia and pathologically by the presence of amyloid β (Aβ)-containing plaques and neurofibrillary tangles.1,2⇓ Aβ is derived from its large precursor protein (amyloid precursor protein [APP]) by sequential proteolytic cleavages.1,2⇓ Recent experimental evidence suggests that Aβ plaques may not underlie the clinical features of AD3,4⇓ but that soluble forms of APP, its fragments, or other Aβ assemblies may be harmful. Although apolipoprotein E (apoE) was originally described as a transport molecule for lipoproteins,5 apoE has also been demonstrated to play a critical role in the central nervous system integrity, function, and repair after injury.6–8⇓⇓ In addition, the e4 allele of apoE is the main known risk factor for the most common sporadic form of AD,9 whereas inheritance of the more frequent e3 allele, or rare e2 allele, lowers the likelihood of developing AD.10,11⇓
Both in vitro12 and in vivo13–15⇓⇓ data suggest that apoE influences Aβ deposition. Lack of apoE dramatically reduces plaque load in the APPV717F transgenic mouse model of AD.13 Interestingly, when these mice were crossed with mice expressing human apoE4 and apoE3 on a murine null apoE background, the human apoE isoform–carrying double transgenic mice showed fewer Aβ immunoreactive deposits than did mice deficient in apoE.14 As these mice age, however, a human apoE isoform–dependent deposition of Aβ is found.15 Aged APPV717F mice expressing apoE4 have substantially more fibrillar amyloid deposits than do APPV717F mice expressing apoE3. However, the effect of apoE isoforms on neuronal survival in AD or transgenic models of AD is unknown.
In addition to the association of the apoE4 allele with AD, several studies have shown that it may also have an effect on poor outcome from various forms of acute brain injury, including head injury,16 intracerebral hemorrhage17 and stroke.18 Consistent with the clinical observations, transgenic mice expressing the human apoE4 isoform have been reported to develop larger infarcts after focal ischemia than human apoE3–carrying mice.19
It has been proposed that there is a relationship between AD and ischemic brain injury.20 A history of cerebral infarct increases the risk of AD.21 Furthermore, dementia and AD pathology are increased in patients who meet the neuropathological criteria for AD with coexisting evidence of cerebral infarcts.22,23⇓ On the other hand, elderly people with low cognitive abilities are at increased risk for later development of stroke.24 Evidence of the involvement of APP or Aβ in sensitizing the brain to ischemic damage is increasing. Zhang et al25 reported increased susceptibility to permanent focal brain ischemia in transgenic mice overexpressing the Swedish mutant APP, which was possibly due to Aβ40-mediated disturbance in endothelium-dependent vascular reactivity.26 Also, the mice expressing human wild-type APP751 have increased neuronal vulnerability to focal brain ischemia that is due to enhanced inflammation and the p38 mitogen-activated protein kinase pathway.27
In the present study, we report that overexpression of human wild-type APP751, which results in diffuse Aβ deposition and plaque-independent, specific, and progressive learning impairment in transgenic mice,3,28–30⇓⇓⇓ increases ischemic vulnerability also in the absence of murine apoE. In addition, the expression of human apoE3 and apoE4 isoforms reduces the APP751-induced ischemic susceptibility.
Materials and Methods
The methods for obtaining the glial fibrillary acidic protein (GFAP):apoE3 and GFAP:apoE4 mice have been reported earlier.31 Once GFAP:apoE3 and GFAP:apoE4 mice were produced in the C57Bl6/CBA background, they were bred to mouse apoE knockout (mapoE KO) mice and backcrossed >10 times onto the C57Bl6 background (The Jackson Laboratory, Bar Harbor, Me). Transgenic mice expressing the 751–amino-acid isoform of human wild-type APP (neuron-specific enolase [NSE]:APP751 mice,28 F10 pedigree in inbred JU background) were also bred to mapoE KO mice (The Jackson Laboratory) and backcrossed to the C57Bl6 strain to obtain mice expressing NSE:APP751 that lack both endogenous apoE alleles. Mice heterozygous for GFAP:apoE3 or GFAP:apoE4 and NSE:APP751 in the homozygous mapoE KO background were further bred with each other to generate mice expressing no transgenes (mapoE KO mice), mice expressing only the NSE:APP751 transgene, and double transgenic mice expressing both NSE:APP751 and GFAP:apoE3 or NSE:APP751 and GFAP:apoE4.
To study the effects of murine apoE and human e3 and e4 isoforms of apoE on a previously described APP751-induced susceptibility to focal brain ischemia,27 the above-described littermates with a similar genetic background were used: mapoE KO mice (n=15), mice expressing the human APP751 in the mouse apoE KO background (APP751/mapoE KO, n=14), and double transgenic mice expressing either the human apoE3 or apoE4 isoform and APP751 in the mapoE KO background (APP751/apoE3/mapoE KO, n=14; APP751/apoE4/mapoE KO, n=12). Tail biopsies were taken, and the presence of human apoE3 or apoE4 and APP751 transgenes in the DNA was confirmed by polymerase chain reaction by using primers specific for NSE:APP751 (CACTGGCCTCAGGCTCCACCC and TCAGTGGGTACCTCC AGCGCC) and GFAP:apoE (CCAGGGG- GTGTTGCCAGGGGCACC and TCCAGTTCCGATTTGTAGGCCTTCAACTCC), as described in detail elsewhere.28,31⇓ The presence of an equal amount of apoE3 and apoE4 protein in the brain homogenates of the double transgenic mice was confirmed by immunoblotting with the use of rabbit anti-human apoE antibody (Southern Biotechnology Associates, 1:5000 dilution). The blots were stripped and reprobed with rabbit anti-rat apoE antibody (Southern Biotechnology Associates, 1:2000 dilution) to confirm the absence of mouse endogenous apoE.
All animal experiments were conducted in accordance with National Institutes of Health and institutional guidelines. Male 14- to 18-month-old mice were housed in a standard temperature and light-controlled environment with ad libitum access to food and water.
Induction of Focal Cerebral Ischemia
Mice were anesthetized with a subcutaneous injection of a 1:4 dilution of Hypnorm (Roche) and Dormicum (Janssen Pharmaceutica) solution given at 0.05 to 0.075 mL/10 g (1.25 mg/mL midazolam, 0.079 mg/mL fentanyl citrate, and 2.5 mg/mL fluanisone). The middle cerebral artery (MCA) was exposed as described previously.27 Briefly, the left temporoparietal region of the head was shaved, and a midline incision was made between the orbit and ear to expose the skull. A small burr hole was drilled in the temporal bone just above the MCA, and saline was applied to the area throughout the procedure to prevent heat injury. The inner layer of the skull and the dura were carefully removed with fine forceps, and the MCA was elevated and cauterized permanently without damaging the brain surface. The body temperature was maintained at 36°C to 37°C during the surgery with a heating pad.
Determination of CBF by Laser Doppler Flowmetry
For monitoring of the relative changes in local cerebral blood flow (CBF) produced by the MCA occlusion, a laser Doppler probe (OxyFlo, Oxford Optronix Ltd) was placed in the center (1 mm caudal to the bregma, 3.5 mm lateral to the midline) of the ischemic territory. Data were acquired at the speed of 40 data points per second for 10 minutes before the MCA occlusion and 20 minutes after the MCA occlusion and then analyzed with PowerLab System software (ADInstruments Pty Ltd). The duration of anesthesia in all animals was <60 minutes, and the animals were allowed to regain full consciousness on a heating pad before returning to the cage.
Magnetic Resonance Imaging
Previous studies have demonstrated that the volume of hyperintense area in T2-weighted images correlates well with the lesion size determined by histological staining methods,32,33⇓ justifying MRI as a quantification method of infarcts. A 4.7-T horizontal magnet (Magnex Scientific) equipped with actively shielded field gradients (Magnex Scientific) interfaced to an SMIS console (Surrey Medical Imaging Systems) was used for MRI experiments. For the determination of the infarct volume, the mice were anesthetized with halothane at 24 hours after the onset of the MCA occlusion, when the lesion volume is maximal in this model. A T2-weighted multislice (repetition time 2100 ms, echo time 55 ms, matrix size 256×128, field of view 20 mm, slice thickness 0.6 mm, and slices 25) single spin-echo method was used. Infarct volume (mm3) was calculated by manually delineating hyperintense areas from the T2-weighted multislice data set with the use of SMIS image software. Results are expressed as lesion volume percentage from the volume of the contralateral hemisphere.
Measurement of Physiological Parameters and Determination of Vascular Anatomy
Physiological parameters were measured from ischemic mice (24 hours after the MCA occlusion, n≥4 for all groups) or nonischemic mice under Hypnorm-Dormicum anesthesia. A separate set of animals (n≥4 for all groups) was used for the determination of preischemic physiological parameters. A catheter was placed into the right common carotid artery to monitor blood pressure (Cardiocap MDSU II, Datex-Ohmeda Division Instrumentarium) and to withdraw blood. Arterial blood samples (75 μL) were analyzed for pH, partial pressure of oxygen, and carbon dioxide by using a blood gas/pH analyzer (ABL-5, Radiometer Medical). Blood glucose was determined by using a One Touch Basic Analyzer (LifeScan). The mice used for the determination of the ischemic lesion volume were then killed by cervical dislocation, and the similarity of the blood vessel anatomy in the ischemic hemispheres was confirmed by drawing the image detected through the preparation microscope. The images were standardized by using the site of cauterization (just proximal to the inferior cerebral vein) as a reference point.
A separate set of mice (n=4 per group) was deeply anesthetized with pentobarbital (150 mg/kg) and perfused transcardially with 0.01 mol/L PBS, followed by 4% paraformaldehyde 24 hours after the onset of MCA occlusion. Cryosections (10 μm thick) were reacted for 48 hours at 4°C with a rat anti-mouse F4/80 antibody (Serotec, diluted 1:50) to visualize microglial cells. After incubation with a biotinylated goat anti-rat IgG (Santa Cruz Biotechnology, diluted 1:200) and avidin-biotin complex (Vectastain Elite kit, Vector Laboratories), the complex was visualized by H2O2 and nickel diaminobenzidine. Four sections from each mouse were examined by an Olympus AX70 microscope (Olympus Optical) equipped with a digital camera and used for quantitative image analysis. Light microscopic images were obtained from cortical areas surrounding the ischemic lesions at the level of −0.1 to 0.1 mm from Bregma by using a ×40 objective lens. Images were opened with image analysis software (Image Pro-plus, Media Cybernetix), and the percentage of area occupied by F4/80 immunopositive cells in the outlined area of interest was measured.
Data are expressed as mean±SD. Infarct volumes, CBF data, and physiological parameters were compared between the groups with 1-way ANOVA followed by a Newman-Keuls post hoc test. Two-group comparisons were evaluated by 2-tailed Student t tests. Significance was assumed at a value of P<0.05. All statistical analyses were performed with SPSS for Windows software (SPSS Inc).
The permanent occlusion of the MCA produced infarcts restricted to the cerebral cortex. MRI data revealed that overexpression of APP751 significantly increased the susceptibility to permanent focal brain ischemia also in the absence of apoE (Figure 1). APP751/mapoE KO mice developed significantly larger infarcts than did mapoE KO mice or mice expressing APP751 in the presence of either of the human apoE isoforms (F3,22=5.9182, P<0.01 by ANOVA). Lesion volumes of mapoE KO, APP/apoE3/mapoE KO, and APP/apoE4/mapoE KO mice did not significantly differ from each other.
Cerebral Blood Flow
Local relative CBF was measured by laser Doppler flowmetry to confirm that the severity of CBF reduction produced by the occlusion of MCA was similar in all experimental groups. Baseline flow recorded before MCA occlusion was constant during the 10-minute recording period in all animals. Immediately after the onset of MCA occlusion, CBF dropped by ≈96%, showing that the probe was placed in the ischemic core. The magnitude of CBF reduction was identical in all mouse groups (Table 1). CBF recordings were continued for 20 minutes after the MCA occlusion. CBF remained at the same level during the postocclusion recording period, and there were no statistically significant differences between the groups (Table 1).
Physiological Variables and Vascular Anatomy
Physiological parameters of ischemic mice are presented in Table 2. There were no significant differences between the groups for any of the parameters studied. Preischemic values did not differ from the postischemic values, and there were no differences between the groups (data not shown). Importantly, the MCA territory was of similar size and had approximately the same distribution for all mouse groups used in the present study (data not shown), excluding the possibility that vascular anatomy contributes to the differences in stroke outcome.
F4/80 antigen immunohistochemistry revealed resting microglia in nonischemic brain tissue and activated ameboid microglial cells in areas adjacent to the ischemic lesion (Figure 2). There were no significant differences in the extent of F4/80 antigen immunoreactivity between the human apoE3 and apoE4 carriers also expressing APP751. However, MCA occlusion resulted in more severe microgliosis in APP751/mapoE KO mice compared with APP751/apoE3 or APP751/apoE4 double transgenic mice (Figure 2B and 2C). The relative area occupied by F4/80 immunoreactive ameboid microglia in the peri-infarct area was significantly higher in APP751-overexpressing mice lacking apoE compared with mice expressing APP751/human apoE (P<0.05 by t test, Figure 2D).
Previous studies have demonstrated that endogenous mouse apoE is protective against transient focal and global brain ischemia and traumatic brain injury.34–36⇓⇓ Although human apoE isoforms markedly delay the appearance of Aβ deposition in APPV717F transgenic mice,14 apoE ultimately facilitates the neuritic degeneration associated with amyloid deposits and with conversion of Aβ into mature fibrillar amyloid,15 which is thought to be neurotoxic.37 In the present study, we show that neuronal overexpression of human APP751 at the level detected in early AD and Down’s syndrome28 dramatically increases the susceptibility to ischemic brain damage in the absence of apoE. Moreover, expression of human apoE3 or apoE4 in astrocytes significantly reduces the neuronal vulnerability to ischemia in APP751 transgenic mice. These results indicate that even though apoE may be a contributory factor in Aβ-induced toxicity in AD, it has a beneficial role against APP751-induced ischemic vulnerability. Altogether, our results suggest that the increased vulnerability to ischemia in APP751 transgenic mice may be brought about by mechanisms other than direct neuronal toxicity of Aβ.
The exact mechanism of how human apoE isoforms reduce the ischemic damage in aged APP751 transgenic mice remains to be investigated. Perfusion parameters measured by bolus-tracking MRI and the vascular reactivity to endothelium-dependent vasodilator acetylcholine are similar in APP751 transgenic and wild-type JU mice.27 Also, previous studies using [14C]iodoantipyridine autoradiography in a global cerebral ischemia model have shown that apoE deficiency does not significantly alter CBF in C57Bl6 mice.35 In the present study, the vascular anatomy of the occluded MCA and the relative drop of CBF in the ischemic core did not reveal differences between the groups. However, we cannot exclude the possibility that the baseline CBF or the recovery of the flow in the penumbra region is altered in apoE3- or apoE4-expressing mice.
ApoE isoforms have been reported to have antioxidant and anti-inflammatory functions.38–40⇓⇓ In our model of permanent MCA occlusion in APP751 transgenic mice, no increase in the markers of oxidative stress were detected in the ischemic territory,27 as can be expected in a model of ischemia without free radical–producing reperfusion. Instead, it has been previously demonstrated that microglial cells in aged APP751 transgenic mouse brains have increased activity of p38 mitogen-activated protein kinase,27 which is a kinase needed for the induction of proinflammatory genes, such as inducible NO synthase, cyclooxygenase-2, tumor necrosis factor-α, and interleukin-1β.41–44⇓⇓⇓ On permanent MCA occlusion, APP751 transgenic mice show enhanced microglial activation and inflammatory response, which can be counteracted by anti-inflammatory treatments.27 Our finding that ischemia-induced microglial reaction in the peri-infarct areas was attenuated by human apoE expression supports the anti-inflammatory role of apoE. This anti-inflammatory hypothesis is favored also by previous studies showing that apoE downregulates microglial activation and cytokine production in vivo and in vitro.40 In addition, Barger and Harmon45 have demonstrated that a secreted derivative of APP (sAPP-α), the production of which is expected to be increased by neurons overexpressing wild-type APP, activates microglial cells at nanomolar concentrations to produce neurotoxic factors and that this activation is regulated by human apoE3 but not apoE4. Therefore, at least the protective effect of human apoE3 may be partially due to the prevention of direct sAPP-α activation of microglial cells, whereas the possible anti-inflammatory effect of apoE4 could be restricted to inflammation induced secondarily to brain injury.
A large body of evidence has indicated that apoE4 is less protective than the apoE3 isoform against ischemic or traumatic brain injury and that compared with apoE3, apoE4 is associated with impaired neuronal plasticity, neurite outgrowth, and anti-inflammatory response.31,39,46,47⇓⇓⇓ The present study, which involves the ischemic vulnerability induced by permanent focal brain ischemia in APP751/apoE transgenic mice, suggests that apoE in general may be beneficial for neuronal survival in AD, especially when it coexists with cerebrovascular disease.
This study was supported by the Saastamoinen Foundation, Finland, the Sigrid Juselius Foundation, Finland, and the Academy of Finland.
- Received December 7, 2001.
- Revision received April 1, 2002.
- Accepted April 9, 2002.
- ↵Selkoe D. Alzheimer’s disease: genes, proteins and therapy. Physiol Rev. 2001; 81: 741–766.
- ↵Koistinaho M, Ort M, Cimadevilla JM, Vondrous R, Cordell B, Koistinaho J, Bures J, Higgins LS. Specific spatial learning deficits become severe with age in β-amyloid precursor protein transgenic mice that harbor diffuse β-amyloid deposits but do not form plaques. Proc Natl Acad Sci U S A. 2001; 98: 14675–14680.
- ↵Westerman MA, Cooper-Blacketer D, Mariash A, Kotilinek L, Kawarabayashi T, Younkin LH, Carlson GA, Younkin SG, Ashe KH. The relationship between Aβ and memory in the Tg2576 mouse model of Alzheimer’s disease. J Neurosci. 2002; 22: 1858–1867.
- ↵Mahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science. 1988; 240: 622–630.
- ↵Arendt T, Schindler C, Bruckner MK, Eschrich K, Bigl V, Zedlick D, Marcova L. Plastic neuronal remodeling is impaired in patients with Alzheimer’s disease carrying apolipoprotein epsilon 4 allele. J Neurosci. 1997; 17: 516–529.
- ↵Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, Pericak-Vance MA. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science. 1993; 261: 921–923.
- ↵Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, Myers RH, Pericak-Vance MA, Risch N, van Duijn CM. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease: a meta-analysis: APOE and Alzheimer Disease Meta Analysis Consortium. JAMA. 1997; 278: 1349–1356.
- ↵Holtzman DM, Bales KR, Tenkova T, Fagan AM, Parsadanian M, Sartorius LJ, Mackey B, Olney J, McKeel D, Wozniak D, et al. Apolipoprotein E isoform-dependent amyloid deposition and neuritic degeneration in a mouse model of Alzheimer’s disease. Proc Natl Acad Sci U S A. 2000; 97: 2892–2897.
- ↵Slooter AJ, Tang MX, van Duijn CM, Stern Y, Ott A, Bell K, Breteler MM, Van Broeckhoven C, Tatemichi TK, Tycko B, et al. Apolipoprotein E ε4 and risk of dementia with stroke: a population based investigation. JAMA. 1997; 227: 818–821.
- ↵Kokmen E, Whisnant JP, O’Fallon WM, Chu CP, Beard CM. Dementia after ischemic stroke: a population-based study in Rochester, Minnesota (1960–1984). Neurology. 1996; 46: 154–159.
- ↵Zhang F, Eckman C, Younkin S, Hsiao KK, Iadecola C. Increased susceptibility to focal brain damage in transgenic mice overexpressing the amyloid precursor protein. J Neurosci. 1997; 17: 7655–7661.
- ↵Koistinaho M, Kettunen MI, Goldsteins G, Keinänen R, Salminen A, Ort M, Bures J, Liu D, Kauppinen RA, Higgins LS, et al. β-Amyloid precursor protein transgenic mice that harbor diffuse Aβ deposits but do not form plaques show increased ischemic vulnerability: role of inflammation. Proc Natl Acad Sci U S A. 2002; 99: 1610–1615.
- ↵Higgins LS, Rodems JM, Catalano R, Quon D, Cordell B. Early Alzheimer disease-like histopathology increases in frequency with age in mice transgenic for β-APP751. Proc Natl Acad Sci U S A. 1995; 92: 4402–4406.
- ↵Sun Y, Wu S, Bu G, Onifade MK, Patel SN, LaDu MJ, Fagan A, Holtzman DM. Glial fibrillary acidic protein-apolipoprotein E (apoE) transgenic mice: astrocyte-specific expression and differing biological effects of astrocyte-secreted apoE3 and apoE4 lipoproteins. J Neurosci. 1998; 18: 3261–3272.
- ↵Sheng H, Laskowitz DT, Mackensen GB, Kudo M, Pearlstein RD, Warner DS. Apolipoprotein E deficiency worsens outcome from global cerebral ischemia in the mouse. Stroke. 1999; 30: 1118–1124.
- ↵Lorenzo A, Yankner BA. Beta-amyloid neurotoxicity requires fibril formation and is inhibited by congo red. Proc Natl Acad Sci U S A. 1994; 91: 12243–12247.
- ↵Da Silva J, Pierrat B, Mary JL, Lesslauer W. Blockade of p38 mitogen-activated protein kinase pathway inhibits inducible nitric-oxide synthase expression in mouse astrocytes. J Biol Chem. 1997; 272: 28373–28380.
- ↵Guan Z, Buckman SY, Pentland AP, Templeton DJ, Morrison AR. Induction of cyclooxygenase-2 by the activated MEKK1→SEK1/MKK4→p38 mitogen-activated protein kinase pathway. J Biol Chem. 1998; 273: 12901–12908.
- ↵Bhat NR, Zhang P, Lee JC, Hogan EL. Extracellular signal-regulated kinase and p38 subgroups of mitogen-activated protein kinases regulate inducible nitric oxide synthase and tumor necrosis factor-alpha gene expression in endotoxin-stimulated primary glial cultures. J Neurosci. 1998; 18: 1633–1641.