Differential Expression of Cholesterol Hydroxylases in Alzheimer's Disease
2004; Elsevier BV; Volume: 279; Issue: 33 Linguagem: Inglês
10.1074/jbc.m402324200
ISSN1083-351X
AutoresJames A. Brown, Catherine Theisler, Simone Silberman, Debra J. Magnuson, Numa R. Gottardi-Littell, John M. Lee, Debra Yager, Janet Crowley, Kumar Sambamurti, Mohammad Mizanur Rahman, Allison B. Reiss, Christopher B. Eckman, Benjamin Wolozin,
Tópico(s)Drug Transport and Resistance Mechanisms
ResumoCholesterol is eliminated from neurons by oxidization, which generates oxysterols. Cholesterol oxidation is mediated by the enzymes cholesterol 24-hydroxylase (CYP46A1) and cholesterol 27-hydroxylase (CYP27A1). Immunocytochemical studies show that CYP46A1 and CYP27A1 are expressed in neurons and some astrocytes in the normal brain, and CYP27A1 is present in oligodendrocytes. In Alzheimer's disease (AD), CYP46A1 shows prominent expression in astrocytes and around amyloid plaques, whereas CYP27A1 expression decreases in neurons and is not apparent around amyloid plaques but increases in oligodendrocytes. Although previous studies have examined the effects of synthetic oxysterols on the processing of amyloid precursor protein (APP), the actions of the naturally occurring oxysterols have yet to be examined. To understand the role of cholesterol oxidation in AD, we compared the effects of 24(S)- and 27-hydroxycholesterol on the processing of APP and analyzed the cell-specific expression patterns of the two cholesterol hydroxylases in the human brain. Both oxysterols inhibited production of Aβ in neurons, but 24(S)-hydroxycholesterol was ∼1000-fold more potent than 27-hydroxycholesterol. The IC50 of 24(S)-hydroxycholesterol for inhibiting Aβ secretion was ∼1 nm. Both oxysterols induced ABCA1 expression with IC50 values similar to that for inhibition of A β secretion, suggesting the involvement of liver X receptor. Oxysterols also inhibited protein kinase C activity and APP secretion following stimulation of protein kinase C. The selective expression of CYP46A1 around neuritic plaques and the potent inhibition of APP processing in neurons by 24(S)-hydroxycholesterol suggests that CYP46A1 affects the pathophysiology of AD and provides insight into how polymorphisms in the CYP46A1 gene might influence the pathophysiology of this prevalent disease. Cholesterol is eliminated from neurons by oxidization, which generates oxysterols. Cholesterol oxidation is mediated by the enzymes cholesterol 24-hydroxylase (CYP46A1) and cholesterol 27-hydroxylase (CYP27A1). Immunocytochemical studies show that CYP46A1 and CYP27A1 are expressed in neurons and some astrocytes in the normal brain, and CYP27A1 is present in oligodendrocytes. In Alzheimer's disease (AD), CYP46A1 shows prominent expression in astrocytes and around amyloid plaques, whereas CYP27A1 expression decreases in neurons and is not apparent around amyloid plaques but increases in oligodendrocytes. Although previous studies have examined the effects of synthetic oxysterols on the processing of amyloid precursor protein (APP), the actions of the naturally occurring oxysterols have yet to be examined. To understand the role of cholesterol oxidation in AD, we compared the effects of 24(S)- and 27-hydroxycholesterol on the processing of APP and analyzed the cell-specific expression patterns of the two cholesterol hydroxylases in the human brain. Both oxysterols inhibited production of Aβ in neurons, but 24(S)-hydroxycholesterol was ∼1000-fold more potent than 27-hydroxycholesterol. The IC50 of 24(S)-hydroxycholesterol for inhibiting Aβ secretion was ∼1 nm. Both oxysterols induced ABCA1 expression with IC50 values similar to that for inhibition of A β secretion, suggesting the involvement of liver X receptor. Oxysterols also inhibited protein kinase C activity and APP secretion following stimulation of protein kinase C. The selective expression of CYP46A1 around neuritic plaques and the potent inhibition of APP processing in neurons by 24(S)-hydroxycholesterol suggests that CYP46A1 affects the pathophysiology of AD and provides insight into how polymorphisms in the CYP46A1 gene might influence the pathophysiology of this prevalent disease. Alzheimer's disease (AD) 1The abbreviations used are: AD, Alzheimer's disease; LXR, liver X receptor; APP, amyloid precursor protein; PMA, phorbol 12-myristate 13-acetate; PKC, protein kinase C; MOPS, 4-morpholinepropanesulfonic acid; OHC, hydroxycholesterol; RXR, retinoid X receptor; Aβ, β-amyloid. 1The abbreviations used are: AD, Alzheimer's disease; LXR, liver X receptor; APP, amyloid precursor protein; PMA, phorbol 12-myristate 13-acetate; PKC, protein kinase C; MOPS, 4-morpholinepropanesulfonic acid; OHC, hydroxycholesterol; RXR, retinoid X receptor; Aβ, β-amyloid. is characterized by a widespread loss of neurons and an even greater loss of synapses (1Callahan L.M. Coleman P.D. Neurobiol. Aging. 1995; 16: 311-314Crossref PubMed Scopus (81) Google Scholar). The neurodegeneration that occurs in AD is associated with restructuring of the neuropil. Degenerating neurons retract their processes and reduce the number of synapses (1Callahan L.M. Coleman P.D. Neurobiol. Aging. 1995; 16: 311-314Crossref PubMed Scopus (81) Google Scholar). Neurons with reduced arborization have less surface area and correspondingly less plasma membrane. Cholesterol is the major sterol present in the plasma membrane. Because cholesterol is so abundant, the synaptic degeneration occurring in AD requires that neuron dispose of the excess cholesterol. Cholesterol is catabolized through three main routes. Cholesterol can be esterified and stored within neurons as cholesterol esters (2Russell D.W. Biochim. Biophys. Acta. 2000; 1529: 126-135Crossref PubMed Scopus (304) Google Scholar). Alternatively, cholesterol can be oxidized at the 24 or 27 position to form 24(S)-hydroxycholesterol (24-OHC) or 27-hydroxycholesterol (27-OHC), which are termed oxysterols (2Russell D.W. Biochim. Biophys. Acta. 2000; 1529: 126-135Crossref PubMed Scopus (304) Google Scholar, 3Javitt N.B. Biochem. Biophys. Res. Commun. 2002; 292: 1147-1153Crossref PubMed Scopus (32) Google Scholar). Both oxysterols are more soluble than cholesterol and are excreted from cells to associate with lipoprotein particles in serum or extracellular fluid. Oxysterols serve a dual function because they are both cholesterol catabolites and ligands of the nuclear transcription factor LXR (4Lehmann J.M. Kliewer S.A. Moore L.B. Smith-Oliver T.A. Oliver B.B. Su J.L. Sundseth S.S. Winegar D.A. Blanchard D.E. Spencer T.A. Willson T.M. J. Biol. Chem. 1997; 272: 3137-3140Abstract Full Text Full Text PDF PubMed Scopus (1032) Google Scholar). The oxysterols are particularly intriguing because cholesterol 24-hydroxylase (CYP46A1), one of the enzymes that oxidizes cholesterol, is largely brain-specific, whereas the other enzyme, cholesterol 27-hydroxylase (CYP27A1), is expressed more broadly (5Lund E.G. Guileyardo J.M. Russell D.W. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7238-7243Crossref PubMed Scopus (521) Google Scholar). Multiple studies indicate that levels of 24-OHC increase in plasma of patients with neurodegenerative illness, including AD (6Lutjohann D. Papassotiropoulos A. Bjorkhem I. Locatelli S. Bagli M. Oehring R.D. Schlegel U. Jessen F. Rao M.L. von Bergmann K. Heun R. J. Lipid Res. 2000; 41: 195-198Abstract Full Text Full Text PDF PubMed Google Scholar, 7Papassotiropoulos A. Lutjohann D. Bagli M. Locatelli S. Jessen F. Rao M.L. Maier W. Bjorkhem I. von Bergmann K. Heun R. Neuroreport. 2000; 11: 1959-1962Crossref PubMed Scopus (127) Google Scholar, 8Papassotiropoulos A. Lutjohann D. Bagli M. Locatelli S. Jessen F. Buschfort R. Ptok U. Bjorkhem I. von Bergmann K. Heun R. J. Psychiatr. Res. 2002; 36: 27-32Crossref PubMed Scopus (194) Google Scholar); however, it is unclear whether increases in 24-OHC affect the pathophysiology or modulate APP metabolism. Recent studies show that the processing of APP is sensitive to cholesterol levels. Cleavage of APP by β- and γ-secretases produces Aβ. These cleavage events take place in the high cholesterol environments of lipid raft domains. Reducing cholesterol levels inhibits Aβ production and increases secretion of APP (9Simons M. Keller P. De Strooper B. Beyreuther K. Dotti C. Simons K. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 6460-6464Crossref PubMed Scopus (1079) Google Scholar, 10Fassbender K. Simons M. Bergmann C. Stroick M. Lutjohann D. Keller P. Runz H. Kuhl S. von Bergmann K. Hennerici M. Beyreuther K. Hartmann T. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 5856-5861Crossref PubMed Scopus (1025) Google Scholar, 11Kojro E. Gimpl G. Lammich S. Marz W. Fahrenholz F. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 5815-5820Crossref PubMed Scopus (722) Google Scholar). The dependence of Aβ production on cholesterol has generally been explained by localization of β- and γ-secretases to lipid rafts (12Burns M. Gaynor K. Olm V. Mercken M. LaFrancois J. Wang L. Mathews P.M. Noble W. Matsuoka Y. Duff K. J. Neurosci. 2003; 23: 5645-5649Crossref PubMed Google Scholar, 13Cordy J.M. Hussain I. Dingwall C. Hooper N.M. Turner A.J. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 11735-11740Crossref PubMed Scopus (310) Google Scholar, 14Tun H. Marlow L. Pinnix I. Kinsey R. Sambamurti K. J. Mol. Neurosci. 2002; 19: 31-35Crossref PubMed Scopus (79) Google Scholar, 15Wahrle S. Das P. Nyborg A.C. McLendon C. Shoji M. Kawarabayashi T. Younkin L.H. Younkin S.G. Golde T.E. Neurobiol. Dis. 2002; 9: 11-23Crossref PubMed Scopus (361) Google Scholar). The relationship between Aβ and cholesterol has led to speculation that statins, a type of medicine that lowers cholesterol, might reduce Aβ production and be useful as a therapeutic agent for AD (16Wolozin B. Kellman W. Ruosseau P. Celesia G.G. Siegel G. Arch. Neurol. 2000; 57: 1439-1443Crossref PubMed Scopus (1336) Google Scholar, 17Jick H. Zornberg G.L. Jick S.S. Seshadri S. Drachman D.A. Lancet. 2000; 356: 1627-1631Abstract Full Text Full Text PDF PubMed Scopus (1575) Google Scholar). Cholesterol catabolites also modulate Aβ production. Cholesterol esters increase Aβ production, and inhibiting the enzyme acyl-coenzyme A:cholesterol acyltransferase that synthesizes these enzymes reduces Aβ production (18Puglielli L. Konopka G. Pack-Chung E. Ingano L. Berezovska O. Hyman B. Chang T. Tanzi R. Kovacs D. Nat. Cell Biol. 2001; 3: 905-912Crossref PubMed Scopus (384) Google Scholar). In contrast, recent studies show that the synthetic oxysterol 22-hydroxycholesterol inhibits Aβ production in some neuroblastoma cell lines (19Koldamova R.P. Lefterov I.M. Ikonomovic M.D. Skoko J. Lefterov P.I. Isanski B.A. DeKosky S.T. Lazo J.S. J. Biol. Chem. 2003; 278: 13244-13256Abstract Full Text Full Text PDF PubMed Scopus (217) Google Scholar, 20Sun Y. Yao J. Kim T.W. Tall A.R. J. Biol. Chem. 2003; 278: 27688-27694Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar). Understanding how production of 24(S)-hydroxycholesterol affects APP metabolism has taken on particular relevance because of two reports suggesting that polymorphisms in CYP46A1 are associated with an increased load of Aβ or earlier onset of AD (21Kolsch H. Lutjohann D. Ludwig M. Schulte A. Ptok U. Jessen F. von Bergmann K. Rao M.L. Maier W. Heun R. Mol. Psychiatry. 2002; 7: 899-902Crossref PubMed Scopus (126) Google Scholar, 22Papassotiropoulos A. Streffer J.R. Tsolaki M. Schmid S. Thal D. Nicosia F. Iakovidou V. Maddalena A. Lutjohann D. Ghebremedhin E. Hegi T. Pasch T. Traxler M. Bruhl A. Benussi L. Binetti G. Braak H. Nitsch R.M. Hock C. Arch. Neurol. 2003; 60: 29-35Crossref PubMed Scopus (192) Google Scholar). In the present study we compared the expression of CYP24 with CYP27A1 in the brains of patients with AD and age-matched controls. We also examined how the oxysterol products of these two enzymes, 24-OHC and 27-OHC, affect APP processing in primary neurons. We observed that the distribution of both enzymes change in the brains of subjects with AD. CYP46A1 is selectively expressed in degenerating neurites around senile plaques, whereas CYP27A1 expression is increased in white matter oligodendrocytes. We showed that 24- and 27-OHC inhibit Aβ production and protein kinase C (PKC)-stimulated APP secretion, but 24-OHC is more potent that 27-OHC. Inhibition of APP secretion can be explained by indirect inhibition of PKC activity in neurons, which suggests a role for 24-OHC in signal transduction. Tissue—Cortical tissue (fixed and frozen) was obtained from seven subjects with AD (mean age 77.0 ± 2.2 years, mean post-mortem interval 16.9 ± 6.0 h) and seven normal subjects (mean age 73.3 ± 3.1 years, mean post-mortem interval 6.3 ± 2.4 h). The tissue was obtained from the Loyola University Medical Center Brain Bank. All AD subjects had moderate to severe dementia. Primary Cortical Neurons—Neurons were obtained from Sprague-Dawley rat pups (E18), cultured on poly-l-lysine-coated 75-cm2 flasks using 9 million cells/flask, and plated with neurobasal medium with B27 supplement (Invitrogen), which contains retinoic acid. At day 3, the neurons were treated with 50 μm araC. After 24 h the medium was removed, and fresh medium containing neurobasal, B27 supplement, and 10% fetal bovine serum was added. Six days after plating, the cells were treated with APP adenovirus (multiplicity of infection = 10) for 6 h (23Yuan H. Zhai P. Anderson L.M. Pan J. Thimmapaya B. Koo E.H. Marquez-Sterling N.R. J. Neurosci. Methods. 1999; 88: 45-54Crossref PubMed Scopus (10) Google Scholar). The next day, cells were treated with an oxysterol in fresh medium consisting of neurobasal, B27 supplement, and 10% fetal bovine serum for 24 h. Medium was collected for measuring Aβ. The cells were then incubated in fresh serum-free neurobasal/B27 medium containing oxysterol. After 1 h, the medium was replaced with fresh serum-free neurobasal/B27 medium containing oxysterol for 1 h, and the medium from this second incubation was collected to measure basal APP secretion. The neurons were then incubated in fresh serum-free neurobasal/B27 medium containing oxysterol with and without 1 μm phorbol 12-myristate 13-acetate (PMA) for 1 h, and the medium was again collected. Secreted APP was concentrated from the medium samples by heparin-agarose immunoprecipitation, and APP was analyzed by immunoblot (24Palacino J. Berechid B. Alexander P. Eckman C. Younkin S. Nye J. Wolozin B. J. Biol. Chem. 2000; 275: 215-222Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). Cells were collected, APP in the lysates was analyzed by immunoblot, and PKC was assayed as described below. For analysis of ABCA1, the neurons were lysed in buffer containing 250 mm sucrose, 10 mm HEPES (pH 7.4), plus protease inhibitors. After quantification of protein by BCA assay, the lysates were incubated in SDS loading buffer for 30 min, separated by electrophoresis on a SDS-6% polyacrylamide gel, and then immunoblotted (25Fukumoto H. Deng A. Irizarry M.C. Fitzgerald M.L. Rebeck G.W. J. Biol. Chem. 2002; 277: 48508-48513Abstract Full Text Full Text PDF PubMed Scopus (140) Google Scholar). Adenoviral Treatment—Adenoviral expression of APP was achieved by infecting neurons at a multiplicity of infection of 10 as described previously (23Yuan H. Zhai P. Anderson L.M. Pan J. Thimmapaya B. Koo E.H. Marquez-Sterling N.R. J. Neurosci. Methods. 1999; 88: 45-54Crossref PubMed Scopus (10) Google Scholar). Antibodies—Antibodies 3046 and 3047 were rabbit polyclonal antibodies produced against amino acids 226–242 of human CYP46A1. The third CYP46A1 antibody, T623, was produced against amino acids 254–270 of murine Cyp46A1 and has been described previously (5Lund E.G. Guileyardo J.M. Russell D.W. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7238-7243Crossref PubMed Scopus (521) Google Scholar). Anti-CYP27A1 antibody recognizes amino acids 15–28 of mature CYP27A1 and is similar to one described previously by Cali and Russell (26Cali J.J. Russell D.W. J. Biol. Chem. 1991; 266: 7774-7778Abstract Full Text PDF PubMed Google Scholar). LN27, a monoclonal antibody directed against the N terminus of APP, was used to detect APPs (Zymed Laboratories Inc.). Holo-APP and C-terminal fragments were detected using antibody 0443, as described previously (27Marlow L. Cain M. Pappolla M.A. Sambamurti K. J. Mol. Neurosci. 2003; 20: 233-240Crossref PubMed Scopus (79) Google Scholar). ABCA1 was detected with rabbit anti-ABCA1 (1:1000, Novus). Immunocytochemistry—Free-floating fixed 40-μm tissue sections were used for immunocytochemistry. The samples were washed, permeabilized by incubation for 30 min with 0.2% Triton X-100, blocked with 5% dry milk, 1% goat serum, and phosphate-buffered saline, washed, incubated overnight in primary antibody in 2% bovine serum albumin, 0.1% Triton X-100, and phosphate-buffered saline, washed, incubated for 3 h in secondary antibody in 2% bovine serum albumin, 0.1% Triton X-100, and phosphate-buffered saline, washed, and then cover-slipped. Immunoblotting—Transfers to polyvinylidene difluoride (Bio-Rad) were carried out overnight at 4 °C at 0.05 amps/gel in transfer buffer. The immunoblot was blocked in 0.2% I-block (Tropix) in Tris-buffered saline with 0.1% Tween 20 for 1 h at room temperature with shaking. We then incubated blots overnight at 4 °C in primary antibody at appropriate the concentration in 5% bovine serum albumin in Tris-buffered saline with Tween 20. Blots were then washed three times for 10 min each and incubated for 3 h in secondary antibody (1:5000, Jackson Laboratories) in I-block at room temperature. Blots were washed three times and developed using a chemiluminescent reaction (PerkinElmer Life Sciences). Thioflavine T Staining—Following the immunocytochemical treatment, sections were incubated with 0.5% thioflavine T for 8 min, washed three times in 80% ethanol, washed once in H2O, and then mounted. Aβ Enzyme-linked Immunosorbent Assays—Secreted Aβ40 and Aβ42 levels were analyzed by sandwich enzyme-linked immunosorbent assay system as described previously using the following antibodies: Aβ40, MM27 33.1.1 capture, MM32 13.1.1 for detection; Aβ42, MM27 33.1.1 capture, MM26 4.1.3 for detection (28Suzuki N. Cheung T. Cai X. Odaka A. Otvos L. Eckman C. Golde T. Younkin S. Science. 1994; 264: 1336-1340Crossref PubMed Scopus (1344) Google Scholar). PKC Assay—PKC activity was measured by monitoring the phosphorylation of a PKC substrate peptide (QKRPSQRSKYL) using a PKC assay kit (Upstate Biotechnology, Lake Placid, NY). Lysates from brain or cortical neurons were harvested in PKC lysis buffer (150 mm phosphate-buffered saline (pH 7.4), 50 mm Tris HCl, 1% Nonidet P-40, 1 mm EDTA, plus 0.02 V protease and phosphatase inhibitor mixtures (Sigma)). Reactions were carried out in a 50-μl volume at 30 °C for 10 min in the presence of 20 mm MOPS (pH 7.2), 1 mm CaCl2, 25 mm β-glycerolphosphate, 1 mm sodium orthovanadate, 1 mm dithiothreitol, 0.5 mg/ml phosphatidylserine, 0.5 mg/ml diacylglycerol, and 30–100 μg of sample. 10 μl of an Mg2+/ATP mixture (75 mm magnesium chloride, 500 μm ATP, 1–2 μCi of [32P]ATP (3000 Ci/mmol)) was added to start the reaction. Twenty microliters of reaction mixture were spotted onto a 1-cm square of phosphocellulose P-81 paper at the end of the incubation period. The squares were washed five times with 0.75% phosphoric acid and once in acetone and counted for radioactivity. Samples containing no substrates were included in every experiment to determine basal levels of radioactivity, and the basal radioactivity was subtracted from the experimental counts. Statistics—All results are presented as mean ± S.E. When analysis of variance showed significant differences, pairwise comparisons between the means are analyzed by Newman-Keuls post-hoc testing. IC50 values were calculated using the Prism program (GraphPad). Production of Antibodies to CYP46A1—The ability of oxysterols to modulate APP processing suggests that their pattern of expression in the brain in AD could influence production of Aβ. To explore this issue, we used immunochemistry to analyze the expression patterns of CYP46A1 and CYP27A1, which produce 24- and 27-OHC, respectively. We used three different antibodies. Antibodies 3046 and 3047 were generated against amino acids 226–242 of human CYP46A1. Both antibodies identified CYP46A1 by immunoblotting (Fig. 1, A and B). Antibodies 3046 and 3047 both recognized the 70-kDa band characteristic of CYP46A1. Preadsorption of both antibodies with the immunizing peptide eliminated all reactivity (Fig. 1D). A third antibody, which was produced against amino acids 254–270 of murine CYP46A1, gave a similar banding pattern (Fig. 1C) (5Lund E.G. Guileyardo J.M. Russell D.W. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 7238-7243Crossref PubMed Scopus (521) Google Scholar). Immunocytochemistry of CYP46A1 and Expression in Neurons, Astrocytes, and Senile Plaques—Next, we examined the distributions of CYP46A1 and CYP27A1 in human frontal cortex from donors without neurological disease. Antibody 3047 gave the least background staining. CYP46A1 expression was detected in pyramidal neurons (Fig. 2A) as well as in occasional astrocytes (Fig. 2B). No staining was observed with preimmune serum (Fig. 2D). The relative amount of expression in astrocytes and neurons did not show a correlation with post-mortem interval, sex, or age, although a larger sample size would be required to obtain definitive statistical information on this question. Staining with the anti-CYP46A1 antibody also showed reactivity in neurons and astrocytes in frontal cortex tissue from an AD brain, with more neurons evident in tissue from neurologically normal samples (Fig. 2E). The striking feature of CYP46A1 staining in the Alzheimer brain tissue was the presence of CYP46A1 among neuritic plaques (Fig. 2C). Staining of neurons and astrocytes in AD cortex was similar to that seen in normal cortex, although the number of neurons expressing CYP46A1 was decreased, possibly because of disease-related neuronal loss (Fig. 2E). Preadsorption removed staining of CYP46A1 in neurons, astrocytes, and neuritic plaques (data not shown). To determine whether CYP46A1 staining localized to the amyloid core or the neuritic periphery, we performed double staining with thioflavine T (Fig. 3). Double staining with thioflavine T and anti-CYP46A1 antibody 3047 showed that CYP46A1 reactivity was present in over 90% of the plaques identifiable by thioflavine T (Fig. 3). The CYP46A1 reactivity was present in the periphery of neuritic plaques but did not co-localize with the amyloid core, suggesting that the CYP46A1 was not directly associated with compact, aggregated Aβ (Fig. 3). The peripheral staining around neuritic plaques was consistent with staining in degenerating neurites. In contrast to the strong concordance between thioflavine staining and CYP46A1 staining, we did not observe CYP46A1 staining in any tangles that were thioflavine-positive (data not shown). A similar pattern of reactivity was observed for anti-CYP46A1 antibody 3046 (data not shown).Fig. 3CYP46A1 is located in and around the amyloid core of neuritic plaques.A, immunocytochemical staining of CYP46A1 (red) in the frontal cortex of Alzheimer tissue identifies neuritic plaques. B, thioflavine T (green) staining identifies neuritic plaques and exhibits intense reactivity in the amyloid core. C, merging panels A and B shows that CYP46A1 (red) reactivity is located around the amyloid core (green), suggesting that the enzyme is present in the neuritic periphery of the plaque. Bar, 30 μm.View Large Image Figure ViewerDownload (PPT) Next, we examined the pattern of CYP27A1 expression. The protein CYP27A1 was analyzed using an anti-CYP27A1 antibody recognizing amino acids 15–28 of mature CYP27A1, similar to one described previously by Cali and Russell (26Cali J.J. Russell D.W. J. Biol. Chem. 1991; 266: 7774-7778Abstract Full Text PDF PubMed Google Scholar). In the normal cortex, CYP27A1 was present in many neurons where it showed strong reactivity in both the soma and proximal dendrites (Fig. 4A). Little staining of CYP27A1 was observed in the white matter of normal brain (Fig. 4B). The pattern of CYP27A1 staining in the AD brain differed from that in normal brain. Neuronal staining with anti-CYP27A1 antibody was greatly reduced (Fig. 4, C and E). In contrast, there was a large increase of staining in the white matter of the AD brain (Fig. 4, D and E). The CYP27A1 reactivity present in the white matter occurred in the soma of small round cells, the morphology that is classic for oligodendrocytes (Fig. 4D). In the AD brain, quantification showed a 66% (p < 0.05) reduction in the number of cortical neurons identified per high power field and a 4.6-fold (p < 0.01) increase in the number of oligodendrocytes identified per high power field (Fig. 4E). The staining did not show a correlation with post-mortem interval, sex, or age, although a larger sample size would be required to obtain definitive statistical information on this question. Quantification of CYP46A1 Levels in AD and Normal Brains—Immunocytochemical studies show that CYP46A1 and CYP27A1 proteins are present in multiple cell types in the AD and normal brain. To understand how the complex changes in expression affect the total expression of each protein, we performed immunoblots on cortical tissue from normal and AD brain. The CYP46A1 reactivity was apparent as a band at 70 kDa (Fig. 5A, arrow). There was a large variation in the total expression of CYP46A1 among the brains. Although quantification of band intensities showed a modest increase in staining in the AD brain, the increase was not statistically significant (normalized OD: control, 0.17 ± 0.15; AD, 0.34 ± 0.20; n = 7 each, p = 0.094). The CYP27A1 reactivity was apparent as a band at 47 kDa (Fig. 5B, arrow). CYP27A1 band intensity also showed variation among subjects, which resulted in no significant change in band intensity in AD brain samples (normalized OD: control, 0.35 ± 0.21; AD, 0.21 ± 0.11; n = 7 each, p = 0.13). The absence of disease-related differences in total CYP46A1 and CYP27A1 detected by immunoblot might occur because the decreased neuronal staining balances the increased staining in neuritic plaques (CYP46A1) or oligodendrocytes (CYP27A1). 24-OHC Inhibits APP Processing in Cortical Neurons Grown in Primary Cultures—Recent studies show that the synthetic oxysterol 22-OHC inhibits Aβ secretion in cell lines, but the modulation of APP processing by the natural oxysterols 24- and 27-OHC has yet to be investigated (19Koldamova R.P. Lefterov I.M. Ikonomovic M.D. Skoko J. Lefterov P.I. Isanski B.A. DeKosky S.T. Lazo J.S. J. Biol. Chem. 2003; 278: 13244-13256Abstract Full Text Full Text PDF PubMed Scopus (217) Google Scholar, 20Sun Y. Yao J. Kim T.W. Tall A.R. J. Biol. Chem. 2003; 278: 27688-27694Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar). In addition, modulation of APP processing by oxysterols in primary cultures of neurons also has not been investigated. To allow detection of Aβ, cortical neurons were infected with adenovirus, driving the expression of human APP (23Yuan H. Zhai P. Anderson L.M. Pan J. Thimmapaya B. Koo E.H. Marquez-Sterling N.R. J. Neurosci. Methods. 1999; 88: 45-54Crossref PubMed Scopus (10) Google Scholar). Two days later, the neurons were treated with 10 μm 24-OHC for up to 24 h to determine whether oxysterols exerted rapid or delayed actions on Aβ or APP secretion. At this point, medium was collected for Aβ and APP analysis. We observed that 24-OHC inhibited both Aβ secretion and PMA-stimulated APP secretion (Fig. 6, A and B). 24-OHC showed slightly stronger inhibition of Aβ and APP secretion after 24 h than after 6 h (Fig. 6A). By 24 h, 24-OHC inhibited Aβ secretion by up to 68.5 ± 1.6% (p < 0.001) and APP secretion by over 65% (at 6 h, 68.3 ± 14.9% inhibition, p < 0.05; at 24 h, 87.1 ± 10.4 inhibition, p < 0.005) (Fig. 6, A and B). No changes were observed in steady state levels of cellular APP, although the levels of α-C-terminal fragments were decreased in the presence of 24-OHC, suggesting that the changes in APP processing are the result of decreased cleavage of APP by secretases (Fig. 6C). Next we examined the dose-response relationship of 24-OHC after 24 h treatment. 24-OHC inhibited secretion of Aβ40 and Aβ42, with an IC50 of ∼7 nm, based on a plateau for inhibition at 40% control values (Fig. 6, D and E). 24-OHC also prevented PKC from stimulating APP secretion, exhibiting an IC50 for the inhibition of APP secretion similar to that observed for the inhibition of Aβ secretion (Fig. 6F). To determine whether the changes in APP processing correlated with LXR activity, we analyzed the expression of ABCA1, which is induced by LXR activity. Primary neurons were treated with 24-OHC as described above, and then ABCA1 expression in the lysates was determined by immunoblot. Expression of ABCA1 was stimulated by treatment with 1 μm 24-OHC, consistent with LXR activation (Fig. 6G). Finally, we investigated whether 24-OHC was causing toxicity in the neuronal cultures, because toxicity could reduce Aβ secretion. Analysis of toxicity by an assay using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide showed no evidence of neurotoxicity over 24 h of OHC treatment at doses below 5 μm 24-OHC (data not shown). These data indicate that 24-OHC inhibits both Aβ and APP secretion and that the inhibition is accompanied by LXR activation. 27-OHC Inhibits Aβ and APPs Secretion—Next we examined the effects of 27-OHC on APP processing. We observed that 27-OHC also inhibited APP processing, although with lower potency. Cortical neurons grown in primary culture were exposed for 24 h to 1–15 μm 27-OHC. The medium and cellular lysates were then collected, and APP processing was examined. 27-OHC inhibited secretion of Aβ40 and Aβ42 with an IC50 of ∼2 μm, based on a plateau for inhibition at 62% of control values (Figs. 6, A and D, and 7A). 27-OHC also inhibited PKC stimulated APP secretion (Fig. 7C). Immunoblots of the lysates with anti-ABCA1 antibody indicated that the induction of ABCA1 paralleled the inhibition of APP processing (Fig. 7D). This suggests that the inhibition of Aβ by 27-OHC was less potent than th
Referência(s)