Elevated Expression of a Regulator of theG2/M Phase of the Cell Cycle, NeuronalCIP-1-Associated Regulator of Cyclin B,in Alzheimer’s Disease

Xiongwei Zhu,1 Andrew McShea,2 Peggy L.R. Harris,1 Arun K. Raina,1

Rudy J. Castellani,3 Jens Oliver Funk,4 Sapan Shah,1 Craig Atwood,1,5

Richard Bowen,6 Robert Bowser,7 Laura Morelli,8 George Perry1 and Mark A. Smith1*1Institute of Pathology, Case Western Reserve University, Cleveland, Ohio2Combimatrix Corp., Mukilteo, Washington3Michigan State University Clinical Center, East Lansing, Michigan4Laboratory of Molecular Tumor Biology, Department of Dermatology, Friedrich-Alexander-Universityof Erlangen-Nuremberg, Erlangen, Germany5Section of Geriatrics and Gerontology, Department of Medicine, Univeristy of Wisconsin, Madison, Wisconsin6Voyager Pharmaceutical Company, Raleigh, North Carolina7Department of Pathology, University of Pittsburgh, Pittsburgh, Pennsylvania8Instituto de Quımica y Fisicoquımica Biologicas (IQUIFIB), Universidad de Buenos Aires-CONICET,Buenos Aires, Argentina

Adult neurons are generally accepted to be in a quies-cent, nonproliferative state. However, it is becoming in-creasingly apparent that, in Alzheimer’s disease (AD),alterations in cell cycle machinery, suggesting an attemptto reenter cell cycle, relate temporally and topographi-cally to degenerating neurons. These findings, togetherwith the fact that neurons lack the necessary compo-nents for completion of mitosis, have led to the notionthat an ill-regulated attempt to reenter the cell cycle isassociated with disease pathogenesis and, ultimately,neuronal degeneration. To understand better the role ofsuch cell cycle abnormalities in AD, we undertook astudy of CIP-1-associated regulator of cyclin B (CARB), aprotein that associates with two key proteins, p21 andcyclin B, involved in cellular checkpoints in the cell cycle.Our results show that there are increases in CARB local-ized to intraneuronal neurofibrillary tangles and granulo-vacuolar degeneration in susceptible hippocampal andcortical neurons in AD. By marked contrast, CARB isfound only at background levels in these neuronal pop-ulations in nondiseased age-matched controls. Our datanot only provide another line of evidence indicative of cellcycle abnormalities in neurons in AD but also lend furthercredence to the hypothesis that susceptible neurons maybe arrested at the G2/M phase of the cell cycle beforethey die. Therefore, therapeutics targeted toward initia-tors of the cell cycle are likely to prove of great efficacyfor the treatment of AD. © 2004 Wiley-Liss, Inc.

Key words: Alzheimer’s disease; cell cycle; CIP-1-associated regulator of cyclin B cytoskeleton; pathogen-esis; phosphorylation; tau

Several recent findings have highlighted the similar-ities between neurogenesis during development and neu-rodegeneration during Alzheimer’s disease (AD). Indeed,neuronal populations that are known to degenerate in AD,despite their postmitotic status, exhibit phenotypicchanges characteristic of cells reentering the cell divisioncycle (Bowser and Smith, 2002). Evidence for cell cycleprocesses and AD includes the activation of related signaltransduction pathways and cell-phase-dependent kinasesand transcriptional activation that lead to cytoskeletal al-teration and DNA replication (Arendt et al., 1996; Vin-cent et al., 1996, 1997; McShea et al., 1997, 1999a,b;Nagy et al., 1997; Busser et al., 1998; Perry et al., 1999;Zhu et al., 1999, 2000a–c, 2001a,b, 2003; Yang et al.,2001, 2003; Ogawa et al., 2003a,b). The consequences ofsuch a reentrant phenotype are now beginning to beunderstood (McShea et al., 2000b), and, given the linkbetween loss of cell cycle control and apoptotic neuronaldeath during development, attempts by neurons to prolif-erate may be a central event in the process of neurode-

Contract grant sponsor: Alzheimer’s Association; Contract grant number:NIRG-02-3923; Contract grant sponsor: Voyage Pharmaceutical Corpo-ration.

*Correspondence to: Mark A. Smith, PhD, Institute of Pathology, 2085Adelbert Road, Case Western Reserve University, Cleveland, OH 44106.E-mail: mark.smith@case.edu

Received 6 October 2003; Revised 11 December 2003; Accepted 12December 2003

Journal of Neuroscience Research 75:698–703 (2004)

© 2004 Wiley-Liss, Inc.

generation. Because a number of markers in the G2 phaseof the cell cycle have been detected in AD neurons, suchas induction of cyclin B (Vincent et al., 1997), elevatedcdc2 kinase activity (Vincent et al., 1997), and the appear-ance of polo-like kinase 1 (Harris et al., 2000), we hy-pothesized that susceptible neurons enter G2 phase andarrest at the G2/M checkpoint, which leads to their ulti-mate cell death. However, there is a paucity of informa-tion regarding cell cycle checkpoint markers in susceptibleneurons of AD, so in this study we investigated markersassociated with the G2 phase of the cell cycle. The iden-tification of CIP-associated regulator of cyclin B (CARB),a novel regulator of the G2/M transition of the cell cycle(McShea et al., 2000a), prompted us to determine whetherits expression levels were altered in AD neurons. CARBwas identified in a protein interaction trap with the cellcycle checkpoint gene p21. CARB localization was foundto be partially located proximal to the centrosomal in thecell line (HCT116) and also colocalized with anotherantigen known to be centrosome in origin, cyclin B(McShea et al., 2000a). The interaction of p21 and cyclinB with CARB seemed to be mutually exclusive, suggest-ing that this molecule may be involved in regulatoryprocesses occurring at the G2 phase of the cell cycle, whenboth cyclin B and p21 are present. CARB also has someother interesting features; it has been identified as a mem-ber of the syntaxin family and christened syntaxin-8, basedon a partial homology to syntaxin-6 (McShea et al.,2000a). Although the significance of this homology isunclear, it does suggest that this protein might be involvedin its subcellular localization and/or the transport of pro-teins or vesicles. CARB also associates with pericentrin, acritical component of the mitotic spindle machinery thatregulates the equal segregation of genetic material duringmitosis. Given both the proposed roles of cell cycle reac-tivation (Arendt et al., 1996; Vincent et al., 1996, 1997;McShea et al., 1997, 1999a,b; Nagy et al., 1997; Busser etal., 1998; Zhu et al., 1999, 2000a,c; Yang et al., 2001,2003; Ogawa et al., 2003a,b) and the possibility that ADneuropathology could be related to cell cycle defects at theG2/M phase of the cell division cycle, we investigated thestatus of CARB in diseased and control brains.

MATERIALS AND METHODS

Tissue

Hippocampal and neocortical tissue from cases of AD(n � 9, ages 59–96 years; Table I) and aged-matched controls(n � 8, ages 62–81 years; Table I) with similar post-mortemintervals (AD, 2–47 hr; controls, 4–30 hr), were fixed in 10%buffered formalin or methacarn (methanol:chloroform:aceticacid; 60:30:10) at 4°C overnight. No difference in immunocy-tochemical results was noted between different fixatives in thisstudy. AD cases that met CERAD criteria for AD (Khachatu-rian, 1985; Mirra et al., 1991) were further assessed post-mortemand confirmed to be corresponding to Braak stages V–VI (Braakand Braak, 1991). Control cases were also assigned by CERADcriteria and, in some instances, showed only age-related patho-logical structures identified with phosphorylated tau (AT8).

After fixation, tissue was dehydrated through ascending ethanoland embedded in paraffin, and 6-�m sections were placed onsilane-coated slides.

Antibody Preparation

The rabbit polyclonal antibody against CARB was firstused for and is briefly described in our previous publication(McShea et al., 2000a). Briefly, prior to raising an antibody,12 rabbits were screened by Western blotting with “prebleed”serum at 1:100 against whole-cell lysates. The two rabbits withthe least reactivity (none in the range 2–100 kDa) were chosenfor subsequent immunization. Full-length CARB was expressedas a His-tagged protein from pET16b (Novagen) in Escherichiacoli (strain BL21 DE3), purified on Ni-NTA resin (Qiagen)under denaturing conditions in 8 M urea, and step (8, 7, 6, 5, 4,3, 2, 1, 0.5 M urea) dialyzed into buffered saline for immuni-zation. A specific antiserum was raised in rabbits againsthexahistidine-tagged protein expressed in E. coli. The antiserumused in this study was from the third, fourth, or fifth immuni-zation.

Immunocytochemistry

Tissue sections were deparaffinized in xylene, then hy-drated through descending ethanol. Endogenous peroxidase ac-tivity was reduced by a 30-min incubation in 3% H2O2 inmethanol, followed by a 30-min incubation in 10% normal goatserum to block nonspecific binding sites. Tissue sections wereimmunostained with either rabbit antisera to CARB (1/100) ormonoclonal antiserum to phosphorylated tau (AT8; 1:2,000),followed by the peroxidase-antiperoxidase method with 3,3�-diaminobenzidine as cosubstrate as modified from previouslydescriptions (Sternberger and Sternberger, 1986). Control ex-

TABLE I. Cases Used in This Study*

CaseAge

(years)PMI(hr) Sex

Dementia(years)

Cause ofdeath

AD cases (n � 9)AD1 59 7 F 9 EAD2 69 8 M 8 NAAD3 76 47 F NA NAAD4 81 10.5 M 3 RAD5 89 8 F NA NAAD6 91 2 F 10 NAAD7 93 7.5 F NA HAD8 93 22 F 9 NAAD9 96 13.5 F 8 NA

Controls (n � 8)C1 62 19 M CC2 63 22 M HC3 67 18 M HC4 69 4 F CC5 74 30 F SDC6 78 24 F RC7 78 13 M KC8 81 22 M H

*C, cancer; E, emaciation; F, female; H, heart attack; K, kidney failure; M,male; NA, not available; PMI, post-mortem interval, R, respiratory illness;SD, sudden death.

CARB in Alzheimer’s Disease 699

periments included omission of primary antisera or adsorption ofCARB antisera with the same CARB protein immunogen(10 mg/ml) overnight at 4°C prior to application to the tissuesections.

Immunoblotting

Gray matter tissues from hippocampus or frontal cortexfrom AD (n � 4) and control (n � 4) cases were homogenizedin 10 volumes of Tris-buffered saline (TBS) with 0.02% sodiumazide, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate(SDS), 1% NP-40, 1 mM phenylmethylsulfonyl fluoride(PMSF), 1 �g/ml aprotinin, and 1 �g/ml antipain. Ten micro-grams of protein were loaded per lane and separated by poly-acrylamide gel electrophoresis. Protein was transferred toImmobilon-P (Millipore, Bedford, MA) and probed with anti-CARB, followed by affinity-purified goat anti-rabbit immuno-globulin peroxidase conjugate preadsorbed to eliminate humancross-reactivity (StressGen Biotechnologies, Victoria, BritishColumbia, Canada) and developed using ECL (Santa CruzBiotechnology, Santa Cruz, CA). Blots were stripped in strip-ping buffer (2% SDS, 62.5 mM Tris-HCl, 100 nM�-mercaptoethanol, pH 6.8) for 30 min at 60°C and thenprobed with anti-heme oxygenase-2 (HO-2), a constitutivelyexpressed protein (Zhu et al., 2000b).

RESULTSIn the brains of individuals affected by AD, CARB

showed intense colocalization with neurofibrillary tangles(NFTs) throughout the sections of medial temporal tissue,including hippocampus, subiculum, entorhinal cortex, andisocortex (Fig. 1A). Dystrophic neurites within neuriticplaques were occasionally stained. Immunostaining forCARB differed slightly from AT8 (phospho-tau) immu-nocytochemistry, in that neuropil threads were generallyunstained with CARB (not shown). Extracellular amyloid

deposits within senile plaques also did not stain withanti-CARB immunocytochemistry. CARB staining ofpathologically unaffected neurons, reactive astrocytes, andblood vessels with and without mural amyloid depositswas also present; however, the intensity of staining ap-proached background levels encountered in control tissue.Also noteworthy was the intense anti-CARB immunola-beling of neuronal cytoplasmic granules of granulovascuo-lar degeneration (Fig. 1A). Not surprisingly, this latterfinding was most conspicuous within Sommer’s sector(CA-1), where granulovacuolar degeneration is generallymost pronounced (not shown). In aged-matched controlcases (Fig. 1B), CARB was found only at backgroundlevels in similar pyramidal neurons, but CARB did labelNFTs in those control cases that contained NFTs. Nonuclear staining was seen in either the AD cases or thecontrols. Of importance is that there is no differencebetween AD and age-matched control cases in cerebellum(data not shown), an area that is unaffected in AD. Ad-sorption of anti-CARB with CARB protein prior toapplication to the section abolishes the immunolabeling,demonstrating the specificity of the antibody (Fig. 2).

Immunoblotting revealed an anti-CARB reactiveband at the appropriate molecular weight of 27 kD in ADand control (Fig. 3A), which further demonstrated thespecificity of the antibody. Quantification of the blot,which was normalized with the constitutively expressedHO-2 protein, confirms our immunocytochemistry find-ings and shows that CARB protein is significantly in-creased more than tenfold in AD compared with controlcases (P � .02; Fig. 3B).

DISCUSSIONIn this study, we show that CARB, a G2/M phase

regulator, is specifically increased in the cytosolic com-

Fig. 1. Immunocytochemical localization of CARB in hippocampal neurons in AD patients (A; therepresentative case AD3) and control patients (B; the representative case C7). A: CARB is localizedto neurofibrillary tangles and senile plaques (inset), with some neurons containing CARB restrictedto granulovacuolar degeneration in AD (arrows). B: Only background levels of CARB immuno-staining are found in control neurons. Scale bars � 50 �m.

700 Zhu et al.

partment of vulnerable neurons in AD, in particular thoseneurons affected by intracellular NFT formation andgranulovacuolar degeneration. Similar neuronal popula-tions in control cases showed only background levels ofimmunoreactivity. Confirming our immunocytochemicalstudies, immunoblot analyses revealed statistically signifi-

cantly (P � .02) higher levels of CARB in homogenates ofAD compared with control brain. Given the role ofCARB in cyclin B regulation and G2/M phase transition(McShea et al., 2000a), our findings lend further credenceto the hypothesis that susceptible neurons may be arrestedat the G2/M phase, which ultimately leads to cell death.

Fig. 2. Neuronal staining in the hippocampus with CARB antibody in AD (A; the representative caseAD2) is abolished completely by absorption with CARB protein (B). Scale bar � 100 �m.

Fig. 3. A: Representative results of immunoblots of cortical gray matter from AD and control patientsprobed with antisera against CARB show a strong band at the expected molecular weight of about27 kDa in AD (AD) and weaker in control (con) samples. The purified recombinant CARB proteindetected by the same antibody is shown as positive control (PC). B: Quantification of CARB,normalized to the constitutive HO-2 protein, shows a significant increase of CARB intensity in AD(*P � .02). Result is shown as average � SEM.

CARB in Alzheimer’s Disease 701

Indeed, CARB affects the steady-state levels of cyclin B1by preventing proteolytic degradation, and CARB over-expression leads to increased levels of cyclin B1 in CARB-transfected cells (McShea et al., 2000a). Therefore, theelevation of CARB observed in AD neurons in this studymay contribute to the aberrant elevation of cyclin B1observed in susceptible neurons in AD (Vincent et al.,1997). Insofar as translocation of cyclin B into the nucleusis a prerequisite for the activation of cdc2 and the onset ofthe prophase of mitosis, CARB sequestration of cyclin Bin the cytoplasm likely regulates the transit through theG2/M phase of the cell cycle. Therefore, it is conceivablethat elevated levels of CARB in AD may cause excessivesequestration of cyclin B1 in the cytoplasm, as was foundin many susceptible neurons in AD (Vincent et al., 1997),and contribute to cell cycle arrest. That cyclin B1 was alsofound in the nucleus in some AD cases (Vincent et al.,1997) may be due to the concurrent increase of p21 CIP-1in AD (McShea et al., 1999a; Engidawork et al., 2001),which releases cyclin B1 from interaction with CARB.

Of significance is the observation that most of theCARB-positive NFTs were intraneuronal rather than ex-traneuronal. A similar observation was reported previouslyfor the active cdc2, presumably associated with cyclin B,and MPM-2-reactive phosphoepitopes (Vincent et al.,1996, 1997). Given that cyclin B1 indirectly associateswith CARB in a multiprotein complex (McShea et al.,2000a), this colocalization suggests that cdc2 may also beassociated with CARB as part of the multiprotein complexthrough the interaction with cyclin B1. Therefore, it isconceivable that CARB can further provide docking sitesto sequester active mitotic kinase (cdc 2/cyclin B1 pairs)from entering the nucleus. This not only directly leads tocell cycle arrest but also results in the generation of MPMepitopes observed in the cytoplasm, which may contributeto NFT formation (Vincent et al., 1996, 1997).

The presence of a growing number of control ele-ments of cell cycle machinery in degenerating neurons ofAD suggests a crucial role for temporally ectopic cell cyclereentry driving proximal pathophysiology in AD. Giventhat successful mitosis has not been reported for neuronalpopulations in AD, it is plausible that neurons becomearrested at some point between G1 and M phases. De-pending on where cells are arrested, they can either returnto G0 and redifferentiate or die via apoptosis. For example,if a terminally differentiated cell, for any reason, reacheslate G1 or even later, when cyclin A is expressed, arrestleads to cell death. Therefore, the characterization of theexact phase when arrest occurs will help to determine themechanism by which neurons degenerate. The presence ofcyclin E/cdk2 complex indicates that neurons have passedG1. Coordinated DNA replication suggests that the sus-ceptible neurons may complete S phase, and the aberrantexpression of cyclin B1/cdc2 complex indicates that de-generating neurons in AD may reach G2 phase. Ourfinding that CARB is aberrantly elevated in degeneratingneurons in AD not only supports the latter notion thatdegenerating neurons in AD reach G2 phase but also

suggests that neurons may arrested at G2/M phase, in thatactive mitotic kinases (cdc2/cyclin B1 pair) were seques-tered by CARB in cytoplasm. Arrest at G2/M phase willinevitably lead to neuronal degeneration.

What factor or factors stimulate the neurons to re-enter the cell cycle remains an important unansweredquestion. Insofar as AD-specific forms of hyperphospho-rylated tau are also highly expressed in the brain duringfetal life, it is notable that fetal development is under thecontrol of numerous growth factors, including manyhypothalamic-pituitary-gonadal (HPG) hormones. Nota-bly, two of the HPG hormones, luteinizing hormone andactivins, become elevated during senescence because ofthe age-related decline in reproductive function (Neaveset al., 1984), and we have shown that luteinizing hormoneis more highly expressed in AD brain (Bowen et al., 2000)and colocalizes to the brain areas most susceptible to ADneuropathology (Bowen et al., 2002). Activins and bonemophogenetic protein 4 bind to the same receptors, andthe latter has been shown to regulate p21 in neuronal stemcells (Gomes and Kessler, 2001; Israsena and Kessler,2002). Taken together, these findings suggest that, duringsenescence, changes in HPG hormone concentrations,resulting from the body’s attempt to maintain reproduc-tive function, cause dysregulation of cell cycle events andthat this cell cycle dysregulation is responsible for theneuropathological changes associated with AD.

REFERENCESArendt T, Rodel L, Gartner U, Holzer M. 1996. Expression of the

cyclin-dependent kinase inhibitor p16 in Alzheimer’s disease. Neurore-port 7:3047–3049.

Bowen RL, Isley JP, Atkinson RL. 2000. An association of elevated serumgonadotropin concentrations and Alzheimer disease? J Neuroendocrinol12:351–354.

Bowen RL, Smith MA, Harris PL, Kubat Z, Martins RN, Castellani RJ,Perry G, Atwood CS. 2002. Elevated luteinizing hormone expressioncolocalizes with neurons vulnerable to Alzheimer’s disease pathology.J Neurosci Res 70:514–518.

Bowser R, Smith MA. 2002. Cell cycle proteins in Alzheimer’s disease:plenty of wheels but no cycle. J Alzheimers Dis 4:249–254.

Braak H, Braak E. 1991. Neuropathological staging of Alzheimer-relatedchanges. Acta Neuropathol 82:239–259.

Busser J, Geldmacher DS, Herrup K. 1998. Ectopic cell cycle proteinspredict the sites of neuronal cell death in Alzheimer’s disease brain.J Neurosci 18:2801–2807.

Engidawork E, Gulesserian T, Seidl R, Cairns N, Lubec G. 2001. Expres-sion of apoptosis related proteins in brains of patients with Alzheimer’sdisease. Neurosci Lett 303:79–82.

Gomes WA, Kessler JA. 2001. Msx-2 and p21 mediate the pro-apoptoticbut not the anti-proliferative effects of BMP4 on cultured sympatheticneuroblasts. Dev Biol 237:212–221.

Harris PL, Zhu X, Pamies C, Rottkamp CA, Ghanbari HA, McShea A,Feng Y, Ferris DK, Smith MA. 2000. Neuronal polo-like kinase inAlzheimer disease indicates cell cycle changes. Neurobiol Aging 21:837–841.

Israsena N, Kessler JA. 2002. Msx2 and p21(CIP1/WAF1) mediate theproapoptotic effects of bone morphogenetic protein-4 on ventricularzone progenitor cells. J Neurosci Res 69:803–809.

Khachaturian ZS. 1985. Diagnosis of Alzheimer’s disease. Arch Neurol42:1097–1105.

702 Zhu et al.

McShea A, Harris PL, Webster KR, Wahl AF, Smith MA. 1997. Abnormalexpression of the cell cycle regulators P16 and CDK4 in Alzheimer’sdisease. Am J Pathol 150:1933–1939.

McShea A, Wahl AF, Smith MA. 1999a. Re-entry into the cell cycle: amechanism for neurodegeneration in Alzheimer disease. Med Hypoth52:525–527.

McShea A, Zelasko DA, Gerst JL, Smith MA. 1999b. Signal transductionabnormalities in Alzheimer’s disease: evidence of a pathogenic stimuli.Brain Res 815:237–242.

McShea A, Samuel T, Eppel JT, Galloway DA, Funk JO. 2000a. Identifi-cation of CIP-1-associated regulator of cyclin B (CARB), a novel p21-binding protein acting in the G2 phase of the cell cycle. J Biol Chem275:23181–23186.

McShea A, Vincent I, Linford NJ, Smith MA, Dorsa DM, Shapiro RA.2000b. Is Alzheimer’s disease cancer of the neurons? Modeling oncogenicstimulation of primary neurons initiates Alzheimer’s neuropathology. SocNeurosci Abstr 26:2290.

Mirra SS, Heyman A, McKeel D, Sumi SM, Crain BJ, Brownlee LM,Vogel FS, Hughes JP, van Belle G, Berg L. 1991. The Consortium toEstablish a Registry for Alzheimer’s Disease (CERAD). Part II. Standard-ization of the neuropathologic assessment of Alzheimer’s disease. Neu-rology 41:479–486.

Nagy Z, Esiri MM, Cato AM, Smith AD. 1997. Cell cycle markers in thehippocampus in Alzheimer’s disease. Acta Neuropathol 94:6–15.

Neaves WB, Johnson L, Porter JC, Parker CR Jr, Petty CS. 1984. Leydigcell numbers, daily sperm production, and serum gonadotropin levels inaging men. J Clin Endocrinol Metab 59:756–763.

Ogawa O, Lee HG, Zhu X, Raina A, Harris PL, Castellani RJ, Perry G,Smith MA. 2003a. Increased p27, an essential component of cell cyclecontrol, in Alzheimer’s disease. Aging Cell 2:105–110.

Ogawa O, Zhu X, Lee HG, Raina A, Obrenovich ME, Bowser R,Ghanbari HA, Castellani RJ, Perry G, Smith MA. 2003b. Ectopic local-ization of phosphorylated histone H3 in Alzheimer’s disease: a mitoticcatastrophe? Acta Neuropathol 105:524–528.

Perry G, Roder H, Nunomura A, Takeda A, Friedlich AL, Zhu X, RainaAK, Holbrook N, Siedlak SL, Harris PL, Smith MA. 1999. Activation ofneuronal extracellular receptor kinase (ERK) in Alzheimer disease links

oxidative stress to abnormal phosphorylation. Neuroreport 10:2411–2415.

Sternberger LA, Sternberger NH. 1986. The unlabeled antibody method:comparison of peroxidase-antiperoxidase with avidin-biotin complex by anew method of quantification. J Histochem Cytochem 34:599–605.

Vincent I, Rosado M, Davies P. 1996. Mitotic mechanisms in Alzheimer’sdisease? J Cell Biol 132:413–425.

Vincent I, Jicha G, Rosado M, Dickson DW. 1997. Aberrant expression ofmitotic cdc2/cyclin B1 kinase in degenerating neurons of Alzheimer’sdisease brain. J Neurosci 17:3588–3598.

Yang Y, Geldmacher DS, Herrup K. 2001. DNA replication precedesneuronal cell death in Alzheimer’s disease. J Neurosci 21:2661–2668.

Yang Y, Mufson EJ, Herrup K. 2003. Neuronal cell death is preceded bycell cycle events at all stages of Alzheimer’s disease. J Neurosci 23:2557–2563.

Zhu X, Raina AK, Smith MA. 1999. Cell cycle events in neurons.Proliferation or death? Am J Pathol 155:327–329.

Zhu X, Raina AK, Boux H, Simmons ZL, Takeda A, Smith MA. 2000a.Activation of oncogenic pathways in degenerating neurons in Alzheimerdisease. Int J Dev Neurosci 18:433–437.

Zhu X, Rottkamp CA, Boux H, Takeda A, Perry G, Smith MA. 2000b.Activation of p38 kinase links tau phosphorylation, oxidative stress, andcell cycle-related events in Alzheimer disease. J Neuropathol Exp Neurol59:880–888.

Zhu X, Rottkamp CA, Raina AK, Brewer GJ, Ghanbari HA, Boux H,Smith MA. 2000c. Neuronal CDK7 in hippocampus is related to agingand Alzheimer disease. Neurobiol Aging 21:807–813.

Zhu X, Castellani RJ, Takeda A, Nunomura A, Atwood CS, Perry G,Smith MA. 2001a. Differential activation of neuronal ERK, JNK/SAPKand p38 in Alzheimer disease: the ‘two hit’ hypothesis. Mech Ageing Dev123:39–46.

Zhu X, Raina AK, Rottkamp CA, Aliev G, Perry G, Boux H, Smith MA.2001b. Activation and redistribution of c-Jun N-terminal kinase/stressactivated protein kinase in degenerating neurons in Alzheimer’s disease.J Neurochem 76:435–441.

Zhu X, Sun Z, Lee HG, Siedlak SL, Perry G, Smith MA. 2003. Distribu-tion, levels, and activation of MEK1 in Alzheimer’s disease. J Neurochem86:136–142.

CARB in Alzheimer’s Disease 703