Ž .Brain Research 769 1997 233–244

Research report

Neonatal transection of the infraorbital nerve increases the expression ofproteins related to neuronal death in the principal sensory nucleus of the

trigeminal nerve

Michael W. Miller a,b,c,), Peter E. Kuhn c

a Departments of Psychiatry and Pharmacology, UniÕersity of Iowa College of Medicine, Iowa City, IA 52242-1057, USAb Research SerÕice, Veterans Affairs Medical Center, Iowa City, IA 52246-2208, USA

c Program in Cell and DeÕelopmental Biology, Rutgers UniÕersity, Piscataway, NJ 08854-1059, USA

Accepted 28 May 1997

Abstract

Neonatal lesion of the primary afferents in the infraorbital nerve causes the death of one-third of the neurons in the second-orderŽ .target, the principal sensory nucleus of the trigeminal nerve PSN . We examined the expression of two candidate ‘death’ proteins, p53

and the antigen recognized by the antibody ALZ-50, in the normal and deafferented PSN. In addition, the effect of neonatal transection ofŽ .the infraorbital nerve a major component of the trigeminal nerve on protein expression was examined. The expression of c-fos in the

developing PSN was also studied as an index of metabolic activity. Protein expression was measured using quantitative analyses ofimmunoblots and immunohistochemical preparations. The expression of p53- and ALZ-50-immunoreactivity in the normal PSN peakedduring the first postnatal week. Transection of the infraorbital nerve directly affected the expression of p53 and the ALZ-50-positiveantigen. The immunoblots showed that whereas p53 amounts were unaffected by the lesion, ALZ-50 expression was significantlyupregulated in the ipsilateral PSN 2 h and 2 days postlesion. The density of p53- and ALZ-50-immunoreactive neurons was significantly

Ž .higher in the ventral ipsilateral PSN i.e., the target of the transected infraorbital nerve than in the contralateral PSN. c-fos expressionselectively and transiently rose in the ventral ipsilateral PSN within 2 h of the lesion. Thus, both p53 and the ALZ-50-positive antigen areinvolved in neuronal death. In light of data suggesting that ALZ-50 recognizes a phosphorylated form of p53, we conclude that neuronaldeath in the developing nervous system involves the post-translational modification of an existing protein, p53. The increase in ALZ-50expression apparently occurs during a catabolic phase of neuronal death, as indicated by the increase in c-fos expression. q 1997 ElsevierScience B.V.

Keywords: ALZ-50; Apoptosis; c-fos; Cell death; p53

1. Introduction

Naturally occurring neuronal death is a widespreadphenomenon in the developing central nervous system. Ithas been hypothesized that this death is an active process

w xduring which novel proteins are generated 26,27 . Onesuch protein may be the antigen identified by the mono-

w xclonal antibody ALZ-50 1,31 . ALZ-50 immunoreactivityincreases in primary cultured hippocampal neurons thatdegenerate following exposure to high levels of glutamate

) Corresponding author. Department of Psychiatry-M.E.B., Universityof Iowa College of Medicine, Iowa City, IA 52242-1000, USA. Fax: q1Ž .319 353-3003; E-mail: michael-w-miller@uiowa.edu

w x12,28 . Furthermore, neonatal transection of the infraor-bital nerve transiently increases the number of ALZ-50-positive second-order neurons in the principal sensory

Ž . w xnucleus of the trigeminal nerve PSN 30 .Immunoprecipitation experiments indicate that the anti-

body ALZ-50 recognizes a form of the ‘housekeeping’w xprotein p53 20 . Both proteins can conjugate with ubiqui-

w xtin 9,20,43,51 and contain a number of serines that can bew xphosphorylated 29,46,49,52 . The phosphorylation of the

serines in p53 is key to the degenerative process. Forexample, ionizing radiation and ultraviolet rays, i.e., agentsthat damage DNA, induce the death of lymphocytesw x10,19,20,24,44,56 . This death depends upon the phospho-

w xrylation of p53 by raf 18,23,33 or protein kinase CŽ . w xPKC 37,47 . Not only is PKC activity increased by

0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved.Ž .PII S0006-8993 97 00713-0

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244234

irradiation, but PKC inhibitors block the induction of p53w x20,36,42,53 . Increased p53 expression also occurs during

w xneuronal degeneration 39 .The data on the ALZ-50-positive antigen and p53 sug-

gest that naturally occurring neuronal death is not charac-terized by the production of novel proteins, rather thisdegeneration involves a post-translational modification ofproteins that are used during normal processing. In the

Ž .present report, we tested the hypotheses: 1 that duringneuronal degeneration, there are parallel increases in theexpression of the ALZ-50-positive antigen and p53, andŽ . Ž2 that these changes are an active process as measured

.by alterations in c-fos expression . The focus of the studywas the PSN, a site where ALZ-50 is expressed by neuronsthat die naturally and in response to a deafferenting lesionw x31 . PSN neurons die naturally during the first postnatal

w xweek 30 . In fact, more than half of the neurons generateddie. Transection of the infraorbital nerve on the day of

w xbirth causes additional PSN neurons to die 31 . Whereasnaturally occurring neuronal death takes place throughoutthe entire nucleus, lesion-induced death is confined to theventral ipsilateral PSN. Thus, the PSN offers an idealstructure to study protein expression during naturally oc-curring and lesion-induced neuronal death.

2. Materials and methods

2.1. Animals and experimental manipulations

Pregnant, Long–Evans rats were obtained fromŽ .Harlan–Sprague–Dawley Altamount, NY on gestational

Ž .day G 4. The first day that the dams exhibited a sperm-positive vaginal plug was designated as G1. Animals werehoused in a humidity- and temperature-controlled facilitywith a fixed 12:12 h lightrdark cycle. The rats were fedchow and water ad libitum. Pups were born on G22, a day

Ž .also referred to as postnatal day P 0.Two series of studies were performed. In the first, the

expression of various proteins in the pons of developingrats was determined with Western immunoblots. FetusesŽ . ŽG16 and G19 and pups P0, P3, P6, P9, P12, P15 and

.P90 were harvested during the periods of neuronal genera-w xtion, migration, synaptogenesis and death 2,3,5,30 . For

each prenatal age, two fetuses were taken from each of twoŽlitters. Pregnant dams were anesthetized 60 mgrkg ke-

tamine and 7.5 mgrkg xylazine administered intraperi-.toneally and pups were delivered by Cesarean section.

Fetuses were decapitated and fresh pontine tissue wasprepared for immunoblots. For each postnatal time point,

Ž .four pups each from a different litter were anesthetizedand decapitated.

In the second study, the effects of transecting theinfraorbital nerve in the neonate on protein expression wasdetermined. Lesions were placed on P0 because the num-ber of neurons in the PSN peaked on the day of birth

w x5,30 . Within 4 h of birth, neonates were anesthetized byhypothermia. An incision was placed caudal to the whiskerpad on the right side. Soft tissues were reflected, theinfraorbital nerve was visualized, and the nerve was cut.The pups were warmed and after a short recuperative

Ž .period 30 min or less , they were returned to their moth-ers. The pups were anesthetized 2 h, 2 days, 12 days or 21days postlesion and killed either by decapitation or byperfusion with an aldehyde fixative. This 21-day periodencompassed the period of naturally occurring neuronal

Ž . w xdeath in the rat PSN P0–P5 5,30 . Four pups per timepoint were processed immunochemically and an additionalfour pups per time point were used in the immunohisto-chemical analyses. Each of the pups used for a particulartime point was taken from a different litter.

2.2. Biochemical studies

2.2.1. Tissue preparationEach unfixed brainstem was taken from the cranium;

the cerebellum was removed, and the pons was dissected.One block from each side of the pons was isolated. Eachblock was defined as all of the pontine tissue lateral to aparasagittal cut at the lateral limit of the pyramidal tracts.This block included the PSN and the motor trigeminalnucleus.

For each experiment, the block was homogenized incold, modified RIPA buffer containing detergent, proteaseand phosphatase inhibitors to prevent protein degradationŽ Ž .10 mM sodium phosphate buffer pH 7.2 , 150 mMsodium chloride, 1.0% Nonidet P-40, 1.0% sodium deoxy-cholate, 0.1% sodium dodecylsulfate, 1.0 mM dithio-threitol, 50 mM leupeptin, 10 mg of phenylmethylsulfonyl

. w xfluoride per ml, and 50 mM sodium orthovanadate 55 .The DNA and protein concentrations of the homogenatewere determined by quantifying the absorbance at 260 nmw x Ž40 and with a kit obtained from BioRad Burlingame,

.CA , respectively.

2.2.2. Preparation of blotsŽThe amount of tissue in a sample 10 mg for the

determinations of p53 and c-fos content, and 25 mg for the.assays of ALZ-50 immunoreactivity was defined by the

DNA content. These aliquots contained an amount of theALZ-50-positive antigen, p53, or c-fos that was within thelinear range of detectability. Protein content was not used

Žas our standard because the size of PSN cells thus, the.amount of protein per cell changes dramatically during

w xdevelopment 30 . Samples were applied to 12% polyacryl-Ž . w xamide gels PAG containing 0.50% SDS 22 . The pro-

teins in the samples were separated by one-dimensionalelectrophoresis and then transferred onto a nitrocellulose

Ž . w xfilter 30 V, overnight at 48C 48 . A set of samples wasapplied to each PAG. For the first study, each set includednine samples covering the ages from G16 to P90. For thesecond study, each set included the paired samples from

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244 235

the left and right pons of an animal that represented eachof the four time points. Three replicates of the homogenatefrom each animal were used in the immunoblot analyses.

The immunochemical reactions were carried out bystandard procedures. Non-specific immunoreactivity was

Žblocked by immersing a blot in 5.0% non-fat dry milk for. Žthe assays of c-fos or in 5.0% goat serum for the assays

.of p53 or the ALZ-50 immunoreactivity for 30 h. Theblots were incubated overnight at 48C with an antibody

Ždirected against p53 either pantropic anti-p53 Ab-2 or.anti-mutant p53 Ab-3; Oncogene, Uniondale, NY , the

Žmonoclonal antibody ALZ-50 generously provided by Dr.Peter Davies, Albert Einstein College of Medicine, Bronx,

. Ž .NY , or an anti-c-fos antibody Ab-1; Oncogene . Eachantibody was diluted to 1:50 in 0.50% Tween-20 with 0.10

Ž .M phosphate buffered saline T-PBS . The immunoreac-tive proteins were detected with a secondary antibody

Žcrosslinked to alkaline phosphatase Vector, Burlingame,.CA . The sizes of the immunoreactive proteins were deter-

mined relative to the sizes of standard proteins coveringŽ .the range from 20.5 to 206 kDa BioRad or from 97 to

Ž .292 kDa Sigma, St. Louis, MO .Control immunoblots, in which either the primary or

secondary antibody was omitted from the processing, weregenerated. In each case, these controls were negative.

2.2.3. AnalysisThe relative amount of a protein in the immunoblots

was determined with the Bioquant Image Analysis SystemŽ .R&M Biometrics, Nashville, TN . Using this system, therelative density and the size of an immunolabeled bandwere determined. The product of these values was taken asan estimate of the amount of a protein in the blot. Theamount of a particular protein was measured in each of thethree replicates and the means of the triplicates for eachside per animal per time point were calculated. In turn, the

Ž .grand means "S.D. for all age-matched subjects weredetermined. The statistical significance of the changes inrelative protein expression was assessed using a one-way

Žanalysis of variance followed by a post-hoc analysis Stu-.dent–Newman–Keuls test .

2.3. Anatomical studies

2.3.1. Preparation of tissue sectionsŽ .Pups 0, 2, and 12 days old were anesthetized by

intraperitoneal injection of 60 mgrkg ketamine and 7.5mgrkg xylazine and transcardially perfused with an alde-hyde fixative. For the p53 immunolabeling, pups wereperfused with Bouin’s fixative for 15 min. For the ALZ-50and c-fos immunolabeling, pups were perfused with 4.0%

Ž .paraformaldehyde in 0.10 M phosphate buffer pH 7.4 for15 min. Each brain was removed from the cranium andthen immersed in fresh fixative overnight. The brainstemwas blocked and the block was cryoprotected by immer-sion in 10% sucrose in phosphate buffer for 4–8 h and in30% buffered sucrose for up to 2 days. Subsequently, thebrainstem was frozen, cut into a complete series of 20 mmcoronal sections with a cryostat, and the sections weremounted on slides.

2.3.2. Immunohistochemical proceduresNon-specific binding of the sections was blocked by

immersing the sections with 5.0% goat serum for 1 h atroom temperature. After rinsing in PBS, the sections wereincubated in a 1:50 dilution of a primary antibody in PBSŽthe same antibodies as those used in the immunoblotting

.procedure; see above . Immunopositive cells were detectedby conjugating the primary antibody with an avidin–bio-

Ž . w xtin–horseradish peroxidase complex Vector 17 . Theperoxidase was reacted with 0.50% hydrogen peroxide inthe presence of the chromogen diaminobenzidine.

Two controls for non-specific activity were performed.Both controls involved the omission of an antibody fromthe processing. In one control the sections were not ex-posed to a primary antibody and in the other control, thesecondary antibody was omitted from the processing. In allcases, the results of the two controls were consistentlynegative.

2.3.3. AnalysisThe sections were examined microscopically to identify

spatiotemporal changes in the distribution of immunoreac-

Fig. 1. Immunoblots of the principal bands recognized by anti-p53, ALZ-50, and anti-c-fos antibodies in the pons between G16 and P90. Top: thedevelopmental expression of the 58-kDa protein identified by anti-p53 antibodies in the lateral pons is shown. Middle: ALZ-50 identified a 56-kDa proteinthat was transiently expressed in the immature pons. Peak expression occurred during the first postnatal week. Bottom: the anti-c-fos antibody labeled twoproteins, a 75-kDa peptide and an 85-kDa peptide, in lateral pontine tissue.

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244236

Fig. 2. Temporal change in the expression of p53-, ALZ-50-, andc-fos-immunoreactivity in the developing pons. Top: the relative amountof the 58-kDa protein was quantified using an image analysis system.Although the expression of p53 was evident throughout development, thegreatest expression occurred during the first and second postnatal weeks.Middle: in the pons of a normal, unoperated rat, ALZ-50 expression firstappeared on the day of birth, peaked during the first postnatal week, anddropped in the second postnatal week. Bottom: the relative expression ofthe two c-fos-positive proteins is described. The 75-kDa peptide wasconsistently expressed from G16 to P90. No significant differences weredetected throughout this period. The 85-kDa protein, however, was most

Ž .common prenatally. Its expression was significantly P -0.05 lowerduring the first 3 postnatal weeks and fell to virtually undetectable levelsin the mature pons.

tive elements. The relative density of immunopositive cellsin the ventral or dorsal PSN on each side of the pons were

w xanalyzed using a stereological method 15,31,32,45 thataccounts for overestimation due to counting somatic frag-ments as whole cell bodies. Accordingly, the cross-sec-tional area was determined for each labeled neuron that

Ž .had its nucleus in a counting box 125=125 mm . Suchmeasures were taken from five sections per animal. Thisprocedure assumes that the counted elements are spheres.This assumption was appropriate for PSN neurons since

Žthe mean eccentricity of their cell bodies the quotient of

the length of the minor axis divided by the length of the. Žmajor axis was 0.89"0.04. The eccentricity of a circle

.is 1.00 . The diameters of the cell body fragments wereŽ .used to estimate the mean maximal diameter D using an

Ž .iterative formula. The neuronal density N was calculatedsw Ž .xby the formula, N sN tr tqDy2k , where N wass a a

the number of neurons counted in the box, t was theŽ .section thickness 20 mm , and k was the height of the

smallest recognizable labeled soma with a cut nucleus. Themean density of labeled neurons per animal was used to

Ž .calculate a grand mean "S.E.M. for the four animalsexamined at each time point.

The density of labeled neurons in the motor nucleusŽ . Ž .MoV and the mesencephalic nucleus MeV of thetrigeminal nerve were examined. The infraorbital nervedoes not innervate the MoV and it does not contain anyperipheral processes of the neurons with cell bodies in theMeV. Therefore, the data from these nuclei were used as anegative control for the effects of the nerve transection ofthe infraorbital nerve.

ŽDifferences between sides, dorsal vs. ventral location,.and age were assessed with an analysis of variance and

post-hoc tests.

3. Results

3.1. p53

3.1.1. Biochemical studiesThe expression of p53 was examined using a quantita-

tive immunoblotting method. Both anti-p53 antibodiesŽ .identified a 58-kDa protein Fig. 1 and weakly labeled

other peptides, e.g., 28-, 75-, or 85-kDa peptides. The

Fig. 3. Differential effects of unilateral, neonatal transection of theinfraorbital nerve on p53, ALZ-50, and c-fos expression. Top: the expres-

Žsion of p53 in the lateral pons including the principal sensory nucleus ofŽ .the trigeminal nerve PSN was qualitatively similar on the side ipsilateral

Ž . Ž .right and contralateral left to the lesion. Middle: the expression of the56-kDa protein identified by ALZ-50 was increased by neonatal transec-tion of the infraorbital nerve. Bottom: the expression of the two c-fos-positive proteins was transiently higher on the right side, i.e., the sideipsilateral to the lesion.

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244 237

58-kDa protein was expressed in the pons throughoutŽ .development, i.e., the period from G19 to P90 Fig. 2 .

The peak expression was achieved during the first postna-tal week and maintained through the second postnatalweek.

The relative amount of p53-immunoreactivity in thepons was unaffected by transection of the infraorbital

Ž .nerve Figs. 3 and 4 . No significant differences betweenthe sides were detected in p53 expression on P0, P2, P12or P21. No differences in p53 expression were identified intissue from the left and right somatosensory cortices. Nev-

Žertheless, the amount of p53 detected in the left pons the

.side unaffected by the lesion was similar to the amountsdetermined from the intact, age-matched rats.

3.1.2. Anatomical studiesp53 was expressed by many components of the trigemi-

nal system. PSN neurons exhibited p53-immunoreactivityon the day of birth and P2 as puncta located at the origins

Ž .of primary dendrites Fig. 5 . There appeared to be onlyone puntum in each labeled cell body. By P12, p53immunoreactivity was evident throughout the perikaryonand extended into proximal dendrites. The MoV and theMeV also expressed p53-immunoreactivity. Unlike the

ŽFig. 4. Effect of infraorbital nerve transection on p53, ALZ-50, and c-fos expression. Top: the relative amount of 58 kDa was measured in the pons left. Ž .graph and somatosensory cortex right graph . Regardless of the time postlesion, no significant differences were detected between the right or left side.

Ž .Middle: ALZ-50 immunoreactivity rose significantly P-0.05 within 2 h of transecting the infraorbital nerve; ALZ-50 expression was 1.87-fold greateron the right side than in the left pons. This elevation was maintained for at least 2 days at which time the right pons had 141% the amount of ALZ-50expression of that measured on the left side. No significant differences between the right and left somatosensory cortex were detected at any time

Ž . Ž .postlesion. Bottom: The 75- and 85-kDa proteins were significantly P-0.05 increased in the right pons relative to the left pons 2 h postlesion. Thisincrease was transient for 2 days postlesion no differences between the sides were detected. Interestingly, c-fos expression in the somatosensory cortex wasalso increased following the neonatal transection. As in the pons, the increase in the cortex was transient; however, in the case of the 75-kDa protein,

Ž .expression was significantly P-0.05 elevated for at least 2 days.

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244238

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244 239

Fig. 6. Effect of infraorbital nerve transection on the density of p53-posi-tive neurons in the trigeminal system. The number of immunoreactivecells in a box 125=125 mm was determined stereologically. There wasno acute difference in the density of immunolabeled cells. Two dayspostlesion, however, the density of p53-positive cells on the side ipsilat-

Ž) .eral to the lesion was significantly greater P -0.05 than the numberŽ .on the contralateral control side. This difference was gone 12 days

postlesion. No lateralized differences in p53 expression were evident inthe motor nucleus and the mesencephalic nucleus of the trigeminal nerveŽ .MeV , regardless of time postlesion.

PSN neurons, the entire somata of neurons in the MoV andMeV were labeled by P0.

p53 expression in the PSN was affected by the periph-eral nerve lesion as noted by the spatiotemporal-specificchange in the distribution of immunoreactive neurons. Twohours after the lesion was placed, there was no significantchange in the number of p53-positive cells in the PSN,MoV, or MeV. Two days postlesion, i.e., P2, however, thedensity of p53-positive elements was increased, but only in

Ž .the ventral PSN ipsilateral to the lesion Fig. 6 . ThisŽ .increase was significant P-0.01 relative to the dorsal,Ž .ipsilateral PSN or to the dorsal or ventral contralateral

PSN. Interestingly, the immunolabeled cell bodies tendedto located within regions in which the background staining

Žwas slightly higher than it was in unlabeled regions e.g.,areas such as the tissue intervening between the PSN and

.the MoV . These regions were arranged in rows suggesting

Fig. 7. Effect of infraorbital nerve transection on the density of ALZ-50-positive neurons in the trigeminal system. Within 2 h postlesion, thedensity of ALZ-50-immunoreactive neurons was 271% greater in theventral PSN ipsilateral to the lesion. This difference was statistically

Ž) .significant P -0.01 . Even 2 days after placing the lesion, the rightventral PSN had 233% more ALZ-50-positive neurons than did the leftventral PSN. On P12, no ALZ-50-immunoreactive neurons were detectedon either side of the brainstem. Furthermore, the lesion had no effect onALZ-50 expression in the motor nucleus or the mesencephalic nucleus atany time postlesion.

that they were barreloids. No lesion-induced differences inany of the trigeminal nuclei were observed on P12 or P21.

Regardless of the time postlesion, there were no signifi-cant changes in the density of p53-positive neurons eitherin the MoV or the MeV.

3.2. ALZ-50

3.2.1. Biochemical studiesThe 56-kDa protein recognized by ALZ-50 was tran-

Ž .siently expressed in the developing pons Figs. 1 and 2 .ALZ-50-immunoreactivity was evident during the first andsecond postnatal weeks; however, it was virtually absentprenatally and in animals 12 days old and older.

ALZ-50-immunoreactivity increased in the right ponsafter a lesion was placed in the right infraorbital nerve on

Ž .P0 Figs. 3 and 4 . In fact, an increase in the 56-kDa

Fig. 5. p53-immunoreactive neurons in the pons. Two days after placing the neonatal transection, a marked increase in p53 immunolabeling was evident inŽ .the ipsilateral ventral principal sensory nucleus of the trigeminal nerve PSN . Immunoreactive elements were located within regions in which the

Ž . Ž . Ž .immunolabeling was greater than background areas inside dashes . A: left control PSN. B: left ventral PSN. C: right deafferented PSN. D: right ventralPSN. MoV, trigeminal motor nucleus; s5, sensory tract of the trigeminal nerve. A,C, =110; B,D, =225.

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244240

peptide was evident within 2 h of when the lesion wasplaced. Thus, the elaboration of the ALZ-50-positive anti-gen occurred in considerably less time than the 3 days

w xpreviously described 31 . The infraorbital nerve transec-tion had no effect on ALZ-50-immunoreactivity in thesomatosensory cortex.

3.2.2. Anatomical studiesIn the neonates, the density of ALZ-50-positive neurons

Ž .was higher in the ipsilateral right PSN and specifically inthe ventral half. The density of labeled neurons in theventral PSN on the right side was 121% greater than it was

Ž .for the left, ventral PSN Fig. 7 , a difference that wasŽ .statistically significant P-0.01 . This increase was sus-

tained until P2 when the density of ALZ-50-immunoreac-tive neurons in the ipsilateral PSN was 133% higherŽ .P-0.01 than it was in the contralateral PSN. By P12,however, no ALZ-50-labeled neurons were detected oneither side of the brainstem.

During the first postnatal week, ALZ-50-immunoreac-

tive neurons were identified in the MoV and MeV. Never-theless, 2 h and 2 days postlesion, the density of ALZ-50-positive neurons in the MoV and MeV was similar on bothsides of the brain. On P12 and P21, no labeled neuronswere detected in either nucleus.

3.3. c-fos

3.3.1. Biochemical studiesThe anti-c-fos antibody identified two proteins in pon-

Ž .tine tissue, 75- and 85-kDa peptides Fig. 1 . The expres-sion of the 75-kDa peptide was steady throughout the

Ž .period of pontine development Fig. 2 . According to aStudent–Newman–Keuls analysis, variation between G16and P90 was not statistically significant. On the other

Žhand, the amount of the 85-kDa peptide significantly P-.0.01 declined from G16 to P6. After P6, c-fos expression

was maintained at a low level.The expression of both c-fos-positive peptides in the

Ž .neonatal pons was significantly P-0.01 affected by

Fig. 8. c-fos-immunoreactive neurons in the pons. c-fos-positive cells were evident on both sides of the brainstem. There was no effect of the lesion in theŽ . Ž . Ž . Ž .dorsal PSN on the left A and right B sides. On the other hand, the right ventral PSN D had more of these cells than the left ventral C PSN. In

Ž .addition, the right ventral PSN contained among the most intensely labeled cells indicated by arrows in the brainstem. =250.

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244 241

Fig. 9. Effect of infraorbital nerve transection on the density of c-fos-posi-tive neurons in the trigeminal system. The density immunolabeled cellswas significantly greater in the right ventral PSN 2 h postlesion than itwas in the left ventral PSN. The transection had no effect on the densityof c-fos-positive cells in the dorsal PSN. Two days postlesion, theincrease in c-fos-positive cells had dissipated.

Ž .lesion of the infraorbital nerve Fig. 3 . The amount of the75- and 85-kDa peptides on the right side was 83% and171% greater, respectively, than it was on the left sideŽ .Fig. 4 . This effect was short-lived, however, because byP2, no significant differences in the amount of the 75- or85-kDa peptide were evident. It should be noted that the

Žamount of the two peptides expressed by the left unaf-.fected pons was not significantly different from that de-

tected in the pons of intact animals.Interestingly, the amount of c-fos in the somatosensory

cortex was transiently affected by the peripheral nervelesion. Within 2 h of transecting the nerve, the expression

Ž .of the 75- and 85-kDa peptides was significantly P-0.01Ž .higher on the affected contralateral side than on the right

side. This elevation was 138% for the 75-kDa peptide and123% for the 85-kDa peptide. The increased expression ofthe 75-kDa peptide was maintained for at least 2 dayspostlesion. On P2, the amount of the 75-kDa peptide on

Ž . Ž .the left side was significantly P-0.01 more 225%than that detected on the right side. This lesion-inducedincrease was transient for by P12, no differences betweenthe sides were observed.

3.3.2. Anatomical studiesc-fos-immunolabeled cells were identified in the PSN,

Ž .MoV, and MeV Fig. 8 . One hour post-lesion, c-fos was

found in the nuclei of cells throughout the PSN. Not onlyŽwas the density of immunoreactive cells significantly P-

. Ž .0.01 greater 75% in the right ventral PSN than in theŽ .right dorsal PSN Fig. 9 , but the intensity of the immuno-

label in the c-fos-positive cells was greater on the rightside than on the left. No differences between the sides inthe density of labeled cells was evident on P2, P12 or P21.

The density of c-fos-immunoreactive cells in the MoVand MeV was similar on the two sides of the brainstem.

4. Discussion

4.1. Expression of ALZ-50 and c-fos immunoreactiÕity

Neonatal transection of the infraorbital nerve affects theexpression of the ALZ-50-positive antigen and c-fos. Thepresent data document an increase in the ALZ-50-positive56-kDa peptide, an increase that is evident within 2 h ofthe damage. Thus, the expression of the ALZ-50-positiveantigen is a rapid response to the peripheral nerve lesion.This rapid increase is matched by a parallel, albeit ashorter lasting increase in the expression of the 75- and85-kDa proteins that are recognized by the anti-c-fos anti-body. These biochemical changes reflect increases in thenumbers of immunohistochemically labeled neurons. Al-though ALZ-50- and c-fos-immunoreactive neurons aredistributed throughout the neonatal PSN, following theinfraorbital nerve lesion the densities of ALZ-50- andc-fos-immunoreactive neurons are significantly higher inthe affected segment of the PSN. Therefore, like thebiochemical changes, the anatomical changes are also evi-dent within 2 h of lesion placement.

The findings of rapid lesion-induced changes in ALZ-50-immunoreactivity and c-fos expression lead to four

Ž .conclusions. 1 The increased expression of ALZ-50-im-munoreactivity 2 h postlesion does not result from alteredtranscription, rather the lesion affects a translational andror

Ž .post-translational event. 2 Since the amount of c-fos risestransiently, it appears that lesion-induced changes in the

Ž .second order neurons are metabolically demanding. 3ALZ-50-positivity may be related to impending neuronal

w xdeath 1,31,54 . If so, then the sequence of degenerationcommences shortly after the damage has occurred. Whether

Ž .or not this sequence is irreversible is unknown. 4 Sur-vival of the PSN neurons depends upon an interactionbetween the infraorbital nerve and the second-order neu-rons, and likely from peripheral nerve-derived information.To our knowledge, however, no such data on acute, trans-synaptic changes in second-order trigeminal neurons areavailable.

4.2. Expression of p53

On superficial inspection, the biochemical data on p53expression do not wholly complement those from the

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244242

associated anatomical studies. Both the biochemical andanatomical studies show that deafferentation does not ef-fect the amount of p53 expressed in the pons 2 h postle-sion. On the other hand, the data of p53 expression 2 dayspostlesion are paradoxical. Whereas the biochemical studyshows that p53 expression is unaffected by neonatal tran-section, the anatomical data show that the number ofp53-positive cells is increased. The spatiotemporal changesin the number of p53-immunoreactive elements suggeststhat p53 expression is directly affected by transection ofthe infraorbital nerve. The increase in the number of

Ž .p53-positive cells: 1 is restricted to the deafferentedŽ . Žregion of the PSN; and 2 is not immediate about 2 days

.are required before p53 expression rises . This implies thatthe lesion promoted p53 transcription.

Despite the differences in the temporal expression ofthe ALZ-50-positive antigen and p53, these proteins ap-pear to be intimately related. Data from an immunoprecip-itation study show that ALZ-50 recognizes a form of p53w x21 . Whereas ALZ-50 identifies only a phosphorylated

w xprotein 49,52 , the anti-p53 antibody recognizes phospho-rylated and dephosphorylated forms. Accordingly, ALZ-50labels only some of the p53. This differential antigenicityexplains why the biochemical experiments show that theamount of p53 expressed in the normal pons umbrellas thatfor the ALZ-50-positive antigen. Furthermore, the lack ofa lesion-induced change in total p53 content 2 h post-le-sion implies that the nerve transection does not induce p53synthesis. Rather, the lesion induces a post-translational

Ž .change phosphorylation in p53 so that the modifiedŽ .phosphorylated protein is recognized by ALZ-50.

We hypothesize that the phosphorylation of p53 is acritical and early step in the death of the deafferentedneurons. After all, the number of p53-positive cells in-crease in the deafferented site where neuronal death iscommon. This hypothesis is further supported by various

Ž .evidence. 1 ALZ-50 identifies phosphorylated p53w x Ž .21,49,53 . 2 ALZ-50-immunoreactivity is expressed dur-

w x Ž .ing neuronal degeneration 1,31,50,54 . 3 Neonatal tran-section of the infraorbital nerve causes neuronal death as

w xexemplified by an increase in pyknotic 5 and ALZ-50-w xpositive cells 30 and a reduction in neuronal number

w x30,31 during the first postnatal week.Data from non-neural tissues support the role of p53 in

cell death. p53 is used to repair the breaks in the DNA ofw x w xcolon carcinoma cells 44 and mouse fibroblasts 35

caused by g-irradiation. The upregulation of p53 expres-sion begins quickly after the trauma occurs; significantchanges are detected within 1 h. This rapid elevation ofp53 may not be the result of de novo synthesis, rather itmay be due to a decrease in the degradation of the protein

w xandror a release of the inhibition of p53 translation 25,35 .The translation of p53 begins in response to changes inDNA replication. p53 binds to its mRNA and blocksfurther translation. This blockage is released either by

w xUV-dependent damage or DNA replication 35 .

4.3. Phosphorylation of p53

The regulation of p53 during neuronal death may becontrolled by the phosphorylation of different segments ofp53. Two important domains of p53 are its carboxy andamino termini.

The carboxy terminus of p53 regulates the interactionw xbetween p53 and the mRNA for p53 35 . Cells transfected

with modified p53 lacking its carboxy sequence do notexhibit protein–mRNA interactions. Thus, the carboxy-endmust be pivotal for initiating a cascade of transcriptionalevents. Equally important is that the carboxy terminus

w xcontrols the transcription of p53-responsive genes 6,7 .The carboxy region of p53 can bind to the ends of

Ž .single-stranded DNA. The hydrophobic middle region ofp53 can bind to the middle of a segment of single-strandedDNA. These interactions are critical for p53 to bring two

w xbroken pieces of DNA together for reannealing 36 . DNAexhibiting double-stranded breaks, however, can bind onlyat the carboxy terminus of p53 and the interaction betweenp53 and DNA with double-stranded breaks leads to DNA

w xdegradation 6,7,41 . Therefore, the p53 can subserve oneŽ .of two opposing actions repair or degradation depending

upon the presentation of the DNA break.The carboxy region of p53 can be phosphorylated by

Ž . w xprotein kinase C PKC 8,20,47 . Radiation-induced celldeath depends upon the phosphorylation, and hence the

w xactivation, of p53 by PKC 20 . Phosphorylation of theŽcarboxy end of p53 increases after the DNA is broken be

.it by a damaging agent or prior to transcription . Thus, p53is a focal point for complex cascades that are critical forrepair and degenerative processes.

The amino end of p53 can be phosphorylated by variouskinases including PKC, DNA-directed protein kinase raf-1,

Ž . w xcasein kinase I, and jun kinase jnk1 4,18,20,23,33,34 .Each kinase can phosphorylate serines in different posi-tions along the amino acid sequence of p53. For example,DNA-directed protein kinase is activated by breaks inDNA and phosphorylates mouse p53 on the serine residues

w xin positions 7 and 18 23 . Casein kinase 1 phosphorylatesw xserines 7, 9, and 12 33 . Since these residues are close to

other phosphorylated serines, casein kinase I may alter thew xactivity of other kinases 16 . One example is jnk1; ultra-

Ž .violet UV irradiation stimulates jnk1 to phosphorylatew xserine 37 of p53 34 .

DNA damage by UV irradiation leads to the transactiva-w xtion of p53 and the upregulation of p53 synthesis 10,24,44 .

UV-irradiation induces the phosphorylation of both thew xcarboxy and amino termini of p53 8,20,34,47 . These

modifications are necessary to transcribe p53-sensitivew xgenes 11,13,14,38 . Thus, the phosphorylation of p53 sets

into motion a complex set of events that ultimately lead tocell death. Martin and colleagues hypothesized that as theydevelop, each immature neuron reaches a critical stagewhen it will elaborate proteins that foster its further differ-entiation and survival or novel proteins, ‘thanatins,’ that

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244 243

w xlead to its death 26,27 . The data from the present studydo not contradict this hypothesis; however, they do supporta corollary hypothesis. Neuronal death is an active processresulting in the post-translational modification and in-

Ž .creased synthesis of an already existing protein s . Further-more, we suggest that the phosphorylation of p53 is thefulcrum that determines whether neurons will survive ordie.

Acknowledgements

The present study was funded by the National InstitutesŽ .of Health AA06916, AA07568, AA09611 and DE 07734

and by the Department of Veterans Affairs.

References

w x1 W.M. Al-Ghoul, M.W. Miller, Transient expression of Alz-50 im-munoreactivity in developing rat neocortex: a marker for naturally

Ž .occurring neuronal death?, Brain Res. 481 1989 361–367.w x2 W.M. Al-Ghoul, M.W. Miller, Orderly migration of neurons to the

principal sensory nucleus of the trigeminal nerve of the rat, J. Comp.Ž .Neurol. 330 1993 464–475.

w x3 W.M. Al-Ghoul, M.W. Miller, Development of the principal sensorynucleus of the trigeminal nerve of the rat and evidence for atransient synaptic field in the trigeminal sensory tract, J. Comp.

Ž .Neurol. 330 1993 476–490.w x4 C.W. Anderson, DNA damage and the DNA-activated protein ki-

Ž .nase, Trends Biochem. Sci. 18 1993 433–437.w x5 K.W.S. Ashwell, P.M.E. Waite, Cell death in the developing trigem-

Ž .inal nuclear complex of the rat, Dev. Brain Res. 63 1991 291–295.w x6 G. Bakalkin, G. Selivanova, T. Yakovleva, E. Kiseleva, E. Kashuba,

K.P. Magnusson, L. Szekely, G. Klein, L. Terenius, K.G. Wilman,p53 binds single-stranded DNA ends through the C-terminal domainand internal segments via the middle domain, Nucleic Acids Res. 23Ž .1995 362–369.

w x7 G. Bakalkin, T. Yakoleva, G. Selivanova, K.P. Magnusson, L.Szekely, E. Kiseleva, G. Klein, L. Terenius, K.G. Wilman, p53binds single-stranded DNA ends and catalyzes DNA renaturation

Ž .and strand transfer, Proc. Natl. Acad. Sci. USA 91 1994 413–417.w x8 J. Baudier, C. Delphin, D. Grunwald, S. Khochbin, J.J. Lawrence,

Characterization of the tumor suppressor protein as a protein kinaseC substrate and a S100b-binding protein, Proc. Natl. Acad. Sci.

Ž .USA 89 1992 11627–11931.w x9 A. Ciechanover, D. Shkedy, M. Oren, B. Bercovich, Degradation of

the tumor suppressor protein p53 by the ubiquitin-mediated prote-olytic system requires a novel species of ubiquitin-carrier protein,

Ž .E2, J. Biol. Chem. 269 1994 9582–9589.w x10 A.R. Clarke, C.A. Purdie, D.J. Harrison, R.G. Morris, C.C. Bird,

M.L. Hooper, A.H. Wyllie, Thymocyte apoptosis induced by p53-Ž .dependent and independent pathway, Nature 362 1993 849–852.

w x11 A. Deffie, H. Wu, V. Reinke, G. Lozano, The tumor suppressor p53Ž .regulates its own transcription, Mol. Cell. Biol. 13 1993 3415–

3423.w x12 E.M. Elliott, M.P. Mattson, P. Vanderklish, G. Lynch, I. Chang,

R.M. Sapolsky, Corticosterone exacerbates kainate-induced alter-ations in hippocampal tau immunoreactivity and spectrin proteolysis

Ž .in vivo, J. Neurochem. 61 1993 57–67.w x13 S. Fan, W.S. el-Diery, I. Bae, J. Freeman, D. Jondie, K. Bhatia, A.J.

Fornace Jr., I. Magrath, K.W. Kohn, P.M. O’Connor, p53 genemutations are associated with decreased sensitivity of human lym-

Ž .phoma cells to DNA damaging agents, Cancer Res. 54 19945824–5830.

w x14 S. Fields, S.K. Jang, Presence of a potent transcription activatingŽ .sequence in the p53 protein, Science 249 1990 1046–1049.

w x15 S. Floderus, Untersuchungen uber den Bau der menschlichen Hy-pophyse mit besonderer Berucksichtigung der quantitativen mikro-morphologischen Verhaltnisse, Acta Pathol. Microbiol. Scand. Suppl.

Ž .153 1944 1–276.w x16 H. Flotow, P.R. Graves, A. Wang, C.J. Fiol, R.W. Roeske, P.J.

Roach, Phosphate groups as substrate determinants for casein kinaseŽ .I action, J. Biol. Chem. 265 1990 14264–14269.

w x17 S.-M. Hsu, L. Raine, H. Fanger, Use of avidin–biotin–peroxidaseŽ .complex ABC in immunoperoxidase techniques, J. Histochem.

Ž .Cytochem. 29 1981 577–580.w x18 S. Jamal, E.B. Ziff, Raf phosphorylates p53 in vitro and potentiates

p53-dependent transcriptional transactivation in vivo, Oncogene 10Ž .1995 2095–2101.

w x19 M.B. Kastan, Q. Zhan, W. el-Diery, F. Carrier, T. Jacks, W.V.Walsh, B.S. Plunkett, B. Vogelstein, A.J. Fornace Jr., A mammaliancell cycle checkpoint pathway utilizing p53 and GADD45 is defec-

Ž .tive in ataxia-telangiectasia, Cell 71 1992 587–597.w x20 K.K. Khanna, M.F. Lavin, Ionizing radiation and UV induction of

p53 protein by different pathways in ataxia-telangiectasia cells,Ž .Oncogene 8 1993 3307–3312.

w x21 P.E. Kuhn, M.W. Miller, An immunoprecipitation study of therelationship of p53 and the ALZ-50-positive antigen, Neurosci. Lett.,submitted.

w x22 U.K. Laemmli, Cleavage of structural proteins during the assemblyŽ .of the head of bacteriophage T4, Nature 227 1970 680–685.

w x23 S.P. Lees-Miller, K. Sakaquchi, S.J. Ullrich, E. Appella, C.W.Anderson, Human DNA-activated protein kinase phosphorylates ser-ine 15 and 37 in the amino-terminal transactivation domain of

Ž .human p53, Mol. Cell. Biol. 12 1992 5041–5049.w x24 S.W. Lowe, E.M. Schmitt, S.W. Smith, B.A. Osborne, T. Jacks, p53

is required for radiation-induced apoptosis in mouse thymocytes,Ž .Nature 362 1993 847–849.

w x25 W. Maltzman, L. Czyzyk, UV irradiation stimulates levels of p53cellular tumor antigen in nontransformed mouse cells, Mol. Cell.

Ž .Biol. 4 1984 1689–1694.w x26 D.P. Martin, R.E. Schmidt, P.S. DiStefano, O.H. Lowry, J.G. Carter,

E.M. Johnson Jr., Inhibitors of protein synthesis and RNA synthesisprevent neuronal death caused by nerve growth factor deprivation, J.

Ž .Cell Biol. 106 1988 829–844.w x27 D.P. Martin, T.L. Wallace, E.M. Johnson Jr., Cytosine arabinoside

kills postmitotic neurons in a fashion resembling trophic factordeprivation: evidence that a deoxycytidine-dependent process may

Ž .be required for NGF signal transduction, J. Neurosci. 10 1990184–193.

w x28 M.P. Mattson, Antigenic changes similar to those seen in neurofibril-lary tangles are elicited by glutamate and Ca2q influx in cultured

Ž .hippocampal neurons, Neuron 2 1990 105–117.w x29 D.W. Meck, W. Eckhart, Phosphorylation of p53 in normal and

simian virus 40-transformed NIH 3T3 cells, Mol. Cell. Biol. 8Ž .1989 461–465.

w x30 M.W. Miller, W.M. Al-Ghoul, Numbers of neurons in the develop-ing principal sensory nucleus of the trigeminal nerve: enhancedsurvival of early-generated neurons over late-generated neurons, J.

Ž .Comp. Neurol. 330 1993 491–501.w x31 M.W. Miller, W.M. Al-Ghoul, M. Murtaugh, Expression of Alz-50

immunoreactivity in the developing principal sensory nucleus of thetrigeminal nerve: affect of transecting the infraorbital nerve, Brain

Ž .Res. 560 1991 132–138.w x32 M.W. Miller, S.J. Muller, Structure and histogenesis of the principal

sensory nucleus of the trigeminal nerve: effects of prenatal exposureŽ .to ethanol, J. Comp. Neurol. 282 1989 570–580.

w x33 D.M. Milne, R.H. Palmer, D.G. Campbell, D.W. Meek, Phosphory-lation of the p53 tumour-suppressor protein at three N-terminal sites

( )M.W. Miller, P.E. KuhnrBrain Research 769 1997 233–244244

Ž .by a novel casein kinase I-like enzyme, Oncogene 7 1992 1301–1369.

w x34 D.M. Milne, L.E. Campbell, D.G. Campbell, D.W. Meck, p53 isphosphorylated in vitro and in vivo by an ultraviolet radiation-in-duced protein kinase characteristic of the c-jun kinase, jnk-1, J.

Ž .Biol. Chem. 270 1995 5511–5518.w x35 J. Mosner, T. Mummenbrauer, C. Bauer, G. Sczakiel, F. Grosse, W.

Deppert, Negative feedback regulation of wild-type p53 biosynthe-Ž .sis, EMBO J. 14 1995 4442–4449.

w x36 P. Oberosler, P. Hloch, U. Ramsperger, H. Stahl, p53-catalyzedannealing of complementary single-stranded nucleic acids, EMBO J.

Ž .12 1993 2389–2396.w x37 F. Ojeda, M.I. Guarda, C. Maldonado, H. Folch, H. Diehl, Role of

protein kinase C in thymocyte apoptosis induced by irradiation, Int.Ž .J. Radiat. Biol. 61 1992 663–667.

w x38 J.A. Pietenpol, T. Tokino, S. Thiagalingam, W.S. el-Deiry, K.W.Kinzler, B. Vogelstein, Sequence-specific transcriptional activationis essential for growth suppression by p53, Proc. Natl. Acad. Sci.

Ž .USA 91 1994 1998–2002.w x39 S. Sakhi, A. Bruce, N. Sun, G. Tocco, M. Baudry, S.S. Schreiber,

p53 induction is associated with neuronal damage in the centralŽ .nervous system, Proc. Natl. Acad. Sci. USA 91 1994 7525–7529.

w x40 J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning: A Labo-ratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor,NY, 1992.

w x41 L. Schaeffer, R. Roy, S. Humbert, V. Moncollin, W. Vermeulen,J.H. Hoeijmakers, P. Chambon, J.M. Egly, DNA repair helicase: a

Ž .component of BTF2 TFIIH basic transcription factor, Science 260Ž .1993 58–63.

w x42 V.V. Shaposhnikova, O.R. Debrovinskaya, L.K. Eidus, Y.N. Korys-tov, Dependence of thymocyte apoptosis on protein kinase C and

Ž .phospholipase A , FEBS Lett. 348 1994 317–319.2w x43 D. Shkedy, H. Gonen, B. Bercovich, A. Ciechanover, Complete

reconstitution of conjugation and subsequent degradation of thetumor suppressor protein p53 by purified components of the ubiqui-

Ž .tin proteolytic system, FEBS Lett. 348 1994 126–130.w x44 M.L. Smith, I.-T. Chen, Q. Zhan, P.M. O’Conner, A.J. Fornace Jr.,

Involvement of the p53 tumor suppressor in repair of UV-type DNAŽ .damage, Oncogene 10 1995 1053–1059.

w x45 A.J. Smolen, L.L. Wright, T.J. Cunningham, Neuron numbers in thesuperior cervical sympathetic ganglion of the rat: a critical compari-

Ž .son of methods for cell counting, J. Neurocytol. 12 1983 739–750.w x46 T. Soussi, C.C. de Fromental, P. May, Structural aspects of the p53

Ž .protein in relation to gene evolution, Oncogene 5 1990 945–952.w x47 I. Takenaka, F. Morin, B.R. Seizinger, N. Kley, Regulation of the

sequence-specific DNA binding function of p53 by protein kinase CŽ .and protein phosphatases, J. Biol. Chem. 270 1995 5405–5411.

w x48 H. Towbin, T. Staehelin, J. Gordon, Electrophoretic transfer ofproteins from polyacrylamide gels to nitrocellulose sheets: procedure

Ž .and some applications, Proc. Natl. Acad. Sci. USA 76 19794350–4354.

w x49 K. Ueda, E. Masliah, T. Saitoh, S.L. Bakalis, H. Scoble, K.S. Kosik,Alz-50 recognizes a phosphorylated epitope of tau protein, J. Neu-

Ž .rosci. 10 1990 3295–3304.w x50 F. Valverde, L. Lopez-Mascaraque, J.A. de Carlos, Distribution and

morphology of Alz-50-immunoreactive cells in the developing vi-Ž .sual cortex of kittens, J. Neurocytol. 19 1990 662–671.

w x51 I.J. Vincent, P. Davies, ATP-induced loss of Alz-50 immunoreactiv-ity with the A68 proteins from Alzheimer brain is mediated by

Ž .ubiquitin, Proc. Natl. Acad. Sci. USA 87 1990 4840–4844.w x52 I.J. Vincent, P. Davies, Phosphorylation characteristics of the A68

Ž .protein in Alzheimer’s disease, Brain Res. 531 1990 127–135.w x53 G.E. Woloschak, C.-M. Chang-Liu, P. Shearin-Jones, Regulation of

Ž .protein kinase C by ionizing radiation, Cancer Res. 50 19903963–3967.

w x54 B.L. Wolozin, A. Scicutella, P. Davies, Reexpression of a develop-mentally regulated antigen in Down syndrome and Alzheimer dis-

Ž .ease, Proc. Natl. Acad. Sci. USA 85 1988 6202–6206.w x55 X. Yang, R. Martinez, J. Le Beau, O. Wiestler, G. Walter, Evolu-

tionary expression of the neuronal form of the src protein in theŽ .brain, Proc. Natl. Acad. Sci. USA 86 1989 4751–4755.

w x56 Q. Zhan, F. Carrier, A.J. Fornace Jr., Induction of cellular p53activity by DNA-damaging agents and growth arrest, Mol. Cell.

Ž .Biol. 13 1993 4242–4250.