You are on page 1of 12
Experimental Neurology 190 (2004) 192 – 203 www.elsevier.com/locate/yexnr Alzheimer’s pathology in human temporal
Experimental Neurology 190 (2004) 192 – 203 www.elsevier.com/locate/yexnr Alzheimer’s pathology in human temporal

Experimental Neurology 190 (2004) 192 – 203

Experimental Neurology 190 (2004) 192 – 203 www.elsevier.com/locate/yexnr Alzheimer’s pathology in human temporal

www.elsevier.com/locate/yexnr

Alzheimer’s pathology in human temporal cortex surgically excised after severe brain injury

Milos D. Ikonomovic a,b , Kunihiro Uryu c , Eric E. Abrahamson a , John R. Ciallella a , John Q. Trojanowski c , Virginia M.-Y. Lee c , Robert S. Clark d , Donald W. Marion e , Stephen R. Wisniewski f , Steven T. DeKosky a,b, *

a Department of Neurology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA b Department of Psychiatry, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA c Department of Pathology and Laboratory Medicine, Center for Neurodegenerative Disease Research, Institute on Aging, University of Pennsylvania, Philadelphia, PA 19104, USA d Department of Critical Care Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA e Department of Neurosurgery, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA f Epidemiology Data Center, Graduate School of Public Health, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA

Received 22 January 2004; revised 20 May 2004; accepted 10 June 2004 Available online 2 September 2004

Abstract

Traumatic brain injury (TBI) is a risk factor for the development of Alzheimer’s disease (AD). This immunohistochemical study determined the extent of AD-related changes in temporal cortex resected from individuals treated surgically for severe TBI. Antisera generated against A h species (total A h, Ah 1 42 , and Ah 1 40 ), the C-terminal of the A h precursor protein (APP), apolipoprotein E (apoE), and markers of neuron structure and degeneration (tau, ubiquitin, a -, h -, and g-synuclein) were used to examine the extent of Ah plaque deposition and neurodegenerative changes in 18 TBI subjects (ages 18 – 64 years). Diffuse cortical Ah deposits were observed in one third of subjects (aged 35 – 62 years) as early as 2 h after injury, with only one (35-year old) individual exhibiting ‘‘mature’’, dense-cored plaques. Plaque-like deposits, neurons, glia, and axonal changes were also immunostained with APP and apoE antibodies. In plaque-positive cases, the only statistically significant change in cellular immunostaining was increased neuronal APP ( P = 0.013). There was no significant correlation between the distribution of Ah plaques and markers of neuronal degeneration. Diffuse tau immunostaining was localized to neuronal cell soma, axons or glial cells in a larger subset of individuals. Tau-positive, neurofibrillary tangle (NFT)-like changes were detected in only two subjects, both of more advanced age and who were without A h deposits. Other neurodegenerative changes, evidenced by ubiquitin- and synuclein-immunoreactive neurons, were abundant in the majority of cases. Our results demonstrate a differential distribution and course of intra- and extra-cellular AD-like changes during the acute phase following severe TBI in humans. Ah plaques and early evidence of neuronal degenerative changes can develop rapidly after TBI, while fully developed NFTs most likely result from more chronic disease- or injury-related processes. These observations lend further support to the hypothesis that head trauma significantly increases the risk of developing pathological and clinical symptoms of AD, and provide insight into the molecular mechanisms that initiate these pathological cascades very early during severe brain injury. D 2004 Elsevier Inc. All rights reserved.

Keywords: Alzheimer’s disease; A h plaque; Glial cell

Introduction

Traumatic brain injury (TBI) is an environmental risk factor for the development of chronic neurodegenerative

* Corresponding author. Department of Neurology, University of Pittsburgh School of Medicine, 3471 Fifth Avenue, Suite 811, Pittsburgh, PA 15213. Fax: +1-412-692-4526. E-mail address: DeKoskyST@upmc.edu (S.T. DeKosky).

0014-4886/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.expneurol.2004.06.011

disorders, including Alzheimer’s disease (AD). TBI corre- lates with accelerated cognitive decline in head-injured, aged subjects (Luukinen et al., 1999), and there is an increased risk for AD dementia in patients who, in adulthood, sustained a severe head injury (Guo et al., 2000; McKenzie et al., 1994; Nemetz et al., 1999; Plassman et al., 2000; Roberts et al., 1991, 1994, 1993; Schofield et al., 1997) . One of the detrimental consequences of TBI involves altered processing of amyloid precursor protein (APP), which potentially con-

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

193

tributes to AD-like pathological cascades (DeKosky et al., 1998; McIntosh et al., 1998) . In AD, amyloid h peptide (A h ) overproduction and deposition induce inflammatory cyto- kine infiltration with microglial activation and oxidative stress (Altstiel and Sperber, 1991; Butterfield et al., 2002; Buxbaum et al., 1992; Forloni et al., 1992; Griffin et al., 1989, 1995) , the sequelae of which are also involved in trauma-related neurological dysfunction (Rothwell et al., 1997) . TBI studies in both animal models (Ciallella et al., 2002; Masumura et al., 2000; Murakami et al., 1998; Pierce et al., 1996; Smith et al., 1999; Uryu et al., 2002) and post- mortem human tissue (Gentleman et al., 1993; Graham et al., 1995; McKenzie et al., 1994; Roberts et al., 1991, 1994) demonstrated that increased APP production and/or accumu- lation can develop after injury. However, evidence of AD- like extracellular A h deposition has been described only in post-mortem human brain tissue so current knowledge of amyloidogenic APP metabolism and A h accumulation after human TBI derives almost exclusively from studies of victims of severe, lethal TBI (Roberts et al., 1994) . Notably, AD-like neurodegenerative changes as well as intracellular accumulation of neurofibrillary tangle (NFT)-like patholo- gies were not found in these patients, with exception of subjects with repetitive head injury (Geddes et al., 1999; Schmidt et al., 2001) . In the present study, we collected freshly resected human brain tissue from subjects who underwent surgical treatment within hours of TBI and assessed the extent of AD-like extracellular and intracellular changes during this acute post-injury period. We further examined possible associations between these changes and variables such as subjects’ age, primary lesion character- istics, hypothermia treatment, and functional outcome.

Materials and methods

Subjects

The study examined 18 patients (age range 18 –64 years), admitted to the University of Pittsburgh Medical Center (UPMC) between June 1999 and May 2001 for surgical treatment of severe closed head injuries (Glasgow Coma Scale, GCS score < 9). Patients’ demographic and clinical variables are listed in Table 1 . Data on the initial status (GCS; Teasdale and Jennett, 1974 ) and functional outcome at 3, 6, and 12 months after injury (Glasgow Outcome Scale, GOS; Jennett et al., 1981 ) were obtained for all TBI patients as described previously (Clark et al., 1999; Marion et al., 1997) . Time intervals from the initial head injury to surgical excision of traumatized brain tissue ranged from 2 to 19 h, except for two subjects who were admitted 58 and 74 h after TBI (Table 1) . Half of the patients had been randomized to 48 h of hypothermia (Clifton et al., 1993) , the other half was normo- thermic. Subjects in the hypothermic group were maintained at 32– 33 j C for 48 h, then rewarmed to 37 –38 j C over 12 h. Systemic neuromuscular paralytic medication (vecuronium

bromide) was used to prevent shivering during hypothermia. All normothermic patients also received neuromuscular par- alytic medication for a minimum of 24 h post injury. CT scans taken at the hospital admission were read and classified as described in Marshall et al. (1991) . A h plaque immunostaining from TBI subjects was com- pared to that in temporal cortex obtained post-mortem from subjects who died of causes not related to head trauma. This group included 10 non-demented subjects (age range 50– 75; M/F = 7/3; APOE q4 1/10) and 10 patients with end-stage Alzheimer’s disease (AD; age range 76 – 90, M/F = 3/7; APOEq4 6/10) who had been enrolled and followed in the University of Pittsburgh Alzheimer’s Disease Research Cen- ter (ADRC). Neuropathological confirmation of AD diagno- sis was made according to established criteria (Anonymous, 1997; Khachaturian, 1985; Mirra et al., 1991) by ADRC Brain Bank neuropathologists. TBI and non-demented con- trol subjects had no concomitant neurological disorders and no history of AD. In addition to control and AD samples, comparison of TBI data was made with surgery samples resected from temporal lobe epilepsy patients. These were processed with antibodies to tau and Ah , to support the specificity of these antibodies in surgery samples. The data on these patients were published previously (Matsuo et al., 1994) .

Surgical acquisition of tissue specimens

Brain specimens were obtained under the approval of the University of Pittsburgh Institutional Review Board for the Clinical Core of the University of Pittsburgh’s Brain Trauma Research Center and for the University of Pittsburgh ADRC. All head-injured patients underwent decompressive craniec- tomy, either to relieve intractable cerebral swelling or for removal of severely traumatized brain tissue. Under the Clinical Trauma Center protocol for management of life- threatening intracranial hypertension, temporal cortex tissue samples were taken in the operating room. Only tissue samples removed for decompression of the brain, and that would normally be discarded, were used; no tissue was removed solely for this study. Written informed consent was obtained from family members for each patient. Equal numbers of subjects ( n = 9) contributed tissue samples from the left or right temporal cortex (Table 1) .

Tissue processing for immunohistochemistry and histology

Cortical tissue was immersion-fixed in 2% paraformalde- hyde in phosphate buffer (PB) for 4 h at 4 j C. Tissue was cryoprotected in graded sucrose and embedded in OCT. Tissue sections were cut at 30 A for free-floating immuno- histochemistry (IHC) with antibodies against A h (10D5, A h40, A h 42), APP, and apoE (Table 2) , as well as for Thioflavin-S (Thio-S, Sigma, St. Louis, MO) and X-34 (Styren et al., 2000) histochemistry. Before IHC, but not histochemistry, tissue was pre-treated with 99% formic acid

194

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

Table 1 Case demographics, injury/lesion specifics, and functional outcomes

Plaque-positive cases (n = 8)

Case #

With amyloid-beta

 

No amyloid-beta*

 

123

 

45

6

7

8

Age

35

60

62

36

45

39

24

21

Gender

M

F

F

F

M

M

M

M

APOE

34

34

33

34

33

33

33

33

Injury type

n/a

Fall

Fall

A

Fall

MV

AT

n/a

h

post-TBI

2

16

16

10

9.2

12

2.8

4.4

Side/region GCS Lesion type Lesion side/region Lesion density Lesion volume CT delay (days) H/N treatment GOS 3 months GOS 6 months GOS 12 months

L/T

R/T

R/T

R/T

R/T

R/T

L/T

R/T

4

9

9

7

5

3

3

3

EC

IC

IC

IC

IC

EC

EC

EC

L

R/T

R/T

L/T

R/T

L

L

L/F

high

high

high

high

high

high

high

low

27.5

14.7

24.5

10.5

47.3

67.2

86.4

21.6

none

none

none

none

none

none

none

18 d

N

N

N

N

H

N

H

N

2

2

4

4

4

1

2

2

2

3

5

4

4

1

2

3

3

3

5

1

4

1

2

3

Plaque-negative cases (n = 10)

 

Case #

9

10

11

12

13

14

15

16

17

18

Age Gender APOE Injury type Hours post-TBI Tissue side/region GCS Lesion type Lesion side/region Lesion density Lesion volume CT delay (days) H/N treatment GOS 3 months GOS 6 months GOS 12 months

37

42

59

30

64

23

51

40

n/a

18

M

F

M

F

F

M

M

M

M

F

33

33

33

34

33

34

23

33

33

n/a

n/a

Fall

Fall

MC

n/a

MV

HO

Fall

Fall

MV

74

2.7

7

19

2.2

7.5

3.3

9.3

3.1

58

L/T

L/T

L/T

R/T

R/T

L/T

L/T

R/T

L/T

L/T

8

4

6

7

4

3

7

7

4

6

n/a

n/a

IC

EC

EC

EC/IC

IC

EC/IC

EC

EC

n/a

n/a

R/BG

R

R

R/F

L/BG

R/F

L

R

n/a

n/a

high

med

med

high/high

high

high/med

high

low

n/a

n/a

165.4

49.4

69.42

27.4/20.2

18.7

24.7/23.1

53.3

16.4

none

none

none

3d

none

none

none

none

16 d

21 d

H

H

H

H

N

N

H

H

N

H

2

1

1

n/a

1

2

1

4

1

2

3

1

1

n/a

1

2

1

4

1

3

4

1

1

4

1

2

1

4

1

2

Abbreviations: A = assault; APOE = apolipoprotein E genotype; AT = all terrain vehicle accident; BG = basal ganglia; d = days; EC = extracerebral; F = fro ntal cortex; GCS = Glasgow Coma Score; H/N = hypothermia/normothermia; HO = hit with object; IC = intracerebral; L = left; MC = motorcycle accident; MV = motor vehicle accident; NFTs = neurofibrillary tangles; R = right; T = temporal cortex.

* Plaque-positive cases marked as ‘‘no amyloid-beta’’ were positive for APP and/or apoE plaques.

Table 2 Antibodies used for immunohistochemistry

Antibody

Epitope

Host

Dilution

Source

10D5

A h 1 16

mouse

1:3,000

Elan Pharm.

A h 1 40

A

h

40

rabbit

1:500

Chemicon

A h 1 42

A

h

42

rabbit

1:500

Chemicon

ApoE

apoE

goat

1:10,000

Genzyme

Anti-6

APP 590 695

rabbit

1:10

Athena Neurosci.

17026

pan-tau

rabbit

1:10,000

CNDR

PHF-1

PS 396 /pS 404

mouse

1:1,000

P. Davies

NFL

NF-L Syn a h g Syn a (conf.) Ubiquitin

rabbit

1:5,000

CNDR

202

mouse

1:20,000

CNDR

303

mouse

1:500

CNDR

Ubiquitin

mouse

1:1,000

Chemicon

A h = amyloid h protein; apoE = apolipoprotein E; PHF = paired helical

filaments; NFL = neurofilament; Syn = synuclein.

(Sigma) for 5 min. The IHC procedure for human tissue samples has been described in detail (Ikonomovic et al., 1997). Briefly, tissue sections were rinsed between all steps with 0.1% Triton X-100 in 0.1 M PBS. Sections were blocked with normal serum and incubated with primary antibodies for 16 h at 4 j C. The antigen–antibody reaction was visualized by incubating sections with appropriate affinity-purified, biotinylated secondary antibodies (Vector Laboratories, Bur- lingame, CA) for 1 h at room temperature, followed by an avidin –biotin complex (ABC kit, Vector Laboratories), and imidazole acetate buffer (IAB) containing 3,3 V -diaminoben- zidine (DAB, Sigma), nickel ammonium sulfate, and hydro- gen peroxide. Immunostained sections were rinsed in IAB and mounted from PB onto gelatin-coated glass slides, cover- slipped, and examined with standard bright field microscopy.

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

195

IHC with markers of neuronal degeneration, including anti-tau, anti-neurofilament (NF) low Mr subunit (NFL), anti- ubiquitin, and anti-synuclein Syn202 and Syn303 (confor- mation-dependent) antibodies (Table 2) , followed the same staining protocol on 6- A-thick formalin-fixed, paraffin- embedded sections from a portion of excised temporal cortex tissue adjacent to that processed in paraformalde- hyde for A h staining. After visualizing immunoreaction with DAB as chromogen, tissue sections were counter- stained with hematoxylin. Additional tissue sections were processed as described but without the primary antibody as a control for nonspecific immunostaining. In no instance was immunostaining observed in such control sections. For each antibody, tissue sections from all cases were pro- cessed simultaneously, including tissue from control and confirmed AD cases from the ADRC, which served as negative and positive controls, respectively, for A h plaque and NFT staining (see Fig. 2 ). Nissl-stained adjacent sections were used to determine tissue cytoarchitecture. Three tissue sections from each case were also stained for aggregated Ah using fluorescent compounds Thio-S and X-34, and examined on an Olympus microscope with a fluorescent attachment.

Semi-quantitative estimation of pathological markers

In two randomly chosen coded IHC tissue sections from each case, frequencies of A h, apolipoprotein E (apoE), APP, and of neurodegenerative markers were estimated (Table 3) . These semi-quantitative evaluations reflected area density of immunoreactive (IR) plaques or neuronal/glial elements. That amount of surgical tissue was limited, and obtained only in one brain region (temporal cortex), precluded more quantitative analysis of neuropathological changes, and prevented full evaluation of plaques by CERAD criteria that require analysis of multiple brain regions. Two inves- tigators blinded to diagnosis and other clinical findings performed analyses of both plaques and neuronal changes in TBI, controls, AD, and epilepsy samples. In the case of plaques, sections free of deposits were given a score of ‘‘0’’; those showing ‘‘mild’’ plaque pathology with only focal collections of deposits separated by uninvolved cortex were scored ‘‘1’’; ‘‘moderate’’ plaque deposition, characterized by diffuse cortical involvement with plaques estimated to cover 30 – 50% of cortical area were scored as ‘‘2’’; while diffuse cortical involvement with estimated 50% or more of the cortical area covered with plaques was characterized as ‘‘severe’’ plaque pathology with a score of ‘‘3’’. For semi- quantitative assessment of tau-, NFL-, ubiquitin-, and syn- uclein-IR elements, the absence of immunoreactivity was scored ‘‘0’’, cases with an infrequent number of profiles

Notes to Table 3:

* Plaque-positive group > plaque negative group for APP neurons ( P =

0.0133).

** Neurofibrillary tangles.

Table 3 Semi-quantitative analysis of immunohistochemical markers of Alzheimer’s changes in temporal cortex from TBI cases

Case # Plaque-positive cases

Plaque-negative cases

 

1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18

Ab total

Plaques 3

2

3

1

2

2

0

0

00

0

0

0

00000

Neurons 0

1

0

0

0

0

0

1

00

1

0

0

10102

Glia

0

0

0

0

0

1

0

0

11

0

0

0

00001

Axons

0

0

0

0

1

0

0

0

00

0

0

0

00102

Ab 1 – 40

Plaques 2

1

0

1

1

1

0

0

00

0

0

0

00000

Neurons 0

0

0

0

0

1

0

1

11

0

1

0

10001

Glia

0

1

0

1

1

1

0

1

11

0

1

0

10001

Axons

0

0

0

0

0

0

0

0

00

0

0

0

00000

Ab 1 – 42 Plaques 3

2

3

1

2

2

0

0

00

0

0

0

00000

Neurons 0

1

0

0

0

1

0

1

11

1

0

0

00002

Glia

0

1

0

1

1

1

2

2

11

1

1

0

10000

Axons

0

0

0

0

0

0

0

0

00

0

0

0

00000

apoE

Plaques 3

1

3

1

2

1

0

1

00

0

0

0

00000

Neurons 0

3

1

0

1

0

1

0

10

0

1

0

10301

Glia

2

3

2

3

3

3

2

3

22

2

3

1

33331

Axons

0

0

0

0

0

0

0

0

00

0

0

0

00000

APP c-term*

 

Plaques 2

1

2

0

0

1

1

1

00

0

0

0

00000

Neurons 3

3

3

0

0

3

3

3

12

1

2

1

00103

Glia

0

0

0

1

0

0

0

0

00

1

0

0

10110

Axons

3

0

2

0

0

1

0

0

10

0

0

0

00023

Tau

Neurons 0

0

0

0

0

0

0

0

0 0

1** 0

1** 01001

Glia

1

0

0

0

0

0

0

0

00

0

0

0

00011

Axons

0

1

1

1

1

1

1

1

00

0

1

1

01101

NFL

Neurons 2

3

3

1

1

2

3

3

23

3

2

2

23131

Glia

0

0

0

0

0

0

0

0

00

0

0

0

00000

Spheroid 1

0

2

0

0

0

1

2

00

2

1

0

00000

Axons

0

1

2

0

0

0

2

1

11

0

0

1

02011

Ubiquitin

Neurons 2

2

0

3

2

0

2

3

33

3

2

2

33332

Glia

2

2

0

2

2

0

2

2

30

0

1

0

22322

Spheroid 2

0

0

1

0

2

0

1

13

2

2

2

21122

Axons

0

0

0

0

0

0

0

0

00

0

0

0

00000

Syn202

Neurons 0

0

2

1

1

2

2

1

01

1

0

1

01220

Glia

0

1

2

1

1

1

2

1

11

1

0

1

01211

Spheroid 0

0

0

0

0

0

0

0

00

0

0

0

00000

Axons

0

0

0

0

0

0

0

0

00

0

0

0

00000

Syn303

Neurons 0

0

3

2

2

2

2

1

02

2

2

1

01122

Glia

0

0

2

2

2

2

2

1

03

2

2

1

02122

Spheroid 0

0

0

0

0

0

2

0

20

0

0

0

00000

Axons

0

0

0

0

0

0

0

0

00

0

0

0

00000

196

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

were scored ‘‘1’’, cases with a moderate number of profiles were scored ‘‘2’’, and cases with clustered or widespread profiles were scored ‘‘3’’.

Statistical analysis

For statistical analysis, TBI cases were divided into two groups based on the presence (plaque-positive) or absence (plaque-negative) of plaque deposits containing Ah and/or related proteins (e.g., APP, apoE). The two groups were compared based on their demographic (age, gender, APOE genotype), clinical (time to surgery, GCS, GOS, hypother- mia treatment), and lesion (type, side, volume, density) characteristics using Fisher’s Exact tests, Wilcoxon tests and two-sample t tests. The two groups were also compared based on their semi-quantitative scores for immunostained cellular elements using Fisher’s Exact Statistic.

Results

AD plaque markers after TBI

Exact Statistic. Results AD plaque markers after TBI In eight of the 18 subjects suffering severe

In eight of the 18 subjects suffering severe TBI, temporal cortex contained AD-like plaque deposits (plaque-positive cases) of A h and/or related proteins (e.g., APP, apoE). There were no such plaques detectable in the remaining 10 (plaque-negative) TBI subjects, al- though some of these cases showed immunoreactive

pyramidal neurons, axons, and glia ( Table 3 , Figs. 1D –

F ). Plaque-positive and plaque-negative groups were sim-

ilar when compared by demographic variables and APOE genotype (percent of APOE q4 allele carriers). The two groups were also not statistically different with respect to clinical variables shown in Table 1 , including lesion characteristics analyzed on CT scans, hypothermia treat- ment, initial severity (GCS), or functional outcome scores on the GOS scale. For each subject, the specific pattern of plaque and cellular staining with antibodies against total A h, A h 1 40 , Ah 1 42 , apoE, and APP is shown in Table 3 and described below. All AD subjects had numerous cortical plaques detected with all five markers and tau- positive NFT (score of ‘‘3’’ for all), consistent with their neuropathological diagnosis. In contrast, only two con- trols, both of advanced age ( z74 years), had scarce, diffuse A h deposits (score of ‘‘1’’) and no (score of ‘‘0’’) NFT. There was no positive immunostaining with markers of plaques or neuronal changes in surgery sam- ples from epilepsy patients described in the study by Matsuo et al. (1994) .

Total Ab (10D5-positive) immunostaining

Temporal cortex in six of the eight plaque-positive TBI cases contained 10D5-IR A h plaques (Table 3) . These plaques were identified as extracellular deposits of A h in

Fig. 1. 10D5-immunoreactive plaques and cellular deposits in surgically excised temporal cortex from subjects who sustained severe traumatic brain injury. Compact amyloid beta (Ah ) deposits are detected as early as 2 h (A; case #1) after injury and are characterized by an irregular ‘‘fleecy’’ morphology. Plaques are also present at 16 h after injury (B; case #3), with comparable abundance, and better-defined borders. A h deposits in TBI are also present as large and dispersed neuropil accumulates (C; case #4). Intracellular A h deposits are observed in the cytoplasm of cortical pyramidal neurons (D) and in axons (E) of a plaque-negative case (case #18). Sporadic, Ah -positive glial cells are seen in another plaque-negative case (F; case #10). Scale bar 200 A (A –C); 100 A (D – F).

the neuropil, not associated with vascular elements. There was a considerable inter-subject variability in the abun- dance, size, and morphology of 10D5-IR plaques. They were most abundant in subjects #1 (35 year old; 2 h post- TBI; Tables 1 and 3 ) and #3 (62 year old; 16 h post-TBI; Tables 1 and 3 ), and appeared as well-defined circular deposits in the neuropil (Figs. 1A and B) . Similar, less frequent deposits were also observed in cases #2, #5, and #6 (Table 3) . The distribution of A h plaques in TBI subjects contrasted with that in our AD cases, which served as ‘‘positive staining controls’’ and consistently showed abun- dant 10D5-IR A h plaques throughout the temporal cortex (not shown). In one A h plaque-positive TBI subject (#4), 10D5-IR plaques were scarce, and characterized by diffuse and irregular accumulations of A h (Fig. 1C) , similar to those found in two aged control brains (not shown). Two of the plaque-positive TBI subjects (#7 and #8) had no A h deposits, despite the presence of APP and/or apoE plaques (Table 3) . In these cases, we found only an overall increase of 10D5 immunostaining in neuropil (not shown). Plaque- positive and plaque-negative groups were not statistically different regarding the presence of 10D5-IR pyramidal neurons, axons, and glial cells.

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

197

Ab 40 and Ab 42 immunohistochemistry

Immunohistochemical staining for Ah 1 40 and Ah 1 42 was performed to determine the proportion of A h species in plaques after TBI. There was a close correspondence in the abundance and morphology of A h42- and 10D5-IR plaques (Table 3) . When comparing the Ah42- and Ah40-IR depos- its in TBI cases, A h40-IR plaques were smaller and less numerous (Figs. 2A and B) . Plaques in TBI cases showed a restricted regional organization, whereas in AD cases, they were more abundant and distributed throughout the entire cortical region (Figs. 2E and F) . Ah 40-IR plaques were present in all but one of the 10D5/A h42 plaque-positive TBI cases (case #3, Table 3 ); notably, this case was also characterized by elevated and diffuse A h40 neuropil immu- noreactivity. There were no vascular A h deposits detected

noreactivity. There were no vascular A h deposits detected Fig. 2. Temporal cortex from a subject

Fig. 2. Temporal cortex from a subject who sustained severe TBI, (case #1, A – D) and an AD patient (E – H), immunostained for Ah 42 (A, E), A h40 (B, F), X-34 (C, G), or apoE (D, H). In the temporal cortex of the TBI case, A h42 plaques (A) are more abundant than A h40 plaques (B). Frequency of plaques stained with X-34 (C) is closer to that of A h40, while the distribution of apoE plaques (D) is similar to A h 42 plaques. AD temporal cortex shows a profusion of all plaque types (E – H). Not only plaques, but also NFT, neuropil threads, and vascular amyloidosis are abundantly stained in AD cases (G). Scale bar 200 A.

with A h40 and Ah 42 antibodies or with the general marker of A h (10D5 immunostaining) in temporal cortex of TBI cases. Neuronal and glial cellular elements were less fre- quently immunostained (Table 3) .

Thio-S/X-34 histochemistry

Thio-S and X-34 staining of aggregated A h in plaques was detected in one TBI subject only (case #1, Fig. 2C). NFT and neuropil threads were not histochemically detected in any TBI subjects, including case #1. Every AD case showed abundant thio-S and X-34 fluorescent plaques, NFT, neuropil threads, and vascular elements (Fig. 2G) , while no such staining was present in the control cases or the epilepsy surgical samples.

Apolipoprotein E immunohistochemistry

ApoE was localized to plaques in a subset of TBI cases. All cases with Ah -IR plaques also showed apoE-IR plaques (Table 3) . One subject (#8) without Ah -IR plaques had a very small number of apoE- and APP-IR deposits. ApoE-IR deposits were readily detectable as relatively small, compact formations that resembled Ah 40-IR plaques in their size and shape, and 10D5-IR or A h42-IR plaques in frequency and distribution (Table 3 , Fig. 2D). As with other plaque markers in TBI cases, apoE-IR plaques were similar, although less numerous, to those in AD cases (Figs. 2D and H) . There were no apoE-IR plaques in control subjects. In addition to plaques, apoE immunoreactivity was observed in neurons and glial cells after TBI. Glial apoE immunoreactivity was the most consistent feature of plaque-negative cases. There were no significant differences in apoE-IR cell elements between plaque-positive and plaque-negative groups.

C-terminal APP fragments immunohistochemistry

The immunohistochemical profile of neuropil deposits was characterized further by examining the distribution of C-terminal APP (APP 675 695 , anti-6 antibody) fragments. TBI subjects showed a complex pattern of anti-6 immunos- taining in neurites, neurons, and glia (Fig. 3) . APP-IR plaques were seen in almost every A h plaque-positive case, as well as in two subjects without Ah plaques (Table 3) . These APP deposits appeared as darkly stained aggregations of neuritic elements or as more diffuse neuropil accumu- lations with poorly defined borders (Fig. 3A) . Similar deposits were seen in AD cases, but were absent in controls (not shown). In TBI cases, the frequency of APP-IR plaques was generally low, comparable to that of A h40-IR plaques. In addition to plaques, the majority (13/18) of TBI cases showed APP-IR neurons, axons, and to lesser extent glia (Table 3) . Large pyramidal neurons were particularly darkly immunostained in their soma and apical dendrites, and were often accompanied by numerous APP-IR neuritic processes scattered throughout the neuropil (Fig. 3B) . These processes

198

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

et al. / Experimental Neurology 190 (2004) 192–203 Fig. 3. Immunostaining with antibodies against the

Fig. 3. Immunostaining with antibodies against the C-terminus of APP in temporal cortex from subjects who sustained severe traumatic brain injury. A number of APP-immunoreactive plaques, of inconsistent shape and size, are seen in case #1 (A). Intense APP immunostaining is observed in soma and processes of many pyramidal neurons from case #2 (B). Most of these cells appear dysmorphic, as do many APP-immunoreactive corkscrew- shaped neurites crossing the neuropil. APP immunostaining is similarly dense in axonal processes within the white matter, seen as bulbous axonal swellings (case #1, C and D). Scale bar 200 A (A, C); 100 A (B, D).

were intensely stained and frequently characterized with a ‘‘twisted,’’ corkscrew-like morphology. APP immunostain- ing was also localized to numerous, darkly stained axonal processes and bulbous swellings in the white matter of TBI tissue samples (Figs. 3C and D) . This pattern of axonal staining was not observed in control brains. There was a

statistically significant difference in the presence of APP- IR neurons when comparing plaque-positive and plaque- negative cases ( P = 0.013); frequencies of APP-IR axons and glia were not statistically different between the two groups.

AD neuropathology markers in TBI

Cellular pathology was further evaluated by examining the distribution of tau, synuclein, and ubiquitin immunos- taining in TBI cases. The majority of TBI subjects showed tau (PHF-1)-IR dystrophic axons in the white matter (Table 3) , consistent with diffuse axonal injury (Fig. 4A) . In addition, fine tau-IR deposits were dispersed in white matter of some cases, likely the remnants of degenerated axons. Light, diffuse tau immunoreactivity in neuronal somata/ dendrites was detected in only two cases (cases #15, 51 years and #18, 18 years; Table 3 ). These cases were negative for our plaque markers (A h , apoE, or APP). In two other plaque-negative cases, pyramidal cell soma showed a unique pattern of tau immunoreactivity that resembled NFTs (cases #11, 59 years and #13, 64 years; Table 3 ), appearing as dense somatodendritic staining. Consistent with the appearance of NFTs, some of these PHF-IR inclusions were also thio-S-positive (not shown). Normal neuronal cell bodies were NFL-IR in all cases examined, without evi- dence of pathological alterations at the level of the perikar- ya. Irregular axonal NFL staining resembling classic diffuse axonal injury was observed in 10 cases, and six cases showed NFL-IR axonal spheroids (data not shown).

six cases showed NFL-IR axonal spheroids (data not shown). Fig. 4. Temporal cortex from TBI subjects

Fig. 4. Temporal cortex from TBI subjects immunostained using antisera generated against paired helical filaments (A, PHF-1), synuclein (B, Syn202 ; C, Syn303), or ubiquitin (D), and counterstained with hematoxylin. Positive immunoreaction product is detected in axons (A) and cell bodies (B – D). Sca le bar 20 A.

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

199

Synuclein Syn202 (which recognizes alpha-, beta-, and gamma-synuclein) immunostaining was observed in cortical neuropil and in the cytoplasm of small clusters of neurons (Fig. 4B) . In addition, pericyte and muscle cells in the vascular wall showed intense Syn202 staining (not shown). The Syn303 antibody (which recognizes oxidized alpha- synuclein) showed cytoplasmic staining in neurons and some glial cells, while punctuate immunoreaction product was also detected in the neuropil (Fig. 4C) . Ubiquitin antibodies also labeled small intracellular elements in neu- ronal and glial cells (Fig. 4D) . Ubiquitin-IR pericytes and other cells, presumably of microglial or macrophage line- age, were detected near blood vessels (not shown). There were no significant differences in the presence of these markers of neuronal degeneration between plaque- positive and plaque negative groups. Further, none of the abnormal neurodegenerative pathologies described above were detected in control tissue samples or in surgical temporal lobe epilepsy samples studied earlier with anti- bodies to tau and Ah (Matsuo et al., 1994) .

Discussion

This analysis of surgically resected temporal cortex from survivors of severe TBI reveals that the formation of immunohistochemically detectable extracellular Ah depos- its can occur as early as 2 h after injury, demonstrating the extraordinary rapidity with which AD-like pathology can develop after acute neuronal insult. Extracellular deposits of A h, APP, and apoE were documented in 30% of subjects examined, corroborating and extending several previous post-mortem analyses (Horsburgh et al., 2000; Roberts et al., 1991, 1994) . In addition, these biopsy data are not confounded by hard-to-control variables such as agonal state and post-mortem interval, providing compelling evi- dence that amyloidogenic changes observed after TBI are not due to post-mortem artifacts. We remain, however, confronted with a critical issue, that of whether the observed AD-like pathogenic changes were preexistent (e.g., age- or pre-symptomatic disease-related) or injury-induced. Several compelling lines of evidence argue against age-related amyloidosis being a confounding factor in the interpretation of our results. First, the emergence of Ah plaques in non- demented, non-injured subjects is clearly age dependent, with the incidence of detectable A h deposition being virtually absent until the sixth decade of life (Braak and Braak, 1997) . Second, there are distinct differences in the morphological characteristics of presumed injury-induced plaques described in the current study, the diffuse amyloid- osis of aged control subjects, and the extensive A h pathol- ogy characteristic of AD. Third, there is considerable evidence that TBI induces elevated amyloidogenic APP processing in experimental models of TBI (Ciallella et al., 2002; Masumura et al., 2000; Murakami et al., 1998; Pierce et al., 1996; Smith et al., 1999; Uryu et al., 2002) .

A h plaques are virtually absent in populations younger than 50 years (Dayan, 1970; Mackenzie, 1994), and only after the eighth decade of life they are detected in 50% of subjects (Crystal et al., 1988; Delaere et al., 1990; Dickson et al., 1991; Katzman et al., 1988) . The extensive non- selective autopsy investigation by Braak and Braak (1997) rarely detected Ah plaques in younger subjects. None of the 58 cases examined in the 31 –35 years age range, and only two of the 83 subjects in the 36 –40 years age range had some Ah plaques (Braak and Braak, 1997) . Furthermore, A h deposits were detected in only 20 of the 207 people (<10%) in their fifth decade of life (Braak and Braak, 1997) . Notably, non-selective studies do not provide participants’ clinical history, thus one must be cautious in ruling out the potential contribution of older brain injuries and/or disease processes as causative factors in the formation of A h deposits even in the ‘‘normal’’ older subjects in the report by Braak and Braak (1997) . An additional large-scale human TBI study by Roberts et al. (1994) did not detect A h deposits in a normal control (non-injured) population under the age of 60. None of the younger plaque-positive subjects in this study had a family history of early-onset AD. Thus, it is highly unlikely that our TBI patients had preexisting plaque pathology that could account for the observations made in the current report. Several important similarities as well as differences exist between Ah plaques in AD and those detected after TBI. AD cases in our study consistently had numerous, compact and hard core Ah deposits, as well as abundant neurofibril- lary tangles (NFT) and neuropil threads. In contrast, TBI subjects were virtually devoid of dense-cored plaques, NFT, and neuropil threads. In agreement with this, a high inci- dence of diffuse plaques after head injury has been reported previously (Clinton et al., 1991) . In the present study, only one of our TBI subjects, a 35-year-old male, had detectable dense-cored plaques. Notably, this case was characterized by the most abundant A h40 deposits. In contrast, one plaque-positive TBI subject (without dense-cored plaques) exhibited abundant Ah 42 plaques in the absence of Ah 40 deposits. Overall, Ah 42 deposits were more abundant than A h40 deposits in the remaining plaque-positive TBI cases, suggesting that, similar to plaque genesis in AD, Ah 42 deposition may precede that of Ah 40 after TBI, and the formation of dense-cored plaques might require a substantial accumulation of A h40 species. The observed preponderance of Ah 42 over A h40 plaques in TBI subjects is in agreement with previous post-mortem analyses of TBI (Gentleman et al., 1997; Horsburgh et al., 2000) and AD (Iwatsubo et al., 1994, 1995) . Further studies using very early post-injury brain pathology and animal models are needed to under- stand the mechanisms underlying such rapid histopatholog- ical changes after TBI, including the sequence of deposition and/or contribution of the different A h species in the formation of more mature AD-like lesions. ApoE was also localized to plaques after TBI. In contrast to the post-mortem study by Horsburgh et al. (2000) , we

200

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

found apoE deposits in all of our A h plaque-positive TBI subjects, as well as in one individual with no detectable Ah deposits. Moreover, apoE and A h42 plaques were similarly abundant, with one case exhibiting numerous apoE and Ah 42 plaques in the absence of Ah 40 deposits. Thus, the deposition of apoE could, together with A h42, represent an initial plaque component or, alternatively, reflect a role for apoE in Ah clearance from the neuropil. It is possible that age-related changes in efficacy of apoE-mediated A h clearance could contribute to apoE-dependent amyloidogenesis after TBI, as it might occur in AD (Gomez-Isla et al., 1996; McNamara et al., 1998; Schmechel et al., 1993) . In support of this notion, the APOE q 4 genotype is associated with increased A h load and risk for developing AD (Mayeux et al., 1993a; McNa- mara et al., 1998) , as well as increased A h deposition after injury in a mouse model of AD (Hartman et al., 2002) and in human TBI (Horsburgh et al., 2000; Mayeux et al., 1993b, 1995; Nicoll et al., 1995) . Although we did not observe such a relationship in our TBI cases, greater numbers of subjects should be examined for this and other possible correlations, considering that a higher prevalence of the APOE q4 allele appears to be linked with Ah deposits even in some aged control subjects (Arai et al., 1999). Our results appear to contradict previous findings by Nakagawa et al. (1999), (2000) , who reported that brain trauma can diminish existing Ah plaques in transgenic, APP over-expressing mice. However, severe injury in those studies created a large cavity at the site of impact and analysis. Thus, it is possible that the absence of plaques was related to significant loss of APP-producing neurons due to lesion formation. Another study of trans- genic mice reported that head injury causing a minor scar, but not cavity-related neuronal loss, lead to accelerated plaque formation (Uryu et al., 2002) . The latter observa- tion is more comparable to our present study, which was of surgical samples that were not the site of the primary injury, and were not affected directly by necrotic lesion and associated neuronal loss. Thus, remaining injured neurons, and axons, are likely source of increased APP and A h production after TBI. Increased numbers of APP-immunoreactive neuronal cell bodies were the only statistically significant cellular feature distinguishing the plaque-positive TBI group. The dystrophic morphology of intensely APP-labeled neurons indicates an early neurode- generative process that could result in a surge in A h production, as demonstrated experimentally in traumatized APP transgenic mice (Smith et al., 1998) . Furthermore, the distribution of amyloidogenic APP fragments (c-terminal APP) and Ah in cortical plaques following TBI suggests that plaque-deposited Ah could be derived in part from APP in axonal ends of injured neurons projecting to this brain region. C-terminal fragments of APP are increased in experimental models of TBI (Ciallella et al., 2002; Stone et al., 2000) and in AD brains where they likely represent an initial step in A h production (Hardy, 1992; Wilson et al., 1999) . The observation that increased APP processing and

accumulation of c-terminal APP fragments were detected (albeit to a significantly lesser extent) in plaque-negative cases, and were not present in normal controls, suggests that they might precede A h plaque formation after TBI. Although our surgically extracted tissue samples were not the sites of primary mechanical injury, they exhibited a variety of neurodegenerative changes, including ubiquitin, synuclein, and PHF tau-IR neuronal and axonal elements, as well as spheroid inclusions. A similar pattern of immunos- taining was described previously in the brains of patients with various neurodegenerative diseases associated with alpha-synuclein pathology (Trojanowski and Lee, 2002) , and alpha synuclein-IR axonal spheroids were present in cases of TBI (Newell et al., 1999) . Furthermore, recurrent TBI in dementia pugilistica is associated with neuropatho- logical PHF-tau profiles (Geddes et al., 1999; Schmidt et al., 2001), indicating that in chronic or repetitive brain injury tau can undergo pathological phosphorylation and neuronal accumulation similar to that observed in AD brains (Lee et al., 2001; Roberts, 1988; Schmidt et al., 2001) . Notably, both A h plaque-positive and -negative TBI cases had tau-, ubiq- uitin-, and synuclein-positive cell profiles. While evidence of such lesions developing within hours after a single injury is important for our understanding of neuronal vulnerability to TBI, the present study could not establish a clear link between the expression of these markers of neurodegenera- tive changes and Ah plaques. Subjects who showed NFT- like profiles were relatively older, cautioning against a cause –effect relationship between acute TBI and NFT formation. Unlike A h plaque deposition, intracellular NFT formation may be the result of a slower and more chronic neurodegenerative process. Thus, in contrast to A h deposi- tion, AD-like NFT changes either require longer evolution, or are not pathological consequences of a single acute TBI. Neither A h deposition nor neuronal degeneration in temporal cortex in TBI cases correlated significantly with subjects’ age, severity of brain injury, post-injury interval, or APOE genotype. We cannot exclude the possibility that A h or other AD-like changes have developed in some other cortical areas that we were not able to examine. Unlike post- mortem studies, we were limited in our investigation only to the single brain region (temporal cortex) that was the target of the surgical intervention. However, post-mortem studies indicate that temporal cortex is the area most often affected and most likely to show A h changes after TBI (Roberts et al., 1994) . In conclusion, our study provides new evidence that acute brain injury is frequently associated with AD-like pathology, and that such changes can evolve very rapidly (i.e., within hours). These data and the extensive literature strongly argue that early AD pathology, as a pre-existing condition, contributes little, if at all, to these observations. How dynamic these changes are over the long term, and their possible relationship with functional and pathological outcomes, remains to be determined in greater number of cases. Future studies focusing on additional plaque-associ-

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

201

ated molecules (e.g., acute phase molecules ACT, IL-1 h, in addition to apoE) will be critical for a better understanding of regulatory factors underlying A h deposition and/or clearance from cortical tissue following TBI.

Acknowledgments

We thank Yetta I. Wilbur, Barbara A. Isanski, and Daniel Martinez for expert technical assistance. We are grateful to the University of Pittsburgh Alzheimer’s Disease Research Center (AG05133), the Brain Trauma Research Center (NS30318), and Ava Puccio for brain tissue processing. We also thank Dr. Peter Davies for providing PHF-1 antibodies. Supported by NINDS NS30318, NIA AG05133, NIA AG11542, MH18273.

References

Altstiel, L.D., Sperber, K., 1991. Cytokines in Alzheimer’s disease. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 15, 481 – 495. Anonymous, 1997. National Institute on Aging and Reagan Institute work- ing group on diagnosis criteria for the neuropathological assessment of Alzheimer’s disease. Consensus recommendations for the postmortem diagnosis of AD. Neurobiol. Aging 18, S1 – S3. Arai, T., Ikeda, K., Akiyama, H., Haga, C., Usami, M., Sahara, N., Iritani, S., Mori, H., 1999. A high incidence of apolipoprotein E epsilon4 allele in middle-aged non-demented subjects with cerebral amyloid beta pro- tein deposits. Acta Neuropathol. 97, 82 – 84. Braak, H., Braak, E., 1997. Frequency of stages of Alzheimer-related lesions in different age categories. Neurobiol. Aging 18, 351 – 357. Butterfield, D.A., Griffin, S., Munch, G., Pasinetti, G.M., 2002. Amyloid beta-peptide and amyloid pathology are central to the oxidative stress and inflammatory cascades under which Alzheimer’s disease brain exists. J. Alzheimer’s Dis. 4, 193 – 201. Buxbaum, J.D., Oishi, M., Chen, H.I., Pinkas-Kramarski, R., Jaffe, E.A., Gandy, S., Greengard, P., 1992. Cholinergic agonists and interleukin-1 regulate processing and secretion of the Alzheimer ß/A4 amyloid pro- tein precursor. Proc. Natl. Acad. Sci. U. S. A. 89, 10075 – 10078. Ciallella, J.R., Ikonomovic, M.D., Paljug, W.R., Wilbur, Y.I., Dixon, C.E., Kochanek, P.M., Marion, D.W., DeKosky, S.T., 2002. Changes in expression of amyloid precursor protein and interleu- kin-1beta after experimental traumatic brain injury in rats. J. Neu- rotrauma 12, 1555 – 1567. Clark, R.S.B., Kochanek, P.M., Chen, M., Watkins, S.C., Marion, D.W., Chen, J., Hamilton, R.L., Loeffert, J.E., Graham, S.H., 1999. Increases in Bcl-2 and cleavage of caspase-1 and caspase-3 in human brain after head injury. FASEB 13, 813 – 821. Clifton, G.L., Allen, S., Barrodale, P., Plenger, P., Berry, J., Koch, S., Fletcher, J., Hayes, R.L., Choi, S.C., 1993. A phase II study of moderate hypothermia in severe brain injury. J. Neurotrauma 10, 263 – 271. Clinton, J., Ambler, M.W., Roberts, G.W., 1991. Post-traumatic Alzheim- er’s disease: preponderance of a single plaque type. Neuropathol. Appl. Neurobiol. 17, 69 – 74. Crystal, H., Dickson, D., Fuld, P., Masur, D., Scott, R., Mehler, M., Mas- deu, J., Kawas, C., Aronson, M., Wolfson, L., 1988. Clinico-pathologic studies in dementia: nondemented subjects with pathologically con- firmed Alzheimer’s disease. Neurology 38, 1682 – 1687. Dayan, A.D., 1970. Quantitative histological studies on the aged human brain. I. Senile plaques and neurofibrillary tangles in ‘‘normal’’ patients. Acta Neuropathol. 16, 85 – 94. DeKosky, S.T., Kochanek, P.M., Clark, R.S.B., Ciallella, J.R., Dixon, C.E.,

1998. Secondary injury after head trauma: subacute and long-term mechanisms. Semin. Clin. Neuropsychiatry 3, 176 – 185. Delaere, P., Duyckaerts, C., Masters, C., Beyreuther, K., Piette, F., Hauw, J.J., 1990. Large amounts of neocortical beta A4 deposits without neu- ritic plaques nor tangles in a psychometrically assessed, non-demented person. Neurosci. Lett. 116, 87 – 93. Dickson, D.W., Crystal, H.A., Mattiace, L.A., Masur, D.M., Blau, A.D., Davies, P., Yen, S.H., Aronson, M.K., 1991. Identification of normal and pathological aging in prospectively studied nondemented elderly humans. Neurobiol. Aging 13, 179 – 189. Forloni, G., Demicheli, F., Giorgi, S., Bendotti, C., Angeretti, N., 1992. Expression of amyloid precursor protein mRNAs in endothelial, neuro- nal, and glial cells: modulation by interleukin-1. Brain Res., Mol. Brain Res. 16, 128 – 134. Geddes, J.F., Vowles, G.H., Nicoll, J.A., Revesz, T., 1999. Neuronal cyto- skeletal changes are an early consequence of repetitive head injury. Acta Neuropathol. 98, 171 – 178. Gentleman, S.M., Graham, D.I., Roberts, G.W., 1993. Molecular pathology of head trauma: altered h APP metabolism and the aetiology of Alz- heimer’s disease. In: Kogure, K., Hossmann, K.-A., Siesjo, B.K. (Eds.), Progress in Brain Research, vol. 96. Elsevier Science Publishers B.V., New York, pp. 237 – 246. Gentleman, S.M., Greenberg, B.D., Savage, M.J., Noori, M., Newman, S.J., Roberts, G.W., Griffin, W.S., Graham, D.I., 1997. A beta 42 is the pre- dominant form of amyloid beta-protein in the brains of short-term survi- vors of head injury. NeuroReport 8, 1519 – 1522. Gomez-Isla, T., West, H.L., Rebeck, G.W., Harr, S.D., Growdon, J.H., Locascio, J.J., Perls, T.T., Lipsitz, L.A., Hyman, B.T., 1996. Clinical and pathological correlates of apolipoprotein E epsilon 4 in Alzheimer’s disease. Ann. Neurol. 39, 62 – 70. Graham, D.I., Gentleman, S.M., Lynch, A., Roberts, G.W., 1995. Distri- bution of B-amyloid protein in the brain following severe head injury.

Neuropathol. Appl. Neurobiol. 21,

27 – 34.

Griffin, W.S.T., Stanley, L.C., Ling, C., White, L., Macleod, V., Perrot, L.J.,

White III, C.L., Araoz, C., 1989. Brain interleukin 1 and S-100 immu- noreactivity are elevated in Down syndrome and Alzheimer’s disease. Proc. Natl. Acad. Sci. U. S. A. 86, 7611 – 7615. Griffin, W.S., Sheng, J.G., Roberts, G.W., Mrak, R.E., 1995. Interleukin-1 expression in different plaque types in Alzheimer’s disease: significance in plaque evolution. J. Neuropathol. Exp. Neurol. 54, 276 – 281. Guo, Z., Cupples, L.A., Kurz, A., Auerbach, S.H., Volicer, L., Chui, H., Green, R.C., Sadovnick, A.D., Duara, R., DeCarli, C., Johnson, K., Go, R.C., Growdon, J.H., Haines, J.L., Kukull, W.A., Farrer, L.A., 2000. Head injury and the risk of AD in the MIRAGE study. Neurology 54, 1316 – 1323. Hardy, J., 1992. An ‘‘anatomical cascade hypothesis’’ for Alzheimer’s disease. Trends Neurosci. 15, 200 – 201. Hartman, R.E., Laurer, H., Longhi, L., Bales, K.R., Paul, S.M., McIntosh, T.K., Holtzman, D.M., 2002. Apolipoprotein E4 influences amyloid deposition but not cell loss after traumatic brain injury in a mouse model of Alzheimer’s disease. J. Neurosci. 22, 10083 – 10087. Horsburgh, K., Cole, G.M., Yang, F., Savage, M.J., Greenberg, B.D., Gen- tleman, S.M., Graham, D.I., Nicoll, J.A., 2000. h -amyloid (Ah )42(43), ah 42, ah 40 and apoE immunostaining of plaques in fatal head injury. Neuropathol. Appl. Neurobiol. 26, 124 – 132. Ikonomovic, M.D., Mizukami, K., Davies, P., Sheffield, R., Armstrong, D.M., 1997. The loss of GluR2(3) immunoreactivity precedes neurofi- brillary tangle formation in the entorhinal cortex and hippocampus of Alzheimer brains. J. Neuropathol. Exp. Neurol. 56, 1018 – 1027. Iwatsubo, T., Odaka, A., Suzuki, N., Mizusawa, H., Nukina, N., Ihara, Y., 1994. Visualization of A h42(43) and A h 40 in senile plaques with end- specific Aß monoclonals: evidence that an initially deposited species is A h 42(43). Neuron 13, 45 – 53. Iwatsubo, T., Mann, D., Odaka, A., Suzuki, N., Ihara, Y., 1995. Amyloid protein (A) deposition: A 42(43) precedes A 40 in Down syndrome. Ann. Neurol. 37, 294 – 299. Jennett, B., Snoek, J., Bond, M.R., Brooks, N., 1981. Disability after severe

202

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

head injury: observations on the use of the Glasgow Outcome Scale. J. Neurol., Neurosurg. Psychiatry 44, 285 – 293. Katzman, R., Terry, R., DeTeresa, R., Brown, T., Davies, P., Fuld, P., Renbing, X., Peck, A., 1988. Clinical, pathological and neurochemical changes in dementia: a subgroup with preserved mental status and numerous neocortical plaques. Ann. Neurol. 23, 138 – 144.

Khachaturian, Z.S., 1985. Diagnosis of Alzheimer’s disease. Arch. Neurol. 42, 1097 – 1105. Lee, V.M., Goedert, M., Trojanowski, J.Q., 2001. Neurodegenerative tauo-

pathies. Annu. Rev. Neurosci. 24,

1121 – 1159.

Luukinen, H., Viramo, P., Koski, K., Laippala, P., Kivela, S.L., 1999. Head injuries and cognitive decline among older adults: a population-based study. Neurology 55, 557 – 562. Mackenzie, I.R., 1994. Senile plaques do not progressively accumulate with normal aging. Acta Neuropathol. 87, 520 – 525. Marion, D.W., Penrod, L.E., Kelsey, S.F., Obrist, W.D., Kochanek, P.M., Palmer, A.M., Wisniewski, S.R., DeKosky, S.T., 1997. Treatment of traumatic brain injury with moderate hypothermia. N. Engl. J. Med. 336, 540 – 546. Marshall, L.F., Gautille, T., Klauber, M.R., 1991. The outcome of severe

closed head

injury. J. Neurosurg. 75, S28 – S36.

Masumura, M., Hata, R., Uramoto, H., Murayama, N., Ohno, T., Sawada, T., 2000. Altered expression of amyloid precursors proteins after trau- matic brain injury in rats: in situ hybridization and immunohistochem- ical study. J. Neurotrauma 17, 123 – 134. Matsuo, E.S., Shin, R.-W., Billingsley, M.L., Van de Voorde, A., O’Con- nor, M., Trojanowski, J.Q., Lee, V.M.-Y., 1994. Biopsy-derived adult human brain tau is phosphorylated at many of the same sites as Alz- heimer’s disease paired helical filament tau. Neuron 13, 989 – 1002. Mayeux, R., Stern, Y., Ottman, R., Tatemichi, T.K., Tang, M.-X., Maestre, G., Ngai, C., Tycko, B., Ginsberg, H., 1993a. The apolipoprotein E4 allele in patients with Alzheimer’s disease. Ann. Neurol. 34, 752 – 754. Mayeux, R., Ottman, R., Tang, M.X., Noboabauza, L., Marder, K., Gur- land, B., Stern, Y., 1993b. Genetic susceptibility and head injury as risk factors for Alzheimer’s disease among community-dwelling elderly persons and their 1st-degree relatives. Ann. Neurol. 33, 494 – 501. Mayeux, R., Ottman, R., Maestre, G., Ngai, C., Tang, M.-T., Ginsberg, H., Chun, M., Tycko, B., Shelanski, M., 1995. Synergistic effects of trau- matic head injury and apolipoprotein-E4 in patients with Alzheimer’s disease. Neurology 45, 555 – 557. McIntosh, T.K., Saatman, K.E., Raghupathi, R., Graham, D.I., Smith, D.H., Lee, V.M., Trojanowski, J.Q., 1998. The Dorothy Russell Memorial Lecture. The molecular and cellular sequelae of experimental traumatic brain injury: pathogenetic mechanisms. Neuropathol. Appl. Neurobiol. 24, 251 – 267. McKenzie, J.E., Gentleman, S.M., Roberts, G.W., Graham, D.I., Roy- ston, M.C., 1994. Increased numbers of h APP-immunoreactive neuro- nes in the entorhinal cortex after head injury. NeuroReport 6, 161 – 164. McNamara, M.J., Gomez-Isla, T., Hyman, B.T., 1998. Apolipoprotein E genotype and deposits of Abeta40 and Abeta42 in Alzheimer disease. Arch. Neurol. 55, 1001 – 1004. Mirra, S.S., Heyman, A., McKeel, D., Sumi, S.M., Crain, B.J., Brownlee, L.M., Vogel, F.S., Hughes, J.P., van Belle, G., Berg, L., 1991. The consortium to establish a registry for Alzheimer’s disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzhei- mer’s disease. Neurology 41, 479 – 486. Murakami, N., Yamaki, T., Iwamoto, Y., Sakakibara, T., Kobori, N., Fush- iki, S., Ueda, S., 1998. Experimental brain injury induces expression of amyloid precursor protein, which may be related to neuronal loss in the hippocampus. J. Neurotrauma 15, 993 – 1003. Nakagawa, Y., Nakamura, M., McIntosh, T.K., Rodriguez, A., Berlin, J.A., Smith, D.H., Saatman, K.E., Raghupathi, R., Clemens, J., Saido, T.C., Schmidt, M.L., Lee, V.M.-Y., Trojanowski, J.Q., 1999. Traumatic brain injury in young, amyloid-beta peptide overexpressing transgenic mice induces marked ipsilateral hippocampal atrophy and diminished Abeta deposition during aging. J. Comp. Neurol. 411, 390 – 398. Nakagawa, Y., Reed, L., Nakamura, M., McIntosh, T.K., Smith, D.H.,

Saatman, K.E., Raghupathi, R., Clemens, J., Saido, T.C., Lee, V.M.-Y., Trojanowski, J.Q., 2000. Brain trauma in aged transgenic mice induces regression of established Abeta deposits. Exp. Neurol.

163, 244 – 252. Nemetz, P.N., Leibson, C., Naessens, J.M., Beard, M., Kokmen, E., Annegers, J.F., Kurland, L.T., 1999. Traumatic brain injury and time

to onset of Alzheimer’s disease: a population-based study. Am. J. Epi-

demiol. 149, 32 – 40. Newell, K.L., Boyer, P., Gomez-Tortosa, E., Hobbs, W., Hedley-White, E.T., Vonsattel, J.P., Hyman, B.T., 1999. Alpha-synuclein immuno- reactivity is present in axonal swellings in neuroaxonal dystrophy and acute traumatic brain injury. J. Neuropathol. Exp. Neurol. 58,

1263 – 1268.

Nicoll, J.A., Roberts, G.W., Graham, D.I., 1995. Apolipoprotein E epsilon 4 allele is associated with deposition of amyloid beta-protein following head injury. Nat. Med. 1, 135 – 137. Pierce, J.E.S., Trojanowski, J.Q.T., Graham, D.I., Smith, D.H., McIn- tosh, T.K., 1996. Immunohistochemical characterization of altera- tions in the distribution of amyloid precursor proteins and B- amyloid peptide after experimental brain injury in the rat. J. Neu- rosci. 16, 1083 – 1090. Plassman, B.L., Havlik, R.J., Steffens, D.C., Helms, M.J., Newman, T.N., Drosdick, D., Phillips, C., Gau, B.A., Welsh-Bohmer, K.A., Burke, J.R., Guralnik, J.M., Breitner, J.C., 2000. Documented head injury in early adulthood and risk of Alzheimer’s disease and other dementias. Neu- rology 55, 1158 – 1160. Roberts, G.W., 1988. Immunocytochemistry of neurofibrillary tangles in dementia pugilistica and Alzheimer’s disease: evidence for common genesis. Lancet 2, 1456 – 1458. Roberts, G.W., Gentleman, S.M., Lynch, A., Graham, D.I., 1991. hA4 amyloid protein deposition in brain after head trauma. Lancet 338,

1422 – 1423.

Roberts, G.W., Nash, M., Ince, P.G., Royston, M.C., Gentleman, S.M.,

1993. On the origin of Alzheimer’s disease: a hypothesis. Clin. Neuro-

sci. Neuropathol. 4, 7 – 9. Roberts, G.W., Gentleman, S.M., Lynch, A., Murray, L., Landon, M., Graham, D.I., 1994. h amyloid protein deposition in the brain after severe head injury: implications for the pathogenesis of Alzheimer’s disease. J. Neurol., Neurosurg. Psychiatry 57, 419 – 425. Rothwell, N., Allan, S., Toulmond, S., 1997. The role of interleukin 1 in acute neurodegeneration and stroke: pathophysiological and therapeutic

implications. J. Clin. Invest. 100, 2648 – 2652. Schmechel, D.E., Saunders, A.M., Strittmatter, W.J., Crain, B.J., Hulette, C.M., Joo, S.H., Pericak-Vance, M.A., Goldgaber, D., Roses, A.D.,

1993. Increased amyloid beta-peptide deposition in cerebral cortex as

a consequence of apolipoprotein E genotype in late onset Alzheimer disease. Proc. Natl. Acad. Sci. U. S. A. 90, 9649 – 9653.

Schmidt, M.L., Zhukareva, V., Newell, K.L., Lee, V.M.-Y., Trojanowski, J.Q., 2001. Tau isofom profile and phosphorylation state in dementia pugilistica recapitulate Alzheimer’s disease. Acta Neuropathol. 101,

518 – 524.

Schofield, P.W., Tang, M., Marder, K., Bell, K., Dooneief, G., Chun, M.,

Sano, M., Stern, Y., Mayeux, R., 1997. Alzheimer’s disease after remote head injury: an incidence study. J. Neurol., Neurosurg. Psychiatry 62,

119 – 124.

Smith, D.H., Nakamura, M., McIntosh, T.K., Wang, J., Rodriguez, A., Chen, X.-H., Raghupathi, R., Saatman, K.E., Clemens, J., Schmidt, M.L., Lee, V.M.-Y., Trojanowski, J.Q., 1998. Brain trauma induces massive hippocampal neuron death linked to a surge in beta-amy- loid levels in mice overexpressing mutant amyloid precursor protein. Am. J. Pathol. 153, 1005 – 1010. Smith, D.H., Chen, X.H., Nonaka, M., Trojanowski, J.Q., Lee, V.M., Saat- man, K.E., Leoni, M.J., Xu, B.N., Wolf, J.A., Meaney, D.F., 1999. Accumulation of amyloid beta and tau and the formation of neurofila- ment inclusions following diffuse brain injury in the pig. J. Neuropa- thol. Exp. Neurol. 58, 982 – 992. Stone, J.R., Singleton, R.H., Povlishock, J.T., 2000. Antibodies to the C-

M.D. Ikonomovic et al. / Experimental Neurology 190 (2004) 192–203

203

terminus of the beta-amyloid precursor protein (APP): a site specific marker for the detection of traumatic axonal injury. Brain Res. 871, 288 – 302. Styren, S.D., Hamilton, R.L., Styren, G.C., Klunk, W.E., 2000. X-34, a fluorescent derivative of Congo red: a novel histochemical stain for Alzheimer’s disease pathology. J. Histochem. Cytochem. 48, 1223 – 1232. Teasdale, G., Jennett, B., 1974. Assessment of coma and impaired con- sciousness. A practical scale. Lancet 2, 81 – 84. Trojanowski, J.Q., Lee, V.M.-Y., 2002. Parkinson’s disease and related

synucleinopathies are a new class of nervous system amyloidoses. Neu- rotoxicology 23, 457 – 460. Uryu, K., Laurer, H., McIntosh, T., Pratico, D., Martinez, D., Leight, S., Lee, V.M., Trojanowski, J.Q., 2002. Repetitive mild brain trauma accel- erates Abeta deposition, lipid peroxidation, and cognitive impairment in a transgenic mouse model of Alzheimer amyloidosis. J. Neurosci. 22, 446 – 454. Wilson, C.A., Doms, R.W., Lee, V.M., 1999. Intracellular APP processing and A beta production in Alzheimer disease. J. Neuropathol. Exp. Neu- rol. 58, 787 – 794.