SEARCH

SEARCH BY CITATION

Keywords:

  • Creutzfeldt–Jakob disease;
  • dementia;
  • Gerstmann–Sträussler–Scheinker syndrome;
  • hippocampus;
  • prion

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

Aim:  The hippocampus can be very sensitive to damage in the scrapie-infected mouse, a well-established animal model of prion diseases. Terminally ill scrapie-infected animals exhibit nearly complete loss of cornu ammonis (CA) 1 pyramidal neurons, but few studies have focused on the neuropathological lesions of the human hippocampus in autopsied brain tissue; in particular, few findings on differences in severity of pathology between the hippocampal and parahippocampal formations have been obtained. The aim of the present paper is to evaluate the human hippocampus of prion disease through neuropathological examination.

Methods:  A systemic, detailed neuropathological study throughout the subdivisions of the hippocampus was carried out in 23 autopsied cases of prion diseases. Prion protein immunohistochemistry was performed in serial brain sections to determine the topography of prion deposits.

Results:  Compared to lesions in other brain regions, hippocampal lesions were mild, despite numerous prion deposits. The distribution of prion deposits did not appear to be correlated with neuropathological changes. The present findings differed from the hippocampal pathology observed in scrapie-infected mice. In addition, differences in neuropathological severity were observed within the hippocampal formation.

Conclusion:  The human hippocampus may be protected from the neurotoxic effects of prion deposits.

PRION DISEASES ARE a group of closely related neurodegenerative disorders of humans and animals that includes scrapie in sheep, bovine spongiform encephalopathy (BSE) in cattle, and Creutzfeldt–Jakob disease (CJD) and Gerstmann–Sträussler–Scheinker syndrome (GSS) in humans.1 All human prion diseases have in common the distinctive clinical features of subacute, progressive development of symptoms of dementia. Studies using experimental models of scrapie have shown that severe neuronal loss develops in the hippocampal pyramidal cell layer.2–10 Under some circumstances, vacuolation and more destructive degeneration occur in the hippocampus, with pyramidal cell necrosis accompanied by glial reactions, which can progress to severe hippocampal sclerosis.8 The hippocampal lesions in human prion diseases, however, have been little investigated.

Although a few studies focusing on the neuropathology of the hippocampus have been performed using autopsied brain tissue,11–13 they appear to lack a sufficient number of cases, examination of the different types of prion diseases, and more importantly, prion immunostaining. A few prion-immunostaining studies based on a large number of cases14–16 have summarized hippocampal pathology simply by stating that neuropathological changes in the hippocampus were very mild compared with those in the cerebral cortex. No detailed studies on the hippocampal pathology of prion diseases have clarified the differences in severity of the basic lesions throughout all subdivisions of the hippocampal formation.

We performed a systematic, detailed neuropathological study of 23 cases of prion disease, including sporadic CJD, CJD after transplantation of the dura, familial CJD, and GSS. We report here the distribution of prion deposits in the hippocampal formation, and the relationship between prion deposits and neuropathological lesions throughout the subdivisions of the hippocampus.

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

A total of 23 cases of prion disease, including 12 cases of sporadic CJD, seven cases of familial CJD of E200K type (codon 200 glutamic acid to lysine change in the prion protein gene), three GSS cases including two of P102L (codon 102 proline to leucine mutation) and one of P105L (codon 105 proline to leucine mutation), and one infectious case associated with cadaveric dura mater were examined. Clinical, neuropathological, and genetic data of the cases examined are summarized in Table 1. In the present study we did not classify the sporadic cases into type 1 or 214 on western immunoblotting with antiprion antibody, that is, the pathological isoform of the prion protein (PrPSc) consisting of two non-glycosylated bands of either approximately 21 kDa (type 1) or approximately 19 kDa (type 2).15 This research was conducted following the guidelines of the Ethics Committee on Clinical Experiments of Shinshu University School of Medicine.

Table 1.  Prion disease subjects examined
No.SexAge at onsetAge at deathDuration (months)Brain weight (g)Clinical diagnosisPathological diagnosisMutationCodon 129 genotype
  1. CJD, Creutzfeldt-Jakob disease; GSS, Gerstmann–Sträussler–Scheinker syndrome; MM, methionine homozygous; MV, methionine/valine heterozygous; N/A, no information available, PSP, progressive supranuclear palsy.

1F656622726CJDCJDSporadic 
2MN/AN/AN/A1200CJDCJDSporadic 
3F575920825CJDCJDSporadic 
4F697031000Cerebellar ataxia, PSP, CJDCJDSporadic 
5M606218600CJDCJDSporadic 
6MN/AN/AN/A1200CJDCJDSporadic 
7F515543700CJDCJDSporadic 
8F77785890CJDCJDSporadic 
9F687025560CJDCJDSporadic 
10FN/A60N/A920CJDCJDSporadic 
11F666816840CJDCJDSporadic 
12FN/AN/A19770CJDCJDSporadic 
13M4045511160GSSGSSP102LMM
14M4656120930Cerebellar ataxiaGSSP102LMM
15F465384930Frontal atrophyGSSP105LMV
16F505113N/ACJDCJDE200KMM
17M656561300CJDCJDE200KMM
18M737471190CJDCJDE200KMM
19F585871060CJDCJDE200KMM
20M6970101300CJDCJDE200K 
21M62639N/ACJDCJDE200KMM
22F78784950CJDCJDE200KMM
23M323533910CJDCJD associated with a cadaveric dura matter  

Formalin-fixed, paraffin-embedded brain sections from these patients were stained with HE, Luxol fast blue, or Holzer's stain. Immunohistochemistry for prions in the brain sections was performed using established and validated procedures.17 The sections were pretreated with 2 mmol/L HCl for 15 min at 121°C. After being washed with distilled water, the sections were incubated with 96% formic acid for 5 min at room temperature. After more washes, the sections were treated with 4 mol/L guanidine thiocyanate for 2 h at 4°C. Thereafter, the sections were incubated with an appropriate dilution of 3F4, a mouse monoclonal antibody that recognizes amino acids 109–112 of prion protein (Biodesign International, Saco, ME, USA). The sections were rinsed and incubated with horseradish peroxidase-linked antimouse IgG (Nichirei, Tokyo, Japan). The peroxidase substrate reaction was done using a diaminobenzidine peroxidase substrate tablet set (Sigma Chemicals, St Louis, MO, USA). The distribution of prion deposits was evaluated qualitatively, and the severity of lesions was evaluated semiquantitatively.

This study used unusual autopsied brain tissue of patients with prion diseases, in that, in Japan, unlike Western countries, the lifetime of patients was often artificially prolonged by enteral feeding, even after the clinical diagnosis of prion disease had been made. Duration of disease was therefore relatively long (mean 26 months), as indicated in Table 1.

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

The neuronal loss, astrogliosis and spongiform change observed in cornu ammonis (CA) 1, CA2, CA4, subiculum, entorhinal cortex, and cerebral cortex are summarized in Table 2.

Table 2.  Neuropathology for 23 cases of prion disease
No.CA4CA2CA1SubiculumEntorhinal cortexCerebral cortex
abcabcabcabcabcabc
  1. Neuronal loss: 3+, severe, almost all nerve cells are lost; 2+, moderate, approximately two-thirds of nerve cells are lost; +, mild, approximately one-third of nerve cells are lost; ±, little neuronal loss; –, no neuronal loss.

  2. Astrogliosis: 3+, severe, >20 abnormal astroglias are seen in a field of ×200 magnification; 2+, moderate, 10–20 abnormal astroglias are seen in a field of ×200 magnification; +, mild, <10 abnormal astroglias are seen in a field of ×200 magnification; ±, few abnormal astroglias; –, no astrogliosis.

  3. Spongiform change: 3+, severe; almost all of the region is changed into vacuolation and loosening; 2+, moderate, two-thirds of the region is changed into vacuolation and loosening; +, mild, one-third of the region is changed into vacuolation and loosening; ±, little vacuolation; –, no vacuolation.

  4. a, neuronal loss; b, astrogliosis; c, spongiform change; CA, cornu ammonis; cerebral cortex, inferior temporal cortex; NA, not available.

1+2+±±±+±±+++2+2+2+3+2+3+
2+2+±+±2+2+2+3+2+3+
3+2+±±±+±++±+2+3+2+3+3+3+
4±±±±±+±+2++2+
5+3+±±±±±±±3+3+3+3+3+3+
6++±±±±2++3+3+2+3+
7++±+±+++2+++2+2++2+3+2+3+
8++±±±+++2+2+2+
9++±±±±2+2+2+3+3+3+
10+++±±±+±2++±2+2+2+2+3+3+3+
11++±±±±±±2+2+2+3+3+3+
12+2+±±±±±3+3+2+3+3+3+
13±+±+±++++2++2+3+3+3+3+3+3+
14±+±+±++++2++2+3+3+3+3+3+3+
15±+±±±++±+++
16NANANANANANA±±±±±2+2+2+2+2+
17±±±±±+++±2++±2+2+2+3+
18++±±+±+±±+2+2+2+3+3+3+
19±+±±±±±+++2+3+3+3+3+3+3+
20+2+±++±+2+++++2+3+2+3+3+3+
21NANANA±±±+±++2+2+2+3+3+3+
22±±±±±±++2+2+2+2+
23±+±±±±±±±3+3+3+3+3+3+

Neuropathology of the hippocampus

Although the main lesions in cerebral cortex were severe in terms of neuronal loss, astrogliosis, and spongiform change, the hippocampal lesions were generally very mild (Fig. 1). In the hippocampus, the lesions with the highest level of severity were found in the subiculum, where small spongiform vacuoles appeared. Astrogliosis and mild neuronal loss without spongiform change were found in CA4 in almost all cases. Basic lesions were continuous in severity, and pronounced in the entorhinal cortex, although less so in CA. Fibrillary gliosis without spongiform change was observed in the stratum radiata, stratum lacunosum, and CA. Slight spongiform change was observed in the stratum moleculare. In each of the different types of prion disease examined, the pattern of hippocampal lesions was similar.

image

Figure 1. Representative photomicrographs showing the cerebral hemisphere and hippocampal formation of a patient with sporadic Creutzfeldt–Jakob disease (CJD). Note that the hippocampus is relatively preserved compared with cerebral cortex. Compared to cerebral cortex, cornu ammonis (CA)1, CA2, CA4, and the subiculum exhibit no major loss of neurons. Entorhinal cortex displays astrogliosis and mild spongiform change, and the subiculum exhibits only mild spongiform change. Entorhinal cortex, subiculum: HE stain, bar, 200 μm; cerebral cortex, CA1–4: HE stain, bar, 50 μm.

Download figure to PowerPoint

Topography of prion accumulation in the hippocampus

Considerable accumulation of abnormal prion protein was observed in the hippocampus, especially in the stratum moleculare, subiculum, and CA1–4 (Fig. 2a). There were two main types of prion deposits, diffuse and granular (Fig. 2b,d). The diffuse type was seen in the CA, the strata oriens, moleculare, radiatum, and lacunosum, and in the vestigial hippocampal sulcus (Fig. 2a). The granular pattern was observed only in the stratum moleculare. In the GSS cases, many plaque-like patterns of prion deposits were observed in the stratum moleculare (Fig. 2c,d).

image

Figure 2. Distribution of prion deposits in the hippocampus: (a,b) sporadic Creutzfeldt–Jakob disease (CJD); (c) Gerstmann–Sträussler–Scheinker syndrome (GSS; P105L-MV); (d) GSS (P102L-MM). Prion protein staining is pale and diffuse, and granular synaptic deposits appear in the molecular and fiber matrix. (a) The vestigial hippocampal sulcus and stratum radiatum are paler in color. (b) There are widespread fine granular and linear deposits with occasional coarse granular accentuation. Intraneuronal cytoplasmic deposition is rare but present, especially in the strata moleculare and lacunosum, and the cornu ammonis. (c) Plaque deposits are observed in the stratum moleculare. (d) Spongiform change is typically seen in the stratum moleculare. Amyloid plaques and granular prion deposits are also observed. Bars: (a,c) 1 mm; (b) 50 μm; (d) 100 μm. MM, codon 129: methionine homozygote; MV, codon 129: methionine/valine heterozygote.

Download figure to PowerPoint

Relationship between prion deposits and pathological lesions

Unexpectedly, the distribution of prion deposits did not correlate with the neuropathological changes in the brain. Figure 3 summarizes the neuropathological and immunohistochemical findings. For example, in the pyramidal cell layer of CA1–3, considerable amounts of prion deposits were observed without major neuronal loss (Figs 1,2). In contrast, the entorhinal cortex adjoining the subiculum was moderately damaged by more or less the same amounts of prion deposits (Fig. 1).

image

Figure 3. (a) Qualitative distribution of prion protein deposits. Densely colored regions, indicating more amounts of prion deposits, are the strata oriens, radiatum and moleculare. (b) Basic lesions consisting of neuronal loss, astrogliosis, and spongiform change. The basic lesions are seen in entorhinal cortex, subiculum and cornu ammonis (CA)1–4. CA4 is composed of much more astrogliosis. The strata lacunosum and moleculare show spongiform change predominantly, and the stratum radiata and polymorphic layer show fibrillary gliosis predominantly. The distribution of prion deposits did not correlate with the neuropathological changes in the hippocampus. In pyramidal cell layer of CA1–3, considerable amounts of prion deposits were observed without major neuronal loss.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

The hippocampus plays an important role in learning and memory. It is markedly affected from the initial stages of many dementing illnesses, such as Alzheimer's disease. In the scrapie-infected mouse, an animal model of prion disease, the hippocampus can be very sensitive to damage.2–10 Vacuolar degeneration, with pyramidal cell loss accompanied by glial reactions, can progress to severe hippocampal sclerosis.8 The timing and nature of the neuropathology in the hippocampus of the ME7/CV scrapie model have been described in detail.5,18,19 Surprisingly, in human prion diseases known to present with presenile progressive dementia, the hippocampal formation is spared until the final stages of disease. This raises the question of what the mechanism of resistance of the human hippocampus to prion diseases is. We expected that build-up of abnormal prions would be absent or minimal in the hippocampus, resulting in only mild neuropathological changes. Unexpectedly, however, considerable prion deposits were observed in the hippocampus. The topography of prion accumulation was thus not clearly correlated with neuropathological damage in the brain.

The findings of the present in situ study suggest the following possibilities: (i) the biochemical characteristics of prion deposits in the hippocampus may differ from those of deposits found at other sites in the brain; or (ii) the hippocampus is protected against the neurotoxic effects of prion deposits by an as-yet unknown mechanism. Findings of neurotoxicity should in general be interpreted cautiously, in view of the interaction between abnormal deposits at a given site in the brain. Elucidation of the molecular pathological relationship between the accumulation of abnormal prion protein and the basic lesions in prion disease is beyond the scope of the present study. Given the similarities between the distribution of some cell adhesion factors and the distribution of the hippocampal lesions in prion disease, we are planning to examine the relationship between cell adhesion factor expression and the resistance of the hippocampus to accumulation of abnormal prion protein.

Recently, biochemical studies have shown that hyperphosphorylated protein deposits in neurons or glial cells, such as tau and α-synuclein, play central roles in the pathophysiology of Alzheimer's disease and dementia with Lewy bodies, respectively.20 Methods including molecular genetics, protein chemistry, and protein engineering are now universally accepted, and widely used as models of not only other neurodegenerative diseases but psychiatric disorders, as well. But biochemical studies have yielded little insight into the location in brain from which tau and α-synuclein originate, and many studies do not specify this information. The brain is an elaborate composite, in which different types of neurons exist 1 μm apart, and may produce com pletely different substances. We previously reported that the bed nucleus of the anterior commissure, a very small nucleus within the rat hypothalamic region, expresses the mRNA for brain-derived neurotrophic factor during the perinatal period.21 Within the hippocampal formation, the different subdivisions play different roles.22 Study of specimens from the cerebral cortex and from the hippocampus might lead to entirely different results. Thus, it is important for research to be performed with reference to the specific site of the brain being examined. We are convinced that similar accumulations of abnormal prion protein have different toxic effects depending on the location in the brain. In the study of neuropsychiatric disorders, in vitro and in situ techniques, such as immunohistochemistry and in situ hybridization histochemistry, should complement one another.

There are several limitations to the present study. Only one infectious case was examined, and some significant strains, such as variant CJD, were not examined. Despite these drawbacks, we believe the present findings are of great value to the study of prion diseases, as a bridge between biochemistry and neuropathology.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

This work was supported in part by Grants-in-Aid from the Ministry of Health, Labor, and Welfare (to N.A) and the Ministry of Education, Science, and Technology, Japan (to N.S).

REFERENCES

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES