Aluminosilicates and the ageing brain: implications for the pathogenesis of Alzheimer's disease

Senile plaques are a neuropathological feature of the ageing brain and consist of abnormal neuritic and glial processes surrounding an extracellular core of material with fibrillary ultrastructure. Present at low densities in the cerebral cortex of most aged individuals, they occur in large numbers in Alzheimer's disease, the major form of senile dementia. Energy-dispersive X-ray microprobe analysis of isolated cores and plaques in situ from patients with Alzheimer's disease or Down's syndrome and from normal controls has shown co-localization of high concentrations of aluminium (4-19%) and silicon (6-24%) at the centre of the core. The presence of these elements as aluminosilicates has been confirmed using solid-state 27Al nuclear magnetic resonance. These findings provide a link with the other major neuropathological feature of Alzheimer's disease, the neurofibrillary tangle-bearing neurons, where high intracellular levels of Al and Si have also been reported. The focal deposition of these elements may be an early and essential factor in the pathogenesis of Alzheimer-type changes, reflecting an increased exposure to aluminium.

Aluminium accumulation, beta-amyloid deposition and neurofibrillary changes in the central nervous system

Deposition of beta-amyloid and the formation of neurofibrillary tangles (NFTs) are central to the aetiopathogenesis of Alzheimer's disease (AD). The possible effects of aluminium on these processes have been investigated in patients with renal failure who are exposed chronically to high blood levels of aluminium. Focal accumulation of aluminium was observed in neurons with high densities of transferrin receptors, indicating transferrin-mediated uptake, in regions such as cortex and hippocampus which are selectively vulnerable in AD. Increased staining for the beta-amyloid precursor protein (APP) in cortical pyramidal neurons was evident in the majority of renal patients and immature senile plaques were present in 30% of cases, suggesting that aluminium may induce or accelerate beta-amyloid deposition. The absence of neurofibrillary changes in this group of renal patients indicates that aluminium does not directly cause the formation of NFTs. The brain aluminium content was not raised in neuropathologically assessed cases of AD and we have been unable to confirm claims of defective transferrin binding in this disorder. If aluminium contributes to the development of sporadic AD, it must do so indirectly, perhaps via effects on the synthesis or metabolism of APP, or by contributing generally to the age-related attrition of neurons and thus reducing the threshold for deficits produced by more specific disease-related processes.

Individual dopaminergic neurons show raised iron levels in Parkinson disease

OBJECTIVE

Evidence suggests that abnormal iron metabolism is associated with Parkinson disease (PD), with raised iron levels found in pathologically affected areas in PD. It is unknown if this elevated iron is actually associated with neurons or reactive glia, and we therefore addressed this issue by determining if raised iron was present in single dopaminergic neurons.

METHODS

We used unfixed frozen sections from postmortem tissue of PD patients and elderly normal individuals to avoid metal contamination and translocation. Levels of iron and other elements were measured using sensitive and specific wavelength dispersive electron probe x-ray microanalysis coupled with cathodoluminescence spectroscopy in individual substantia nigra dopaminergic neurons.

RESULTS

We identified raised intraneuronal iron in single defined substantia nigra neurons in PD (mean neuronal iron 2,838 vs 1,611, p < 0.0001) but not in other movement disorders such as Huntington disease. These findings were unrelated to the density of remaining neurons.

CONCLUSIONS

Primary changes in neuronal iron could lead to neurodegeneration in Parkinson disease.

Does the mitochondrial genome play a role in the etiology of Alzheimer’s disease?

We report here the analyses of complete mtDNA coding region sequences from more than 270 Alzheimer's disease (AD) patients and normal controls to determine if inherited mtDNA mutations contribute to the etiology of AD. The AD patients and normal individuals were carefully screened and drawn from two populations of European descent in an effort to avoid spurious effects due to local population anomalies. Overall, there were no significant haplogroup associations in the combined AD and normal control sequence sets. Reduced median network analysis revealed that the AD mtDNA sequences contained a higher number of substitutions in tRNA genes, and that there was an elevated frequency of replacement substitutions in the complex I genes of the control sequences. Analysis of the replacement substitutions indicated that those arising in the AD mtDNAs were no more deleterious, on average, than those in the control mtDNAs. The only evidence for the synergistic action of mutations was the presence of both a rare non-conservative replacement substitution and a tRNA mutation in 2 AD mtDNAs, from a total of 145, whereas such a combination of mutations was not observed in the control sequences. Overall, the results reported here indicate that pathogenic inherited mtDNA mutations do not constitute a major etiological factor in sporadic AD. At most, a small proportion of AD patients carry a pathogenic mtDNA mutation and a small proportion of cognitively normal aged individuals carry a mtDNA mutation that reduces the risk of AD.

The Institute for Ageing and Health, University of Newcastle, UK

The Institute For Ageing And Health (IAH) is the largest cross-disciplinary research grouping within Newcastle University's Faculty of Medicine, which recently obtained the highest 5 or 5* ratings in all fields evaluated in the UK Research Assessment Exercise 2001. The IAH was set up in 1994 to bring together clinical, basic and social scientists in partnership with colleagues in the National Health Service. It builds upon a long tradition of outstanding clinical research on age-related disorders, particularly in the field of dementia where the pioneering studies of Tomlinson and Roth in the 1960s first showed Alzheimer's disease to be the commonest cause of cognitive decline in later life. The clinical research of the IAH now extends to both neurodegenerative and vascular dementia in a joint Medical Research Council-University Development for Clinical Brain Ageing, and to studies in many other areas including depression in later life, falls and neurovascular instability, stroke and ischaemic brain disease, and health services research on the medical and social care of older people. These diverse areas of clinical investigation are now complemented by strong research on the basic biology of ageing within the new Department of Gerontology with its programmes on the genetics of ageing and longevity; molecular mechanisms of cellular ageing, including oxidative stress, DNA damage and genomic instability, telomere reduction and regulation, mitochondrial DNA mutations, and accumulation of aberrant proteins; and theoretical models of the ageing process. An ambitious strategy for future research on ageing and age-related disorders is based on the synergy between these complementary approaches.

Multiple substrates of late-onset dementia: implications for brain protection

Age is the single most important risk factor for progressive dementia in populations worldwide. In developed countries the prevalence of dementia is estimated to be 3-5% at age 65 years and expected to double every decade thereafter. Although there is ageing-related attrition of neural tissue accompanied by profound changes in brain glia, marked neuronal loss and severe cognitive impairment are associated with pathological changes. Accelerated somatic ageing of the vasculature comprising endothelial and smooth muscle cells and slowed glial replacement are also likely to pre-dispose to degenerative processes. Approximately 90% of patients with late-onset dementia have neuropathological features of Alzheimer's disease (AD), dementia with Lewy bodies (DLB), or vascular dementia (VaD), alone or in combination. Both AD and DLB reveal extensive amyloid beta deposition within senile plaques. Neurofibrillary tangles evident as tau pathology are much reduced in DLB where symptoms may be more related to cholinergic transmitter abnormalities than structural pathology. Depletion of brain acetylcholine is also encountered in VaD, which like AD and DLB may respond to cholinergic therapy. Cerebrovascular pathology, ischaemic brain damage and neurovascular instability resulting in cerebral hypoperfusion appears fundamental in the pathogenesis of late-onset dementia. The apolipoprotein E epsilon 4 allele, a major genetic susceptibility factor for AD also associated with cardiovascular pathology, may contribute to neurodegenerative changes through vascular mechanisms. The interrelationships of these multiple substrates of late-onset dementia have major implications for neuroprotective and disease slowing therapies. Measures that improve cardiovascular function and increase brain perfusion would be useful to attenuate cognitive decline.

Nitric oxide synthase gene polymorphisms in Alzheimer's disease and dementia with Lewy bodies

Evidence suggests that vascular and inflammatory components may be important in the aetiology of dementia and genetic risk factors affecting these processes may therefore influence disease development. Recently, polymorphisms in the endothelial constitutive nitric oxide synthase 3 (NOS3) and also the inducible nitric oxide synthase gene (NOS2A) have been suggested to lead to increased risk of Alzheimer's disease (AD) or dementia with Lewy bodies. We have studied the relationship of both these NOS gene polymorphisms to development of AD and dementia with Lewy bodies and find no evidence for association with either condition. We conclude that NOS gene polymorphisms do not alter disease risk in the majority of late-onset dementia cases.

Quantification and characterization of fluctuating cognition in dementia with Lewy bodies and Alzheimer's disease

Fluctuating cognition (FC) is a common and important symptom in dementia, particularly dementia with Lewy bodies (DLB), although it has not been empirically quantified or characterised. Forty subjects (15 DLB, 15 AD, 10 elderly controls) were evaluated using a clinical FC severity scale, as well as receiving measures of variability in attentional performance and slow EEG rhythms across 90 s, 1 h and 1 week. DLB patients had significantly more severe FC and more severe variability in attentional and slow electrocortical measures than either AD patients or normal controls in all time frames. Attentional and EEG variability also correlated significantly with independent clinical ratings of FC. Clinical quantification and measures of attention and EEG variability can therefore make an important and standardised contribution to the assessment of FC in dementia, facilitating future treatment studies with important implications for the potential causative mechanisms and differential diagnosis.

Mitochondrial DNA haplogroups and susceptibility to AD and dementia with Lewy bodies

The authors analyzed the relationship between nuclear genetic risk factors (apolipoprotein E genotype) and mitochondrial DNA (mtDNA) sequence variants in pathologically proved cases of AD (n = 185), dementia with Lewy bodies (DLB; n = 84), and control subjects (n = 179). Specific European mtDNA haplogroups and the A4336G mutation were not associated with an increased risk of AD. mtDNA haplogroup H was overrepresented in the DLB patients when compared with control subjects. Additional studies are needed to clarify the significance of the association.

Brain oestradiol and testosterone levels in Alzheimer's disease

Epidemiological studies indicate that oestrogen improves memory and may delay the onset of Alzheimer's disease (AD) in postmenopausal women. Furthermore, evidence from experimental studies suggests beneficial effects of oestrogen on several pathogenic mechanisms implicated in AD. We have therefore measured the levels of oestradiol and testosterone in control and AD brains. The results show that in control brain, oestradiol levels are 3.5 fold higher in females than males, though testosterone levels are equivalent. In AD, oestradiol levels were not significantly increased compared to those in control brain, while testosterone levels were unaffected in AD. The results do not support the hypothesis that a lack of oestrogen is a contributory factor in AD.

Absorption of aluminium-26 in Alzheimer's disease, measured using accelerator mass spectrometry

Although chromosomal abnormalities underpin some early onset cases of familial Alzheimer's disease (AD), most cases are sporadic and not associated with such abnormalities. Aluminium (Al) is a significant but controversial risk factor for sporadic AD, and studies have reported associations between Al and the principal pathological features of AD, senile plaques and neurofibrillary tangles. The present study measured gastrointestinal (GI) absorption of Al under normal dietary conditions using (26)Al tracer and accelerator mass spectrometry (AMS). Following overnight fast, 13 AD patients (aged 63-76 years) and 13 age-matched controls (aged 62-76 years) ingested a fruit drink containing 27 ng (26)Al. Plasma samples were obtained before and 1 h after the drink and from these the fraction of (26)Al absorbed across the GI tract was estimated. The GI tract rigorously excludes Al with only 0.06-0.1% of the ingested Al being absorbed. The mean fraction absorbed by AD subjects exceeded controls by a factor of 1.64 (p</=0.05, Anova). AMS is capable of determining <10(-16) g of (26)Al with many orders of magnitude more sensitivity than other techniques. Using this sensitivity, we have shown, under normal physiological conditions, that the ability of the GI tract to exclude Al is reduced in AD, possibly leading to greater systemic exposure to Al. Public health measures to limit Al dietary uptake or bioavailability may decrease the prevalence of AD in the community and should be considered.

Lack of association of the alpha2-macroglobulin locus on chromosome 12 in AD

OBJECTIVE

Analysis of AD has revealed that the apolipoprotein E locus (APOE) cannot account for all of the genetic risk associated with AD. Whole genome scanning in AD families suggests that a chromosome 12 locus may contribute significantly to disease development. The alpha2-macroglobulin gene (A2M) has been suggested as a candidate locus for AD based on analysis of familial AD.

METHOD

We determined, in 195 neuropathologically verified AD cases and 107 age-matched control subjects, the association of two common polymorphisms in A2M (a pentanucleotide deletion 5' to the bait domain exon, and a valine-1000-isoleucine polymorphism in the thiolester site of the protein).

RESULTS

Evidence was observed for linkage disequilibrium between the deletion and Ile1000 polymorphisms. No evidence was observed for an association between the thiolester polymorphism and AD alone or when accounting for the APOE-epsilon4 allele. No alteration in the frequency of the bait domain deletion was observed, although a small excess (4%) of deletion homozygotes was found in the AD group, which were absent in the control population.

CONCLUSIONS

The A2M deletion polymorphism at most accounts for a small fraction of the genetic contribution toward AD, and this is small compared with APOE. Furthermore, reverse transcriptase PCR of A2M RNA from the brains of patients homozygous for the deletion polymorphism showed that the bait domain exon still is present in the RNA. This suggests that the A2M deletion polymorphism may be nonfunctional and that the chromosome 12 AD locus is situated elsewhere.

Risk for Alzheimer's disease in older late-onset cases is associated with HLA-DRB1*03

The allele frequency of the HLA-DRB1 gene was compared between groups of 48 clinically diagnosed elderly Alzheimer's disease (AD) cases and 44 pathologically confirmed elderly control cases. Specific primers were used to PCR amplify the highly polymorphic second exon of HLA-DRB1 using DNA extracted from blood samples or frozen brain tissue. The allele type was identified using sequence specific oligonucleotide probes. The results showed an increased frequency of DRB1*03 (P < 0.006) and decreased frequency of DRB1*09 (P < 0.049) in the AD cases compared with the controls. The results suggest that DRB1*03 is associated with an increased risk and DRB1*09 a possible decreased risk for the development of late-onset AD with first detectable clinical symptoms occurring at age 75 years or greater.

Alpha2-macroglobulin polymorphisms in Alzheimer's disease and dementia with Lewy bodies

Dementia with Lewy bodies (DLB) is the second most common cause of dementia in the elderly after Alzheimer's disease (AD). The apolipoprotein E gene (APOE) is a major risk factor, but can only account for approximately 50% of AD cases. Whole genome scanning in late-onset AD families has suggested that a locus on chromosome 12 may contribute significantly to disease development. Recently the alpha2-macroglobulin gene (A2M) on chromosome 12 has been suggested as a candidate locus for AD. We therefore determined the influence of two polymorphisms in A2M, a pentanucleotide deletion 5' to the bait domain exon, and a valine to isoleucine polymorphism in the thiolester site of the protein, in AD and DLB cohorts. No evidence was observed for an association between the thiolester or deletion polymorphisms and AD or DLB alone or when accounting for the APOE epsilon4 allele. We did, however, identify a non-significant excess of deletion homozygotes in the AD and DLB groups. This genotype accounted for 4% of disease cases but was absent in the control population. Given that the A2M deletion polymorphism is non-functional, the chromosome 12 AD/DLB locus may be situated elsewhere and not with these A2M polymorphisms.

Frequency of HLA-A and B alleles in early and late-onset Alzheimer's disease

The frequency of various allele types of the class I Major Histocompatibility Complex (MHC) genes HLA-A and HLA-B were compared between pathologically confirmed groups of late and early-onset Alzheimer's disease (AD) and a control group. DNA was extracted from frozen brain tissue and the highly polymorphic second and third exons of the HLA-A and HLA-B genes were independently PCR amplified using specific primers. Individual allele types were identified using sequence-specific oligonucleotide probes. The results showed that the main frequency differences occurred between the late-onset AD and the control group however none of these reached statistical significance.

CYP2D6 is associated with Parkinson's disease but not with dementia with Lewy Bodies or Alzheimer's disease

The similarities between the clinical and pathological findings of dementia with Lewy Bodies (DLB) with Alzheimer's disease and Parkinson's disease are complex, and their significance for pathogenesis is unresolved. It is likely that DLB shares common disease determinants with both Alzheimer's disease and Parkinson's disease. Clinically DLB shows the presence of dementia similar, though not identical, to that found in Alzheimer's disease. A parkinsonian movement disorder is present in a proportion of DLB cases. Pathologically DLB shows senile plaques, as with Alzheimer's disease, and also substantia nigra neurone loss and Lewy bodies, as with Parkinson's disease. At a genetic level, DLB shows an elevated Apolipoprotein E epsilon4 frequency as described in Alzheimer's disease, but this is absent in Parkinson's disease. An elevated frequency of the CYP2D6*4 allele has been found in Parkinson's disease and we have therefore genotyped a large series of clinically and neuropathologically confirmed cases of DLB, Alzheimer's disease, Parkinson's disease and age-matched control individuals for the CYP2D6*4 allele. Whilst an elevated frequency of the CYP2D6*4 allele was found in Parkinson's disease, no such elevations were found in DLB or Alzheimer's disease. Stratification of the CYP2D6*4 allele with respect to the Apolipoprotein E epsilon4 also did not show any significant associations with the CYP2D6*4 allele. The CYP2D6*4 allele is not a major genetic determinant of DLB and the results place DLB with Alzheimer's disease rather than Parkinson's disease on a genetic level.

No association between the K variant of the butyrylcholinesterase gene and pathologically confirmed Alzheimer's disease

The polymorphic K variant of the butyrylcholinesterase ( BCHE-K ) gene recently has been demonstrated to have an elevated frequency in Alzheimer's disease (AD) patients carrying the epsilon4 allele of the apolipoprotein (APO E) gene when compared with a control population. We therefore genotyped a large series of pathologically confirmed AD patients and controls to confirm this association. We found no change in the frequency of this genetic variant, either in the AD group as a whole or in early- or late-onset patients when compared with age-matched controls. Stratification of these groups with reference to the APO E epsilon4 allele also showed no difference between AD and control groups. To determine if a biological effect were present, we also looked at senile plaque and neurofibrillary tangle densities in the frontal, temporal, parietal and occipital cortices in AD patients either carrying or not carrying a copy of the K variant. We found no difference in plaque or tangle load between these two groups in either the total, late-onset or early-onset AD subjects. Stratification of the total AD group in terms of APO E epsilon4 allele possession, and further comparison of plaque and tangle load between carriers and non-carriers of BCHE-K still failed to disclose a relationship between BCHE-K and AD. We conclude that in the population studied here there is no association between BCHE-K and AD, or that if such a relationship exists it is precluded by another, as yet unknown factor.

Gastrointestinal absorption of aluminium and citrate in man

Aluminium (Al) is an abundant terrestrial element, but toxic to tissues, including brain. The body is largely protected because systemic Al absorption is very low and in normal individuals almost all absorbed Al is excreted from the body. However gastrointestinal (Gl) absorption is enhanced by organic acids, including citrate. Aluminium and citrate Gl absorption was measured in three healthy males, aged 40-46. After overnight fast, subjects drank a 100 ml fruit drink containing 280 mg Al and 3.2 g citrate (104 and 167 mM, respectively). Al was measured in timed blood and urine samples by GFAAS and serum citrate by enzymatic assay. Blood Al peaked by an increase of 13 +/- 2.1 micrograms/l after 87 +/- 19 min then fell slowly over 24 h. Plasma citrate peaked after 32 min, returning to baseline by 90 min. Al was excreted at a constant rate for the first 24 h, 0.4% of the dose being excreted in urine by this time. It is unlikely that Al is absorbed as Al citrate because the blood citrate peak preceded the Al peak by 45-60 min.

Apolipoprotein E and alpha-1 antichymotrypsin polymorphism genotyping in Alzheimer's disease and in dementia with Lewy bodies. Distinctions between diseases

The possibility of gene interactions in Alzheimer's disease (AD) has been suggested by the finding of an association of the AA genotype of the alpha-1 antichymotrypsin (AACT) gene and the apolipoprotein E (apoE) epsilon 4/4 genotype in AD. We tested this possibility by genotyping a large series of clinically and neuropathologically confirmed cases of AD and a series of cases with dementia with Lewy bodies (DLB) with a matched control group for the AACT locus and apoE. ApoE genotyping showed the established finding of an increased frequency of the apoE epsilon 4 allele in AD and in DLB. The AD and DLB groups differed between each other with a higher epsilon 2 allele frequency and a reduced incidence of the epsilon 4/4 genotype in DLB. Differences in the apoE frequencies may account for some of the differences between the two diseases. No association was found for the AACT A allele in AD or DLB in the groups as a whole or when stratified with respect to apoE, with the exception of a trend showing an increased incidence of the apoE epsilon 4/4 AACT AA genotype combination in AD patients (chi 2 = 3.18, p = 0.07), although in DLB this was not apparent (chi 2 = 0.0, p = 1.0). The AACT A allele is not a major risk factor for late-onset AD or DLB.

Gastrointestinal absorption of aluminum is increased in Down's syndrome

Individuals with Down's Syndrome (DS) develop the neuropathological features of senile dementia of the Alzheimer's type (SDAT) by early middle age. Because of recent evidence that gastrointestinal (GI) aluminum (Al) absorption is increased in patients with SDAT, and that Al may contribute to associated neuropathological changes, we have investigated the GI uptake of Al in patients with DS by two methods. The first measured the absorption of 27Al at concentrations associated with antacid use, in the presence of citrate, using atomic absorption spectrometry. There was no difference between basal blood concentrations of 27Al in 15 DS subjects (36-46 years) and 15 age-matched controls. The mean increase in 27Al blood concentrations 60 minutes after the dose of Al was four times greater in the DS group than in controls (p < 0.001). The second measured GI absorption of 26Al under normal dietary conditions using accelerator mass spectrometry. With 26Al the mean Al absorption in DS subjects (n = 5) exceeded that of controls (n = 4) by a factor of 6 (p < 0.02). Although the mechanisms of enhanced absorption are unknown, the data indicate that similar abnormalities in the GI handling of Al occur in both SDAT and DS suggesting that it may be advisable to minimize dietary exposure to Al in subjects at risk of developing Alzheimer-type pathology.

Effect of aluminium on expression and processing of amyloid precursor protein

The environmental agent aluminium has been extensively investigated for a potential role in the aetiology of Alzheimer's disease. Despite many investigations there is at present no definite proof for any involvement. If aluminium is involved it is possible that its action is mediated through interaction with the synthesis or processing of amyloid precursor protein (APP). The present study compared aluminium loaded IMR-32 neuroblastoma cells and rat brains with control cells and brains to determine if aluminium affected APP expression and/or processing. In the IMR-32 model system aluminium had no effect on steady-state APP mRNA levels or on the ratio of individual isoforms. It also had no quantitative or qualitative effect on APP-immunoreactive bands detected in protein extracts from conditioned medium of these cells. In total cell extracts, aluminium reduced the intensity of APP-immunoreactive bands between 120-105 kDa but had no effect on a 9 kDa band. In rat brains, aluminium had no effect on APP-immunoreactive bands from soluble or insoluble-membranous extracts. The results, in general, provide no evidence for any effect of aluminium on APP expression or processing.

Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): report of the consortium on DLB international workshop

Recent neuropathologic autopsy studies found that 15 to 25% of elderly demented patients have Lewy bodies (LB) in their brainstem and cortex, and in hospital series this may constitute the most common pathologic subgroup after pure Alzheimer's disease (AD). The Consortium on Dementia with Lewy bodies met to establish consensus guidelines for the clinical diagnosis of dementia with Lewy bodies (DLB) and to establish a common framework for the assessment and characterization of pathologic lesions at autopsy. The importance of accurate antemortem diagnosis of DLB includes a characteristic and often rapidly progressive clinical syndrome, a need for particular caution with neuroleptic medication, and the possibility that DLB patients may be particularly responsive to cholinesterase inhibitors. We identified progressive disabling mental impairment progressing to dementia as the central feature of DLB. Attentional impairments and disproportionate problem solving and visuospatial difficulties are often early and prominent. Fluctuation in cognitive function, persistent well-formed visual hallucinations, and spontaneous motor features of parkinsonism are core features with diagnostic significance in discriminating DLB from AD and other dementias. Appropriate clinical methods for eliciting these key symptoms are described. Brainstem or cortical LB are the only features considered essential for a pathologic diagnosis of DLB, although Lewy-related neurites, Alzheimer pathology, and spongiform change may also be seen. We identified optimal staining methods for each of these and devised a protocol for the evaluation of cortical LB frequency based on a brain sampling procedure consistent with CERAD. This allows cases to be classified into brainstem predominant, limbic (transitional), and neocortical subtypes, using a simple scoring system based on the relative distribution of semiquantitative LB counts. Alzheimer pathology is also frequently present in DLB, usually as diffuse or neuritic plaques, neocortical neurofibrillary tangles being much less common. The precise nosological relationship between DLB and AD remains uncertain, as does that between DLB and patients with Parkinson's disease who subsequently develop neuropsychiatric features. Finally, we recommend procedures for the selective sampling and storage of frozen tissue for a variety of neurochemical assays, which together with developments in molecular genetics, should assist future refinements of diagnosis and classification.

Apolipoprotein E genotype and Alzheimer's disease in an elderly Norwegian cohort

Apolipoprotein E (Apo E) genotyping was performed on an autopsy cohort of neuropathologically verified non-demented controls and subjects with Alzheimer's disease (AD) resident in nursing homes in the Oslo area. AD was associated with a significantly increased frequency of the Apo E epsilon 4 allele; the frequency of the epsilon 2 and epsilon 3 alleles was lower in AD but not significantly so. Age at death in the control group and the AD group did not differ significantly; neither did age at death nor age at onset of dementia in AD vary according to Apo E genotype, though tendencies towards an earlier age at death was seen in individuals with epsilon 4/4 and earlier age at onset dementia in the presence of an epsilon 4 allele and a later age of onset the presence of an epsilon 3 allele were seen. Possession of an epsilon 2 allele had no effect on age at onset of dementia or age at death. Among the possible genotypes there was a trend towards a progression of earliest onset epsilon 4/4, epsilon 2/4, epsilon 3/4, epsilon 3/3, epsilon 2/3 latest onset of dementia and longest duration epsilon 2/4, epsilon 4/4, epsilon 3/4, epsilon 3/3, epsilon 2/3 to shortest duration of dementia.

Content of brain aluminum is not elevated in Alzheimer disease

Several studies have reported that the bulk aluminum (Al) concentration is increased in the brain in Alzheimer disease (AD), while other studies have failed to demonstrate an increase. Most of these investigations have had one or more methodological deficiencies, including lack of adequate neuropathological assessment; failure to age-match the control samples; small sample sizes, lacking statistical power; and geographical heterogeneity in the AD and control populations. The present population-based study of 92 clinically and histopathologically diagnosed AD patients and normal elderly nursing home residents was designed to avoid these potential biases. When a subsample of AD cases with the most severe brain pathology was compared with controls having no or minimal pathology, no statistically significant differences were found in the bulk aluminum concentration measured by graphite furnace atomic absorption spectrometry in frontal cortex (1.8 +/- 0.7 vs. 1.7 +/- 0.7 micrograms/g dry wt), temporal cortex (1.4 +/- 0.3 vs. 1.5 +/- 0.5 micrograms/g dry wt), liver (2.0 +/- 1.3 vs. 2.0 +/- 1.2 micrograms/g dry wt), or head of femur (2.4 +/- 1.6 vs. 2.2 +/- 1.0 micrograms/g ash wt). Within the whole series of 92 cases, there was no difference in the bulk aluminum concentration of the frontal cortex between individuals diagnosed as definite, probable, and possible cases of AD using the CERAD (Consortium to Establish a Registry for Alzheimer's Disease) criteria. The density of senile plaques and neurofibrillary tangles in frontal and temporal cortex showed no correlation with the bulk aluminum concentration. Logistic regression analyses, which controlled for age and sex, did not influence outcome for any of the comparisons. The data show conclusively that in AD, bulk aluminum concentration is not increased in two cortical brain regions that are selectively vulnerable to the neuropathological changes associated with this disorder.

Molecular biology of APO E alleles in Alzheimer's and non-Alzheimer's dementias

Current research into the aetiology of the dementias is focused upon genetic factors which give rise to the disease process. Recently the Apolipoprotein E gene (APO E) and in particular the epsilon 4 allele has been shown to be a risk factor for late onset Alzheimer's disease (AD) where there is an increased frequency of the epsilon 4 allele. The epsilon 4 allele has also been shown to reduce the age at onset of dementia in AD in a dose dependent manner, with the epsilon 2 allele having an opposing effect. We have genotyped a large series of clinically and neuropathologically confirmed cases of AD and found the expected increase in the Apolipoprotein epsilon 4 allele frequency when compared to a control population. Similarly, in Lewy Body Dementia (LBD) an increased epsilon 4 frequency is also found though a normal epsilon 2 frequency exists, unlike in AD where the epsilon 2 frequency is reduced. No changes in APO E allele frequencies were found in presenile AD, Parkinson's disease with or without dementia, or in Down's syndrome. No association was found between any of the APO E alleles and the histopathological indices of AD, cortical senile plaques and neurofibrillary tangles, in any disease category. Neurochemical indicators of AD, loss of choline acetyltransferase activity was also unaffected by APO E genotype. Whilst their appears to be a strong association between the APO E allele and AD and also in LBD, other related neurodegenerative disorders associated with dementia do not show such a linkage. Changes in the epsilon 2 allele frequency may indicate a genetic difference between AD and LBD. The epsilon 4 allele does not appear to influence the burden of AD type pathology and this is particularly relevant given the relative lack of NFT in LBD indicating that factors other than SP or NFT may govern the onset of dementia.

Effects of apolipoprotein E genotype on cortical neuropathology in senile dementia of the Lewy body and Alzheimer's disease

Apolipoprotein E (APO E) genotypes were determined in a UK population of neuropathologically confirmed control cases, and in cases of Lewy body dementia (SDLT) and late onset Alzheimer's disease (AD). APO E epsilon 4 allele frequency was significantly elevated in both SDLT and AD groups with a concomitant reduction in the APO E epsilon 3 allele frequency. The epsilon 2 allele frequency in the AD group was only 25% of the control population, though because of the relatively small sample size this reduction was not significant; the epsilon 2 allele frequency in the SDLT group was normal. No significant association was found between senile plaque density and neurofibrillary tangle density in the neocortex and APO E allele dose in either SDLT or AD. Although the possession of APO E epsilon 4 is associated with an increased risk of developing SDLT and AD, actual APO E genotype does not appear to affect the burden of pathology.

Effect of apolipoprotein E genotype on Alzheimer's disease neuropathology in a cohort of elderly Norwegians

A cohort of elderly Norwegians dying in nursing homes in the Oslo region have been genotyped for the Apolipoprotein E (ApoE) gene. Alzheimer's disease (AD) cortical neuropathology and clinical evidence of dementia were used to assign cases without evidence of other confounding neuropathology. Senile plaque (SP) and neurofibrillary tangle (NFT) densities in frontal, temporal and parietal cortex were then correlated with ApoE genotype to determine any relationship between ApoE genotype and AD pathology. Comparisons with ApoE epsilon 3, epsilon 4 and epsilon 2 allele dosage failed to show any significant effect on cortical SP densities in any cortical area. NFT densities were increased by epsilon 4 allele dosage in the frontal cortex but not in other cortical regions. A reduction was seen in cortical NFT densities with epsilon 2 allele, though again this was not consistently significant in any of the groups. The epsilon 3 allele failed to show any consistent effect on cortical NFT densities. Assessment by individual genotypes showed epsilon 2/3 < epsilon 2/4 < epsilon 3/3 < epsilon 3/4 < epsilon 4/4 which had highest cortical NFT densities in all areas. By genotype, SP densities were generally of the order epsilon 2/4 < epsilon 2/3 < epsilon 3/3 < epsilon 4/4 < epsilon 3/4 though in none of the groups was this significant. Duration of disease showed no consistent effect on neuropathological burden. ApoE genotype may have an effect on determining whether individuals suffer from AD and the age at onset of disease but may only have a minimal effect on pathology burden.

Risk factors in clinically diagnosed presenile dementia of the Alzheimer type: a case-control study in northern England

STUDY OBJECTIVE

To investigate the relationship between presenile dementia of the Alzheimer type (PDAT) and family history, medical history, cigarette smoking, and exposure to aluminum.

DESIGN

A case-control study in which 109 cases of clinically diagnosed PDAT and 109 controls matched for age and sex were compared for exposure to the risk factors. Odds ratios (ORs) were calculated using McNemar's test.

SETTING

The northern health region of England.

PATIENTS

Cases comprised those under 65 years diagnosed as having dementia by specialist services, who met clinical algorithm criteria for Alzheimer's disease (AD). Cases were confirmed at interview.

MAIN RESULTS

Comparing cases with controls, (ORs) significantly greater than unity were obtained when there was a first degree relative with dementia (OR 2.5, 95% confidence interval 1.05, 6.56), any relative with dementia (OR 2.1, 95% CI 1.01, 4.55), and any relative aged less than 65 with dementia (OR 8.0, 95% CI 1.07, 348). Exposure to moderate levels of cigarette smoking (cumulative) was not significant; nor was exposure to aluminum in drinking water, diet, and medicinal sources.

CONCLUSION

In this study of modest statistical power, a family history of dementia was confirmed as a risk factor in PDAT. No significant relationship between exposure to aluminium in water supplies, tea, and antacids was found. What is important, however, is the bioavailability of all dietary aluminium, determined by the concentrations of dissolved silicon in water: this requires further investigation.

Neuropathological diagnoses in elderly patients in Oslo: Alzheimer's disease, Lewy body disease, vascular lesions

Neuropathological changes in elderly residents of Oslo, Norway were characterised with respect to the cerebral substrates of dementia. Ninety-two brains were examined, representing 41% of all deaths occurring in 10 nursing homes during a 9-month period. The autopsy cohort showed a similar mean age (85 years) and sex ratio (73% female) and proportion of demented patients (75%) compared to all the patients resident in these homes who died during the same period. Clinical data was compiled retrospectively. Diagnosis was made using the CERAD protocol, and criteria for the diagnosis of Lewy body dementia. Lewy body formation was present in 20% and cerebral infarction in 21% of patients. In the demented group (69 patients) 90% fulfilled CERAD criteria for definite or probable Alzheimer's disease. Eight demented cases had absent neocortical neurofibrillary tangles and 6 other cases showed Lewy body dementia (9% of demented patients). A further 8 of these demented cases had brain stem Lewy bodies with only minimal cortical involvement. Thirteen cases (19% of the sample) had cerebral infarcts but these were considered to be clinically significant in only 4 (6%). In the non-demented patients (23) 4 patients had brain stem Lewy bodies and 6 had cerebral infarcts. Despite inclusion criteria biased towards the collection of Alzheimer's disease and normal patients, both Lewy body dementia (7%) and cerebral infarcts contributing to dementia (6%) were frequent.

Non-haem iron histochemistry of the normal and Alzheimer's disease hippocampus

Increased free radical production may occur in Alzheimer's disease (AD). In view of the central role of iron in free radical reactions we have investigated the distribution of non-haem iron in the hippocampal formation in normal control individuals and in patients with AD. In controls, non-haem iron was principally associated with glial elements and the neuropil, with highest levels in the stratum radiatum/lacunosum, fimbria, alveus and oriens layers. Except for the subiculum, the pyramidal cell layers and the granule cell layer showed little or no non-haem iron staining. Intensity of staining was in the order of subiculum >> CA2 and adjacent regions of CA3 > CA4 > the remainder of CA3 > CA1. In the hippocampus in AD, iron was associated with cells containing neurofibrillary tangles (NFT) and was present in glial cells and neurites of senile plaques (SP). These findings were most pronounced in CA1 pyramidal cell layer and subicular complex although not all NFT or SP were stained. Since the NFT and SP staining for non-haem iron appears to be associated with glial cells, the association of non-haem iron with the pathological stigmata of AD may be a secondary response of glial cells, in particular microglia, to neuronal damage.

Iron histochemistry of the substantia nigra in Parkinson's disease

Raised tissue iron levels in the substantia nigra in Parkinson's disease (PD) suggests that altered iron homeostasis may underly the disorder. We have therefore investigated the distribution of non-haem iron in the normal and PD substantia nigra, using a sensitive histochemical procedure, to assess the pathogenic potential of this metal. In control cases non-haem iron staining was highest in the substantia nigra zona reticulata (SNr) and associated with the neuropil, oligodendrocytes, astrocytes and non-pigmented neurones. The substantia nigra zona compacta (SNc) showed lower non-haem iron staining than the SNr, with generalized impregnation of the neuropil and occasional non-haem iron-positive oligodendrocytes and astrocytes. The pigmented dopaminergic neurones were unstained, often present in areas of neuropil with low iron reactivity. In PD the SNc showed increased iron staining of the neuropil with many iron-positive microglial cells associated with extracellular melanin. The remaining dopaminergic neurones were unstained, though many of the non-pigmented neurones of the SNr were iron-positive.

Human IMR-32 neuroblastoma cells as a model cell line in Alzheimer's disease research

The present study investigated expression and processing of amyloid precursor protein by neuronally differentiated IMR-32 neuroblastoma cells. APP mRNA in these cells was found to consist of approximately 58% APP695, 38% APP751, and < 4% APP770. APP-immunoreactive bands detected in western blots of cellular protein extracts were only detected by anti-APP antibodies to peptides with strong homology to APLP2, suggesting that these bands represent APP-like proteins and not APP itself. This result suggests that previous studies claiming immunodetection of cellular forms of APP may have to be re-evaluated. Four main species of C-terminal truncated, secreted APP were detected in blots of protein extracts from medium conditioned by these cells. The immunoreactive profile of these bands suggested a cleavage site N-terminal to the Lys16-Leu17 bond of alpha-secretase. This, together with differences in number and molecular mass of APP-immunoreactive bands between secreted APP from IMR-32 cells and that from the commonly used PC-12 cells, suggests differences in APP processing between these two neuronally differentiated cell lines. In theory, IMR-32 cells being of human neuronal origin may be a more appropriate cell line to study APP-processing in relation to Alzheimer's disease than the rat phaeochromocytoma PC-12 cell line. Therefore, these detected differences warrant further investigation. Additionally IMR-32 cells under certain tissue culture conditions can form intracellular fibrillary material that reacts with anti-PHF specific antibodies. Neuronally differentiated IMR-32 cells could therefore be used as a model system to investigate possible interactions between APP-processing and PHF formation.

Transferrin receptors in the normal human hippocampus and in Alzheimer's disease

Several lines of evidence suggest that aluminium may play a role in the pathogenesis of Alzheimer's disease (AD). The iron transport protein transferrin is the major transport protein for aluminium, and aluminium gains access to cells by means of a specific cell surface transferrin receptor. We have assessed the distribution of transferrin receptors in the normal and AD hippocampal formation using [3H]-transferrin ([3H]-Tf) binding and tritium film autoradiography, in order to assess the role of the transferrin receptor in AD. In normal brain, [3H]-Tf binding was highest in the pyramidal cell layers with CA2 > dentate gyrus granule cell layer > or = CA1 > CA3 > or = CA4 > subiculum > parahippocampal gyrus. In AD, significant reductions in [3H]-Tf binding were found in CA1, CA2 and CA4 pyramidal cell layers. The reduced [3H]-Tf binding in AD may, however, be due to poor pre-mortem agonal states which correlated with reduced [3H]-Tf binding. The discrepancy between the distribution of transferrin receptors in the hippocampus and those areas which are prone to the formation of senile plaques and neurofibrillary tangles suggests that if transferrin-mediated uptake of aluminium in AD/SDAT is significant in the pathogenesis of this disorder, it is not the only determinant of Alzheimer-type neuropathology.

Transferrin receptors in the parkinsonian midbrain

Several hypotheses have been put forward to explain the pathogenesis of Parkinson's disease (PD) and recently it has been suggested that alterations in iron homeostasis may be implicated. Because of the central role of the transferrin receptor in providing access of iron to cells, we have studied the distribution and density of transferrin receptors using [3H]-transferrin ([3H]-Tf) binding and tritium film autoradiography in the normal and PD midbrain. High levels of [3H]-Tf binding were found in the dorsal raphé, oculomotor nucleus and periaqueductal grey whilst lower levels of [3H]-Tf binding were found in the tegmentum, red nucleus and substantia nigra. Significant reductions in binding were found in the substantia nigra, red nucleus and oculomotor nucleus in PD, the reductions in [3H]-Tf binding being similar to the loss of nigral neurons in PD. The data suggest that the increased iron content of surviving nigral neurons may reflect a compensatory metabolic response rather than abnormal transferrin receptor expression.

Alzheimer's-disease-like changes in tau protein processing: association with aluminium accumulation in brains of renal dialysis patients

Tau protein is a major structural protein of the paired helical filaments (PHFs) found in both neuritic senile plaques and neurofibrillary tangles in Alzheimer's disease (AD). Senile plaques also contain amyloid beta protein (A beta). We did an immunochemical analysis of frontal cortex from 15 dialysis cases, 5 Alzheimer's disease patients, and 6 control cases to see whether AD-like changes in A beta deposition and tau protein were linked to aluminium accumulation. Dialysis patients were used because they are frequently exposed to increased levels of aluminium. 8 of the 15 dialysis cases had insoluble A beta, but there was no association between its presence and the accumulation of aluminium. However, we found AD-like changes in the processing of tau protein. In white matter, truncated tau protein in the PHF-core fraction and endogenously truncated tau in the supernatant fraction were both increased in association with aluminium accumulation in the brain. In grey matter, normal tau protein was depleted and insoluble hyperphosphorylated tau increased in association with aluminium concentration. Protease-resistant PHFs were present in grey matter in 2 dialysis cases, a frequency above that expected for AD in this age group. PHF-core tau in both grey and white matter correlated with decreased levels of normal tau protein in white matter. These findings are consistent with a role for aluminium in the development of AD-like pathology in patients subjected to prolonged aluminium exposure.

Effect of silicon on gastrointestinal absorption of aluminium

The reported geographical association between Alzheimer's disease and levels of aluminium (Al) in water supplies may reflect the inverse relation between Al and silicon (Si) concentrations in water, and the potential for Si to reduce the bioavailability of the metal. We tested this hypothesis using isotopic 26Al tracer administered orally to five healthy volunteers in the presence and absence of Si. Dissolved Si, at a concentration found in some water supplies (100 mumol/L), reduced the peak plasma 26Al concentration to 15% of the value obtained in the absence of Si. The results indicate that dissolved Si is an important factor in limiting the absorption of dietary Al.

Autoradiographic comparison of cholinergic and other transmitter receptors in the normal human hippocampus

The vulnerability of the human hippocampal complex to disease, trauma, and aging indicates the necessity to target this area therapeutically. The distribution and density of transmitter receptors provide a rational basis for this approach, and in this study the topography of 11 different pharmacological sites is compared with the cholinergic innervation, which is particularly vulnerable in dementia. The regional distribution of cholinergic innervation to the normal adult human hippocampus and adjacent cortex, marked by acetylcholinesterase (AChE) fiber and terminal reactivity, is notable for its concentration in CA2/3 of Ammon's horn and the dentate fascia. Neither nicotinic (high-affinity nicotine binding) nor muscarinic ("M1" or "M2") cholinergic receptor binding paralleled this distribution. In Ammon's horn, 5-HT2 and kainate receptor binding more closely resembled the pattern of AChE, being concentrated in CA2-4 compared with CA1. By contrast, muscarinic M1 and M2, 5-HT1A, benzodiazepine (including zolpidem-insensitive binding), NMDA (MK801), and AMPA/QUIS receptors were higher in CA1 and/or subiculum. Kainate binding, like AChE, was high in CA4. 5-HT2 and nicotinic binding partially mimicked the pattern of AChE around the granule layer. In the subicular complex and parahippocampal gyrus, where cholinergic activity is relatively lower, muscarinic, 5-HT1A, and benzodiazepine binding were relatively high and the nicotinic receptor was remarkable for its highest density compared to other areas examined. In stratum lacunosum-moleculare of CA1, which was relatively low in AChE activity, there was a dense band of nicotinic, M2, and benzodiazepine receptor binding.(ABSTRACT TRUNCATED AT 250 WORDS)

Evidence for the localization of haemopexin immunoreactivity in neurones in the human brain

Haemopexin is a 60 kDa serum glycoprotein responsible for the transport of haem to tissues such as liver, by receptor-mediated endocytosis, in an analogous manner to the iron transport protein transferrin, with recycling of intact haemopexin. The immunocytochemical distribution of haemopexin has been investigated, using a monospecific polyclonal antiserum to human haemopexin, in human brain. Neurones in all the brain regions studied showed immunostaining of the soma, axons and dendrites. A few scattered glial cells exhibited positive immunostaining. Oligodendrocytes and choroid plexus epithelial cells lacked haemopexin immunoreactivity. Thus, haemopexin is present within neurones and we propose that this protein may play an important role in haem transport for neuronal iron homeostasis.

Germanium-68 as a possible marker for silicon transport in rat brain

Silicon (Si) is an essential trace element normally present in brain and cerebrospinal fluid, although the mechanism by which it enters and distributes in brain is largely unknown. Due to the short radioactive half-life of 31Si (156 min) we have investigated the use of 68Germanium (68-Ge, half-life 282 days) as a possible marker for Si transport in rat brain over longer periods than are possible with 31Si. Adult male anaesthetised rats were given a bolus of 68Ge I.V. and arterial blood samples taken during experiments that lasted between 5 min and 3 days. At termination, the brain was removed and analysed for radioactivity as were the plasma samples. Data were analyzed by Graphical Analysis which showed that the blood-brain barrier permeability to 68Ge (Kin approximately 10(-4) ml/min/g) is similar to that for many non protein-bound electrolytes in plasma and that 68Ge fluxes across cerebral capillaries are bidirectional. The autoradiographic distribution of 68Ge in brain was homogenous. Our results are in agreement with those of previous studies using 31Si or 68Ge, which suggest 68Ge may be a useful marker for Si when investigating the role of this element in conditions such as neurodegenerative diseases.

Iron and aluminium in relation to brain ferritin in normal individuals and Alzheimer's-disease and chronic renal-dialysis patients

Ferritin has been isolated and its subunit composition, iron and aluminium content determined in the cerebral cortex and cerebellum of normal individuals and in the cerebral cortex of Alzheimer's-disease and renal-dialysis patients. An e.l.i.s.a. for ferritin has been developed and the ferritin, non-haem iron and aluminium content of the parietal cortex were determined in normal individuals and Alzheimer's-disease patients. It was found that ferritin from the cerebral cortex and cerebellum of normal individuals had a high H-subunit content, similar to that of heart ferritin. The subunit composition of ferritin isolated from the cerebral cortex was not significantly altered in Alzheimer's-disease or renal-dialysis patients. Ferritin from the cerebral cortex of normal individuals had only approx. 1500 atoms of iron per molecule and the iron content of ferritin was not significantly changed in Alzheimer's-disease or renal-dialysis patients. Ferritin isolated from the cerebral cortex of normal, Alzheimer's-disease and renal-dialysis patients had less than 9 atoms of aluminium per molecule. The failure to find increased concentrations of aluminium associated with ferritin in dialysis patients, who had markedly increased concentrations of aluminium in the cerebral cortex, shows that aluminium does not accumulate in ferritin in vivo. This has important implications for the toxicity of aluminium, since it implies that cells are unable to detoxify aluminium by the same mechanism as that available for iron. Comparison of the concentrations of ferritin, aluminium and iron in the parietal cortex from normal and Alzheimer's-disease patients showed that, whereas the concentration of aluminium was not increased, both ferritin and iron were significantly increased in Alzheimer's disease.

Matching-to-sample deficits in patients with senile dementias of the Alzheimer and Lewy body types

Using a computerized matching-to-sample task, nonverbal visual recognition memory was studied in two groups of patients suffering from senile dementia of the Alzheimer type or the recently described senile dementia of the Lewy body type. The patients' cognitive abilities had been shown to be similar according to a number of standard psychometric tests. The two groups did not differ with respect to simultaneous matching-to-sample performance, although both were impaired relative to control. The group with senile dementia of the Lewy body type was severely impaired, relative to the group with senile dementia of the Alzheimer type, when delays (delayed matching to sample) were introduced. The findings suggest that short-term mnemonic processes, mediated by temporal lobe structures, could be more severely affected in senile dementia of the Lewy body type.

Distribution of transferrin receptors in relation to cytochrome oxidase activity in the human spinal cord, lower brainstem and cerebellum

Neuronal activity and oxidative energy metabolism are tightly coupled. There is evidence that cytochrome oxidase, the terminal enzyme of the electron transport chain, can serve as a metabolic marker of neuronal activity. All the respiratory chain enzymes have iron containing prosthetic groups and therefore represent an important component of iron utilisation. Since iron entry into cells is mediated by the transferrin receptor, this receptor may also serve as marker of neuronal activity. The histochemical distribution of cytochrome oxidase has therefore been compared with the autoradiographic distribution of the transferrin receptor in the human spinal cord, brainstem and cerebellum. Cytochrome oxidase activity showed a very similar pattern of distribution to the transferrin receptor in the spinal cord, brainstem and cerebellum. The highest levels of cytochrome oxidase activity and transferrin receptor binding were associated with; in the spinal cord, the substantia gelatinosa, laminae II and III and the motor neurones; in the medulla and pons, the spinal trigeminal nucleus, hypoglossal nucleus, dorsal motor nucleus of the vagus, inferior and superior olives, nucleus praepositus, nucleus paramedianus, central grey, superior central nuclei and locus coeruleus; in the cerebellum, the molecular layer. The results suggest that the transferrin receptor may provide a useful marker of total neuronal respiratory activity.