From the Archives
Cambridge
(1) Neurotransmitter-related enzymes and indices of hypoxia in senile dementia and other abiotrophies. By David M. Bowen, Carolyn B. Smith, Pamela White and Alan N. Davison. (From the Miriam Marks Department of Neurochemistry, Institute of Neurology, The National Hospital, Queen Square, London WC1N 3BG). Brain 1976: 99; 459–496. With (2) Chemical pathology of the organic dementias. I. Validity of biochemical measurements on human post-mortem brain specimens. By D. M. Bowen, Carolyn B. Smith, Pamela White, Michele J. Goodhart, J. A. Spillane, R. H. A. Flack and A. N. Davison (From the Miriam Marks Department of Neurochemistry, Institute of Neurology, The National Hospital, Queen Square, London WC1N 3 BG). Brain 1977: 100; 397–426. With (3) Chemical pathology of the organic dementias. II. Quantitative estimation of cellular changes in post-mortem brains. By D. M. Bowen, C. B. Smith, P. White, R. H. A. Flack, L. H. Carrasco, J. L. Gedye and A. N. Davison (From the Miriam Marks Department of Neurochemistry, Institute of Neurology, The National Hospital, Queen Square, London WC1N 3 BG, Department of Pathology, Runwell Hospital, Wickford, Essex, and the Department of Electrical Engineering Science, University of Essex, Colchester, Essex CO4 3SQ). Brain 1977: 100; 427–453.
Neurochemical analysis of the human brain was a relatively new discipline in the 1970s, and one that needed to resolve the issues of how best to preserve and store tissue in a state whereby the signatures of disease, free from the artefacts of decay occurring in the early hours after death or sampling, could reliably be characterized. Further limitations were imposed by the paucity and non-specific nature of reagents available for marking individual components of the cellular and subcellular architecture of brain tissue. David Bowen et al. are concerned first with establishing that it is possible to say anything convincing about the neurochemistry of disease through the study of material obtained at post-mortem. Later, they consider the features of certain disease states. Their primary interests are in hypoxic injury and the dementias. As clinical scientists struggling to steer a course between the vicissitudes of designing informative research protocols and acquiring data that support novel conceptual formulations, their style is by necessity cautious. Twice they come close to nailing the cholinergic hypothesis of senile dementia; and twice they shy away from adopting an unambiguous position, sensing that the samples and reagents available to them, and the state of the tissues they study, prevent definitive conclusions. And yet these pioneering papers set standards for neurochemical analyses that describe brain structure in health and disease from an entirely new perspective to that adopted by morphologists ever since vital dyes and silver stains were introduced in the 19th century.
In the first of their papers from 1977, the authors aim to substantiate the hypothesis that narrowing of the cortical ribbon in senile dementia is due to neuronal loss. From a bank of 56 human brains obtained from various sources and frozen within 0.75 and 3 h of death, 25 represent normal or demented individuals in whom biochemical assessments of cellular and subcellular structures are supplemented by morphological details. First, the effect of secondary glial reactions, age, post-mortem handling and the agonal biochemical state must be established as a baseline. The focus is on the temporal lobes, since these are preferentially involved in dementia. And the method is to study whole lobe homogenates in order to reduce errors arising from oedema, water loss and sampling biases. Senile plaque formation, atrophy, neurofibrillary degeneration and neuronal loss are scored from 0 (none) to 4 (severe). Thirteen brains with no apparent morphological changes (scoring 0) form the ‘controls’, of whom five are considered historically to have ‘oxygen deprivation’ and the rest are normal. The 12 other samples are ‘cases’. So as further to increase the range of comparisons, tissue is studied from one example of subacute sclerosing panencephalitis; one baboon subjected to experimental ischaemia in life; four rats with Eck fistulae; and four normal rats. In order to distinguish the separate cellular components of homogenised temporal lobe, the biochemical measurements are categorized as: glutamate decarboxylase (nerve terminals); acetylcholine esterase (dendritic processes); succinate dehydrogenase and mitochondrial light fractionation protein (mitochondria); ganglioside NANA, ß-galactosidase, guanylcyclase, adenylcyclase and neuronin S-5 (14.3.2) and S-6 (neurons); ß-glucuronidase and carbonic anhydrase (macroglia); cathepsin A (microglia); galactolipid and cyclic nucleotide phosphohydrolyase (myelin); and total protein, RNA and DNA. The results are generally reassuring with respect to the artefacts of tissue handling and, hence, the validity of such measurements in ‘cases’. At least in rodents, these neurochemical signatures are all stable for 3 h but with differential decay thereafter. Other than glutamate decarboxylase, nothing much alters with age. However, succinate dehydrogenase is reduced in association with agonal ‘oxygen deprivation’; and experimental ischaemia in baboons also reduces glutamate decarboxylase and adenylcyclase activities. Therefore, these markers cannot reliably be used to infer the pre-terminal state or number of nerve endings, especially with respect to the basal ganglia by comparison with crests of the anterior and prefrontal gyri.
Against this background, the second paper from 1977 considers the ‘cases’ in more detail. Distinctions are made between senile and vascular dementia. Th, an 84 year old spinster died within 3 weeks of appearing naked outside her house, remaining confused, irrational and unable to care for herself before succumbing to bronchopneumonia; autopsy showed marked atherosclerosis. Ar had an 8 year history of cognitive deterioration with lack of personal care and behavioural changes. Wb, aged 91 years, was deluded and hallucinated for 4 years. Pa required institutional care for 4 years and died soon after a series of strokes. Cl, an ex-comédienne, suffered progressive loss of memory for 7 years before death from bronchopneumonia. Ca became demented over a period of 6 months before dying from pneumonia. Bd had cognitive failure complicating Parkinson's disease of 8 years duration. Each of these individuals is considered to have had vascular or senile dementia. Comparisons are made with single examples of Huntington's disease and multiple sclerosis, and with two coroner's cases about whom little is known although one, aged 97 years, seems to have been demented in life.
So what can be learned about dementia from this eclectic assortment of cases and controls? Strikingly, after pooling the vascular and senile cases, 6/8 of the neurone specific neurochemical markers are much reduced, whereas the indices of brain volume and glial content are normal, and only galactolipid galactose and the ratio of RNA:DNA are unchanged. The previously identified markers of hypoxia suggest that those with dementia had agonal oxygen deprivation. Unwisely, perhaps, the authors allow themselves to firm-up the diagnosis of senile dementia in one of the coroner's cases (Se) on the basis of reduced glutamate decarboxylase, thus allowing an experimental result retrospectively to adjust the morphological classification. Despite the evidence for loss of neuronal markers, estimates of cell content are not much reduced, despite no apparent macroglial overgrowth but perhaps due to not having estimated the number of microglia and over-interpreting the correlation between DNA content and cellularity. In a prescient subsidiary analysis, the authors seek ‘connectivity’ between neurochemical indices. Adopting the procedure of ‘zygology’ (from the Greek, 
ó
—a yoke) they identify a link between ß-galactosidase, ß-glucuronidase and DNA. Whether this reveals a coherent metabolic pathway is less clear but the concept of cluster analysis for individual components in a screening analysis, and the ingenious methods used to reveal and illustrate ‘zygology’ are ahead of their time (see Fig. 1).
|
Crucially, although comparisons between vascular and senile dementia are unrevealing, the latter are distinguished from controls by reduced ß-galactosidase, phosphohydrolase and acetylcholine esterase suggesting that 53% of neurones are lost from the temporal lobe in senile dementia, a rate of acceleration over and above normal aging of
5% per year (see Fig. 2). Although evidently not fully persuaded by the findings, the authors conclude with a discussion of their earlier finding (see below) of selective choline acetyltransferase loss affecting 2/3 of cortical specimens examined in cases of pre-senile and senile dementia. Now they conclude: ‘the activity is not reduced in cortex from all dement brains while in some brains the activity is decreased in regions that are usually less affected by morphological changes (e.g. caudate nucleus). It is unlikely that the change in the activity (of acetylcholine esterase) is the key explanation of the memory loss in dementia’.
|
, patient Se), mixed senile and vascular dementia (
) are shown in comparison with normal controls (
) and the controls with signs of ‘oxygen deprivation’ ( 
). The regression lines were calculated from the values for the normal controls. From Bowen et al. 1977 (2).But David Bowen et al. had already once come close to reaching a more emphatic position on the neurochemical signatures of senile dementia. In their 1976 paper, they had already assessed the feasibility of assaying neurotransmitter-related enzymes in human post-mortem brain tissues, and considered the artefacts to which these measurements might be exposed. They had, as an example, the successful implication of dopaminergic neurones in Parkinson's disease. Their hunch suggested a focus on glutamic acid decarboxylase in senile dementia. The bank of 56 brains included normal controls, cases of oxygen deprivation and several categories of dementia—together representing 14 groups. Group 10 consists of one case showing features of senile dementia with congophilic angiopathy (see pages 2966 and 2977—although this example from an individual aged 93 years is not exactly early onset Alzheimer’s disease). Seven biochemical markers are studied in seven brain regions. Results for only the prefrontal cortex and caudate (as representative sites) are reported. Carefully establishing the stability of measurements and the findings in normal brain and oxygen deprivation from a variety of causes, they turn to enzymes and proteins in the 12 cases of senile dementia. Grouped according to the associated causes of death, there is a reduction in glutamic acid decarboxylase and choline acetyltransferase (see Fig. 3), and in neuronin S-6. The distributions vary between these markers but with the most profound changes seen in the cerebral cortex. This result correlates with the degree of histological change in the brain tissue. Despite overlap between groups, and concern over the potential confound of hypoxia, the authors risk a conclusion: ‘one interpretation of these findings is that the decrease in CAT activity reflects a defect in the cholinergic system in senile dementia’ ... but ... ‘this is a preliminary finding, (since) although a correlation was indicated between CAT activity and senile morphological changes, the activity was markedly reduced in only 3 brains’. And the authors also remain unconvinced whether their marker of primary interest, glutaric acid decarboxylase, is specifically reduced in cortex or merely changes as a result of hypoxia.
|
Thus, having come close on two occasions to nailing cholinergic neurotransmitters as selectively involved in senile dementia, the authors stray from this target; and it was left for others to secure the cholinergic hypothesis to which so much therapeutic endeavour has since been devoted in Alzheimer's disease (P. J. Whitehouse et al. Alzheimer's disease and senile dementia: loss of neurons in the basal forebrain. Science 1982: 215; 1237–1239).
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||