Dear Dr Scott,

Numerous, almost "cacophonous" pathological factors are implicated in the pathogenesis of AD. Probably, the most important thing is series of acquired, "somatic" nuclear DNA mutations that produce various mutated, pathologically active and highly toxic proteins or prion-like peptides (e.g. the famous beta-amyloid peptide), etc. Your question is for a huge review written by some expert, who personally doesn't prefer non of several theories about AD pathogenesis or, in other words, non of numerous conditions and pathways that lead into the AD state of brain tissue and memory functions. For a start, I suggest you to read some of particularly valuable papers on the AD pathology, such as: (a) "Loss of non-phosphorylated neurofilament immunoreactivity in temporal cortical areas in Alzheimer's disease" - free full paper [Neuroscience, 2009 May 5; Vol.160(2), pp.427–433]; (b) "Intermediate Phenotypes Identify Divergent Pathways to Alzheimer’s Disease" - also free paper [PLoS One, June 2010, Volume 5, Issue 6, e11244]; and, (c) "Metabolic Role of Phosphoenolpyruvate Carboxykinase" (allegorically: The Strange Case of PEPCK-M) - free at PubMed central [J Biol Chem 2009 Oct2, Vol.284(40), pp.27025–27029; this paper is necessary to understand enormous importance of the paper under (b)].

Sincerely,

Dr B.D.Stefanovic

Histology Institute

Medical Faculty

Visegradska_26

11000 Belgrade

Serbia (SRB)

Thank you both. I don't think I've used pubmed, but I will check it out.. I am interested in what goes wrong in dementias such as AD, but as a stepping stone, I need a better understading of the mechanisms of normal memory storage/retrieval. I took an introductory neurobiology course as an elective while completing a master's program in biology nearly 20 years ago, so I have much to learn to get up to speed.

Ok, I hope you don't mind if I, as a non-academic member of the group, put my own oar in.

Memory starts with the selective sensory excitation by energy outside the body.

This activates chemical pathways within specialized cells called sensory cells, which in turn, change the chemical reaction into a signal that is transferred to other neurons via the cells axion.

Another type of cell, called a Neuron, is specialized for: Storage, Transport, and Processing of the signals it receives. There are currently more than 200 different types of neurons known to exist within the CNS or Central Nervous System. But most of the differences noted have to do with somatype (shape) and ion channel differences (sensitivity of the synapse to different chemicals).

The majority of the senses (at least in humans) are routed after a small amount of preprocessing, to the Thalamus, where a surface just outside the thalamus can act as a brake on signal propagation. (Nucleus Reticularis Thalami). The signals which seem sensor-modally separated at the thalamus, project to areas of the Cortex, that are specialized for their particular sensory modality, to areas called Primary sensory processing areas, or the "Core" of the sensory modality.

These areas are critical to the detection of information from the sensory inputs, but are significantly volatile in their synaptic storage. Within the cell, transport of the "Action Potential" or signal propagates down the Axion in much the same way as the original sensory neuron, but transmission to another cell is critically sensitive to calcium levels, which are somehow absorbed in the release of neuro-transmitters (The chemicals that transfer signals from one cell to another). If Calcium levels drop too low, the amount of signalling chemical also drops, and the signal is eventually lost in the clutter of other signals. This is called Habitation.

Evidence is that if the signal drops out, the Short Term Memory needs to refresh it in a process called rehearsal, and that any memory that is not refreshed in about 3 seconds has degraded to the point where it is no longer active.

While we are not sure exactly what is happening, there is evidence that a 40 hertz signal arriving at the cell is likely to result in an activation of implicit long-term memory. (from anesthetic experiments). Kandel's work, and recently Sessions work have suggested that what happens is that chemical pathways are triggered either by the calcium, or the CyclicAMP associated with a complex calcium ion channel, such as the S or NMDA ion channels. And, that these chemical pathways can be separately interfered with by new activations before the current activations are completed, depending on the state of the particular chemical pathway. As a result, Short Term Working Memory might retain a memory for a short while, Medium Term memory might retain it for longer, and Long term might retain it for even longer, but the nature of the memory determines how easily the memory can be changed by a new stimulation.

The previous scientist noted that AD sufferers had effectively reduced short term memory, It would be tempting to think therefore that this was the level at which the interference happened. But my own work suggests that it might be actually in the suppression step, in the partitioning of data before loading it into the Working Memory. My theory which is called Similarity Selection, suggests that selection of partitions of the "Data Cloud" created by implicit memory, is required to make sense out of the information involved.

In this case, the suppression probably starts in the prefrontal cortex, and activates part of the Nucleus Accumbens which in turn activates basket cells in the part of the cortex the brain wants to ignore. Basket cells, act to shunt the input of the neuron without affecting the synaptic function (except for the protein kinase C pathway) and as a result, no feedback from the selection affects the input, it just bars it's propagation.

In order to address the areas that are to be ignored, some method of addressing the clusters of neurons that vote for a specific output, needs to happen. This is probably the nature of the serial dependency implied by the "Bottleneck" effect that results in the 3 second limit to implicit memory. Place-Code Addressing requires a completely different organic structure than implicit memory, and it has been proposed that the upper three laminae of the isocortical tissues, might in fact be there, along with the neural group organization, specifically to allow such addressing. A possible connection to the thalamus is implicated in the selection of Neural Groups needed for the addressing to work, and it's operational mode is the pre-activation of selected Neural Groups. An indirect addressing mechanism called a Chunk has been suggested as a likely intermediate data element, and it has been projected that Chunk Analysis is part of the "Belt" area that often lies close to the primary associative area.

Through analysis and association between co-joint inputs, as directed by the limbic salience test, it is possible for the brain to arrive at an experential model of the most interesting aspects of the environment within a half a second or so. One reason AD is so debilitating is that it interferes with the ability of the mind to arrive at this experential model, by reducing the saliency due to flooding of the partitioning it can affect, with the result that Key environmental issues, tend to fail to cue, the required responses from the sufferer. As well, rehearsal of a larger partition at the short term memory level, will reduce the effectiveness of short term memory because the latency of the serial dependency goes up with the complexity of the partitions functional cluster.

It's a beautiful and competent text of Dr Smith about some basic and very important aspects of the information (i.e. electric signals) flow, and the signals "fixation" within the brain via highly regulated "chemical" synapses of certain neuronal cells and their long or short axonal pathways. However and whatever, we all forgot to answer Ron Scott "what goes wrong in dementias such as AD", and it is following: "memory" NEURONS SUFFER and malfunction, and ultimately NEURONS DIE before it's their normal time to die, and their interneuronal synapses disappear as well. It is like someone dies at the age of 60, instead of 90, because of the "subtile" brain reasons, but before that got ABNORMALLY ERASED majority or entire memory about many things or about everything. Death of a fully developed "AD person" is a death without memories because of the enormous loss of neurons and their synapse. Non-acquiring new memories is a little bit less problem for someone who is about to forget own children. Normally, and if we live in a good and stable social environment, such as in Denmark, we loose 10-15% of the cortical neurons - no more, if we are healthy in accordance with the age. Permanent, repeated and heavy stresses highly accelerate neurons dying (starving people in Africa are probably very "happy" about that), but multiple mutations are the worst thing of all (I have already mentioned it). In conclusion, every dementia is the illness due to abnormal and premature NEURON LOSS of the brain and its cortex.

By the way, considering inhibitory GABA-ergic, parvalbumin-positive interneurons (basket & chandelier cells), specialized to control the electrical activity of projection, efferent neurons (e.g. pyramidal & Purkinje cells), their malfunction and loss(!) is a characteristic substrate of schizophrenia, particularly if it happens in the "prefrontal cortex" of the frontal lobe. Besides, frontal lobe needs more arterial blood than other lobes (the "hyperfrontal distribution"), and the noted interneurons are particularly rich in mitochondria...

Quite true, but consider also the Glial function.

Glial cells make up the Blood Brain Barrier keeping heavy metals that are in our environment out of the brain. But being busy little cells, they also help the brain by offering a sort of framework within which neurons can navigate, and do EMR roles for Stress, Pain, and Inflamation. Like any emergency volunteers, when they are volunteering for EMR duty, they aren't available to act as a blood brain barrier, and also they aren't available to act as a navigation framework for dendritic growth. Of some interest to me, is the productivity levels of Stem cells that need to replace the volunteer population while it is wandering. As long as they can keep up with the demand, dementia is held off, but if the demand gets ahead of them, then dementia sets in earlier.

One of the problems with AD is that heavy metal plaques form in the brain, another is that the brain shrinks with age, suggesting a dropping in the population of cells within the brain, that might be preferentially higher for the Glial cells If this is true, then the neurofibratory tangles that are noted in the brains of both schizophrenia and AD sufferers might be caused because of the lack of a navigation framework for new growth. In Kandel's model of the DNA response to synapses, we are looking at a need for the cell to grow dendritic spikes and dendrites without the navigation framework. This will have the likely effect of NOT supporting Long Term Memory Formation, a failure to adequately navigate, and a general reduction in Synaptic population as a result of failure of the growth to compete successfully with the existing LTM connections.

One of the Aspects this echoes with Vascular Dementia, is simply the fact that once a significant number of neurons begin to die off, due to lack of support from the glial population, it is the same as having a stroke, the existing neurons will pump neuro-transmitters into the void, which in large quantites become a neuro-toxin, and kill off the surrounding neurons. If we add to this the tendency of AD neurons to (in some populations within the AD community) poison each other due to the amaloid factor we can see why AD is responsible for such quick reduction in neural capacity.

Thank you, Graeme and Dr. Stefanovic. Your replies have really helped.

Oh, I guess I missed it the first time, but Dr. Stefanovic, please do not address me as Dr. My brother who is an M.D. would not be amused, and I do not have the academic credentials to claim that level of education, having more or less self-educated after first year college.

Self-education, with such a level of understanding so many issues - it's a great thing...

I am widely read, and soundly criticized, but amazingly few people actually attack my work, most of them just want their own work to be better. Whether this is because I talk in circles, or whether it is because they can't be bothered to study on their own, I can't say. But thanks for the vote of confidence.