The following articles are contained below:-

Magnetic Resonance Angiography

The Neuropathology of Alzheimer's Disease

Addiction

Reductionism versus Holistic Approaches in Biology

Ageing

Magnetic Resonance Angiography

Nuclear Magnetic Resonance {NMR} is a spectroscopic technique used by scientists to obtain microscopic chemical and physical information. In particular it can be used to measure and image body tissue, a technique, which is known as Magnetic Resonance Imaging {MRI}. The technique was discovered independently by Bloch and Purcell, both of who were awarded the 1952 Nobel Prize and as the name suggests, involves the resonance of certain nuclei in a magnetic field. Nuclei, which are subjected to an external magnetic field, become excited into a higher resonance state by means of a radio frequency {RF} pulse. The way in which the nuclei subsequently decay to their lower energy state, is monitored and this information can be used to mage the tissue environment of the nucleus. MRI started out as a tomographic technique, (i.e. it produced an image of the NMR in a thin slice through the human body) but has now advanced beyond this to a volume imaging technique. Its advantage therefore is in its ability to produce images in multiple planes, as well as its capability to differentiate tissue structures, which have similar density.

In conventional MRI, tissue vessels appear dark despite the fact that blood itself readily gives rise to an MR signal. This lack of signal is due to motion and such artefact is usefully employed in monitoring blood flow in a technique known as magnetic Resonance Angiography {MRA}. The term angina itself is derived from the Latin for strangling and hence the importance of this technique in detecting hardening of he arteries (atherosclerosis) and associated ischaemia. When a patient is placed in a magnetic field the hydrogen nuclei (protons) that are a common element in organ tissue, can occupy only one of two energy states. By means of a RF signal those protons in the lower energy state, can be excited into a higher energy state and their subsequent response is detected by coils and analysed by computer. However, the response known as dephasing, is not so well behaved where the hydrogen atoms are allowed to flow through the body and hence interact with different regions of the magnetic field, whose value varies spatially. These artefacts, can however be used to reveal information about such flows (eg in the blood or spinal fluid) and hence the condition of the vessels involved. The images are taken in a series of thin slices in which the high velocity flows (in the case of the so-called echo gradient techniques) are the brightest most prominent features.

The volume of the thin slices is then produced by computer algorithms, which extract these high intensity vessels and produce a set of vascular projection images, which can then be viewed from arbitrary angles. These images can be viewed in a cine format to provide dynamic information about blood flow. Scan times add 10 to 20 minutes onto the time of a routine MRI scan (although some intracranial angiograms can be obtained in as little as 5 to 10 minutes) and most units are now sold with an MRA package as an option.

The advantage of using MRA as opposed to using other forms of angiography is in its non-invasiveness, its high contrast between diseased and healthy tissue and its ability to show both vasculature and soft tissue potentially in 3dimensions. The vessels can therefore be projected along any orientation, hence unlike conventional angiography it can eliminate overlap with other structures. It is widely used to assess abnormality in both neurovascular and cardiovascular systems. Although MRA can be used together with an intravenous application (usually a gadolinium compound), techniques involving the use of such contrast media will not be considered in this review

One of the many clinical applications is in the detection of intercranial aneurysm in which MRA was found to have a sensitivity of 95% and a specificity of 100%, compared to conventional angiography.[1] MRA is also used to monitor cardiac flow and can help detect abnormalities, such as ischaemia, valve defects and congenital lesions, and has been successfully employed to locate vessels as candidates for heart bypass surgery. It can also be useful in planning for stereostatic biopsies and operations on the brain.

There are some limitations with MRA however, since the strong magnetic fields do preclude patients with pacemakers and is a hazard for those who have metallic objects medically inserted in their body. Also patients with severe claustrophobia cannot tolerate a prolonged period of time within the bore of the magnet, without the use of a sedative. Furthermore the study itself is expensive (~£500 per hour) and during the scan itself the patient must be co-operative and still, since movement degrades the image.

MRA is classified into two major categories, according to the method employed. The Time Of Flight {TOF} method relies on inflow enhancement of spins entering into the image, which can be analysed and displayed either using 2D- sequential or 3D- volumetric techniques.[2] With a 2D- TOF a flow compensating gradient-echo sequence is employed so as to obtain thin image slices, which can be recombined using techniques such as a maximum intensity projection{MIP}algorithm, to produce a 3D- image of vessels, as in conventional angiograms. A greater spatial resolution can however be obtained with 3D-TOF, in which a volume of images is simultaneously obtained by phase encoding in the slice-select direction. The image can be produced using MIP and several 3D volumes can be combined so as to visualise longer sections of vessels. However with thick volumes and slow moving blood a poor signal is obtained with 3D-TOF.

The other main category is Phase Contrast{PC} TOF, which relies upon a low induced phase shift. Post-processing the data using a MIP algorithm, which involves a complex subtraction of two sets of data with a different amount of flow sensitivity, also produces images. This produces an image with a signal intensity, which depends on flow velocity, and hence quantitative information about the direction, as well as the speed of flow can be obtained. In order to understand the developments that lead to such innovative techniques it is necessary to review the basic physical mechanisms behind conventional MRI, so that we can elaborate certain aspects that are applicable for monitoring blood flow. After describing in detail the imaging techniques employed in MRA, I will then go on to discuss the medical applications.

Fundamental Theory

When a spinning body (such as the earth itself) is subjected to an external couple (e.g. that caused by an external gravitational field) its axis will precess. This is likewise true of a spinning charge whose magnetic dipole moment will precess when subjected to a couple produced by an external magnetic field. Such a precession will therefore occur with nuclei of certain atoms that have a net magnetic moment and also an electron in orbit . In the study of NMR, it is obviously the former that is most important, however the orbit of the electron and it's precession does cause small refinements that are relevant to differentiating chemical environment and tissue imaging.

The magnetic moment m can be resolved into a vertical component mz parallel to B and although the resultant m changes direction mz is constant in magnitude and direction. In a sample containing many spinning hydrogen nuclei the net magnetic moment M will be derived from the vector sum of all the magnetic moments. Classical theory is inadequate to describe fully the behaviour of single spins because of the quantum nature of the nucleus, which is briefly dealt with in the appendix.{I}

In the sample containing many protons we will henceforth consider the resultant magnetic moment M, which represents the excess of nuclei in the lower energy, each having a component mz in the direction of the applied field. By applying a radio frequency (RF) beam each photon of the beam has an energy hw which is exactly that required to excite the proton energy from a lower to a higher state.Macroscopically (i.e. classically) this will correspond to a change in the magnitude of the Mz component. By applying a RF pulse for the correct amount of time, the value can be reduced to zero i.e. the resultant magnetic moment precesses at 90 to the external magnetic field.

This displacement angle between the nuclear magnetisation vector M and the direction of the static field continues to increase for as long as the rotating RF field is applied to the sample and the rate of increase depends on the power of the field. As mentioned above a pulse strong and long enough to rotate M into the X plane is termed a 90 pulse. We know that this has been achieved because the magnetisation vector M continues to precess in the X_Y plane, and in doing so it generates a small voltage, either in the same coil that produced the RF signal or a separate receiver coil. This is due to Faraday's law of electromagnetic induction, which states that an induced emf is proportional to the rate at which magnetic flux is cut i.e.

            emf = -d0/dt

Imaging Techniques

A second effect of an application of the RF pulse is that the protons move in phase with each other along their precessional path. However this coherent phase amongst the proton precession does not persist, since there are subtle variations in the applied magnetic field, chemical environment or surrounding tissue. The protons also resonate and transfer energy to each other and this proton- proton relaxation also causes a loss of phase coherence. Indeed it is these subtle 'frictional forces' that allow us to obtain data upon the surrounding tissue. The net effect of all this is that after a short period of time, the proton precessions fan out in the X-Y plane. This so called transverse relaxation is described by Bloch's equation and is characterised by a time T2.

dMx/dt = -Mx/T2

dMy/dt = -My/T2

For solids T2 is of the order of 10 s whereas for pure water it increases to 1s. In soft biological tissue it ranges from 0.010 to 0.20.[3] Transverse relaxation therefore influences the natural lifetime of the so called 'free induction decay'{FID} signal that is picked up by the receiver coil, since the strength of this signal relies upon the magnetisation in the X-Y plane remaining in phase. When the excitation pulse ends, each proton continues to 'feel' not only the external field but also local fields associated with the magnetic properties of neighbouring nuclei. The nuclei will therefore acquire a range of slightly different precessional frequencies, causing the free induction decay ignal to go out of phase. In a liquid, where atoms and their nuclei are continuously in motion, the internuclear magnetic fields responsible for spin-spin relaxation tend to average out. Consequently, the signal decays much more slowly than in a solid, where the nuclei remain essentially fixed in position. In a liquid T2 can be as long as several seconds whereas in a solid it is usually only a few microseconds. Due to imperfections in the actual applied magnetic field the FID response decays slightly faster than in the ideal case of a perfectly uniform field. The time constant in such an imperfect field is designated T*2 in order to distinguish it from the true relaxation time T2. Another effect that gives us information about the surrounding tissue is the longitudinal relaxation time T1. This involves the return of M back to its original precession angle. This so called spin- lattice relaxation, as the name suggests results from the transfer of energy between the spin assembly and the lattice. In isolation the time taken for a proton to spontaneously emit energy (of RF frequency), and return to its ground state is quite long but due to the frictional effects of the environment, it occurs much quicker due to stimulated emission of radiation.

The fluctuating environment fields give rise to a spectrum of electromagnetic quanta; some of these quanta will be just the right energy to induce transitions between the nuclear spin states and will therefore restore equilibrium. Classically we can picture the M gradually spiralling in 3 dimensions back to its original angle of precession. This is called longitudinal relaxation since it involves a change in Mz and has an associated time constant T1,which is likewise given by the equation.

dMz/dt = (Mo- Mz)/T1

In soft biological tissue T1 values range between 0.1 to 0.6s. Hence T2 times are generally shorter than T1 times, although in liquids this ratio tends to approaches unity. The reason for the range in time scales, for various tissues,is due to the corresponding range in the random Brownian motion compared to the Larmor frequency. In a solid the random atomic motions have time scales of the order of microseconds, the same order as the precessional time period of the spins. The twitching environmental fields in the solids strongly disrupt the motion of their spins because their power spectra contains significant contributions at the resonance frequency W. For low viscosity liquids however, the random motions will be very much faster and the intervals between random collisions will be much shorter (of the order of picoseconds). Consequently the relatively slower motion of the resonating spins, averages over the random temporal variations of the environmental field. Hence, averaged over one Larmor period the net effect of environmental field is very small. A progressive reduction in fluid viscosity leads to a progressive reduction in resonance linewidth and an increase in T1 and T2 - an effect called motional narrowing. The resonating nucleus thus 'rides out the storm like a gyroscope on gimbals!'[4] Although effectively all the protons are involved to varying extents, the relaxation times of strongly bound components are so short that the signal has decayed to almost nothing after a hundredth of a second. Since most MRI systems are designed to operate, (due to engineering constraints) in the 0,01 to 0.5 seconds, this explains why bone and the more tightly bound water molecules are invisible to conventional MRI.

We therefore have 3 main flavours or parameters with which we can measure the structure of tissue/chemical environment in which the specific nuclei (protons) are placed. These correspond to the proton (free water) density, T1 and the T2 relaxation times (the latter of which is itself dependent upon both spin - spin coupling and the inhomogeneity of the field due to the local environment).[5] Images can therefore be T1-weighted, T2-weighted, or proton density weighted by varying the pulse sequence viz. the repetition time{TR} (the interval between repetitions of the 90 pulse sequence) and echo time{TE} (cf. appendix II). As the magnetic components decay either by T1 or T2 relaxation, the change is monitored by the induced voltage in a detection coil {FID}. In practice the same RF coil can be used to excite and measure the decay of the nuclear spin energy. By applying different magnetic field gradients in the X Y Z directions, three-dimensional localisation of the MRI signal can be obtained. The first gradient to be switched on is called the slice gradient. As the gradient is applied the nuclei precess at different frequencies. The RF pulse is then tuned to those precessional frequencies that correspond to the section of interest, hence causing only these protons to be flipped into the transverse plane. The remaining two dimensions of the slice are then spatially encoded by applying one gradient to encode the different precessional frequencies of the nucleus and the other to encode the different phase shifts

As already mentioned, NMR signals are generated by manipulation of the bulk magnetic moment with suitable RF pulses such as a spin-echo pulse sequence, in which each slice receives a 90 degree pulse followed by 180 degree refocusing pulse, before a signal is received. Protons in fast flowing blood that receive the 90 degree excitation pulse, will have flowed out of the slice by the time of the 180 pulse (which only effects the slice of interest) and will not therefore be rephased and hence will appear very dark on the resulting image (turbulence of the spinning protons also contribute to such a loss of signal). Even if the flow of mobile protons is slow compared to the TE (180 pulse), the spins excited by the 90 degree pulse may not be fully replaced by freshly excited spins, in readiness for the rephasing TE pulse and the signal is therefore attenuated by the lack of such excited spins. Hence spins that leave the image slice between the TR and TE pulses are not rephased and the result is a lower signal (dark image).

(fig 1)

If however the TR and TE values are reduced a flow related enhancement can occur resulting in a bright signal being produced by the vessels. This arises since although the protons passing out of the region are unable to be rephased by the 180 pulse, other protons that become excited by the TR pulse replace them. A 90-degree pulse first saturates the slice and the signal undergoes dephasing. If a second 90-degree pulse is applied, any signal will then have come from the blood flowing into the slice, providing that the T1 relaxation is long compared with the time between pulses. The fresh influx of protons produce a stronger signal than the surrounding stationary protons that have been repeatedly exposed to RF pulses within the slice. These so called Time of Flight (TOF) techniques involve only an excitation pulse followed by the signal readout, since the rephasing is achieved by means of a gradient echo which is much aster than the spin-echo method. Magnetic gradients perform the refocusing function before the signal readout, making the image of the blood flow bright.(fig 2) This is contrary to the TE (180 ) refocusing pulse used in traditional spin-echo imaging which, as already stated, causes blood flow to be imaged in black.[6]

Another method of performing TOF, is to employ the 90-degree and 180-degree pulse at different slices (and hence slightly different frequencies). The 90-degree pulse excites spins in one plane and in the absence of flow, the following 180-degree pulse does not produce a spin- echo response, since it is applied to a different slice. However if the correct TE is used, blood from the 90-degree slice flows into the 80-degree plane and an echo is produced. Hence the location of the 180-degree pulse is moved to match that of the blood that experienced the original 90-degree pulse, so that only this blood will produce an echo signal.

Most studies indicate that three-dimensional TOF imaging is somewhat better than 2D-MRA, especially in demonstrating segmental pulmonary arteries and detecting 70% or greater cartotid stenosis. In the case of the later, MRA was more effective than spiral CT, although it did however have a lower specificity compared with colour-flow ultrasound.[7]

Time of Flight techniques, although widely used, do however have limitations regarding the ability to obtain quantifiable data. Phase Contrast Angiography n the other hand is more useful for distinguishing between different rates (and directions) of flow. This method exploits the fact that when the mobile protons in blood pass through magnetic field gradients they obtain a different phase alignment to that which occurs with static spin tissue. In the spin echo sequence, equal gradients on either side of the 180 degree pulse will 'balance' each other, because the static spins experience the same local field (and therefore the same Larmor frequency) before and after the 180 pulse. Hence there is no phase difference. However if the spins move during this time, the gradient field will not 'balance' and such spins acquire a phase shift proportional to the velocity of flow. By keeping track of these phase changes, we can obtain images that are sensitive to such a flow. Images can therefore be obtained in which flow is encoded so that, in one direction it is in black and in the opposite direction it is white and the grey scale can therefore be related to the velocity of flow. In this way a vectorial image of blood flow can be obtained in which both the direction and magnitude can be displayed.

One of the main applications of MRA is achieved when one combines either gradient-echo or phase contrast techniques in synchronicity with the cardiac cycle to produce a movie or cine format. Such techniques are valuable in the assessment of left and right ventricular function and also allows indirect assessment of pulmonary hypertension. Cine gradient methods, which rely on TOF inflow enhancements are useful for imaging dynamic processes such as a beating heart, however such information is qualitative since the signal intensity of the flowing fluid is dependent on complicated factors. Phase contrast (PC) cine magnetic resonance however, can depict quantitative information and its ability to characterise direction and pulsatibility of flow not only improves diagnostic accuracy but can also give insight into pathological processes. Such techniques can be used in areas, which are not easily accessible, by Doppler ultrasound or where the complex flow associated with a disease process may produce confusing data such as in a patient with aortic dissection. MRA can also provide information on pulmonary embolism and has an advantage over spiral CT in that deep venous thrombosis can be accurately assessed.

Another powerful aspect of MRA is in its ability to measure blood flow--providing excellent temporal data throughout the cardiac cycle. Time Of Flight techniques involve tracking a 'plug' of blood, a measurable distance within a vessel, thus allowing the average velocity of the flow to be calculated. These so-called bolus-tracking techniques are limited however by poor edge definition of the tagged (saturated) blood and by errors in measuring cross sectional area. Also with slow moving blood, small displacements of the tagged blood are difficult to measure accurately. However peak flow velocity measurements show close agreement with Doppler methods. Cine Phase Contrast techniques allow pulsatile flow to be measured by dividing up a bolus of blood into many small volumes and using phase changes to calculate the average velocity for each small volume. Measurement then allows the flow rate to be calculated.[8]

MRA was initially found to be most useful in the head and neck, due to the lack of cardiac, respiratory and peristaltic motion and has been a useful tool in  searching for intercranial aneurysms. Image manipulation in 3 planes allows overlapping vessels to be removed from the image and the vascular tree can be repositioned for the optimal viewing of a specific portion of it.{appendix III} Magnetic resonance techniques have the advantage of showing both vasculature and soft tissue and has a high contrast with regards to delineating disease. Intercranial MRA also has some value in observing arteriovenous malformation after surgical repairs or immobilisation and can be used for planning stereotactic biopsies and operations on the brain.[9] With the high level of accuracy in grading carotid artery stenosis, MRA is now routinely used in cerebral arterial occlusive diseases and has in part replaced contrast angiography. It also plays a major role in the diagnosis of dural venous thrombosis. Large neck coils can also be used which can examine the entire length of the carotids and the innominate artery. Some recent studies indicate that MRA provides a good map of the involvement of the great arterial and venous vessels in skull based tumours but that the spatial resolution is not as good as that provided by traditional angiography. In particular the evaluation of small arterial vessels and slow flow venous channels was poor but MRA was considered useful in monitoring the results of radio-chemotherapy[10]

In the abdomen MRA is used to evaluate patients before and after liver transplants and to study the effects of hepatic tumours, as well as allowing evaluation of renal cell carcinoma. It has a reported sensitivity of 100% in screening patients with hypertension for renal artery stenosis, although there is a tendency for overestimating the degree of stenosis. Indeed one of the problems associated with MRA is due to the signal loss caused by post stenosis turbulence and turbulent flow that is produced by internal ulceration. As turbulence (which produces rapid phase dispersion and hence signal void), is thought to occur in atherosclerotic disease this can cause MRA to overestimate the degree of stenosis.[11] However MRA can be used to examine the inferior vena cava and abdominal aorta and can indicate stenosis and occlusion of the superior mesenteric arteries. Also MRA of the pulmonary arteries and veins may be used to check patients for pulmonary embolism.

Regarding the lower limbs, in a recent study upon patients with lower extremity ischaemia, MRA was able to identify all the diseased vessels that showed up on a contrast arteriography but also in nearly half the 51 patients, MRA also revealed patent tibial segments that were not previously seen.[12] Contrast arteriography can fail to opacify tibial vessels in patients with severe multi -level occlusive disease, presumably due to poor inflow and dilution of the contrast.(Both contrast arteriography and MRA methods should not be used as a screening test to confirm the presence of lower extremity peripheral arterial disease.) The iliac arteries are also difficult to image with MRA due to their irregular course and pulsatile, triphasic blood flow. Also peristalsis from adjacent bowel loops in the pelvis can degrade the image and result in artefacts.

One obvious advantage of MRA is the ease with which it can be integrated with a conventional MRI scan. To cite one example, a 9 year old boy showed MRI features of an ischemic event in the left cerebral artery territory. Under a subsequent MRA beading of the artery was revealed, which was consistent with the irregular blood flow associated with turbulence or luminal narrowing. This raised suspicions of fibromuscular dysplasma {FMD} and a MRA was subsequently performed, which likewise revealed beading of the left renal artery, confirming the earlier diagnosis of FMD. In such instances MRA shows itself to be a ' rapid and less invasive technique associated with far less morbidity and mortality as compared with conventional angiography'[13 ]

In conclusion, MRA techniques using either TOF or PC effects are used to produce signals from flowing blood. Signals from surrounding static tissue are either suppressed or subtracted so as to increase the vascular detail, whose structure is then displayed in a projectional format, to simulate a conventional angiogram. Since vessels can be projected in any orientation, MRA eliminates overlap with other structures.The ability to project a rotating image can provide more information than a single or even two plane conventional angiogram. One of its main advantages is however in its non-invasiveness, as there is no need for intravenous contrast agents that can have unpleasant side effects (which are not therefore suitable for patients who are recovering from a stroke or are likely to suffer renal complications). Although initially used to view vessels in the head and neck, MRA is now an excellent method of imaging the thoracic aorta, pulmonary arteries, abdominal aorta and its branches (renal and mesenteric arteries) as well as the runoff vessels into the legs. In addition many vessels that are not accessible by other non-invasive methods (eg. Doppler), can be examined, such as those contained within the Calvarium.

The Phase contrast method also provides instantaneous velocity and blood low profiles with high temporal resolutions that can be superior to other methods available. Although conventional angiography is considered the 'gold-standard' (especially in imaging peripheral vessels), it is likely that as MRA techniques continues to improve, with more sophisticated hardware/software and faster pulse sequences, it will become more reliable and clinically useful throughout the body.

In quantum mechanics the angular momentum L can only have discrete values. Each nucleus with it's particular structure of neutrons and protons, gives rise to a  net magnetic moment, whose strength depends on its particular configuration . The solitary proton of a hydrogen has a net angular moment of h/2 and an associated magnetic dipole momentum of eh/2p where p is the rest mass of the proton of charge e and h is Planck's constant.(Fig 3) The heavy nuclei that are of interest to medicine such as phosphorus and certain carbon isotopes, have different magnetic moments due to their different nuclear structure.

Quantum mechanics shows that when a proton is placed in a magnetic field, it can only take up one of two possible states, parallel to the field or antiparallel.These correspond to low and high energy states respectively. A very small excess of nuclei line up parallel to the field to produce a net magnetic moment. The diagram below shows an assembly of such protons before and after being subjected to an external magnetic field.

Hence the ratio of nuclei in two states N1 and N2 separated by an energy E is given by

N1/N2 = e^-kt

where N1 is the number of nuclei in the higher energy state, k is Boltzmann's constant and all nuclei are in thermal equilibrium with their environment, at a given temperature t .

In quantum mechanics the energy levels of a system take up discrete values hat are related, via Planck's constant, to the natural frequency w associated with the system. These energy levels are given as

Energy E = hw

E =$ h B

Where $ is the ratio of the magnetic dipole moment to the angular momentum of a spinning charge.

Under a magnetic field as large as 1T this gives an energy difference of 10 ev. and at room temperature this produces an excess of protons in a spin up state of ly about 3^ 10 . This explains why a large magnetic field is required in order to cause a noticeable bias in the resultant magnetic alignment. If the excess of nuclei in the lower energy can be excited to a higher state by absorption of the correct amount of energy, namely hw, then the net magnetic moment of the assembly in the y direction ill be zero (Fig 4).

Dephasing of the FID is caused by inhomogeneity in the magnetic field as shown in Figs 5A, B and C. If after a time 't' following the initial 90-degree RF pulse, a second RF pulse of 180-degree is applied, the transverse magnetism will rephase, as shown in Figs 5D, E and F. This is because the 180-degree pulse reverses the phase lag or lead of each of the effected spins, eventually achieving coherence again after a time '2t', (known as the echo time TE) after which they begin to dephase again. This process an be repeated many times, each echo is however reduced in amplitude due to losses caused by spin-spin interaction, as shown in Fig 6

References:

1 Ross Js, Masaryk TJ, Modic MT, Ruggieri PM, Haacke EM, Selman WR: Intracranial aneurysms: Evaluation by MR angiography. AJNR 1990: 172:351-357

2 Ballinger JR: MRI Tutor http://ballinger.xray.ufl.edu/mritutor/flow.htm

3 Guy CN: The Second Revolution in Medical Physics. Contemporay Physics Vol 37 No1 pg15-45

4 Bloembergen N, Purcell EM, Pound RV. Relaxation effects in nuclear magnetic resonance absorption. Phys. Rev., 73: 679-712

5 Elderman RR, Warach S: Magnetic Resonance Imaging. The New England Journal Of Medicine March 11th 1993; 708-714

6 Makow LS. Magnetic Resonance Imaging: A Brief Review of Image Contrast. pg214-218 (S803 CD-ROM)

7 Magarelli N, Scarabino T, Simeone AL, et al. J. Neuroradiology. 1998; 40: 367-373

8 Pelc NJ, Herfkens RJ, Shimakawa A, Enzmann DR: Phase contrast cine magnetic resonance imaging. Magn Resn Q 1991; 7:229-334

9 Edelman RR, Zhao B, Liu C,et al: MR angiography and dynamic flow evaluation of a portal venus system.AJR Am J Roentgenol 1989; 153: 755-760

10 Duca S, Crasto S, Sajzedo E, Bertone P, Petricig P, J. Rivista Di Neuroradiologia. 1997; 10: 152-154

11 Nishimura DG, Macovski A, Pauly JM, Conolly SM,: MR angiography by selective inversion recovery. Magn Reson Med 1987; 4: 193-202

12 Carpenter JP, Owen RS, Baum RA, et al. Magnetic Resonance Angiography of Peripheral Runoff Vessels. J, Vascular Surgery 1992; 16: 807-815.

13 Leventer RJ, Kornberg AJ, Coleman LT ,et al J. Pediatric Neurology, 1998; 18:172-175

14 C.H. McDonnel III, M.D.Roberts,J.Herfkens A.M.Norbash and G.D.Rubin. Magnetic Resonance Imaging And Measurement Of Blood Flow. pg237-241. West J. Med 1994 160:237-242

A.C.Bhalerao, P.Summens. Multi-resolution Flow Feature Extraction From MRA http://carmen.umds.ac.uk/a bhalerao/project/angio/angio.htm

Laissy JP, Delsola B,et al. Magnetic Resonance Angiography: fields of exploration, main indication and limits (translation). Journal Des Maladies.1997; Vol 22, No5 MRA. http://www.genet.com/maven/aurora/mri/mra.htm

THE NEUROPATHOLOGY OF ALZHEIMER’S DISEASE

CONTENTS

INTRODUCTION 

I PRIMARY FEATURES

1. Neurofibrillary Tangles 
2. Senile Plaques

II ASSOCIATED FEATURES

1. Genetics 
2. Microglia and Astrocyte Activation 
3. Alteration of the Glutamate Transport System
4. Cytoskeletal Destabilisation 
5. Neurotransmitters
6. Environmental Factors 

CONCLUSION

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder of the central nervous system and is the most common form of dementia. The disease develops differently amongst individuals, suggesting that more than one pathological process may produce the same outcome. The early symptoms include a loss of short-term memory, which later progresses to a deterioration of a person’s language, perception and motor skills. This is usually accompanied by irritability and mood swings, together with feelings of anger, anxiousness and depression. In advance stages the patient becomes unresponsive and loses mobility and control of body functions, culminating in death, typically between 5 to 10 years of the initial symptoms.

Although rare before the age of 50 Alzheimer’s Disease (AD) afflicts nearly half the population over 85 and is responsible for two thirds of the cases of dementia in patients over 60 years of age. Computerised tomography and magnetic resonance imaging can reveal the brain shrinkage characteristic of the disease but final diagnosis can still only be positively confirmed by post-mortem examination. The gyri of the cerebral cortex become smaller and the sulci between them widen. The ventricles of the brain then expand to fill the void left by the deteriorating brain tissue. This brain atrophy is often most extensive in the frontal and temporal cortex, although there is considerable variation in the general pattern of pathology among patients. Both cell death and dendritic shrinkage are responsible for this cerebral atrophy and these symptoms together with astrocyte and microglia hypertrophy are known to occur with normal brain ageing but are greatly accentuated during the disease

A part of the disease is genetic in etiology and is classified as familial AD, which in turn is further subdivided into early-onset and late-onset forms. The early-onset disease accounts for about 10 percent of such cases and involves patients who are younger than 60 years of age. The majority of cases however develop after the age of 60 (late-onset) and occurs sporadically i.e. in individuals with no family history of the disease, although a genetic factor may predispose these individuals to the disorder.

The disease was originally described in 1906 by the German neurobiologist Alois Alzheimer, during an autopsy he noted the presence of two cellular abnormalities in the brain – senile plaques and neurofibrillary tangles. These are still the primary features that are used to diagnose AD in autopsy, although both structures are found in smaller amounts in the brain of healthy people. However it is not known whether these plaques and tangles are a cause or a consequence of the disease and there are also other correlating features that will be considered later in the review. Plaques and tangles are formed predominantly in the frontal and temporal lobes including the hippocampus and in advanced stages, the pathology extends to parietal and occipital lobes

Senile plaques (SP) are irregular spherical structures of up to a few hundred micrometers in diameter and accumulate in the grey matter of the cerebral cortex and hippocampus, as well as other forebrain structures. They consist of a deformed axon terminal together with an accumulation of amyloid, a class of protein that congregates as tiny fibres in intracellular space. Lipofuscin granules are also present and electron microscope studies reveal that the plaques contain contributions from many different cell types and that no single portion of a cell is completely responsible for their overall structure.

Histochemical methods have revealed that SP display a marked increase in certain enzymes, while the electron microscope has shown that there is a proliferation in the number of mitochondria that are present.[1] Although the increase in metabolic rate (as indicated by the increase of enzyme activity), might play an important role in the formation of SP, their development is not confined to neurons that respond to specific neurotransmitters, since they can form in neurons of many types. There is one enzyme in particular, whose activity is not increased and that is monoamine oxidase, which curiously is the one enzyme that is increased with normal ageing. This emphasises the fact that AD is not simply a speeded up form of the normal ageing process.

Neurofibrillary tangles (NFT) are dense bundles of long unbranched filaments that form in the cytoplasm of some neurons and are made exclusively of microtubular-associated proteins. These pathological webs of neurofilaments are derived from a hyperphosphorylated tau protein. They are most prominent in the large neurons of the forebrain, including the hippocampus and cortical pyramidal cells. They are also prominent in a number of other disorders of the central nervous system, including Parkinson’s disease and dementia pugilisica. Unlike SP, the NFT exhibit a marked lack of enzyme activity, thus indicating a separate, though unknown cause of these two manifestations of AD. ther main feature of the disease that shows up in the hippocampus, is granulovacuolar dgeneration, which are small vacuoles that appear in the cytoplasm of the neuron. These vacuoles are up to 5 micrometers in diameter and each contains a small granule some 0.5-1.5 um across.

The memory loss characteristic of AD appears to be due to the selective damage of the major input and output pathways to the hippocampus. One such primary input is the perforant pathway, a band of fibres that originates in the entorhinal cortex and projects into the hippocampus.[2] The entorhinal cortex in turn receives input information from a variety of cortical sensory and limbic areas and is therefore a critical structure in delivering sensory information to the hippocampus.

In addition, AD also causes damage to the subiculum and adjacent pyramidal cells of the hippocampus, where there is a plethora of NFT. This region provides the primary output pathway from the thalmus, hypothalmus, basal forebrain, amygdala and the cerebral cortex – AD therefore deprives the forebrain of output from the hippocampus. Hence by damaging both the major input and output pathways of the hippocampus, AD effectively isolates this part of the brain from the rest of the central nervous system, causing profound memory loss.    

1. Neurofibrillary Tangles (NFT)

These are twisted fibres of the protein tau, which is found within neural cell bodies. In the normal brain the tau is a low molecular weight protein that promotes microtubular polymerisation and stabilisation. It exists as 6 isoforms derived from alternate splicing of a gene on chromosome 17 and also post-translational modification.[3] When incorrectly processed this tau molecule clumps together forming tangles. Only one tau isoform is found in foetal brain but at present there is insufficient data upon the changes that occur in tau isoforms during ageing or how these might effect the disposition towards AD. The individual neurofilaments are generally tubular in structure and vary in diameter from 10 to 20 nm. Although often found in the same region as senile plaques (SP) their density can be greater (especially in the hippocampus) and they track closely with neuron damage that may not exhibit any obvious plaque formation. Although lipofuscin bodies are often present, the amount is not correlated with the degree of degeneration.

The tangles consist of aggregated paired helical filaments - with an 80 nm half period - that on a larger scale show variations in the tangled morphology depending on the progression of the disease. These filaments accumulate in the neuronal some, as neuropil threads in the neuronal cell processes, and in the dystrophic neurites of the neuritic plaques, as well as producing the full-blown NFT. The early stages of NFT are characterised by delicate fibrillary argentophilic inclusions within the affected nerve cell body. The intermediate stage is characterised by highly argentophilic aggregates of fibres, forming large bundles which ultimately fill the entire neuron body and extend into neuronal processes, producing the classic intraneuronal tangles. In the late stages, thick flame-shaped tangles (so-called ghost tangles) lie free of neuronal somas in the neuropil.[ Fig 1]

The formation of these abnormally twisted fibres is caused by the hyperphosphorylation of the tau protein, which is a major subunit of the paired helical filaments in AD. They probably result from an imbalance in the protein phosphorylation and dephosphorylation system. In addition to the phosphorylation, the paired helical filaments also have other covalent modifications, such as progressive glycation which produces cross-linked and insoluble proteins. This type of non-enzymatic covalent modification has been implicated as a stimulus to oxidant cell stress that may contribute to neurodegeneration. When the neurons die, the paired helical processes are released into the extracellular space where they undergo proteolysis.

In normal elderly brains neuropil threads are absent or are very scarce even in regions which have developed SP. On the other hand neuropil threads are abundant in cortical grey matter of AD and given their widespread distribution, it is becoming recognised that even diffuse amyloid deposits contain these paired helical filaments. Non-neuritic SP are actually uncommon in advanced AD, and frequently the neurites in SP have both dystrophic and paired helical filament type properties, especially in the the limbic grey matter and the amygdala

A number of studies have implicated the enzyme glycogen synthase kinase–3 (GSK-3) in this abnormality. Being present in brain neuron (especially prevalent in AD brains) GSK-3 has been shown in some studies to induce phosphorylation of the tau protein in vitro and in cells transfected with GSK-3, although non proline dependent kinase may also be influential in the development of AD.

About 21 abnormal phosphorylation sites have been identified in paired helical filaments but although ten of these sites are canonical for proline directed protein kinase such as GSK-3, non of these have been reported to be overexpressed or overactive in AD. However in AD brains with its reduced phosphatase activity even normal levels of GSK-3 activity may be sufficient for hyperphosphorylation of the tau protein.[4 ]

2. Senile Plaques (SP)

These neuritic plaques involve neurons of many types and occur in large numbers within the cerebral cortex, the hippocampus, the amygdala and other areas essential for cognitive function. However they often also appear in parts of the brain that seem unaffected and conversely are sometimes not clearly seen in diseased regions. Each plaque is a slowly evolving extracellular structure, which alters the axons and dendrites that surrounds it, eventually causing their disintegration.[Fig 2(a) & (b)] It consists of a deposit of neural fragments surrounding a core of amyloid

beta-protein. This sticky peptide is a heterogeneous collection of overlapping peptides 39 – 43 amino acids long, that are derived from a larger Amyloid Precursor Protein (beta-APP), which is a normal component of nerve cells and is encoded by a gene on chromosome 21.

In addition to the central core of amyloid beta protein and surrounding abnormal neurites, mature plaques contain two types of altered glial cells. In the centre, microglial cells are found (these are scavengers which respond to inflammation), while around the outside of the plaques are astrocyte cells that are associated with injured brain areas. [The contribution of these glial cells will be discussed later]

One proposed scenario for the development of AD, is that excessive deposition of beta-amyloid aggregates along with certain other proteins into senile plaques, which eventually enrage the nearby glial cells that supply neurons with nutrients and structural support.[5] The inflammatory factors released by the glial cells, damage neurons whose lattice of microtubules are ultimately transformed into a wreckage of neurofibrillary tangles. The genetic link with the formation of amyloid beta-proteins will be considered in section II. A less popular theory holds that amyloid deposition is preceded by neuritic dystrophy, a view supported by the fact that such dystrophic neurites can be found in young Down’s syndrome brains before amyloid is detected. On the other hand neuritic changes similar to those observed in SP can be detected in certain locations of aged brains, completely independent of amyloid presence[3].

The vulnerability of neurons to such insults, may be enhanced by increases in intracellular Ca2+ which is itself induced by beta-amyloid.[6] Neurons that have been treated with aggregated beta-amyloid have been found to have increased levels of Ca2+ and hydrogen peroxide, while agents that lower Ca2+ or scavenge reactive oxygen species (ROS), reduce the toxicity of beta-amyloid. [It is believed that when the beta-amyloid fragments binds to the advanced glycogen end-product receptor, it can generate ROS.] Such observations therefore suggest that beta-amyloid can induce neuronal death via a Ca2+-mediated mechanism or by damage resulting from free radical formation.

Recent research has also shown that an excess of the excitatory amino acids glutamate and aspartate can result in neuronal cell death by either causing a rapid increase in Ca2+ concentration or the release of free radicals. Such free radical production is most prolific in the mitochondria where the damage to the DNA is 16 times greater than in nuclear DNA. The continuous mitochondrial replication that occurs throughout the lifetime of a cell, together with a lack of DNA repair mechanism in response to free radical damage could therefore contribute to mutations that may be responsible for the development of AD with age.

There are at least 6 different APP gene transcripts that result from the splicing of 3 of the 19 exons that make up the APP gene. The three prominent ones that occur in the brain, code for APP protein with either 695, 752 or 770 amino acids. The latter two contain the protease inhibitor domain KPI, which has trypsin or chymotrypsin inhibitory activity. In recent studies of brain ageing, there is little evidence for much change in the overall prevalence of APPmRNA but changes in the splicing ratio have been observed.[7 ] In foetal brains APP-695 is most common (70-90%) whereas in ageing brains APP-751/770 are the predominant transcripts (50-90%), while young and middle aged individuals show an intermediate splicing ratio. Studies have also shown that in AD brains there is an increase in the ratio of KPI-APP to that of APP-695, compared to age matched non-AD brains.

The small fragments that make up the beta protein, usually consists of 40 amino acids (ab40), there are however larger species of 42–43 amino acids (ab42/43) that contribute ~ 10% of total amyloid beta-protein. In a study of APP-695 it was found that 28 of these lie just outside the membrane-spanning domain of the precursor protein.[8] This presents a conundrum with regards to how can the enzymes that cut amyloid beta protein out of its larger precursor, gain access to the transmembrain region.

It is thought that a normal function of beta- APP is to act as an inhibitory molecule that regulates the activity of protease enzymes. Alternative sequences of DNA have been found, that contains either one or two extra coding segments at position 289 of the original 695 amino acid and one of these encodes a stretch that has the ability to bind to and inhibit proteases. Such protease, are responsible for cutting protein into smaller fragments however, it has been found that the normal fragmentation of beta-APP occurred at amino acid 16, within the amyloid beta protein region. Consequently, this suggests that the beta-amyloid deposition that occurs in AD must utilise an alternative proteolytic pathway, which results in the beta-APP being cut at the beginning and end of the amyloid protein region.

Genetic defects in APP are however present at birth and this presents a puzzle as to why it takes at least 40 years for AD to develop. It may possibly be that the disease merely reflects an accumulation over the years, which has reached a pathological level. Alternatively it may be that older cells respond differently to younger cells and therefore the ageing mechanism, such as changes to gene expression need to be considered.

There is strong evidence to suggest that the SP is not actively produced by damaged neurites but originates from a more amorphous nonfilamented deposit, that has been found in selected blood vessels of the skin, intestine and certain other tissues. [8] These diffuse or pre-amyloid plaques are found not only in the brain areas, such as the cerebral cortex that are implicated in the symptoms of AD but also in the thalamus and cerebellum and occur in increasing numbers during normal ageing. However they contain few or no degenerating neurites or reactive glial cells and much of the tissue within the plaques is indistinguishable from the surrounding normal brain tissue. Hence the presence of these diffuse plaques that are also connected to many non-neural sites, suggest that the deposition of amyloid beta protein precede any alteration of neurons and other brain cells. For some unknown reason the maturation of these diffuse plaques into dense SP occurs much more readily in the cerebral cortex, rather than in the symptom free cerebellum. The diffuse plaques attract other proteins to it and they begin to have both trophic and toxic effects on the surrounding axons and dendrites which result in them evolving into the neuritic form that is seen in the hippocampus and cerebral cortex.

Support for this hypothesis comes from a study of patients who suffer from a rare affliction called "hereditary cerebral haemorrhage with amyloidosis of the Dutch type". Patients in these afflicted families die in midlife from cerebral haemorrhages caused by sever amyloid deposition in innumerable blood vessels. This particular disease has been traced to a common mutation in the beta-APP gene that causes the substitution of the amino acid glutamine for a glutamic acid at position 22 within the amyloid beta protein. This therefore provides strong indications for the importance of the genetic factors in the development of AD which will be considered next.  

1. Genetics

It is plausible that the normal age related changes in gene expression could initiate and promote the course of AD. Both RNA and protein synthesis decline with age and mutations in somatic cell DNA ultimately result in cellular dysfunction. A decrease in neuronal mRNA with age may however simply reflect the reduced need of the cognitive protein by the atrophied cells rather than being an underlying cause of cell atrophy during ageing. However it is the mitochondrial DNA that is more sensitive to mutations due to the small genome, a paucity in DNA repair function compared to nuclear DNA and an environment where ROS (e.g. superoxide and hydrogen peroxide) are by-products of the mitochondrial physiology of energy release. All forms of mitochondrial DNA damage have been found to accumulate with ageing.[7]

Certain genes however, are associated as having a specific proclivity for AD. The rare familial forms of AD that produce an early-onset of the disease have been associated with three different genetic defects found on chromosomes 1, 14, and 21.[9] Also another gene found on chromosome 19 is thought to play a role in the more common late-onset cases.

The gene on chromosome 21 was first to be identified, although it is only linked to 2 to 3 per cent of all early familial cases. It was significant however, because such an abnormal chromosome (and indeed an extra copy) occurs in patients with Down syndrome, virtually all of who suffer AD, if they live to an age of 35. The defective gene codes for the Amyloid Precursor Protein (APP) and is believed to result in an abnormal cleavage that increases the production and deposition of amyloid beta-protein. However studies have lead to the conclusion that familial AD is genetically heterogeneous and can be caused by many different genetic defects even upon the same chromosome. A total of 6 missense mutations have so far been found on the APP gene, located on chromosome 21, all of which lead to AD. Several families have been identified in which everyone who had AD were found to have a mutation that resulted in a switch of amino acid 642 from a valine to an isoleucine.[8]

A further possibility is that other defects on chromosome 21 can result in either abnormal forms of beta-APP or deregulation of the transcription of the beta-APP gene. In this latter case the DNA that is altered, is that which controls how much or in what form the mRNA is transcribed from the beta-APP gene. In this way DNA defects outside the protein coding region for beta APP may enhance the amount or type of beta APP that is manufactured. As a consequence of this plethora of beta APP, the cells may responds by employing an additional enzymatic pathway, which liberates large fragments that contain the amyloid beta protein (as mentioned in the previous section on SP).

However 70 to 80 % of the early–onset disease have a genetic mutation in chromosome 14, while the remainder have a defect in chromosome 1. At least 41 different mutations have been found in the presenelin-1 gene on chromosome 14 and these account for 30-40 % of presenile AD families. All of these mutations, except one, are missense mutations. Mutations of the presenelin-2 gene on chromosome 1, are much rarer causes of early-onset familial AD (~2%). Two missense mutations with incomplete penetrance have been found so far.[10]

The gene on chromosome 19 codes for apolipoprotein E which is involved in cholesterol transport and metabolism and there is evidence indicating a slight age related increase in such mRNA. Three forms of the allele exist and the presence of one of these (Apo-E4) appears to increase the deposition of amyloid beta- protein in the brain and may also increase the number of NFT. If one of these Apo-E4 is inherited, the risk of developing AD is about four times greater than if you had another allele, and the risk is even greater if you inherit two copies (one from each parent). Conversely the Appo-E2 allele is lower in AD cases than in controls [fig 3]

Also recent evidence has accumulated for an association of an allele (HLA-A2) on chromosome 6 with both late and early-onset AD. Finally on chromosome 12, the gene coding for the low-density lipoprotein receptor-related protein (which is the Apo-E receptor), may be correlated with late-onset AD.

Genetic studies comparing the risk for identical twins and fraternal twins do indicate a genetic influence in AD. The inherited forms of the disease appear to act through a common mechanism, which elevates the level of amyloid beta-protein. However there is no epidemiological data which demonstrates that the sporadic forms which account for > 90% of all AD is caused by a similar proliferation of this beta-protein and environmental factors no doubt play an important role.

2.   Microglia and Astrocyte Activation

Microglial cells are the main components of the brain’s resident immune system, while astrocytes aid in the repair of neurons following brain damage. Senile plaques are also known to be containing microglial cells at their centre, while acrocyte cells accumulate around the outside of the plaques. It is known that astrocytes hypertrophy during ageing and also microglia becomes increasingly reactive or activated during ageing.[7] These morphological changes that occur during ageing, resemble those that are found in the brain after injury.

Activated microglia overexpressing interleukin-1 (IL-1) and activated astrocytes, overexpressing S100b have been implicated in the formation and evolution of tau2-immunoreactive neuritic plaques and neurofibrillary tangles in AD. One such study [11], utilises four distinct classifications of NFT, namely neurons with granular perikaryal tau2-immunoreactivity (stage 0); fibrillar neuronal inclusions (stage 1); dense, neuronal soma-filling inclusions (stage2); and acellular, fibrillar deposits (stage 3, "ghost tangles"). It was found that there was a progressive increase in frequency of association between the tangle stages and the number of 1L-1 alpha+ microglia and S100B+ astrocytes (r = 0.72, p = 0.02; r = 0.73, p = 0.01 respectively), in randomly selected fields of the parahippocampal cortex. It has therefore been proposed that the activation of these glial cells could contribute to neuronal degeneration and tangle progression in AD. Such a dual overexpression of the immune system may also contribute to the transformation of amyloid deposits into neuritic beta-amyloid plaques. Hence both these glial inflammatory responses are thought to participate in the appearance and progression of both of the neuropathological features of AD, namely NS and NFT . So although probably not being a primary event, inflammation mechanisms do play a key role.

The progressive association of microglia, overexpressing IL-1 with the evolution of NFT formation, suggests that glial-neural interactions precede neuronal cell death and are not merely responding to extracellular necrotic cell remnants. It seems plausible that these microglia are responding to neural distress that is associated with the early stages of tangle formation and the IL-1 in turn, is known to attract and activate astrocytes, inducing them to produce S100b. These IL-1 based actions, together with the observation that there is a progressive association of S100b+ astrocytes with progressive stages of tau2+ tangle formation, suggests that such astrocytes were attracted and activated by microglia-derived IL-1.

This scenario is supported by observations that "the frequency of tangle-astrocyte association lags behind that of tangle-microglia association through the progressive stages of tangle formation."[11 ]

The attraction and subsequent activation of microglia overexpressing IL-1 and of astrocytes overexpressing S100b to tau2+ tangle bearing neurons, can have both neurotrophic and neurotoxic consequences. Although IL-1 assists neural survival, it does become toxic at high levels. Also S100b is neurotrophic at low levels, improving neuronal survival and growth processes, however it is potentially neurotoxic at high levels, due to its ability to elevate intraneuronal free calcium levels. This in turn may further promote the formation of abnormal phosphorylation of tau protein. In summary therefore, there is an important pathological role played by both microglia and astrocytes together with their cytokines, in both the evolution of amyloid deposits into neuritic plaques and the formation of NFT that are diagnostic of AD.

3. Alteration of the Glutamate Transporter System

Glutamate is the main excitatory neurotransmitter in the mammalian central nervous system and is usually removed from the synaptic cleft by high affinity, Na+ - dependent uptake transporters that are found in both neurons and glia. Some studies of the cortex and hippocampus have shown that deficient functioning of the glutamate transport system might lead to the neurodegeneration that results in AD. For example it is believed that damage to the GluR2 receptor subunit could make the neuron vulnerable due to destabilisation of the Ca2+ homeostasis. There is some data that supports the notion that GluR2 receptor subunit is lost prior to development of NFT. Also, the glutamate transporters themselves, are thought to be critical in preventing extracellular accumulation of potentially neurotoxic chemicals.

Of the four types of glutamate transporters (GT) that were studied [12], only one (known as EAAT2) was found to be deficient in the frontal cortex in cases of AD. This particular GT is specifically located in astrocyte cells. It is believed that alteration in GT expression occurs due to disturbance at the post-transcription level, since EAAT2-immunoreactivity inversely correlated with the mRNA levels of the GT. Furthermore GT levels were found to be directly correlated with APP695mRNA, which supports the notion that EAAT2 is affected by the abnormal processing or functioning of APP in AD.

There is also evidence to suggest that the inefficiency of GT, leads to an accumulation of excessive neurotransmitter in the synapse, resulting in subsequent neurotoxicity. For example experimentally induced blockade of GTs causes neuronal death, while in case studies of AD, the increased level of brain spectrin degradation products (a marker of excitotoxicity) has been correlated with a decrease in levels of D-[3H] asperate binding (a marker for GT activity). Furthermore in the disease amyotrophic lateral sclerosis, there is a selective loss of GT (type EAAT2) in the motor cortex, while in AD itself, the GT system is decreased by 40 to 50 % in the frontal, parietal and temporal lobes. Hence there is good reason to believe that abnormal functioning of the GT system might be involved in the pathogenesis of the synaptic damage that occurs in AD.

In one study 3 potential mechanisms for the inhibition of glutamate uptake have been proposed.[12] One of these being the inhibition of Na+, K – ATPase, another being the direct oxidative damage to the GT molecule and the third being membrane perturbations by lipid peroxidation. Of these, that of the direct inhibition of the GT is considered to be the most likely, although the direct effects of amyloid beta- protein might also play an important role. However since the greater proportion of astrocytes (that produce the EAAT2) are not in close contact with dense amyloid deposits but are instead diffusely scattered throughout the neuropil, it is unlikely that this is the only mechanism involved. In any case there is significant evidence to support the view that astrogial EAAT is affected in the AD cortex and that abnormal processing and/or functioning of APP might play an important role and that this decreased activity of the GT is associated with excitotoxicity and neurodegeneration.

Another avenue of research has however concentrated on the actual loss of the GluR2(3) receptor sub-unit.[13] Receptors that lack this particular subunit are highly permeable for calcium and it has been suggested that neuronal degeneration in AD may result from a loss of intracellular calcium homeostasis. An overactivation of calcium- permeable channels that are gated by AMPA receptors (of which GluR2 is a subunit) is known to produce glutamate mediated excitotoxicity. When these excitotoxins are applied to human neurons grown in culture, they can induce the formation of paired helical filaments, which are the major constituent of NFT in AD. This conversion of tau protein into the hyperphosphorylated tau, that comprises these neural tangles, is believed to be due to dysfunctions of protein kineses and/or phosphatases that are able to phosphorylate or dephosphorylate respectively, the tau protein. It has therefore been speculated that activation of the glutamate receptors, induces an influx of Ca2+ that in turn either activates specific protein kineses or deactivates phosphatases (or both), to produce tau protein.

However, other mechanisms have been proposed which can cause Ca2+ destabilisation, such as altered gene dosage of superoxide dismutase, which may

increase vulnerability to oxidative injury. Whereas such genetic abnormalities alone

may not affect a young person, whose brain is able to produce neurotrophic factors that oppose such Ca2+ destabilisation, it could be influential when combined with age related alterations of the brain or physical trauma.[14 ] 

4. Cytoskeletal Destabilisation

Studies in human brain autopsies and examinations of primates have shown that dystrophic neurites precede deposition of amyloid in the formation of SP. One proposed scenario [15] attempts to replace the dominance of amyloid deposition, with that of a mechanism that results from a destabilisation of the cytoskeleton, which may be associated with a genetic trait. However this is a less popular theory, since such neuritic changes, similar to those in SP, can be detected in the aged brain in certain locations completely independent of amyloid,[3] such as in the cortex of the limbic lobe and in the amygdala. [Other examples of such sites include the dorsal column nuclei in the lower medulla, the pars reticularis of the substantia nigra and the ventral pallidum.] The following description may not therefore be the actual chain of events that predominate in AD but it does indicate the various possible interactions, that can contribute to the pathological consequences of the disease.

Alzheimer’s destabilisation of the cytoskeleton is observed in the microtubules within neurons and in the paired helical filaments (PHF), which consist of hyperphosphorylated tau proteins. Normal tau protein is essential for the stability of tubules and such destabilisation of the microtubular system in the perikaryon, in turn causes the fragmentation of the Golgi apparatus. This is responsible for assembly and/or packaging of proteins. [Such proteins become enclosed in a vesicle, which can then be moved safely to the cell where they are to be used.] The dispersion of the Golgi would therefore be expected to have an effect on processing of APP and such abnormalities could produce excessive amounts of amyloid beta-protein, leading to its deposition in the extracellular space.

Cytoskeletal destabilisation of the microtubule system would also affect axoplasmic flow, which is dependent on the tubules as a track, along which kinesin acts as the motor. This in turn would lead to a diminished supply of substrate to the distant terminals especially the axon but also in dendrites. The resultant distrophic neurites would then lose their ability to synapse. One effect of this would be to reduce the retrieval of trophic factors from the target area for retrograde transmission, because of the deficient flow system back to the neuronal perikaryon. This would eventually result in the cell death, probably by the apoptotic program.[Fig 4] Degenerating synapses would also activate microglia in the neuropil between plaques. These activated microglia then release lytic cytokines which add to the synaptic destruction as previously described in section 2. This loss of synapse therefore causes topographic disconnections, producing the syndrome that is characteristic of dementia. The loss of neuronal perikarya and the transmitting apparatus lead to a deficiency of neurotransmitter that was first observed in the cholinergic system but has subsequently been recognised in several others.[15]  

5. Neurotransmitters

Since nerve cells synthesise the neurotransmitters necessary for synaptic communication, it is not surprising that AD is associated with diminished levels of these chemicals. In addition to effects upon the glutamate transport system dealt with above, a consistent feature of the disease is the degeneration of the cholinergic innervation of the forebrain by neurons in the nucleus basalis. The loss of these cholinergic projections in the cerebral cortex, is the first sign to appear and the magnitude of the cholinergic loss is a good predictor of clinical deterioration. Indeed in the mid-1970’s, Alzheimer’s was blamed on a brain deficit of acetylcholine, a neurotransmitter considered to play an important role in short-term memory. However drugs that were designed to enhance acetylcholine were subsequently shown to be merely palliatives.

It is known that AD affects many other types of neurons and neurotransmitters, which is one of the reasons why the illness is difficult to treat simply by replacing the deficit neurotransmitters. Other examples of diminished levels of neurotransmitters include norepinephine and serotonin, as well as modulatory neuropeptide molecules that transmit signals between nerve cells. Drugs that enhance the action of glutamate (the chief neurotransmitter of the brain) appear to improve patient’s short-term memory .

Other studies [16] have observed that at least in the hippocampus there was a lowering of the presynaptic vesicle protein synaptophysin which correlated with cognitive decline in AD. The hippocampus has a well established role in memory and learning and is particularly vulnerable to AD. However no significant association has been found in the cortex regions which are also susceptible.

Environmental Factors

In addition to the features consider in sections I and II above it is also important to pursue epidemiological surveys, which might reveal environmental factors that are also likely to influence the onset of AD. There have been some indications that vegetarians have a lower susceptibility but at present there is no compelling evidence that factors such as nutrition, occupation or emotional state can influence the onset of AD. One factor that has been noted in a small minority of cases, is a history of major head injuries,[8] although how trauma can accelerate the deposition of beta-amyloid is unclear. [Boxers carrying the Appo-E4 allele are at a higher risk of developing dementia pugilistica.]

Abnormal concentrations of aluminium have also been found to accumulate in NFT and SP but it is not known if the element plays a causative role in AD. Aluminium has no biological function in the body (not even as a trace element) but in recent years it has become more prominent in the human diet. Its increased use as a cooking utensil has been cited as one possible cause, while it is also an important ingredient in the purification process of water. The fact that many cases of AD occur sporadically, without any known familial predisposition, does suggest that environmental factors in general may influence the onset of the disease. It has also been observed that identical twins may manifest AD at considerably different ages. Unfortunately the search for a causal link with environmental factors such as aluminium have been somewhat unsatisfactory. 

Attempts to understand the biochemical events underlying the structural changes that take place in AD, have lead to a wide diversity of proposals. The main avenues of investigation revolve around attempts to locate relevant genes and the study of behavioural and neurological symptoms. Starting with the most consistent and specific abnormalities, researchers attempt to trace their development backwards, with the aim of identifying molecular change or genetic traits that may be involved in the development of the disease. NFT and SP are the two hallmarks of AD but it is not known whether either of them actually cause the disease or are merely by-products of such neurodegeneration, since other ‘suspects’ are also found at the scene of the crime. Indeed NFT and SP will occur to some extent in most people in their late 70’s and the distinction between normal brain ageing and AD is consequently quantitative rather than qualitative. Also, an excess of these features is also common to a large number of other brain diseases.

It is not even known what changes occur with age that permits the expression of AD. However, mitochondrial function declines with age (due to an accumulation of mutations in their DNA) and the resulting impaired energy metabolism could compromise the ability to maintain cellular homeostasis in response to toxic insult. [A threefold increase in damage to mitochondrial DNA in AD has been reported].

Like many degenerative diseases, AD is no doubt connected with an abnormal level of activity of certain natural metabolic processes and susceptibility is associated with genetic factors. Many of these abnormalities have been identified above, although it is difficult at present to determine which exert the most primary influence. However the outcomes of such research do suggest possible strategies that might be employed to combat the disease.

Two of the reoccurring themes are the roles played by amyloid beta–protein in SP and that of the paired helical filaments that result in the formation of NFT. Hence therapeutic treatment could revolve around inhibiting the hyperphosphorylation of tau protein or blocking delivery to the cerebrum of those beta-APP that are responsible for producing amyloid deposits (providing of course that those proteins do actually arrive predominantly via the bloodstream). Alternatively the protease that liberates the amyloid beta-protein could possibly be inhibited in some way. Another approach would be to retard the apparent maturation of the amyloid beta-proteins into dense neuritic plaques, by interfering with the formation of the amyloid filaments that seem to accompany this change.

It might also be beneficial to interfere with the activities of the microglia, astrocytes and other cells that contribute to the chronic inflammation around the neuritic plaques. Finally one could try to block the action of the molecules on the surface of neurons that mediate the trophic and toxic effects of amyloid beta-protein and other proteins associated with it in the plaques

Research may even reveal a more underlying mechanism behind AD that would illuminate our understanding of associated features, such as the deterioration of the glutamate transport system and cytoskelatal destabilisation. At present however there are many contending theories as to which is the fundamental cause of AD and which are secondary effects and it may take further developments in the field of genetics, before an underlying mechanism is revealed and a suitable cure found. The tragic nature of the disease and the fact that AD affects the most rapidly growing portion of the population, gives added impetus to the search for a cure.

Metcalfe J; The Brain Degeneration, Damage and Disorder: Springer

1. J. Beaty J, Principles of Behavioural Neuroscience: Brown & Benchmark
2. Dickson D.W; The Pathogenesis of Senile Plaques: The Journal of Neuropathology and Experimental Neurology,(April, 1997)
3. J.Pei P, Tanaka T.O, .Tung Y, Braak E, et al; Distribution, Levels,and Activity of Glycogen Synthase Kinase-3 in the Alzheimer’s diseased Brain: The Journal of Neuropathology and Experimental Neurology ( January, 1997)
4. Goldman B; Alzheimer’s Disease: The Tangled challenge. Neuroscience Part II .http://www.signalsmag.com/signalsman.n
5. Selkoe D. J; Physiological production of the beta-amyloid protein and the mechanism of Alzheimer’s disease, TINS, Vol.16, No.10 pp403-409 (1993)
6. Johnson S.A, Finch C.E; Changes in gene expression during brain ageing: a survey. Handbook of the Biology of Ageing, pp300-327. Academic Press (1996)
7. Selkoe D.J; Amyloid Protein and Alzheimer’s Disease: Scientific American (November 1991)
8. Hardy J; Amyloid, the preselins and Alzheimer’s disease, TINS,Vol.20,pp154-159 (1997)
9. Yliopisto K; Molecular Genetics of Alzheimer’s Disease http;//www.fi/laitokset/neuro/44the.html
10. Sheng J.G, Mrak R.E & Griffin W.S; Glial-neuronal Interactions in Alzheimer’s Disease: Progressive Association of IL-1 alpha+ Microglia and S100b+ Astrocytes with neurofibrillary Tangle: The Journal of Neuropathology and Experimental Neurology,(March, 1997)
11. Li S, Mallory M, Alford M, Tanaka S, Masliah E; Glutamate Transporter Alteration in Alzheimer’s Disease is Possibly Associated With Abnormal APP Expression: The Journal of Neuropathology and Experimental Neurology,(August, 1997)
12. Ikonomovic M.D, Mizukami K et al; Loss of GLuR2(3) Immunoreactivity Precedes Neurofibrillary Tangle Formation In The Entorhinal Sortex and Hippocampus of Alzheimer’s Disease: The Journal of Neuropathology and Experimental Neurology,(September, 1997)
13. Mattson M.P, Barger S.W, et al; Beta-APP metabolites and loss of neuronal Ca2+ homeostasis in Alzheimer’s disease; TINS, Vol.16 No.10 pp 409-414.(1993)
14. Terry R.D; The Pathogenesis of Alzheimer’s Disease: An Alternative to the Amyloid Hypothesis: The Journal of Neuropathology and Experimental Neurology
15. Sze C.I, Troncoso J.C, Laudia C.L et al; Loss of the Presynaptic Vesicle Protein Synaptophysin in the Hippocampus Correlates with Cognitive Decline in Alzheimer’s Disease: The Journal of Neuropathology and Experimental Neurology,(August, 1997)

Fig 1 &2. Beaty J; Principles of behavioural neuroscience, pp 489,490. Pub: Brown & Benchmark

Fig3. William J, Ashal F, Goate A M; Molecular pathogenesis of sporadic and familial forms of Alzheimers disease: Molecular Medicine Today,(April 1998)

ADDICTION

Some of the problems that have been associated with the use of the word 'addiction' stem from an inadequacy in its definition. Attempts to simplify its meaning, have at the same time restricted its use (e.g. to drugs) and this has lead to difficulties in understanding the nature of other forms of compulsion or devotion. Difficulties also arise due to the variety of ways by which psychologists and physiologists try to explain the phenomena. Some would contend that the best way to understand addiction, is to study the underlying physiological/neurological activities that produce addiction, whereas others stress the importance of behavioural conditions that lead to addiction. Attempts that have been made to specify the nature of one type of addiction (e.g. heroin), often become inappropriate to use when applied to another form e.g. addiction to shopping.

. Even amongst drugs themselves, some can produce a severe physical dependency (e.g. barbiturates), while others (e.g. cannabis), although not exerting such physiological effects, may be psychologically addictive ('habitual'). In response to this many authorities refer specifically to drug dependence to encapsulate both the terms drug addiction and drug habituation.[1]

DISCUSSION

The word itself derives from the Latin verb addico meaning 'giving over' and can therefore be used in a negative or positive sense. This has therefore led to an ambiguity in the traditional meaning of the word, since an addiction can either be tragic or enviable or in between the two. The more modern association of the word, emerged in the 19th century "as a result of a transformation of social thought"[2] and links addiction with the harmful effects of drugs. This restrictive meaning, did not therefore grow out of scientific or medical discoveries but was born out of the rhetoric of contemporary movements that targeted alcohol and opium and is not therefore as useful as the traditional concept. Such a restrictive connotation, has led to problems with the use of the word, especially in areas of psychology in which addiction may not be associated with drug use. Examples of this would include addiction to food (especially chocolate), gambling, shopping, 'love', television, the Internet or even work. [One could also consider the 'stalker' who is not addicted to a personality due to any chemical dependence but rather the route cause of the problem is psychiatric.] These addictions would therefore have a predominantly behavioural interpretation, compared to the craving for drugs such as opiates or amphetamines, which have a strong chemically induced addiction.

In the context of drug addiction there are four major determinants, namely tolerance, sensitization, withdrawal and craving and the fact that it is difficult to gauge the importance of their contribution in general addiction, has in itself led to a rather imprecise usage of the word. Addiction in the restricted sense, is viewed as a persisting susceptibility to relapse, which is associated with withdrawal symptoms and a sensitization to 'wanting'. However the need for the more traditional use of the word has become apparent, since the word addiction itself can have a positive or at least neutral effect and even drugs need not always be associated with withdrawal. Indeed until relatively recently, withdrawal was assumed to be a defining feature of addiction. However this symptom is neither necessary nor sufficient, since cocaine addiction for example, has no obvious withdrawal symptoms, while opiates lead to withdrawal but not necessary addiction.[3]

In addition, the subjective effects of a drug make it difficult to judge the significance and reliability of an addicts experience. An individuals vulnerability to addiction can be influenced by many factors such as genetic makeup, social background or even the environment (e.g. the prevalence of marijuana during the Vietnam war). A particular personality type may even lead to a drug preference, depending on the mood enhancing effect of that drug. All of this has therefore led to addiction being described differently in different disciplines and being applied differently at "the molecular, neural, psychological, behavioural and sociological levels".[4]

The proliferation in the number of substances that are classified as drugs (and the anomalies in their effects), has also caused the homogeneity of addictions to break down. Psychedelic drugs do not create physical addiction, while the use of cannabis in particular, does not even result in tolerance but rather the opposite, in that regular users come to require less in order to achieve the same effect.[5] However there may develop a psychological dependency for such drugs, which can vary in intensity between individuals. [A person may become sensitized to ‘wanting’ a drug even if they don’t particularly like it.] Defining such an addiction has its problems however, since the psychological state that is used to describe such addictions, might equally occur in relation to a whole variety of things such as music, religion and sex. The question then arises as to "whether this dependency is physically, socially or psychologically harmful."[6] There are some reports for example, of athletes being 'addicted' to their sport as a result of the euphoric experience produced by the release of endorphins during exercise.

As a result of such considerations, since the 1965 report of the World Health Organization committee, the words addiction and habituation in relation to drugs, have been replaced by the terms physical and psychic dependency, even though the use of the latter is subject to a good deal of disagreement. Physical dependence (drug addiction) is often characterised by the development of tolerance and of a withdrawal syndrome after the drug's effects have worn off. Psychological dependency or 'habituation' is present when the compulsion to take a drug is strong, even in the absence of physical withdrawal symptoms.

Although certain operational definitions are useful for the clinical study of drug addiction and provide a good starting point for neurobiological and psychopharmacological investigations, they may not be suitable for sociological, psychological and behavioural investigations of addiction. The so called 'disease model' can explain the compulsive features of addiction but not occasions in which price and punishment reduce consumption. Conversely the 'learning model' of addiction can account for these influences but not compulsive drug taking. The fact that such drug taking behaviour can be 'compulsive' yet require planning, has led to long standing debates as to whether addiction is best classified as an involuntary state, a disease or a preference.[7 ]

CONCLUSION

The use of the word addiction has therefore changed over the years, sometimes in response to social changes but most recently due to advancement in scientific and medical understanding. In particular it has been realised that symptoms such as loss of control, preoccupation and compulsion, are common to many recreational activities outside of drug abuse (e.g. eating, gambling), and this has led to problems in using the restricted meaning of the word. There are various degrees and types of addiction and the underlying nature and cause can vary widely from a psychological to a physical dependency. This ambiguity in relation to its use over a wide variety of chronic relapse conditions, has therefore led to a return to the traditional sense of the word, in which addiction is associated with 'giving over' or devotion.

Such usage recognises the more widespread nature of addiction but although being more useful in terms of diagnosis, it also diversifies the types of treatment available (the correct choice of which is critical in reducing the attrition rate). The nature of addiction can vary from one individual to another and it is difficult to disentangle the psychological, physiological and cultural factors. Behind every addiction there is a reason and there may be as many reasons as there are victims – even the same reason can be sparked off by different events. The word itself, can therefore be used in so many different ways, for different reasons by so many different people, that it is difficult for one definition to embrace all the medical, psychiatric, cultural, ethical, sociological and legal considerations that have an important bearing on the word 'addiction'.

References:

1 Alcohol and Drug Consumption: Encyclopaedia Britannica 1996

2 Defining "Addiction": Bruce K. Alexander and Anton R.F.Schweighofer. Canadian Psychology,Vol 29, pp151-162

3 & 7 Resolving The Contradictions Of Addiction: Gene M.Heyman. Harvard University. gmh@ wjh12.harvard.edu

4 The biological, social and clinical bases of drug addiction: commentary and debate. J. Alterman, B.J.Everitt,S.Glautier, et al. Psychopharmacology, Vol 125, pp285-345.

5 & 6 Psychedelic Drugs: Psychological, Medical and Social Issues. Brian Wells. Penguin Educational. pp30 & pp102

Opiate Addiction: The Case For An Adaptive Orientation. Bruce K. Alexander and Patricia F. Hadaway. Psychological Bulletin, Vol92, pp367-381

Reductionism versus Holistic Approaches in Biology

*** Synopsis of the debate ***

As we reach the end of the millennium, one thing we can all agree upon is the enormous impact science has had upon civilisation. It was during the Enlightenment that people realised that scientific knowledge meant power and such reductionism became regarded as a superior form of knowledge. Indeed there is a strong argument to suggest that the rigour of scientific knowledge has much to offer other disciplines of intellectual activity. Many new found disciplines have applied the suffix ‘science’ to enhance their credibility in the academic community (e.g. political science, social science, management science), -- imitation being the sincerest form of flattery!

However a predominantly reductionist approach, may not always be suitable for other more integrated, holistic based disciplines. Recently certain eminent scholars have openly criticised the program of reductionism, even in the sphere of science. In what follows, I will be consider one longstanding debate between two prominent biologists and asking if the reductionist approach is beginning to exhibit signs of limitations and if so, what are the implications for science.

Far from being just a spectacle of two eccentric academics, indulging in an abstract debate, what is at stake is the very direction in which future medical/biological research will evolve during the next century. On the one hand we have the view exposed by Wolpert, that reductionism isolates the key components making experiments possible and that the holistic approach is the enemy of science.[1] Rose however emphasises the importance of structural unity when trying to understand organisms and believes that the Cartesian framework of science lacks viability when applied to certain biological systems. To him the laws of physics and the reductionist approach are useful for understanding a machine in which the various components are made independently and then assembled. However when considering an organism, composition rarely explains form and one has to adopt a holistic approach in which new properties emerge when the biological systems are observed at a higher stratum. Such epiphenomena, which are so typical in life sciences, cannot be catered for in the reductionist approach, whose ‘genetic imperialism’ Rose fears may dominate biological sciences. Reductionism may explain how an automobile works, but it cannot predict where the car will be driven, since although it can describe, it is unable to explain the totality of the organism such as man.

Wolpert retaliates by emphasising that even the most complex of properties cannot contradict the laws of physics and therefore must be capable of being taken apart and analysed at a lower level without loss of understanding. Both agree upon the need to proceed with research into the understanding of genes (eg the genome project) and to apply the right level of explanation. Rose however attaches considerable importance to the role of the environment in the way genetic traits are ‘interpreted’ and he justifies this through the action of the holistic properties of the biological system. He accepts Wolpert’s view that the components are responsible for many of the features at the next higher level but he also insists that this bottom to top approach must also include the emergence of new properties, which grow increasingly complex and exert a feedback influence on the lower level systems.[1]

Wolpert argues that many macroscopic effects in biology can be explained in terms of the latent power of genes and it is the cascade of their action that ultimately determines what we are.[Genes make enzymes, which make primary products, from which are made the secondary products.] Rose however believes that the genes are merely carried by the organism and it is the emergent properties of the organism, which determine which attributes (and therefore genes) will be selected. He emphasises that genes require the entire orchestrated metabolism of the cell in order to synthesise the proteins of life. Biological systems should not be totally explained in terms of "component molecules alone, but in terms of organising relations between them."[1] In his view we cannot allocate properties such as selfishness to a gene, anymore than we can give a colour or personality to an electron. These attributes are only applicable to a certain level and cannot be explained in terms of its components. Rose therefore claims that we need to work in a vertical as well as a horizontal direction and that the whole is often greater than the sum of the parts

Genes have already been associated with a propensity for obesity, cardiovascular disease and schizophrenia [2] and reductionists believe that this is the correct way forward, if we are to understand these and other degenerative diseases. However the holistic approach attaches much more emphasis upon factors such as post-transcription modification and the effect of the environment and contends the claim that genes totally code for such conditions as alcoholism and criminality.

Even though addictions to certain drugs can be correlated to certain genes, disciples of holism believe that environmental factors play a larger role. Especially in psychology,[3] they object to what they regard as excessive DNA worship and they do not regard genes as independent quasi-cognitive agents that pull the string of minds. Instead they are viewed merely as resources in a hierarchically regulated system. The nomethetic epistemology[4 ] of primary biological processes have limited appeal and they would instead invoke the use of secondary and tertiary modes of human action (hermeneutic and transformational epistemology), that are emergent properties.[Carl Jung was himself interested in the psychology that is incorporated into the Tarot cards, while Taoism has a similar code in its ‘I Ching’ – the book of changes.]

In contrast, the reductionist is against the dualistic model of psychology within which genes influence the body but not the mind – the latter being only influenced by culture and environment. Instead, they give priority to the lower level but would prefer to look upon the gene code, not as a ‘blueprint’[1] but as a program for development.[Somewhat analogous to a musical score, in which the harmony is strongly determined by the notes, although each time it is played there is room for a different interpretation.]

In the holistic approach, primacy is given to the higher level controls, rather than the lowest and the emergent properties that arise in biological systems produce new scientific laws, that supersede the less evolved controls of the components.[5] Hence although both Rose and Wolpert might agree upon the need to understand microdeterminism, where they differ is on whether things are determined exclusively from below, upwards, or whether downward causation is also operating. Rose believes that more importance needs to be given to this top-down, emergent causation, that is macrodeterministic and requires that biological systems be studied as a holistic entity, which unlike reductionism embraces non-additive effects

Much of the Rose/Wolpert debate can be viewed in the context of the biological sciences being divided on ideological rather than scientific grounds, with each personality disliking the idea of domination by the agenda of the other. Wolpert expresses fear that science will be subjugated to that of a social construct, while his opponents dislike the tendency to classify sociology and psychology as branches of biology, which can be considered a branch of chemistry, which in turn can be

explained in terms of physics. In fact, Social science has tended to ignore biological science and there has been reservations over applying biological reductionism, due to its association with racism in Nazi Germany.[6] On the other hand sociology has also suffered from the extreme environmentalism that accompanied the Lysenko regime in the Soviet Union.

The dichotomies between nature versus nurture, determinism vs. free will and realism vs. idealism, have preoccupied scholars throughout the ages and no doubt the debate over reductionism vs. holism will continue for many years to come, as each offers its own insight at the cutting edge of science. This century has witnessed many examples of where new approaches have had to be embraced and some of these give support to the holistic approach but not to the exclusion of reductionism. Godel’s incompleteness theorems show us that it is not always possible to completely understand a system without stepping outside it to a higher order system, and this may have some bearing on the nature of consciousness or other biologically based problems.

Even physics has had to embrace a somewhat holistic attitude (e.g. quantum theory, which persuaded Neils Bohr to choose the yin/ yang symbol as his coat of arms) and indeed the general theory of relativity is a non-local theory. Also chaos theory has revealed the indeterministic nature of even simple systems and as Goodwin states,[7] this gives rise to such systems apparently organising themselves and producing new emergent properties that cannot be predicted or understood at a simpler level.

What is needed therefore is a flexible approach, which is not influenced by preconceived ontological beliefs of atomism (reductionism) or mysticism (‘reality is one’). Rose and Wolpert can agree on ontological grounds that genes are the basic components of biological systems but differ on the epistemological approach that science should proceed with. A more eclectic epistemology could reconcile the various strands of biological research, irrespective of whether they tend to be holistic or reductionist, providing the results are judged on their own merit. [Even Einstein’s philosophical standpoint changed from that of critical positivism in his ‘youth’ to that of rational realism, as he extended his work from the special to general theory of relativity]

Both approaches are deterministic and rely upon theory construction and hypotheses testing, and as with most frontiers of science there will inevitably be contending opinions, until a particular theory advances sufficiently to establish a prevalent view – experiment being the arbitrator of truth. The way forward therefore lies not in adopting isolated stances in order to defend disciplinary boundaries. Instead the future lies in an integrated pluralism, which recognises the complementary nature of such diversity and the opportunity that this offers for constructive exchange

References:

1 Rose S, Walpert L; The Place of Genes: Prospect (December 1997)pp. 16 -19

2. Plomin R., Defries J, McClearn, G.E, Rutter, M; Behavioural Genetics, 3^rd edn. W.H.Freeman and co. New York Ch1
3. Richardson, K (1998) Gene Gods Ch 1 in The Origin of Human Potential; Evolution, Development and Psychology pp1-40
4. Stevens R (1998) Trimodal theory as a model for interrelating perspectives in psychology; Theory of social Psychology pp75-83
5. Sperry R W;(1987) Structures and significance of the consciousness revolution’ The journal of Mind and Behaviour 8, pp37-66
6. Weingart P (1997) General introduction, in Human Nature: Between Biology and the Social Sciences, pp 1-13
7. Goodwin b, Dawkins R (1995) What is an organism?; A discussion, from perspectives in Ethiology, vol. 2, Behavioural design, ed. N S Thompson,pp47-60

AGEING