NR AOEH
AU Brugere,H.; Banissi-Sabourdy,C.; Brugere-Picoux,J.
TI Electrochemical analysis of urine from sheep with scrapie and cows with BSE
QU Transmissible Spongiform Encephalopathies, A consultation on BSE with the Scientific Veterinary Committee of the Commission of the European Communities held in Brussels from 14 to 15 September 1993, European Commission Agriculture, S. 359-368
PT Article
AB Voltammetry is an electrochemical method that allows the characterisation of chemicals exhibiting oxidoreduction properties. It is the basis for many ways of electroanalysis. During the late 1980's, on an impulse by Professor R.Buvet, who made theoretical considerations to apply these methods in the field of medicine, a lot of work was performed to search for applications in the field of toxicology (isolation of drugs in urine) or in pathology. An impressive result was obtained in the field of ageing diseases, when a study including a initial group of 110 patients led to the observation that the voltammogram obtained with urine from patients with Alzheimer's disease had a peculiar peak that was not seen in people suffering from dementia's of vascular origin or in old people in good health. Therefore it was decided to search for possible anomalies of the voltammogram of urine taken from sheep with scrapie. A group of sheep with scrapie at the veterinary school was available for a trial, and it was possible to take urine samples in 16 diseased and 19 control animals. For practical reasons urine was taken by catheterisation of the urinary bladder. The main result of that first animal study was that the peak seen on the voltammograms of Alzheimer patients was also observed in the sheep. Electroanalytical criteria were sufficient to lead to the conclusion that the same compound is present in urine of Alzheimer's patients and in sheep with scrapie. A similar pattern was observed in the histograms of the scores of sheep and humans: there is some overlap between the diseased and control groups: the compound excreted in urine does not depend on an "all or none" process. When we met Ray Bradley in Paris he kindly agreed to collaborate by providing us with urine from BSE cows. The study was conducted fully by the "double blind" method: we received urine vials bearing only a number. The status of the cows was unknown to us. After electrochemical analysis of these samples, it is clear that the same chemical as was present in the sheep with scrapie and human patients with Alzheimer's disease is present in the urine of BSE cows. A statistical analysis of the results of this study shows that the predictive value of a positive result is generally good, but that of a negative result is less reliable. Therefore, there are now two reasons to keep an interest in the field of electrochemical analysis of urine: diagnostic application and possible progress in the physiopathology of neurodegenerative diseases.
VT
INTRODUCTION
Diagnosis of neurodegenerative diseases in humans and in animals still presents problems. At present, the ante-mortem diagnosis remains exclusively dependent on clinical observation. As a specific haematological or biochemical criterion is lacking, we have undertaken to search for a possible marker with a method adapted to reveal, in a unique sample, a great number of chemical substances.
METHODS
Basic principle
The method, voltamrnetry at a graphite paste electrode, is an application of the principles of oxido-reduction reactions: an oxidation is a loss of electrons, a reduction a gain of electrons.
In voltammetry, we use electric current as a reagent in a succession of oxido-reduction reactions. The effect of current on chemicals is the same as in a simple experiment of electrolysis: in a vat containing a watery solution, with two electrodes for electrolysis, the cathode provides electrons and induces reduction of the solute, the anode consumes electrons and induces oxidation. As an example, in an aqueous solution of iron salts, the reaction
Fe++ - e- <> Fe+++
is shifted to the right at the anode and to the left at the cathode.
Special features of cyclic voltammetry
To apply this principle in an analytical procedure, it is necessary to plot the curve i = f (V). A voltage with a linear increase (sweeping or scanning voltage) is applied at the electrodes, and the current intensity is measured. The curve i = f (V) presents a peculiar shape, a wave, with a slope at a maximum when the voltage is at the oxidoreduction potential's value. This method allows us to recognise the redox system present in the same solution (Bard, 1983).
But this simple method allows analysis only of mixtures of a few redox systems present in the solution. To detect substances present in biological fluids, more complex methods are required, for instance cyclic voltammetry. In this method, the linear sweeping voltage is switched alternately: current is successively anodic and cathodic, and the cycles can be repeated as often as necessary.
Figure 1
Typical graphs obtained with cyclic voltammetry. The differences of shapes are the consequence of the reversibility or irreversiblity of the oxido-reduction reactions. a: Quinone/hydroquinone, b: ascorbic acid, c: (MHPG) Methoxy-hydroxy-phenyl-glycol, (Banissi-Sabourdy, 1992).
In performing the first cycle, several events can occur, according to the possibility for the reactions to reverse:
The exchange of electrons between the two forms of salts of iron (Fe++ and Fe +++) is reversible. For that reason, the curves are symmetrical for the anodic and the cathodic sweeps. It is the same for organic molecules giving a reversible oxido-reduction reaction, as for example, the quinone/hydroquinone system (figure 1,a).
However, not all the compounds present in a complex solution, when oxidised, give rise to reversible reactions. If the compound obtained by oxidation is transformed to a compound that does not reduce, the cathodic sweep does not exhibit a peak. As an example ascorbic acid, after being oxidised, produces dehydroascorbate which is destroyed in an aqueous solution (figure 1 ,b).
If the compound does oxidise, and if its derivative is reducible, the cathodic sweep leads to a curve at a different voltage in comparison with the peak of oxidation of the native compound. This cathodic peak occurs generally at a lower potential. An example is uric acid, which gives a peak at 900 mV, and is transformed by the oxidation in alloxane. The cathodic sweep of alloxane produces a peak at a lower potential: 300 mV. Other substances behave the same way, for example Methoxy-hydroxy-phenyl-glycol (MHPG) (Figure 1,c).
A critical point is the choice of the electrodes. As it is the common way in electrochemistry, the device includes three electrodes, the active electrode, the reference, and a counter-electrode. The active electrode is made of graphite powder (Moutet, 1986, Banissi-Sabourdy, 1992, Banissi-Sabourdy et al,.1992). This choice offers the advantage of providing a large dimensional area of contact between the electrode (the particles of carbon) and the solution. The actual area of the active electrode reaches several square meters, and the mass of liquid to submit to electrolysis is spread at the surface of the powder particles as a thin layer film.
This situation is the opposite of conventional conditions, where the electrode has a small surface area (for example 1 square cm), and is dipped in a relatively large amount of fluid (several ml). In this case, the influence of the current on the solution is weak, since the phenomenon to be studied occurs in the molecular layer in front of the electrode.
With the graphite powder electrode, the compounds submitted to electric voltage are quickly consumed and disappear from the solution (Moutet, 1986, Banissi-Sabourdy, 1992, Banissi-Sabourdy et al,.1992).
The method is well suited to urine analysis, because urine is an aqueous solution, and it is possible to submit it to voltammetry direedy, without preliminary processing. Moreover, urine is in man the easiest body fluid to collect, and is a non-invasive method.
STUDY IN HUMANS
With voltammetry, physicians working in hospitals for elderly people showed that a group of Alzheimer patients presented a peak at the lower part of the curve: at 850 mV in the cathodic sweep. This peak was not observed in people suffering from dementia's of vascular origin (Planques, 1990, Planques et al., 1991). The values of the peak for the group of Alzheimer's patient and for normal people are different, but there is some overlap between the two groups.
The main question is to identify the molecule responsible for the "850 mV" peak. A chemical purification and a structural analysis is difficult and expensive. For these reasons, the next step was to define the electrochemical identity of neurotransmitters and metabolites. This must assist in the identification of the molecule. But after a long study (more than one hundred chemicals including neurotransmitters, metabolites, and neurotoxins were examined) the compound responsible for the peak "850 mV" is still unknown. For this reason it is considered that the compound is not a common neurotransmitter or a known metabolite. Concurrently, chemical determination is in progress.
It should be clearly stated that there are no established relationships between Alzheimer's disease (AD) and spongiform encephalopathies. In people, if we compare AD and Creutzfeldt Jakob disease (CID), symptoms may be similar, which creates difficulties in obtaining a clinical diagnosis. The clinical courses of the two diseases are, however, different. Generally clinical diagnosis before death can be made before the death of the patient taking account of the epidemiology, PrP gene sequence, and results of the electroencephalogram. The definitive diagnosis is obtained only after death, however, with microscopic examination of brain tissue, and use of histochemical reactions (for example protein Tau, beta 4 amyloid protein, for AD, and presence of CJD fibrils or PrPsc). With these methods, it has been possible to find that AD and CJD can simultaneously exist in the same patient (Chapon F., Dupuy B., Le Biez E. & Lechevalier B.,1993) However, even if these diseases are different, two points remain common:
- they are neurodegenerative diseases
- they lead to formation of amyloid deposits in the brain tissue.
The value of voltammetry of human urine to differentiate between AD and CJD is beeing explored.
STUDY IN SHEEP
Why use that electrochemical method in transmissible spongiform encephalopathies of animals?
One reason to apply the voltammetry in spongiform encephalopathies of animals is the lack of any laboratory test for disease or infection in the pre-clinical or clinical period. The aim of the study in sheep was not searching for the same peak as in AD. The hypothesis was that the scanning of urine for molecules identifiable by voltammetry could perhaps reveal another peculiarity, specific for transmissible spongiform encephalopathy (scrapie). The used sheep with suspect scrapie submitted to the veterinary school. Sixteen diseased animals were available. The control group comprised 19 normal sheep housed and fed identically. All were females, and urine was taken by catheterisation of the urinary bladder. After death, microscopic examination of the brain was undertaken to confirm the clinical diagnosis. The surprise of this study was that a peak "850 mV" was found in sheep with scrapie. These results were very similar to those of the initial study of AD patients, although there is no connection between the two diseases. Even with a small number of sheep, it was clear that the compound seen in AD was also present in urine of sheep with scrapie (Brugère H., Banissi-Sabourdy C., Brugère-Picoux J., Chatelain 3. & Buvet R., 1991).
STUDY IN COWS
After this study in sheep, many questions were raised: is the compound present in CJD also present in BSE?, and, furthermore, what is the identity of this compound?
CJD is very rare, but two urine samples from three patients exhibited the "850 mV" peak, the third sample being too dilute to give a correct curve (Banissi-Sabourdy, 1992).
In the field of BSE, an opportunity arose to collaborate with the Central Veterinary Laboratory (CVL), of the UK.
A protocol was agreed, to permit examination of urine of BSE cows with voltammetry. The urine was collected from 110 cows at the CVL. The samples were preserved from fermentation by acidification and deep freezing. The bottles were identified only with a number, so that the analysis was operated according to the "double blind" principle. The samples were imported in France with a special authorisafion of the Ministry of Agriculture and Forestry, and they were submitted to voltammetry, and the results sent to CVL The statistical analysis was performed at the CVL, when the status of the cows was established, i.e. after death and completion of histological examination of brain tissue, and in some cases after observation of scrapie associated fibrils. A detailed publication of this study is in preparation (Brugère H., Banissi-Sabourdy C., Brugère-Picoux J., Wilesmith J.W. & Bradley R.).
From the 110 cows, 148 samples were analysed, and 105 cows were included in the statistical analysis. In an overview of the results, it can be seen that a majority of cows with BSE were found to have the "850 mV" peak, which gives a confirmation that there is a peculiar compound in the urine in animals with spongiform encephalopathies.
With the prospect of using the method as an ante-mortem diagnosis Crable I), the evaluation of validity shows that the predictive value of a positive result is generally good, but that the predictive value of a negative value is less reliable. In table I the results from a previous evaluation (analysis of the first results, when 2/3 of the samples were known) are shown in 'A', and the results from the overall evaluation are shown in 'B'.
DISCUSSION
From the point of view of improving the accuracy of the method there are some problems to solve:
- the compound is a normal compound which exists at low level in some normal animals or people. The electrochemical method has not at present an accuracy that allows us to fix the minimum normal level
- the results of the electrochemical method form a continuous series of values, which are difficult to convert into positive and negative classes. The results with intermediate values are difficult to interpret and to classify
- following our experience of AD patients, we know that peaks of great amplitude can occur, followed some weeks or months later by a decrease and presumed elimination of the compound responsible. These spontaneous fluctuations limit the value of the test as an aid to diagnosis throughout the period of illness.
However, it is obvious that something occurs in patients with neurodegenerative disease, and that it is necessary to know the identity of the compound. Perhaps its identification could throw some light on the physiological processes involved in the disease, and also assist in the diagnosis of these diseases.
REFERENCES
Banissi-Sabourdy C., Planques B., David J.P., Jeannin C., Potel M., Bizien A., Di Menza C., Brugère-Picoux J., Brugère H., Chatelain J. & Buvet R. (1992). Electroanalytical characterization of Alzheimer's disease and ovine spongiform encephalopathy by repeated cyclic voltammetry at a capillary graphite paste electrode. Bioelectrochemistry & Bioenergetics, 28:127-147.
Banissi-Sabourdy C. (1992). Exploitation d'une technique d'analyse électrochimique, la voltampérométrie cyclique sur électrode à poudre de graphite, pour la détection de composés électroactifs normaux ou pathologiques dans les liquides biologiques. Th. Doct. Univ. Paris XII, décembre.
Bard J.A. & Faulkner L.R. (1983). Electrochimie : Principes, méthodes et applications, Masson Ed. Paris.
Brugère H., Banissi-Sabourdy C., Brugère-Picoux J., Chatelain J. & Buvet R (1991). Recherche d'un témoin biochimique urinaire de l'infection du mouton par la tremblante. Bull. Acad. Vét. de France, 64:139-145.
Brugère H., Banissi-Sabourdy C., Brugère-Picoux 3., Wilesmith J.W. & Bradley R.. Electroanalytical observations on urine from cows confirmed to have bovine spongiform encephalopathy and healthy controls (in preparation).
Chapon F., Dupuy B., Le Biez E. & Lechevalier B. (1993). Coexistence of CreutzfeldtJakob and Alzheimer's disease in a patient. In "Les précurseurs de la protéine amyloide dans le développement, le vieillissement et la maladie d'Alzheimer", Fondation IPSEN, Lyon, 21 juin.
Moutet M. (1986). Caractérisation et évaluation rapide des composés électrochimiquement oxydables dans les urines normales et pathologiques. Th. Doct. Univ. Paris XII, octobre.
Planques B., Banissi C., Jeannin C., Bizien A.& Buvet R. (1990). Caractérisation électroanalytique des démences de type Alzheimer. Actes du IVè Congits International de Gérontologie, Montréal, Edisem, St-Hyacinthe, Québec, et Masson, Paris, p.262-266.
Planques B. , Brugère H., Banissi-Sabourdy C., Jeannin C., David J.P., Di Menza C., Brugère-Picoux J., Chatelain J., Buvet R. (1991). Mise en évidence par microélectrolyse urinaire d'un marqueur biochimique des affections cérébrales dégénératives primaires humaines et des encéphalopathies spongiformes animales. Réunion francophone sur la maladie d'Alzheimer et les syndromes apparentés, Fondation IPSEN, Toulouse 22-23 nov.1991.
Table I
Statistical analysis of the results of the study in cows. The results at lines "A" were obtained in the course of the study, with 213 of the samples. The results at lines "B" include all cows.
SENSITIVITY SPECIFICITY PREDICTIVE VALUE OF A
% % POSITIVE NEGATIVE
RESULT % RESULT %
ANALYSIS 1
All samples included, uncertain results of voltammetry considered as negative
A 58.1 100 93.1 100
B 60.2 100 91.8 100
ANALYSIS 2
All samples included, uncertain results of voltammetry considered as positive
A 71 80 94 38.5
B 69.9 90.5 94.3 56.2
ANALYSIS 3
Only the last samples are included, uncertain results of voltammetry considered as negative
A 80.6 80 97 43
B 60.2 91.7 92.9 56.1
ANALYSIS 4
Only the last samples are included, uncertain results of voltammetry considered as positive
A 83.9 80 96.6 42.8
B 62.4 90.5 91.6 52.7
IN Das krankheitstypische Redoxsystem kommt auch im Urin anscheinend gesunder Menschen und Tiere vor und die gemessenen Werte bilden eher ein Kontinuum als zwei klar abgrenzbare Gruppen positiver bzw. negativer Resultate. Der Test erlaubt keine Unterscheidung zwischen Alzheimer und der Creutzfeldt-Jakob-Krankheit und die bei einem Patienten gemessenen Werte können sehr stark schwanken. Insgesamt deuten positive Resultate relativ zuverlässig auf Alzheimer, die Creutzfeldt-Jakob-Krankheit hin, er könnte aber durchaus auch auf weitere bisher nicht daraufhin untersuchte Krankheiten reagieren. Negative Testergebnisse sind überhaupt nicht zuverlässig und liegen kaum über Zufallstreffern.
AD
H. Brugère und J. Brugère-Picoux, Ecole Nationale Veterinaire, F94704-Maisons-Alfort, France
C. Banissi-Sabourdy, Université Paris-Val-de Marne, F94010-Créteil, France
SP englisch
OR Prion-Krankheiten 1 und Buch