WO1990012318A1

WO1990012318A1 - Assays using albumin-tetrazolium interaction

Info

Publication number: WO1990012318A1
Application number: PCT/GB1990/000552
Authority: WO
Inventors: Anthony Atkinson; Peter Hammond; Roger James Hinton; Julie Miller; Christopher Phillip Price
Original assignee: Public Health Laboratory Service Board
Priority date: 1989-04-12
Filing date: 1990-04-11
Publication date: 1990-10-18
Also published as: GB2250026B; AU5427990A; KR920701821A; CA2050884A1; GB9121010D0; GB2250026A; GB8908182D0; EP0467938A1; FI914812A0; JPH04504467A; HU903644D0

Abstract

A method for the reduction of a tetrazolium compound containing the cationic nucleus (I), to produce a ring - opened formazan compound containing the group (II), in which the tetrazolium compound is caused to interact with albumin and the tetrazolium compound - albumin ineraction product is reduced with a reducing agent in an aqueous solution at a pH between 6 and 10. The method is applied to the assay of albumin and reducing agents in samples, in which the amount of formazan produced is related to the presence and/or amount of albumin or reducing agent present.

Description

ASSAYS USING ALBUMIN - TETRAZOLIUM INTERACTION

This invention relates to an improved method for reduction of tetrazolium compounds to coloured formazans and to the use of this method in assays (ie detection and/or quantitative analysis) of analytes, in particular albumin.

Tetrazolium salts contain the cationic tetrazolium nucleus:

which is opened fcy reduction to produce a formazan containing the ring-opened group:

The formation of a formazan will produce a colour change, as the formazan group is a strong coloured chromophore, and tetrazolium salts are often colourless. Therefore the presence and/or amount of formazan produced may conveniently be detected visually or

colorimetrically, for example by observing the absorbance change in the region between 505 and 660 nm, especially around 590 nm.

Generally weak reducing agents can only reduce tetrazolium salts to formazans in the absence of proteinaceous materials. This colour change on reduction may be used in assay methods for example as described in W088/08882.

Albumin is a protein, in one form having a relative molecular mass of 67000 and consisting of a single chain of 584 amino acids. It is present in human serum at a higher concentration than any other single protein. The primary functions of albumin in serum include the maintenance of osmotic pressure, and acting as a carrier protein for metabolites such as bilirubin, fatty acids, inorganic compounds such as calcium, hormones and many drugs; it also provides a source of amino acids. For several reasons albumin is often assayed in the clinical biochemistry laboratory. For example it is used as an indicator of liver disease, is an essential requirement in albumin replacement therapy, it can indicate levels of unbound (unconjugated) bilirubin, it can indicate cases of myeloma and may be used in nutritional treatment regimes (Whicher and Spencer, Ann. Clin. Biochem.

24572 (1987)).

There are several known methods of quantitatively measuring serum albumin. It can be measured by electrophoresis, or by immunological methods. However, one of the most commonly used methods is based on dye-binding, using either bromocresol green (BGG) (Rodkey , F.L., Clin Chem. 11 478 (1965)) or bromocresol purple (BCP) (Louderback, A., Mealey, E.H., & Taylor, N.A. Clin. Chem. 14793 (1968)). Bromocresol green methods give low recoveries at high albumin concentrations and high recoveries at low albumin concentrations (Webster, D., Bignell, A.H.C. & Atwood, E.C., Clin. Chim. Acta. 53 101 (1974)), and are generally non-specific (Slater, L., Carter, P.M. & Hobbs, J.R., Ann. Chim. Biochim. 12 33 (1975). Methods based on bromocresol purple are generally more specific (Pinnell, A.E. & Northam, B.E., Clin.

Chem. 2480 (1978)). However, bromocresol purple does not have interspecies albumin specificity; many calibrators contain either bovine or equine albumin standards, which give an absorbance too far below that of human albumin to be of value. Further, heparin is reported to interfere (Perry, B.W. & Douman, B.T., Clin.Chem.

25 1520 (1979)), which precludes the use of heparinised blood.

Bilirubin is also believed to interfere.

Therefore a method based on the formation of a coloured product in response to the presence and concentration of albumin which did not suffer interference from other components which may be present in the sample would be of great practical advantage. It is an object of the present invention to provide a new method which does not suffer the disadvantages described and can be used for the quantitation of albumin by a procedure generating a coloured product. Other objects and advantages of this invention will be apparent from the following description. According to a first aspect of this invention a method for the reduction of a tetrazolium compoun d containing the cationic nucleus:

to produce a ring-opened formazan compound containing the group:

is characterised by the step of causing the tetrazolium compound to interact with albumin, and reducing the tetrazolium compound - albumin interaction product with a reducing agent in an aqueous solution at a pH between 6 and 10. The invention derives from the unexpected discovery that although weak reducing agents can only reduce many tetrazolium salts at pH above 10 and in the absence of albumin, at the lower pH range of the invention the reduction of tetrazolium salts is substantially enhanced by interaction with albumin. The method of the invention is particularly applicable to reduction of the leuco form of 2-(2' - benzothiazolyl) - 5-styryl-3-(4' phthalhydrazidyl)-tetrazolium, (abbreviated herein to "BSPT"), especially the hydrochloride.

BSPT contains the tetrazolium cation:

The interaction between the tetrazolium salt and the a lbumin is of a type that results in enhancement, acceleration or catalysis of the reduction of the tetrazolium salt to the coloured formazan dye.

Although the precise nature of interaction between the tetrazolium salt and albumin in the method of the invention is not fully understood, it is believed that the interaction may involve formation of a complex between the salt and the albumin. The interaction and/or complex formation may result from the presence of the substituent groups on the tetrazolium nucleus. Hence the method of the invention may be particularly applicable to tetrazolium salts in which the nucleus is substituted with one or more of the substituents present in BSPT ie benzothiazolyl, styryl or phthalhydrazidyl groups which may themselves carry substituent(s) on their ring systems, or a combination of such groups. Many reducing agents are suitable for use in the method of the invention, but preferably a weak reducing agent is used. Examples of those found to be suitable include dithiothreitol ("DTT"),

mercaptoethanol eg betamercoptoethanol, cysteine, nicotinamide adenine dinucleotide and aminophenol . A particularly preferred reducing agent is NADH (reduced form of nicotinaminde adenine dinucleotide).

An electron carrier may be used in the reaction medium of the method to enhance the speed of electron transfer between the reducing agent and the tetrazolium salt. A preferred electron carrier is a

phenazinium salt, ie compounds containing the cationic nucleus:

where R is hydrogen or alkyl. A preferred phenazinium salt is methoxy-N-methylphenazinium methyl sulphate ("MPMS"), ie

although other suitable but less preferred phenazinium salts include phenazinium methosulphate ("PMS") and phenazinium ethosulphate ("PES")

Any buffer capable of maintaining the pH between 6 and 10 may be used a preferred buffer being tris hydroxymethylaminomethane HC1 ("Tris- HCL"). A preferred solution pH is 8.5.

It is also preferred to include a surfactant in the reaction solution of the method, as this aids the solubilisation of the formazana produced. A preferred surfactant is a non-ionic surfactant

especially a polyoxyethylene sorbitol ester, in particular Tween 80 (Trade Mark). The degree of reduction of the tetrazolium compound and consequent formation of formazan quantitatively relates to the quantity of albumin and/or reducing agent present in the reaction solution, and therefore when one of these two is the limiting factor in the reaction, ie all the other reactants are present in excess, the method of the invention may be used as the basis of assays for albumin or reducing agents.

Therefore in a second aspect this invention provides a method for the assay of human or mammalian albumin in a sample comprises reacting a tetrazolium salt which is capable of interacting with albumin, with a reducing agent to produce a formazan in an aque ous solution at a pH between 6 and 10 in the presence of the sample, and relating the presence and/or amount of formazan produced to the presence and/or amount of albumin in the sample.

Although other tetrazolium salts which can only be significantly reduced (within a time period acceptable to clinical analysis) by weak reducing agents in the pH range 6 - 10 after interaction with albumin may be used, preferred tetrazolium salts are those of BSPT, especially the hydrochloride, or other tetrazolium salts in which the nucleus is substituted with one or more of the substituents present in BSPT as mentioned above, which may themselves carry substituents on their ring systems or a combination of such groups. The method of assay of albumin is suitable for use with any sample believed to contain albumin provided that at least a proportion of any albumin contained therein can be dissolved to form the necessary aqueous solution if the solution does not already comprise an aqueous solution. The method is therefore most suited to the assay of aqueous samples believed to contain albumin, especially biological samples such as bodily fluids, and in particular blood serum, which may be used directly without any pretreatment, or urine etc. In performing the assay method it is desirable that the quantity of tetrazolium salt and reducing agent present in the reaction mixture should be in excess of that expected to interact with the albumin in the sample so that the quantity of albumin present is the limiting factor.

Suitable and preferred reaction conditions and reagents for the albumin assay method of this invention are as discussed above with reference to the first aspect of the invention, ie the choice of reducing agent, the use of an electron carrier, buffers and pH, and use of a surfactant etc. These parameters are summarised in Table 1 below:

Under these conditions reduction of the tetrazolium salt takes place rapidly enough for the assay to be carried out in a few minutes, a time scale convenient for clinical use.

The method is sensitive to albumin levels below those of clinical significance and the colorimetic analysis has a linear relationship with initial albumin levels up to about 100mg/ml of original serum sample. Serum itself does not significantly absorb over the range of wavelengths used for measurement and so a sample blank is not normally required. The method appears to be specific to albumin, requires no pretreatment of a serum sample and is not significantly subject to interference. The method of estimating albumin of the present invention will have many applications, but will be

particularly useful if a quick and routine method of analysis is required, most especially in the field of clinical chemistry. Other uses may include quality control measurement in albumin production, assessment of albumin content in other blood products and structural investigations of albumin.

Accordingly in a third aspect the invention provides a method for the assay of reducing agents in a sample which comprises reacting a tetrazolium salt which is capable of interacting with albumin, with the sample in the presence of albumin in an aqueous solution at a pH between 6 - 10 and relating the presence and/or amount of formazan produced to the presence and/or amount of reducing agent in the sample, the quantity of tetrazolium salt used being in excess of the amount expected to be reduced by the reducing agent in the sample.

Suitable and preferred reaction conditions and reagents for this method of assay of reducing agents are as discussed above with reference to the first and second aspects, ie the choice of

tetrazolium salt, use of an electron carrier, buffers and pH, and use of a surfactant etc. It is desirable to arrange that the parameters of the reaction solution of the method are within the ranges indicated in Table 1 , if the approximate concentration of reducing agent in the sample can be approximated in advance. An albumin concentration of up to about lOOmg/ml appears to be suitable.

This method of assay of reducing agents may be used to assay many different chemical reducing agents. It is particularly suitable for the assay of reducing agents which are of biochemical clinical or diagnostic importance, such as aminophenols, NADH or co-enzyme A (CO-A) and using this method they may be assayed in samples of bodily fluids such as serum, urine ctcβ As a further modification, the method of assay of reducing agents may be used to assay compounds which can participate in a chemical reaction in which a reducing agent is formed which can be assayed by the method of the invention and related to the presence and/or quantity of the compound.

In such a reaction the compound to be assayed may itself be converted into the reducing agent. For example the drug paracetamol (p-hydroxy acetanilide) may be quantitatively converted into the reducing agent para-aminophenol using for example an aryl acylamidase enzyme in a well known reaction. The para-aminophenol formed may then be assayed and related to the amount of paracetamol.

Alternatively in such a reaction one or more of the participating reagents other than the compound may be converted into a reducing agent in an amount related to the quantity of the compound present. For example the antibiotics chloramphenicol and thiamphenicol, or gentanicins may be acetylated in a reaction in which acetyl Co-A is used as an acetylating reagent and which is consequently converted into Co-A. Such reactions proceed readily under the mediation of the enzymes chloramphenicol acetyl transferase ("CAT") or gentanicin acetyl transferace ("GAT") respectively. In each case the presence and/or amount of the Co-A produced may be quantitatively related to the presence and/or amount of antibiotic.

These assay methods of the invention may be performed either manually ^or automatically in a number of ways. For example an aqueous solution may be made up containing at least the sample, the tetrazolium salt and the reducing agent or albumin as appropriate at the indicated pH, optionally also containing the other reagents referred to above, and incubated at a suitable temperature. The colour produced is then detected and for example compared with a standard or measured colo rimetrically. Although the reaction solution is aqueous it may be necessary to include additional water-miscible solvents to assist in dissolution of all of the reagents, in particular the tetrazolium salt. One preferred example of such a solvent is dimethylformamide ("DMF"). When performing the method of the invention in this way the order in which the reagents and sample are mixed does not appear to be critical.

Alternatively, suitable reagents such as the tetrazolium salt, reducing agent or albumin as appropriate, buffer etc may be

immobilised on a solid support by for example impregnation, adsorption or absorption. Exposure to a solution sample containing albumin or reducing agent as appropriate then causes a colour change in the support which can be detected.

To make the assay methods of the invention convenient for laboratory and clinical use, the invention also provides an assay kit including in combination one or more ready made reagents as described above for the performance of the method. For example a kit may comprise separate solutions containing respectively the tetrazolium salt capable of interacting with albumin, buffer, and reducing agent or albumin as appropriate, at concentrations suitable to enable them to be easily used in the method of the invention, optionally together with the other reagents, eg surfactant, electron carrier referred to above. Such a kit may alternatively comprise a solid substrate having immobilised thereon suitable reagents. The kit may also include standards and instructions which may include a colour comparison chart. In such a test kit the reagents have been found to be stable on storage for at least 30 days, but it is desirable to make up the solution of the tetrazolium salt in an acid solution, eg 0.01 - 0.2M hydrochloric, matic or especially citric acid, at a pH of 3 - 6 for stability on long term storage. Methods of making such kits embodying the methods of the invention will be apparent to those skilled in the art.

Examples illustrating the invention will now be described with reference to:

Fig 1 which shows the hyperchromic shift on reduction of a

BSPT-albumin interaction product.

Fig 2 which shows the absorbance at 590nm versus albumin

concentration for the method of Example 1.

Fig 3 which shows the stability of the reagents on storage. Fig 4 which shows comparison between the method of the

invention and a conventional method used manually.

Fig 5 which shows comparison between the method of the

invention and a conventional method used on an automated apparatus.

Fig 6 which shows comparison between the method of the

invention and a conventional immunological method. Fig 7 which shows absorbance versus Co-A concentration.

Fig 8 which shows absorbance versus paracetamol concentration. Fig 9 which shows absorbance versus chloramphenicol

concentration. EXAMPLE 1 DIAGNOSTIC KIT FOR ALBUMIN (I)

REAGENTS

Reagent A:

Stock solution of tetrazolium salt:

(i) BSPT hydrochloride: 10mg dissolved in 10.5ml

dimethyl formamide.

(ii) 9.4ml 12mmol/L citric acid,

working solution of tetrazolium salt:

an 18ml volume of the stock solution is diluted by adding76ml 12mmol/L citric acid, 4ml 15% (w/v) Tween 80 and 2ml 1mmol/L MPMS.

Reagent B:

20mmol/L Tris HC1.pH 8.5 Reagent C:

5mmol/L nicotinamide adenine dinucleotide (reduced form). METHOD

1. To 0.5ml reagent A, add 0.5ml reagent B and 40 ul of

reagent C.

2. Equilibrate for 3 minutes at room temperature.

3. Add 50 μ1 sample and mix well.

4. Read absorbance at 590nm after 1 minute.

The results for the A590nm given by various initial concentrations of albumin are given in the calibration curve in Figure 2. The colour produced is stable for at least one hour. EXAMPLE 2 DIAGNOSTIC KIT FOR ALBUMIN ( II )

REAGENTS

Reagent A :

12.0 mmol/L citric acid

0.18 mmol/L BSPT Hydrochloride

0.5 mmol/L MPMS

Reagent B:

200 mmoL/L Tris HC1 pH 8.5

35% v/v Tween - 80

Reagent C:

2.5 mmol/L NADH

METHOD

1. Prepare a working reagent by mixing 50ml of Reagent A,

50ml of Reagent B and 4.0ml of Reagent C. This mixture is stable for at least 12 hours.

2. To 1.0ml of working reagent add 5μL serum (or a sample

containing albumin). Read absorbance at 590nm against a reagent blank after 1 minute.

The reagents A, B and C as prepared and stored separately are stable for a minimum of 30 days as demonstrated in Fig 3 by the stability of the absorption at 590nm resulting from addition of various albumin concentrations to a working reagent prepared from reagents stored for the indicated periods.

EXAMPLE 3 ALBUMIN ASSAY WITH OTHER REDUCING AGENTS

Table 2 below demonstrates the use of other reducing agents than NADH at indicated concentrations. The range of albumin concentrations over which the relationship between formazan production and albumin concentrations is linear is also shown, with NADH being used as a comparative standard. In each case the tetrazolium salt was BSPT hydrochloride.

EXAMPLE 4 INTERFERENCE IN ALBUMIN ASSAYS

Interference with albumin assays using the kit of Example 1 or 2 was investigated by comparing test samples containing various potentially interfering materials. The results are indicated in table 3 below:

Therapeutic Agents Concentration Range Interference

Salicylic Acid 0.1 - 0.4 mg/ml None

Theophylline 0.1 - 0.5 mg/ml None

Acetaminophen 0.02 - 1.0 mg/ml None

Lidocaine 0.002-0.01 mg/ml None

Barbital 0.01 -0.1 mg/ml None

Caffeine 0.1 mg/ml None

Blood Constituent Concentration Range Interference

Transferrin 1.67 - 2.86 mg/ml None

Creatinine 50 - 300 mg/ml None

Blood Proteins Concentration (mg/ml) Interference

Globulins Bovine 20 None

Bovine Globulins 20 None

Insulin Bovine 20 None

Blood Protein Concentration (mg/ml) Absorbance 590nm

Albumin (Control) 50 0.541

Insulin Bovine 20 0.556

Globulins Bovine 20 0.542

Bovine Globulins 20 0.542

As shown by above, all the blood constituents gave no interference. The concentrations studied were above the normal range found in blood. Heparin is added to blood to step coagulation at a

concentration of 6.7 units/ml.

The only noticeable interference was obtained with urea.

EXAMPLE 5 CORRELATION WITH "CONVENTIONAL" ALBUMIN ASSAY METHODS The BSPT method of albumin determination was compared to other methods which are commonly in use in hospitals. These were a dyebinding method based on bromocresol green (BCG) and an immunological method. A total of 71 pathological serram samples were tested manually using the BSPT method and compared to an automated BCG method. In these samples the amount of albumin varied between about 15 and 60 mg/ml. This data was then statistically analysed to give a regression equation of Y = 9.268 + 0.7X, indicating good correlation between the BSPT and BCG methods. This data relates to use of the BSPT and BCG methods manually; results are shown in Fig 4.

The methods used above (BSPT and BCG) can also be used in automated instruments. Similar analyses were carried out using the BSPT and BCG methods as automated procedures on the "Multistat III plus" and results are illustrated in Figure 5.

The BSPT method also correlates well with the immunological technique as shown in Fig 6. (Results data not shown)

EXAMPLE 6 ASSAY OF COENZYME-A

A solution was prepared containing BSPT hydrochloride, albumin, MPMS, Tris-HCl buffer and Tween-80 at concentrations within the ranges indicated in Table 1. To aliquots of this was added Co-A over the concentration range 0.25 mM of Co-A final concentration. Fig 7 shows that over this range the absorbance at 590 nm was linearly related to Co-A concentration.

EXAMPLE 7 ASSAY OF PARACETAMOL

Paracetamol was quantitatively cleared to form p-aminophenol using the enzyme aryl acylamidase in a known reaction. A solution was prepared containing BSPT hydrochloride, albumin, MPMS, Tris-HCl buffer and Tween-80 at concentrations within the ranges indicated in Table 1. To aliquots of this was added p-aminophenol. The graph of 4- aminophenol concentration against absorbance of 590 shows an

absorbance change up to a concentration of 3.0 mM final concentration of p-aminophenol. This could be related to paracetamol concentration versus absorbance at 590 nm which showed a good linear relationship up to a concentration of 2.0 mM as shown in Fig 8.

EXAMPLE 8 ASSAY OF CHLORAMPHENICOL

Chloramphenicol was acetylated using acetyl Co-A and the enzyme CAT by a known reaction. This resulted in production of a quantity of Co-A which was quantitatively related to the amount of chloramphenicol originally present. The Co-A produced was assayed using the procedure of example 6. The absorbance at 590 nm against chloramphenicol concentration was linear over the chloramphenicol range 10 - 200 μM as shown in Fig 9.

Claims

1. A method for the reduction of a tetrazolium compound containing the cationic nucleus:

to produce a ring-opened formazan compound containing the group:

characterised by the step of causing the tetrazolium compound to interact with albumin, and reducing the tetrazolium compound-albumin interaction product with a reducing agent in an aqueous solution at a pH between 6 and 10.

2. A method according to claim 1 characterised in that the nucleus is substituted with one or more substituent groups selected from benzothiazolyl, styryl or phthalhydrazidyl, which groups may themselves carry substituents on their ring systems, or a combination of such groups.

3. A method according to claim 2 characterised in that the tetrazolium compound is a salt of the leuco form of 2-(2'-benzothiazolyl)- 5-styryl-3-(4'-phthalhydrazidyl)-tetrazolium (BSPT).

4. A method for the assay of human or mammalian albumin in a sample characterised by reacting a tetrazolium salt which is capable of interacting with albumin, with a reducing agent to form a formazan in an aqueous solution at a pH between 6 and 10 in the presence of the sample, and relating the presence and/or amount of formazan produced to the presence and/or amount of albumin in the sample.

5. A method for the assay of reducing agents in a sample characterised by reacting a tetrazolium salt which is capable of interacting with albumin, with the sample in the presence of albumin in an aqueous sol ution at a pH of between 6 and 10 and relating the presence and/or amount of formazan produced to the presence and/or amount of reducing agent in the sample, the quantity of tetrazolium salt used being in excess of the amount expected to be reduced by the reducing agent in the sample.

6. A method according to claim 5 characterised in that the reducing agent is aminophenol, NADH or coenzyme-A.

7. A method for the assay of a compound which participa tes in a chemical reaction in which a reducing agent is formed, characterised in that the reducing agent formed is assayed using a method as claimed in claim 5 and is then related quantitively to the presence and/or quantity of the compound.

8. A method according to claim 7 characterised in that the compound to be assayed is itself converted into the reducing agent.

9. A method according to claim 8 characterised in that the compund to be assayed is paracetamol (p-hydroxy acetanilide) which is converted into the reducing agent p-aminophenol.

10. A method according to claim 7 characterised in that one or more of the participating reagents other than the compound to be assayed is converted into the reducing agent.

11. A method according to claim 10 chaterised in that the compound to be assyed is chloramphenicol or thiamphenicol or a Gentamicin, which are acetylated in a reaction in which acetyl Co-A participates and is converted into the reducing agent Co-A.

12. A method according to claim 4 characterised in that the reducing agent is NADH.

13. A method according to claim 4 characterised in that the reducing agent is dithiothreitol, mercaptoethanol, cysteine or aminophenol.

14. A method according to claim 4 characterised in that the concentration of tetrazolium salt is in the range 10μM to 100μM and of the reducing agent 1μM to 100mM.

15. A method according to any one of claims 4 to 14 characterised in that the tetrazolium salt is one having a nucleus substituted with one or more substituent groups selected from benzothiazolyl, styryl or phthalhydrazidyl, which groups may themselves carry substituents on their ring systems, or a combination of such groups.

16. A method according to claim 15 characterised in that the tetrazolium salt is a salt of BSPT.

17. A method according to any one of claims 4 to 14 characterised in that the albumin, reducing agent or compound (as appropriate) to be assayed are contained in a bodily fluid,

18. A method according to claim 18 characterised in that tlae bodily fluid is blood serum.

19. A method according to any one of claims 4 to 14 characterised in that the tetrazolium salt, reducing agent or albumin (as appropriate) and a pH buffer are immobilised on a solid support which is exposed to a solution sample containing albumin or reducing agent (as appropriate). 20. A solid substrate characterised by having immobilised thereupon a tetrazolium salt, reducing agent or albumin (as appropriate) and a pH buffer and being suitable for use in a method as claimed in claim

19.

21. An assay kit for the assay of albumin using a method as claimed in claim h characterised in that it includes in combination separate solutions containing (i) a tetrazolium salt capable of interacting with albumin, (ii) a pH buffer solution and (iii) a reducing agent.

22. An assay kit according to claim 21 characterised in that the tetrazolium salt is one having a nucleus substituted with one or more substituent groups selected from benzothiazolyl, styryl or pthalhydrazidyl which groups may themselves carry substituents on thoir ring systems, or a combination of such groups.

23. An assay kit according to claim 22 characterised in that the tetrazolium salt is a salt of BSPT.

24. An assay kit for the assay of a reducing agent or a compound (as appropriate) using a method as clained in claim 5 or 7 characterised in that it includes in combination separate solutions containing (i) a tetrazolium salt capable of interacting with albumin (ii) a buffer solution and (iii) albumin.