NZ236048A - Assay for ketosamines and fructosamines and the fructose-nickel or zinc-lysine standards used - Google Patents

Description

New Zealand Paient Spedficaiion for Paient Number £36048 Patents Form No. 5 No. 236048 Date: 13 November 1990 Patents Act 1953 COMPLETE SPECIFICATION ASSAY METHOD FOR KETOSAMINES AND FRUCTOSAMINES We, CANTERBURY AREA HEALTH BOARD, a body corporate pursuant to the Area Health Boards Act 1983, of 10 Oxford Terrace, Christchurch, New Zealand, hereby declare the invention for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement: The present invention relates to an assay method for measuring l-deoxy 1-amino 2-ketoses ("ketosamines") which is suitable for the quantitative chemical analysis of low molecular weight ketosamines and of ketosamines bound to proteins (ketosaminoacid units, especially fructosaminoacids and other "fructosamines" (i.e. l-deoxy 1-amino 2-fructoses)).

Ketosamines are formed naturally when sugars react with amine groups; for example, when glucose reacts with amine functional groups in proteins to produce fructosamines. The formation of these compounds has significance, e.g. in nutrition (loss of the essential aminoacid, lysine), the food processing industry, and in physiology. This means that there is a need to measure the amount of such compounds in certain circumstances. One particular instance is the measurement of fructosamines in human body fluids or tissues as a means of assessing diabetes, since high glucose levels lead to high fructosamine levels.

Previous methods for measuring these compounds have been based on several chemical principles, the ones most like the present invention depending on the ability of ketosamines to act as strong reducing agents under milder conditions than most common sugars do. A variety of electron acceptors has been used, including ferricyanide ion and tetrazolium salts. Two known methods are: (1) reduction of ferricyanide to ferrocyanide, with measurement of this by conversion to Prussian Blue; and (2) reduction of triphenyltetrazolium (tetrazolium red) to its red formazan.

However, these known methods have disadvantages. The coloured products in each case are insoluble in water, and remain in solution at best only briefly as colloids. This makes their quantification difficult. Another disadvantage is the common need (e.g. in biological materials) to measure ketosamines in the presence of other reducing compounds which, though less reactive, may be present in relatively high concentrations. Furthermore, pure standard ketosamines are not generally available and this has made calibration of quantitative assays very difficult.

A recent attempt in New Zealand Patent No. 199380 to overcome these problems has used a different tetrazolium, nitroblue tetrazolium (NBT) which is converted to a blue formazan. This remains dispersed in aqueous solution long enough to allow measurement of the colour intensity. Further, the reaction with fructosamines in particular was found to be slower than with other important biological reducing compounds and the contribution of these latter was minimized by using the difference in colour intensity between two times after the start of the reaction, rather than the intensity itself, as the measure of fructosamine. The synthetic fructosamine, deoxymorpho-linofructose (DMF), was used as a calibrant.

However this latter approach does not adequately solve the problem: the colour is not stable; it tends to discolour glass surfaces and to be absorbed onto plastics surfaces; the response is not always linearly related to fructosamine concentration; the calibrator has limited validity because it does not respond in the assay in the same way as fructosamines in biological materials. It is not easily adapted for all analytical systems because of the need to subtract a substantial first reading, and different users of the method have obtained widely different n results.

Finally, the methods disclosed are limited to testing samples derived from blood.

In EU 215170 (Isolab) , the above disadvantages of New Zealand Patent No. 199380 are overcome to some extent by the use of suitable oxidizing agents to eliminate non-relevant reducing substances present in blood serum, other than fructosamines. Fructosamine quantities are then measured with one reading and compared with a standard.

However, the disadvantage of this method is that it does not address the problems of interference in the measurements resulting from protein bound thiols, the limited solubility of the coloured product and the calibration difficulties.

An alternative approach can be found in EU 250991A (corresponding to Australian Patent Application no. 87/74526, Boehringer), the non-relevant reducing substances and any turbidity-causing agents are eliminated at a neutral pH. By the appropriate choice of oxidizing agents it was found that disturbances in fructosamine determinations due to the activity from the total protein content of the sample could also be removed. However in this method it is then necessary to rebuffer the sample to a pH of between 10-12 before the colour reagent is added. The sample, however, must be blood or derived from blood.

In the method of the present invention, interfering reducing compounds in a biological sample need not be eliminated by oxidation (as in European 215170 and European 250991A) , but are blocked.

An object of the present invention is the provision of an 4 assay or analytical method for quantifying the amount of ketosamines present in biological or other material which, with a given predetermined biological material, can be adapted to give a negligible instrument response when ketosamines are not present. Other material could include, for example, dried milk or other processed foods.

Another object of the present invention is the provision of an assay or analytical method for quantifying the amount of ketosamines or fructosamines present in a biological or other material which gives a result which closely approximates to being directly proportional to concentration over a range of concentrations that is analytically useful.

A further object of the present invention is the provision of an assay or analytical method for quantifying the amount of fructosamines present in a biological or other material which includes measurement of colour intensity of a compound which remains dispersed in a solution which may be in contact with glass and/or plastics vessels.

A still further object of the invention is the provision of a method of synthesising, totally or partially, fructosamines which can be used to calibrate ketosamine assays, such that said assays can be used to calculate the concentration of ketosamine units present in a selected biological sample.

The present invention provides a method to quantitatively determine ketosamine or fructosamine levels in a biological or other material sample, said method including the steps of: mixing said sample with a thiol blocking agent, a buffer solution with a pH above 10 and at least one anionic ester detergent; adding a start reagent to the mixture, said start reagent including at least one oxidant reagent; measuring the absorbance at a pre-determined wavelength, said wavelength determined by the or each oxidant selected; and comparing said absorbance with absorbance readings from a pre-selected standard.

Preferably said thiol blocking agent, the buffer solution, and the at least one phosphate ester detergent are pre-mixed as a diluent reagent.

Said diluent reagent may also contain a chelating agent.

Alternatively or additionally, the diluent and start reagents may be mixed together immediately before the assay is carried out.

Preferably the or each oxidant selected causes the absorbance reading to be between 500 and 560 nanometres.

Alternatively the measure of the rate of colour production of the solution may be used to convert to concentrations of ketosamines or fructosamines by comparison with the rate of colour production of a suitable pre-selected standard.

Preferably, said pre-selected standard is a secondary standard and is assigned a calibration value by comparison with a synthesized fructosamine as the primary standard.

The method of the present invention uses the following reagents: (1) A thiol blocking agent such as iodoacetate or iodoacetamide. This is essential if quantitative data is to be obtained from protein solutions. (2) At least one, and preferably two, electron accepting reagents, such an anthraquinone-2-sulphonate ("AQS"), 23604 anthraquinone 2, 6 disulphonate and/or a variety of tetrazolium salts. The choice will not be the same for different applications, but a preferred system for blood plasma is to combine AQS with nitroblue tetrazolium ("NBT"), with the final colour being that expected from the reduction of NBT. Sulphonated tetrazoliums such as sulphonated triphenyltetrazolium (the preparation of which is outlined below) have been also found to be successful in practice. (3) A phosphate ester detergent, to ensure the formazan remains dispersed in solution. Suitable phosphate ester detergents may include phosphate esters of alkyl and aryl polyethoxyethanols.

Other anionic detergents may replace the phosphate ester when a urea derivative is present (as described under reagent (5) below). In this instance some alkyl and alkaryl sulphate esters and sulphonic acid detergents have been found successful.

Without the urea derivative it has been found that the same detergent, though dispersing the formazan, greatly decreases the reaction rate. However, in practice it has been found that a small quantity of a sulphonic acid detergent should be included. This decreases the response of one of the more serious interfering substances, urate. The alkyl sulphonate Hostapur SAS 93 (Hoechst) is the most suitable choice found in practice. (4) A buffer solution at a high pH. In strongly alkaline solutions many compounds other than ketosamines react, whereas below pH 9 the reaction may become unacceptably slow for many purposes. The exact choice will depend on the application. For - 1 0\ 7 ' % * ' VOV J99/ J analysis of plasma ketosamines a suitable mixture is monohydrogen phosphate ion and phosphate ion in the molar ratio 4:1. Other successful mixtures have made use of the intrinsic buffering capacity of a phosphate ester detergent by adding an alkali, or have used organic bases. The latter bases may in some cases decrease the efficiency of the blocking agent however. (5) A urea derivative. This is not essential, but for the analysis of high molecular weight fructosamines (such as glycated proteins) high concentrations of urea (5 mol/L or more) increase the reaction rates and bring these more in line with those of low molecular weight fructosamines. Thus inclusion of a urea derivative is recommended for plasma assays. It has been found in practice that 2-imidazolidone ("ethylene urea") is also effective, and other urea derivatives may also be used. (6) A chelating agent such as ethylenediamine- tetraacetate ("EDTA") may be added to improve robustness of the assay method. It has been noted in practice that some of the reduction reactions may be affected by the presence of traces of some metal ions. EDTA is added to avoid differences attributable to variable traces of metal ions and chelators in samples and reagents.

By way of example only preferred assay methods and reagents used therein will follow.

The present invention further provides improvements in the preparation of the primary standard (used to calibrate said preselected standard), fructosamine alpha-carbobenzoxycarbonyl-epsilon-fructose-L-lysine ("Fructose-CBZ-lysine"), the method of synthesis of this being found in Lever et al, Clinical Chemistry, 8 Vol 33, No. 3, 1987, p. 447, and incorporated herein by reference; said improvements to the synthesis method there outlined including the steps of: assisting the elution from Dowex-50 columns by adding 50% by volume ethanol to the sodium acetate eluant, and absorbing the final product onto an XAD-2 column and differentially eluting with 35% by volume ethanol to obtain the cleanest product.

The present invention further provides a method for the preparation of the standards fructose-nickel-lysine, fructose-zinc-lysine and related derivatives, for calibrating said preselected standard, said method providing: reacting lysine monohydrochloride in solution with a solid basic metal carbonate, hydroxide or oxide; adding excess glucose and heating the reaction mixture; isolating the product from the reaction mixture; and removing any glucose present. Only the epsilon group reacts because the alpha amino groups are protected by being chelated to the metal.

A similar technique has been used for making other aminoacid derivatives, using cupric carbonate. This partially fails for fructosamine synthesis because fructosamines reduce cupric complexes under the reaction conditions, and more fructosamine is lost during purification.

Suitable metals are nickel (II), using boiling methanol (i.e. under reflux) as a solvent, and zinc (II) using ethanediol at 60—80°C as a solvent. Cobalt (II) may also be used, and it is possible that other transition metals may be suitable.

Preferably, said method further includes the steps of heating the lysine monohydrochloride with a solvent under reflux, with a suspension of the carbonate or oxide metal for 0.5 to 1 hour, the glucose is then added in an excess of approximately ten-fold, and the solution is further heated for between 1 to 5 hours.

If ethanediol is used as the solvent, 3 volumes of methanol is added for each volume of ethanediol.

A preferred method of isolating the product is to precipitate it from an alkaline solution, taking advantage of the low solubility of the unionised product in alcoholic solutions. The zinc derivative may also be precipitated as a sulphate by adding tetrabutylammonium hydrogen sulphate in methanol.

A preferred method of removing the glucose is by gel filtration using a column of Sephadex G15 with water or more preferably dilute acetic acid as the eluting fluid. This may also partly remove the metal ion, which can (for maximum stability of the product) be restored by prolonged shaking with the basic carbonate or oxide.

The method may be adapted to make other epsilon lysyl ketosamines, such as analogues of those derived from bilirubin glucuronides in liver disease; and, using calcium as the cation, the fructosamine derived from aspatate may be similarly purified by adding calcium to form the chelate after the initial reaction (in this case calcium is not included initially because there is no need to protect the alpha amino group, and the calcium salt is not sufficiently soluble for an exactly parallel procedure to be practicable) .

The present invention further provides a method of preparation of glycated albumin standards, for calibrating said pre-selected standard, said method including the steps of: incubating human serum albumin (50 grams per litre) with 0.5 moles per litre glucose for 3-4 days at 37°C, in the presence of antibiotics to prevent decay; exhaustively dialyzing the solution to remove the glucose and the antibiotics; concentrating and freeze-drying said solution; the material obtained is calibrated by repeated comparisons with the fructoselysine calibrants listed above using either of the respective methods outlined above, in the preferred assay method, but with the reaction allowed to continue for longer times and repeated to check than an endpoint has been approached; or conversion of the standard to hydroxymethylfurfural and the assay of this fluorometrically or by HPLC. (This latter method is as described in Lever et al, Analytical Biochemistry, 1986.) The published HPLC method has been modified by removing the detergents with ion-exchange resin (Dowex-50 in the acid form) after heating at 120°C. By using different heating times (45min to 120min), valyl fructosamine groups are distinguished from lysyl fructosamine groups.

The present invention further provides a reagent for use in a heterogeneous assay system for fructosamines, said reagent being formed in the absence of urea, from mixtures of AQS and NBT which tend to slowly precipitate, leading to a yellow deposit on the inside of plastic vessels or tubing containing the reagent. The coated vessels (or coils of tubing) are used with a reagent containing low concentrations of NBT in detergent to obtain most of the advantages of adding AQS without the disadvantage of reagent instability. 11 EXAMPLES Example 1: General Method: Ketosamine In a Protein Solution.

The sample (0.05 millilitres) is mixed with a diluent reagent (1.5 millilitres) and allowed to stand for between 20 and 100 seconds. A start reagent (0.5 millilitres) is then added to the solution and the solution is allowed to stand for 5 minutes or longer.

A reading of the absorbance, at a wavelength of approximately 530 nanometres, is then taken.

The absorbance readings are converted to concentrations of fructosamines by comparison with absorbance readings of a suitable, pre-selected standard.

Alternatively, the diluent and start reagents may be mixed together immediately before the assay is carried out. The plasma sample (0.05 millilitres) is mixed with 2 millilitres of the combined reagent and allowed to stand for 5 minutes or longer. The absorbance reading is taken, as described above.

The above described methods may be carried out manually or by an automated system, as is desired.

The diluent includes: the buffer; the blocking agent (for example, 20 millimole/litre 2-iodoacetamide or 2-iodoacetate or 50 millimole/litre of their bromo-analogues); and the detergent(s) (for example, 0.5-1% by volume GAFAC RS610, or RE410 or RM510 plus 0-0.5% by volume Hostapur SAS9 3).

The diluent may include: urea (4-6 mole/litre) , or as an alternative, 2- 12 iniidazolidone ("IAZ") at a concentration of 3 mole/litre; and EDTA, (for example 10 millimole/litre of the tetrasodium or tetrapotassium salt) The start reagent contains: a first oxidant reagent (0.5-1 millimole/litre AQS); and may include: a second oxidant reagent (0.25-0.5 millimole/litre NBT to yield the convenient coloured product) ; a sulphonic acid detergent (e.g., 0.5% by volume Hostapur SAS93); and urea (4-6 mole/litre) , or as an alternative, 2-4 mole/litre IAZ.

The detergent and urea (or alternative) aid in avoiding precipitation after a few hours or days.

It is presently preferred to include in the start reagent a low concentration (0.5 % by volume) of a GAFAC detergent to improve solubility and to make the overall start reagent solution slightly acidic, which improves stability.

The choice of GAFAC detergent depends on the compound to be measured. For example, if low molecular weight fructosamino acid residues are to be measured, or only one protein is of interest, then the more soluble and easily handled RS610 is adequate, with 0.5% by volume being sufficient to solubilize the formazan. However, to obtain consistent and comparable results from several different proteins it is preferable to use higher concentrations of the more hydrophobic members of the series, and at concentrations of 1% by volume or even more. In this case the reaction solution may not be optically clear; this can be 13 accepted, though it can be minimized or eliminated by including one or more of urea, IAZ, or niacinamide in the reagent.

Since the diluent and start reagents are light sensitive, they should be kept in dark bottles away from bright light. In the refrigerator they have been found to be stable for at least 3 weeks.

An alternative to AQS as the first oxidant reagent in the start reagent which results in slightly more stable reagents is sulphonated triphenyltetrazolium, prepared as a calcium salt (as outlined below). This may be used at a concentration of at 5-10 millimole/litre but to avoid precipitation of calcium salts it is recommended that excess EDTA is added to the start reagent containing this compound. Another useable alternative to AQS is anthraquinone-2, 6-disulphonate, but this was found to be less suitable in practice.

Generally, AQS is preferred because of it is inexpensive, easily available, and gives good sensitivity with low blank reactions at low concentrations. However, the final reaction mixture containing AQS is light sensitive which makes this reagent unsuitable where the assay must be carried out in a brightly lit room, or using equipment that uses intense light sources. Sulphonated triphenyltetrazolium-based reagents are less susceptible to this problem and are more generally applicable. The anthraquinone disulphonate has similar disadvantages to AQS, with the compensating advantages being less cogent.

For the assay of fructosaraino acid residues bound to haemoglobin it has been found that a modified procedure is 14 preferable (see also Example 2 below) . In order to avoid interference by the haem in the reaction, haem complexing agents are incorporated. The thiol blocking agent may be added before analysis, by incorporating it into the reagent used to lyse red cells. The diluent is modified by including (for example) isonicotinonitrile and nitrite, or niacinamide and cyanide. This formulation allows the more hydrophobic GAFAC series detergents to be used and RM510 is recommended to make calibration more robust (by minimizing rate differences between different proteins).

For routine clinical applications, preferably the secondary standard (for plasma) is a solution of albumin or a plasma sample to which a concentration value has been assigned by comparison with concentrations of pure fructosamines, synthesized as described above. For haemoglobin analysis secondary standards may be haemoglobin solutions similarly calibrated, or albumin with added haemin.

Preparation of Sulphonated triphenvltetrazolium.

This is prepared by a standard sulphonation, as would be used with an aromatic amine, the tetrazolium ring system appearing to be stable enough to remain intact through the procedure. Triphenyltetrazolium chloride is warmed with concentrated sulphuric acid until the hydrogen chloride effervescence ceases. Oleum is then added and the reaction mixture heated to 170-190°C, as in the standard preparation of sulphonamides. The product is isolated as the calcium salt by neutralizing the diluted reaction mixture with calcium carbonate, adding an equal volume of methanol, and filtering. The filtrate is concentrated and then freeze-dried.

Example 2: Glycated Haemoglobin Assay For implementation on a Hitachi 704 or 717 Analyzer. The "Twin Test" and "Calculated Test" functions of the analyzer are used.

Lysates are used as the samples. Lysis is carried out with a special reagent containing a thiol blocking agent, which prevents interference by protein thiols and glutathione, and a complexing agent or agents to complex haem iron. A high detergent concentration and complexing agents convert the haemoglobin (or haem) groups into an inert form. Suitable combinations of complexing agents are cyanide ion with nicotinamide or nitrite ion with isonicotinonitrile (4-cyanopyridine). A mild oxidant (nitrate ion is suitable) is included to ensure the iron is fully oxidized. The haem and fructosamine (called "bound sugar" or BNSG in the method) contents are measured and the ratio reported as the measure of glycation.

The sample is diluted with the diluent reagent which contains alkali, detergent, complexing agents and this solution the haemoglobin pigments are present as an alkaline hemichrome, the absorbance of which is used to quantitate haemoglobin as haem. The preferred system uses nitrite ion and isonicotinonitrile both in the diluent reagent and in the lysing solution; with this combination the thiol blocking agent may be included in the diluent reagent well as (or instead of) in the lysing solution, making the reagent combination also suitable for 16 measuring fructosamines in other samples such as blood plasma. The start reagent is added, and the change in absorbance caused by the reducing activity of the fructosamine units is used to measure glycation.

Table 1: Preferred Hitachi 717 Settings (Isoniotinonitrile system): Haem.

Assay Code Sample Volume R1 Volume R2 Volume Wavelength Calibration Method Std 1 conc - pos Std 2 conc - pos Unit SD Limit Duplicate Limit Sensitivity Limit Absorbance Limit Prozone Limit Expected Value Instrument Factor [1 point & R] : [14]-[20] [15] [15]

[250] [100] [no] [0] [100] [no] [ ] [570] [Linear] [0] [2] [12] [0.00] -[2.26] -[MMOL/L] [0.1]

[400] [0]

[20000] [increase] [0] [lower] [0.8] - [3.0] [1.00] [0] (see notes below) (see notes below) Table 2: Preferred Hitachi Settings: Bound Sugar (Fructosamine) (As with Haem, these values are given as a guide.

Assay Code Sample Volume R1 Volume R2 Volume Wavelength Calibration Method Std 1 conc - pos Std 2 conc - pos Unit SD Limit Duplicate Limit Sensitivity Limit Absorbance Limit Prozone Limit Expected Value Instrument Factor R] [1 point & [15] [15] [250] [100] [no] [80] [100] [no] [ 3 [546] [Linear] [0] [0.0] - [2] [255.5]-[12] [UMOL/L]

[10]

[200] [0]

[32000] [increase] [0] [lower] [60] - [300] [1.00] [35-45] [0] (see notes below) (see notes below) Bound Sugar/Haem = Glycated Haemoglobin The units of glycated haemoglobin are thus mmol sugar/mol haem. This gives a useful scale with high normal levels around 17 50 units, a level which well-controlled diabetics can aspire to, with 100 denoting very poor control which calls for intervention.

Procedure: Red cells are spun down and washed with saline. Packed washed cells (1 volume) are mixed thoroughly (vortex mixer) with a lysing solution (7 volumes) and allowed to stand for at least 10 minutes; longer if the solution is not clear.

Reagents: Lysing Solution: GAFAC RM710 Formamide Iodoacetamide Isonicotinonitrile Sodium nitrate 2% (dilute from 10% stock) 10 mol/L (40% v/v) 2 0 mmol/L 10 mmol/L 10 mmol/L The solution should be made up freshly every 3-4 weeks, and stored in dark bottle at 4°c.

Note that other combinations are possible, such as the cyanide-containing system described below. Other anionic detergents such as Hostapur SAS93 may be used in place of the GAFAC.

Diluent Reagent: 2-Imidazolidone GAFAC RM510 GAFAC RM710 Hostapur SAS93 di-potassium hydrogen phosphate 3 mole/litre 1% by volume 0.5% 0.5% 0.2 mole/litre tri-potassium phosphate 0.075 mole/litre 18 tetra potassium EDTA 5 millimole/litre Iodoacetamide 20 mmol/L Isonicotinonitrile 50 mmol/L Sodium nitrite 50 mmol/L Preparation: to reagents add distilled water to give a volume of approximately 95% of final volume. Place on a mechanical stirrer to mix. Made up to volume with distilled water, and filter through No.541 paper. Store in a brown bottle at 4 degrees celsius.

It may be stored in a dark bottle at 4° for at least eight weeks.

One alternative reagent, containing cyanide and nicotinamide instead of nitrite and isonicotinonitrile, is more sensitive but is incompatible with iodoacetamide, which is in this case included in the lysing solution only. The resulting combination is satisfactory for haemoglobin assays but cannot be used for measuring plasma fructosamines. Note that with this reagent a different wavelength combination (546 nm for Haem, 505 nm for BSNG) is used. 3. Start Reagent: Concentration: 2-Imadazolidone 3 mole/litre Gafac RM 510 0.5% by volume Hostapur SAS 93 0.5% by volume AQS 0.5 millimole/litre NBT 0.5 millimole/litre Instead of the AQS, 5 mmol/L each of sulphonated TPTZ and tetra-sodium EDTA may be used to give a less light sensitive reagent. 19 I Preparation: Mix together: 2-Imadazolidone, 10% Gafac RM 510, and 10% Hostapur SAS 93, add distilled water to bring volume to approximately 80 ml and dissolve. Note: Do not add the colour reagents before solution is complete. Add Anthraquinone sulphonate (first) and finally Nitroblue Tetrazolium. Mix and make volume up to 100 mL.

This reagent may be stored in a dark bottle at 4°C for at least 10 weeks. 4. Standards and Controls: Interim standards and controls consist of "low" and "high" calibrated lysates, diluted with lysing solution, and the assigned values calculated from those used in a previous method (Lever, M. , May, P.c. & Andre, C.M. Anal. Biochem, Vol. 144, 1985, pp 6-14).

Alternative Instruments This method may be adapted for operation on other equipment. As an example, satisfactory results using the same reagents and standards have been produced on the Roche Cobas Mira analyzer. The ratio is calculated as for the Hitachi analyzer.

Table 3: GENERAL Measurement Mode Reaction Mode Calibration Mode Reagent Blank Cleaner Wavelength Decimal Position Unit ANALYSIS Post Dil. Factor Conc Factor Sample Cycle Volume Dilution Name Haem.

Absorb R-S Slope Avg Reag/Sol Before 600 nm 2 mmol/L no no 1 uL H20 Table 4: BNSG Volume:20 uL ReagentCycle:1 Volume:200 uL CALCULATION SampleLimit:no Reac. Direction :Increase Check:Off Convers. Factor:1.00000 Offset:0.00000 Number of Steps:1 Calc. Step A:Endpoint Readings First : CB Last : 8 CALIBRATION Calib. Interval Blank Sol.-Pos Standard Pos.

STD-1 STD-2 STD-3 Each Run 6 2.26 no no GENERAL Measurement Mode Reaction Mode Calibration Mode Reagent Blank Cleaner Wavelength Decimal Position Unit Absorb R-S-SR1 Slope Avg Reag/Sol Before 550 nm 1 umol/L ANALYSIS Post Dil. Factor Conc Factor Sample Cycle Volume Dilution Name Volume Reagent Cycle Volume Start R1 Cycle Volume Dilution Name Volume no no 1 uL H20 2 0 UL 1 150 UL 5 50 UL H20 20 uL CALCULATION Sample Limit Reac. Direction Check Convers. Factor Offset Number of Steps Calc. Step A Readings First Last no Increase Off 1.00000 0.00000 1 Kinetic 16 CALIBRATION Calib. Interval : Each Run Blank Sol.-Pos : 6 Standard Pos. : 5 STD-1 : 255.5 STD-2 : no STD-3 : no Q Example 3a: Manual Plasma or Serum Fructosamine Assay Materials Diluent Reagent (lyophilized or freshly prepared) Start Reagent (lyophilized or freshly prepared) Standard (lyophilized) Reconstruction Instructions for Lyophilized Reagents 21 Fructosamine Standard: Reconstitute with 2 mL of freshly distilled or deionized water. Allow to stand at room temperature for 20 minutes, swirling occasionally. This standard is stable for at least one week when stored at 4°C.

Start Reagent: Same as Example 1.

Diluent Reagent: Same as Example 1.

Sample Material: Serum or heparinized plasma.

Note that the diluent and start reagents described in the glycated haemoglobin example may also be used in the following, provided that the cyanide containing options described there are not selected.

Method Diluent Volume 1.5 mL Sample Volume 50 UL Initial Incubation 90 seconds Temperature 37°C Start Reagent Volume 500 uL Incubation (in dark) minutes Stop Reagent Volume 500 uL Wavelength 525 nm Standard Value 1000 umol/L Set up the required number of test tubes, including a tube for the blank and for the standard, plus any controls used . 2. Add 1.5 mL of diluent to each tube. 3. Add 50 uL of sample to the tube (using distilled water for the blank sample). 4. Mix well and incubate for 90 seconds in a 37°C waterbath.

. Add 500 uL of start reagent to each tube. 22 6. Mix well and return tubes to 37 degree Celsius waterbath for 15 minutes. Cover the waterbath to minimize the presence of light. 7. Add 500 uL of stop reagent and mix well. 8. Read the absorbances at 525 nm using distilled water as a blank to zero the spectrophotometer.

Results To calculate the concentration of fructosamines in umol/L: Sample Absorbance - Blank Absorbance x Standard assigned Standard Absorbance - Blank Absorbance concentration value i.e. : 1000 Example 3b: Hitachi 704 Fructosamine Assay.

Materials, Reconstitution Instructions and Reagents are the same as for Example 3a.

Programming Parameters Type of assay used may be Endpoint or Kinetic.

Table 5: Parameters for the Endpoint Assay: Assay Code Sample Volume R1 Volume R2 Volume Wavelength Calibration Method Std 1 conc - pos Std 2 conc - pos Unit SD Limit Duplicate Limit Sensitivity Limit Absorbance Limit Prozone Limit Expected Value Instrument factor (1 point): (32)- (0) (10) (350)(50)(no) (0)(50)(no) (480)(546) (linear)(0) (0.00) - (1) (1000) - (6) (UMOL/L) (10) (100) (0) (0)(increase) (0)(lower) (0) - (400) (1.00) 23 Table 6: Parameters for the Kinetic Assay: Assay Code Sample Volume R1 Volume R2 Volume Wavelength Calibration Method Std 1 conc - pos Std 2 conc - pos Unit SD Limit Duplicate Limit Sensitivity Limit Absorbance Limit Prozone Limit Expected Value Instrument factor (5 point):(?)-(?) (10) (250)(50)(no) (100)(50)(no) (480)(546) (linear)(0) (0.00) - (1) (1000) - (6) (UMOL/L) (10) (100) (0) (0)(increase) (0)(lower) (0) - (400) (1.00) Note that the endpoint assay uses premixed reagents while the kinetic assay uses two reagents. The kinetic assay is less affected by lipamia. The method has been found to perform well on other Hitachi analyzers such as the 717.

The maximum incubation for Rl, R2 and sample mixed solution is 5 minutes. The full five minutes is given and then one reading per sample is taken after incubation.

Example 3c: Technicon Auto Analyser Fructosamine Assay. Materials, Reconstitution Instructions and Reagents are the same as for Example 3a.

Autoanalyzer Manifold: The sample is diluted with water approximately 9 fold (to make the method transferrable to a SMAC), and the diluted sample is resampled into diluent reagent. After mixing in a 5-turn coil, start reagent is added and mixed (another 5 turn coil) and the reaction allowed to occur in a 37° heating coil for 5-6 minutes. The absorbance is read using a filter in the range 525 to 535 nm in an AAII colorimeter.

Results: Concentration of fructosamines in umol/litre 24 Sample peak height x (Standard assigned concentration Standard peak height value i.e: 1000 umol/litre) Example 3d: Cobas Bio Fructosamine Assay Materials, Reconstitution Instructions and Reagents are the same as for Example 3a.

Table 7: Programming Parameters The type of analysis used may be Kinetic or Endpoint. Note that the endpoint method uses premixed reagents.

Kinetic Endpoint Units umol/L umol/L Calculation 0 0 Std 1 concentration 1000 1000 Std 2 concentration 1000 1000 Limit 0 0 Temperature (Deg C) 37.0 37.0 Type of Analysis 3 5 Wavelength (nm) 525 525 Sample volume (uL) 05 05 Diluent volume (uL) 25 25 Reagent volume (uL) 150 200 Incubation time (sec) 30 0 Start reagent volume (uL) 50 0 Time of first reading (sec) 20 20 Time interval (sec) 20 300 Number of readings 15 02 Blanking mode 1 1 Printout mode 1 1 Method 1. Set up sample plate - adding 200uL of sample to each tube. 2. Fill the tray with the diluent, start and standards. 3. Fit the clean cuvette tray into the machine (secure tightly). 4. Change the pipette tip for reagent (secure very tightly).

. Start the test going i.e.: "TEST NR 15" ENTER START. 6. After the assay has finished a sheet of results will be printed out. 7. Remove samples, tray and cuvettes.

NOTE: Printout mode 1 will give the final result in concentrat ion.

Example 3e: Cobas Mira Fructosamine Assay Materials, Reconstitution Instructions and Reagents are the same as for Example 3a.

Table 8: Program Parameters: GENERAL Measurement Mode Reaction Mode Calibration Mode Reagent Blank Cleaner Wavelength Decimal Position Unit Absorb R-S-SR1 Slope Avg Reag/Dil Before 550 nm 0 umol/L Last CALIBRATION Calib. Interval Standard Pos.

STD-1 STD-2 STD-3 : 14 Each Run 2 1000 umol/L no no ANALYSIS Post Dil. Factor Conc Factor Sample Cycle Volume Dilution Name Volume Reagent Cycle Volume Start Rl Cycle Volume Dilution Name Volume CALCULATION Sample Limit : no Reac. Direction : Increase Check : Off Convers. Factor : 1.00000 Offset : 0.00000 Test Range Low : 0 umol/L High : 3000 umol/L Norm. Range Low : 150 umol/L High : 360 umol/L Number of Steps : 1 Calc. Step A : Kinetic Readings First : 6 no no 1 uL H20 20 uL 1 200 UL 5 70 UL H20 2 0 UL 26 Example 4: Continuous Flow Monitor of Low Molecular Weight Fructosamino Acids This method was developed for following the purification of fructosamino acids, and designed for solutions containing a known low molecular weight fructosamino acid. It has been used in conjunction with other techniques (amino group analysis, osazone formation) in the quality control of batches of synthetic ketosamines used as calibrants.

Standard continuous f low AutoAnalyzer (Technicon, Tarrytown, NY) modules can be used, or their equivalent.

The sample is diluted with water, or with buffer (e.g. 0.5 mole/litre sodium acetate, pH 4.8) if samples of varying pH are to be assayed. This diluted sample is resampled (without desegmentation) into the three channels. The dilution depended on the application: for example, when assessing final preparation of fructosamino acids, 0.2 3 mL/min sample is mixed with 0.6 mL/min water, whereas for monitoring column fractions a larger dilution into buffer is more useful. The sample is mixed with NBT dissolved in trisodium or tripotassium phosphate with a GAFAC series detergent (e.g. RS610, RE410, RA600, all of which appropriate) to improve the continuous flow properties and the specificity of the reaction. This mixture is then either passed through a long reaction coil (coiled polyethlene is suitable) or through a heating module at 30-50° and the absorbance of the solution then read between 520 nm and 560 nm. The exact dilution, heating temperature and time (i.e. length of reaction coil) can be adjusted to change sensitivity, though increased temperature leads to some loss specificity. 27 Reagents: Water as the common diluent gives the highest sensitivity and best precision. This is how the system is used in the quality control of the final products. A concentrated buffer (0.5 mole/litre acetate) is used when monitoring column eluates, which differ widely in pH and ionic strength.

The method uses the same principle as used in Borsook's work (Abrams, A., Lowy, P.H. and Borsook, H. (1955) J. Amer. Chem. Soc. 77, 4794-4796? and Lowy, P.H., Borsook, H. (1956) J. Amer. Chem. Soc. 78, 3175-3176) for monitoring the purification of fructosamino acids, but using the exceptionally sensitive reduction indicator (Johnson, R.N., Metcalf, P.A., and Baker, J.R. (1982) Clin. Chim. Acta 127, 87-95) instead of ferricyanide. Since the formazan derived from the tetrazolium is insoluble in water it rapidly partitions into pump tubing and onto flow cell surfaces unless the reaction is carried out in the presence of detergent. Sodium dodecyl sulphate has the effect of slowing the reaction (which can be compensated for by using a heating module) and improving the specificity (for example, virtually abolishing the response of urate). Phosphate ester detergents are still more advantageous, being more effective at solubilizing the formazan while having a lesser effect on the reaction rate. The chosen reagent is 0.2 mole/litre NBT in 0.05 mole/litre trisodium (or potassium) phosphate with 0.5% w/v GAFAC detergent: tetrazolium is dissolved in water first and mixing this with a phosphate-detergent solution.

Example 5: Application as a Post-Column Reaction for the HPLC Assay of Ketosamines.

The continuous flow monitor for ketosamines can be adapted for use with an HPLC system in at least two ways: 1. Column eluate may be added to the segmented continuous flow analyzer instead of the resampled diluted sample. Absorbance is measured in a continuous flow colorimeter. 2. Using all narrow bore tubing, the reagent may be added to the column eluate using a second metering pump, and the mixture heated in a coil before the absorbance is measured using a standard HPLC detector.

If a calcium sulphonate based column is used it is advisable to incorporate a chelating agent (e.g. tetrasodium EDTA) to avoid calcium phosphate precipitates. 29

Claims (42)

WHAT WE CLAIM IS

1. A method to quantitatively determine ketosamine or fructosamine (each as hereinbefore defined) levels in a biological or other material sample, said method including the steps of: mixing said sample with a thiol blocking agent, a buffer solution with a pH above 10 and at least one anionic ester detergent; adding a start reagent to the mixture, said start reagent including at least one oxidant reagent; measuring the absorbance at a pre-determined wavelength, said wavelength determined by the or each oxidant selected; and comparing said absorbance with absorbance readings from a pre-selected standard.

2. A method as claimed in claim 1 wherein said thiol blocking agent, the buffer solution and the at least one anionic phosphate ester detergent are pre-mixed to form a diluent reagent.

3. A method as claimed in claim 2 wherein said diluent reagent further includes a chelating agent.

4. A method as claimed in either claim 2 or 3 wherein the diluent and the start reagents are mixed together immediately before mixing with said sample.

5. A method as claimed in any either claim 3 or 4 wherein the selection of the or each oxidant causes the predetermined wavelength of the absorbance measurement to be between 500 and 560 nanometres.

6. A method as claimed in any one of claims 3-5 wherein said method further includes measuring the absorbance at least twice; obtaining a value for the rate of colour production; and obtaining a value for the concentration of ketosamine or 30 2360 fructosamine by comparison of said rate of colour production with that of a pre-selected standard.

7. A method as claimed in any one of claims 3-5 wherein said pre-selected standard is a secondary standard and is assigned a calibration value by comparison with a synthesized fructosamine as the primary standard.

8. A method as claimed in any one of claims 3-8 wherein the thiol blocking agent is selected from the group consisting of iodoacetate and iodoacetamide.

9. A method as claimed in any one of the preceding claims wherein the buffer solution includes at least one electron accepting reagent, said reagents being selected from the group consisting of AQS, anthraquinone 2, 6-disulphonate, a tetrazolium salt and any combination of these reagents.

10. A method as claimed in claim 9 wherein said buffer solution includes AQS and nitroblue tetrazolium (NBT).

11. A method as claimed in claim 9 wherein the tetrazolium salts are sulphonated.

12. A method as claimed in claim 11 wherein the tetrazolium salt is sulphonated triphenyltetrazolium.

13. A method as claimed in claim 12 wherein the method of preparation of sulphonated triphenyltetrazolium includes: warming triphenyltetrazolium chloride with concentrated sulphuric acid until the hydrogen chloride effervescence ceases; adding oleum; heating the reaction mixture to between 170°C and 190°C; isolating the product as the calcium salt by diluting and neutralizing the reaction mixture with calcium carbonate; adding an equal volume of methanol; filtering the mixture; and 23604 concentrating and freeze drying the filtrate.

14. A method as claimed in any one of claims 12-13 wherein the at least one anionic ester detergent is selected from a phosphate ester of an alkyl polyethoxyethanol or an aryl polyethoxyethanol.

15. A method as claimed in any one of the preceding claims wherein the buffer solution includes an alkali.

16. A method as claimed in any one of claims 3-14 wherein the buffer solution includes an organic base.

17. A method as claimed in any one of claims 3-16 wherein said chelating agent is EDTA.

18. A method as claimed in any one of claims 3-17 wherein the start reagent further includes a sulphonic acid detergent and urea or a urea derivative.

19. A method as claimed in claim 18 wherein said acid detergent is a sodium salt of dodecane sulphonic acid.

20. A method as claimed in any one of claims 3-18 wherein the fructosamine level to be determined is that of fructosamino acid residues bound to haemoglobin and the method includes the steps of adding a haem complexing agent to the diluent reagent and a red cell lysing reagent.

21. A method as claimed in claim 20 wherein the diluent reagent further includes isonicotrinonitrile at a concentration of 50 millimole/litre and a nitrite salt at a concentration of 50 millimole/litre.

22. A method as claimed in any one of claims 1-16 wherein the sample is blood plasma and the buffer solution includes a 32 0 i' r r /i 1 mixture of a monohydrogen phosphate ions and phosphate ions in a molar ratio of 4:1.

23. A method as claimed in any one of the preceding claims wherein the fructosamine level to be determined is that of a high molecular weight fructosamine and said method further includes adding urea or a urea derivative to the sample prior to measuring the absorbance, said urea being at a concentration of at least 5 mol/L.

24. A method as claimed in claim 23 wherein said urea derivative is 2-imidazolidone, being at a concentration of at least 2 mol/litre.

25. A method as claimed in any one of claims 22-24 wherein said urea or urea derivative is included in the diluent reagent.

26. A method as claimed in any preceding claim wherein said pre-selected standard is calibrated using a primary standard.

27. A method as claimed in any preceding claim wherein said primary standard is selected from the group consisting of fructosamine alpha-carbobenzoxycarbonyl-epsilon-fructose-L-lysine (Fructose-CBZ-lysine), fructose-nickel-lysine, fructose-zinc-lysine and related derivatives.

28. A method for the preparation of the standards fructose-nickel-lysine, fructose-zinc-lysine and related derivatives, as discribed in either claim 26 or 27, said method providing: reacting lysine monohydrochloride in solution with a solid basic metal carbonate, hydroxide or oxide; adding excess glucose and heating the reaction mixture; isolating the product from the reaction mixture; and removing any glucose present. 33

29. A method as claimed in claim 28 wherein the metal used and the solvent are selected from the group consisting of nickel (II) and boiling methanol, zinc (II) and ethanediol at a temperature between 60°C and 80°C, and cobalt (II) and methanol.

30. A method as claimed in either claim 28 or 29 wherein said method further includes the steps of: heating the lysine monohydrochloride with the solvent under reflux with a suspension of the carbonate or oxide metal, for between 0.5 and 1 hour; adding the glucose in an approximately ten fold excess; and heating the reaction mixture for between 1 and 5 hours.

31. A method as claimed in any one of claims 28-30 wherein if ethanediol is the solvent, three volumes of methanol is added for each volume of ethanediol.

32. A method as claimed in any one of claims 28-31 wherein the product is isolated by precipitation from an alkaline solution.

33. A method as claimed in any one of claims 30-31 wherein the metal is zinc (II) and the product is isolated by precipitation as a sulphate by the addition of tetrabutyl ammonium hydrogen sulphate in methanol to the reaction mixture.

34. A method as claimed in any one of claims 28-33 wherein the method further includes the step of removing the excess glucose during the isolation of the product by gel filtration, the eluting fluid being selected from the group consisting of water and dilute acetic acid.

35. A method of preparation of the pre-selected standard, as claimed in any one of the preceeding claims 28-34, said method 34 2360 including the steps of: incubating human serum albumin (50 grams per litre) with 0.5 moles per litre glucose for 3-4 days at 37°C, in the presence of antibiotics to prevent decay; exhaustively dialyzing the solution to remove the glucose and the antibiotics; and concentrating and freeze-drying said solution the material obtained is calibrated by repeated comparisons with fructoselysine calibrants.

36. A method as claimed in claim 35 wherein the material obtained is calibrated by repeated comparisons with fructoselysine calibrants until an endpoint is reached.

37. A method as claimed in claim 35 wherein the material obtained is calibrated by conversion of the standard to hydroxymethylfurfural and assayed fluorometrically.

38. A method as claimed in claim 35 wherein the material obtained is calibrated by conversion of the standard to hydroxymethylfurfural and assayed by HPLC.

39. A reagent for use in a heterogeneous assay system for fructosamines (as hereinbefore defined), said reagent being formed in the absence of urea, from mixtures of AQS and NBT which tend to slowly precipitate, leading to a yellow deposit on the inside of plastic vessels or tubing containing the reagent.

40. A method to quantitatively determine ketosamine or fructosamine (each as hereinbefore defined) levels in a biological or other material sample substantially as hereinbefore described and with reference to any one of the Examples.

41. A method for the preparation of the standards fructose-nickel-lysine , fructose-zinc-lysine and related derivatives substantially as hereinbefore described. ^ L ^ 35 ^ 20NOV19< o / V 23604

42. A method of preparation of glycated albumin standards substantially as hereinbefore described. CANTERBURY AREA HEALTH BOARD by their authorized agents P.L. BERRY & ASSOCIATES per: ,Y A °X 36 V z vy 20 NOV 1991' 14