EP2235543A1

EP2235543A1 - Method to determine lipids

Info

Publication number: EP2235543A1
Application number: EP08864896A
Authority: EP
Inventors: Arne Ludvig Faaren; Frank Frantzen; Arne Kristian Nordhei; Erling Sundrehagen; Lars ÖRNING
Original assignee: Axis Shield ASA
Current assignee: Axis Shield ASA
Priority date: 2007-12-21
Filing date: 2008-12-19
Publication date: 2010-10-06
Also published as: JP2011505866A; GB0725069D0; WO2009081140A1

Abstract

The Invention provides for an enzymatic method for determining the plasma concentration of at least one specific class of lipid or specific class of lipoprotein in a blood sample. The invention also provides for kits used in the same method.

Description

METHOD TO DETERMINE LIPIDS

FIELD OF THE INVENTION

The present invention relates to the measurement of specific classes of lipid and lipoprotein in blood samples. Specifically, the invention relates to the measurement of plasma cholesterol and triglycerides associated with certain classes of lipoprotein, particularly by means of enzymatic assays. Most particularly, the invention relates to diagnostic and prognostic assays for use in automated methods, including those conducted on laboratory or "point-of-care" apparatus.

BACKGROUND TO THE INVENTION

The measurement of components in body samples is a common feature of clinical assessments. Many diagnoses can be made or confirmed, or the risk of many conditions can be assessed, from the concentration of a particular analyte in a body sample, or from the profile of concentrations of several components. The increasing number of known correlations between disease states and the concentrations of one or more analytes makes sample analysis an increasingly valuable tool, and also increases pressure on clinical laboratories to analyse ever larger numbers of samples increasingly quickly. To satisfy this, there is a need for assays which are quicker, higher throughput, simpler and more completely automated.

There is also a growing demand for assays to be carried out at the "point of care".

This is beneficial for the patient, who receives immediate advice and is left with less uncertainty. It is also beneficial for the medical practitioner, who can potentially avoid the need for a multiple appointments and can be more certain of his judgements.

Point of care assays are exceptionally demanding in terms of the need for simple manipulation and rapid results. If an assay takes more than a few minutes then much of the advantage of conducting it at the point-of-care is lost. Furthermore, although the staff conducting such assays are likely to be healthcare professionals, they are not analytical specialists and will not have access to multiple instruments. It is therefore necessary that such assays be designed to rely on the minimum of sample handling.

The most common clinical samples taken for assay are fluids, particularly blood and urine, since these are relatively easy to take and to manipulate. In blood, it is typically the content of the fluid plasma which is analysed. Some of the most common and clinically important measurements made on blood-derived samples relate to the plasma lipid content. The predominant lipids present in blood plasma are phospholipids, triglycerides (TG), and cholesterol (CH). Of these TG and CH are of particular diagnostic interest because of their association with cardiovascular disease, which in turn is one of the most prevalent diseases in the developed world.

Lipids are by their very nature water insoluble and in blood are transported in complex with apolipoproteins, which render them soluble. These complexes, the lipoproteins are classified into five groups, based on their size and lipid-to-protein ratio: chylomicrons, very low density lipoproteins (VLDL), intermediate density lipoproteins (IDL), low density lipoproteins (LDL), and high density lipoproteins (HDL). Chylomicrons are basically droplets of fat, and consist of TG to about 90%. Chylomicrons function as vehicles for the transport of dietary lipids from the ileum to adipose tissue and liver and are present in the general circulation for only a short period after a meal.

The four remaining classes of lipoprotein are produced in the liver. Whereas VLDL, IDL, and LDL are responsible for transporting lipids from the liver to the tissues, the fifth class, HDL is engaged in the reverse transport of superfluous lipids from peripheral tissues back to the liver for further hepatobiliary secretion. VLDL and IDL have short half-lives and deliver mainly TG to the tissues. LDL and HDL have longer half-lives and are the major participants in blood cholesterol homeostasis. On average LDL and HDL combined carry about 95% of the cholesterol present in blood, with LDL carrying about 70% and HDL about 25%.

There are five different types of proteins present in lipoproteins: apolipoprotein (Apo) A, B, C, D, and E, and each type may be further subdivided. The apolipoproteins are important for the formation, secretion, and transport of lipoproteins as well as the enzymatic activities working upon the lipoproteins in the peripheral tissues. Apolipoprotein B (ApoB) is the principal protein in VLDL, IDL, and LDL; in LDL, it is the only protein. HDL is devoid of apoB and its principal protein is apo-Al.

High concentrations of TG are associated with different pathophysiological disorders such as diabetes, cardiovascular disease, hyperlipidaemia, hyperglyceridaemia types I and IV, and nephritic syndromes. Low concentrations are found in hepatic infection and malnutrition.

From many epidemiological studies it is a well established fact that the CH associated with chylomicrons, VLDL, IDL, and LDL is a major risk factor for cardiovascular disease (CVD), with increasing concentrations correlating with increased risk of CVD. The CH associated with LDL particles is considered the main risk factor. CH associated with HDL on the other hand is inversely correlated with risk for cardiovascular disease. The lower the concentration of HDL, the higher the risk for cardiovascular disease. Therefore it is common practice to determine CH associated with LDL and HDL to diagnose and predict cardiovascular disease, as well as in formulating the risk of CVD, potentially in combination with other factors.

Two methods are currently used routinely for quantification of CH. Both methods are enzymatic. In the first method utilises an enzyme chain beginning with cholesterol esterase and cholesterol oxidase to generate a coloured or fluorescent signal by the generation of hydrogen peroxide. The other method substitutes cholesterol dehydrogenase in place of cholesterol oxidase and determines the amount of CH in the sample on the basis of the amount of the produced NADH or NADPH.

HDL associated CH may determined by separating this class of lipoprotein from non-HDL lipoproteins. After making the non-HDL unavailable, HDL associated CH is measured using the enzymatic methods of total CH.

Originally, and still much used, was precipitation of non-HDL by using one of the following: (i) heparin/Mg2+ (Hainline A et al (1982) Manual of laboratory operations, lipid and lipoprotein analysis, 2nd ed. Bethesda, MD: US department of Health and Human

Services, 1982: 151 pp),

(ii) phosphotungstate-Mg2+ (Lopes- Virella MF et al (1977) Clin Chem 23:882-4),

(iii) Polyethylene glycol (PEG) (Viikari J, (1976) Scan J Clin Lab Invest 35:265-8), and

(iv) dextran sulfate-Mg2+ (Finley et al (1978) Clin Chem 24:931-3).

The precipitated non-HDL is then removed by centrifugation. The latter method is still recommended by the Cholesterol Reference Method Laboratory Network as reference method for measurement of HDL associated CH (Kimberly et al (1999) Clin Chem 45:1803-12).

Other methods used separation by electrophoresis (Conlin D et al (1979) 25:1965-9) or chromatography (Usui et al (2000) Clin Chem 46:63-72).

The above methods are effective, but require lengthy separation steps and a number of laboratory instruments. In order to eliminate the laborious sample pretreatment two different routes have been taken. Point-of-care instruments have been developed that integrate the separation and quantification of HDL into the test device, which may be cassettes or reagent-impregnated strips, such as the Cholestech HDL assay device and method (US5213965). For the automatic clinical instruments, homogenous methods were developed which did not require a physical separation of non-HDL lipoproteins to measure the HDL associated CH fraction. The non-HDL particles were blocked by different methods and rendered inaccessible to the CH metabolizing enzymes. The most recent development has been highly specific surfactants that selectively dissolve HDL.

LDL associated cholesterol is commonly determined computationally using the Friedewald equation (Friedewald WT et al (1972) Clin Chem 18:499-501):

LDL = Total CH - (HDL + TG/2)

Although convenient and in most cases sufficiently accurate, it suffers from well- known limitations, in particular the need for the patient to fast before being bled (fasting depletes blood of chylomicrons) and the requirement for TG levels to be below 4 g/L. Therefore the NIH sponsored National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATPIII) guidelines have recommended using direct measurement of LDL associated CH rather than computing it from total CH, HDL associated CH, and TG.

Originally, LDL associated CH was measured using ultracentrifugation (Havel RJ et al, J Clin Invest 1955;34: 1345-53). This is still the most used reference method, but evidently requires sample pre-treatment. Homogenous methods were later developed which did not require a physical separation of non-LDL lipoproteins to measure the LDL CH fraction.

VLDL associated CH was originally measured using ultracentrifugation. This remains a preferred reference method, but during recent years homogenous methods for determining VLDL associated CH have been developed. These include methods from US6986998 and US7208287.

CH associated with IDL (also called "VLDL remnants" or "remnant-like particles") is commonly determined using ultracentrifugation, high performance liquid chromatography or electrophoresis. Two homogenous methods were recently developed (US7272047 and US 2007/0161068) that use specific surfactants to further the selective enzymatic decomposition of IDL associated cholesterol.

TG is determined routinely in a four step enzymatic reaction, in which lipoprotein lipase hydrolyzes TG to unesterified glycerol and free fatty acids. The glycerol is then phosphorylated (glycerokinase) and oxidized (glycerol-3 -phosphate oxidase) to di-hydroxy-acetone-phosphate and hydrogen peroxide, which is used to generate a coloured, fluorescent or chemiluminescent signal.

As for the measurement of CH of different lipoprotein classes, measurement of TG of specific lipoprotein classes may be performed by several methods that exploit different chemical and physical characteristics of the different lipoprotein classes.

However, all current methods, whether for use in the laboratory, at the point-of-care or on automatic clinical instruments, are based on plasma or serum. This is because blood cells interfere with the analyses at several levels: (i) cell associated lipids contribute to the measurement of plasma lipids, (ii) light scattering interference by the cells, and (iii) chromogenic interference of in particular erythrocyte hemoglobin. Blood cells are therefore removed prior to initiating the enzymatic reactions that convert the plasma lipids into measurable products, either in a separate pretreatment step or, as in some point-of-care instruments, in the sample devise.

Generally, a pre-treatment step is used, and blood cell plasma separation is accomplished by centrifugation. Usually the blood sample is transported from its site of collection to the clinical laboratory where centrifugation and sample analysis is performed. The steps of blood collection, transport to a clinical laboratory and centrifugation before performing the intended analyses constitute severe limitations both with respect to labour and time resources required and to risks of infection / contamination. It is an established fact that health-care workers experience an increased risk of acquiring in particular hepatic virus infections. Assays for detecting certain antigens and antibodies, which are practicable on whole-blood, have been proposed in US6143510; US2004/0048397. These methods are, however, Heterogenous methods of sandwich type and are only applicable to blood analytes that can react with pairs of specific binding partners (sbp). In the proposed method, one binding ligand is immobilized and serves to capture the selected blood analyte. The other ligand is labeled and allows detection of the captured analyte. Suitable analytes for these methods are polypeptides, proteins and larger carbohydrates which are in them selves specific binders or contain at least two recognisable epitopes. The methods are not applicable to analytes that are too small to allow binding of at least two specific binding ligands at separate sites, such as vitamins, which are too small, or lipids, for which specific binding partners cannot be raised by known methods.

Some assays for use at the point-of-care allow for whole-blood to be added to the test device. These devices, however, then incorporate separation steps, such as passage through membranes to separate the plasma from the intact blood cells. This is effective, but considerably increases the complexity of the assay device (e.g. cartridge) and therefore the cost and waste associated with the assay. Furthermore, since the separation is conducted directly on the whole blood sample added to the device, and there is a considerable dead-volume involved in such a separation method, the volume of sample required is considerably increased. Ideally, a point- of-care test should use a minimal volume of blood (e.g. less than 20 μl), so that this can be obtained with least discomfort for the patient. In-device separation also requires complex assay devices also stand a greater chance of blocking or generating unreliable results, especially where the separation is conducted at high concentrations of cells. If the discomfort or cost of a point-of-care assay becomes too great or the reliability too low, then the clinician, patient or medical insurer will elect for the assays to be carried out at a specialist laboratory rather than at the point of care.

In view of the above, there is evidently a considerably need for improved assays for one or more plasma lipid components. Preferably this method should be simpler, quicker, more reliable, require a smaller sample and/or require less instrumentation than currently available methods.

The present inventors have now surprisingly established that the perceived need for assays measuring plasma lipid components to be conducted in separated plasma, rather than whole blood, is not in fact an essential requirement of a successful assay, providing that the assay is suitably formulated. In particular, they have established that the lipid content of the blood cell component does not interfere with enzyme assays for lipoprotein associated lipids providing that the assay is conducted under conditions maintaining cell integrity. They have furthermore established that although scattering by whole blood cells increases assay background signals, this does not prevent effective measurement of lipid components by suitable methods. In addition, by appropriate selection of detection wavelength, the interference caused by haemoglobin in the sample can be tolerated.

BRIEF SUMMARY OF THE INVENTION

In a first aspect, the present invention thus provides an enzymatic method for determining the concentration of at least one specific class of lipid or specific class of lipoprotein in the plasma portion of a blood sample comprising plasma and intact blood cells, said method comprising;

i) contacting said sample with a reagent mixture comprising a lipid converting enzyme, wherein said reaction mixture causes the selective reaction of said specific class of lipid or said specific class of lipoprotein and concomitantly generates a reaction product or consumes a reaction substrate;

ii) contacting said sample with a reagent mixture whereby to convert said reaction product or reaction substrate into a detectable indirect product;

iii) detecting said indirect product; iv) relating an amount of said indirect product detected or a rate of formation of said indirect product to the concentration of said specific class of lipid or specific class of lipoprotein in said blood sample;

wherein step i) is conducted under conditions which substantially maintain said blood cells in an intact state.

The conditions of the specific reaction step (step i) should be such as to substantially maintain the blood cells in an intact state. This requires control of the chemical environment of the assay in a way that has not been provided in any previous assay for lipids or lipoproteins. Specifically, the reagents added during step i (or during step i and any additional steps carried out simultaneously) should control the osmotic pressure and surfactant content of the assay so as to substantially avoid cell lysis. Other factors such as the presence of solvents and the reaction pH will have lesser effects. After osmotic strength, which should preferably be approximately isotonic, the use of detergents and/or surfactants will have the next most significant effect upon cell lysis in the system. These are typically included in the assay mixture and should be of a type and concentration such that they do not cause a substantial degree of cell lysis. In low concentrations, surfactants tend to have a protective effect upon the cells, while at higher concentrations they are often lysogenic.

In a further aspect, the present invention additionally provides a kit for use in determining the concentration of at least one specific class of lipid or specific class of lipoprotein in the plasma portion of a blood sample comprising plasma and intact blood cells, said kit comprising;

a) a first reagent mixture formulated to cause the selective reaction of said specific class of lipid or said specific class of lipoprotein whereby to generate a reaction product or consume a reaction substrate; b) a second reagent mixture formulated to cause conversion of said reaction product or reaction substrate into a detectable indirect product;

c) optionally a reagent for causing cell lysis;

d) optionally an inhibitor for inhibiting the generation of said reaction product or reaction of said substrate;

DETAILED DESCRIPTION OF THE INVENTION

As indicated above, the conditions of step i) are chosen such that the cells of the blood sample remain intact, or substantially intact (e.g. less than 20% of the cells, preferably less than 10%, and most preferably less than 5% (e.g. 0.1 to 5%) of the cells suffer lysis during step i). Factors including ionic strength and surfactant type/concentration are particularly important in this respect.

One property of non-ionic surfactants of particular importance for the integrity of cells, is the hydrophil-lipophil balance (HLB). Surfactants with HLB values below 10 and above 13 are in particular suitable for use in the presence of intact cells. Such surfactants are thus preferred for use in the method of the invention (e.g. surfactants with HLB from 1 to 10 or from 13 to 20). However, surfactants with HLB values between 10 and 13 (while less preferred) may also be compatible with maintaining intact cells depending on the concentrations used, and depending on the construction and formulation of the assay, such as reaction times and temperatures used, ionic strength, pH and types of salts used in the assay mixture, and the presence of stabilizing substances such as serum albumin. As will be known to the skilled worker, shorter reaction times; higher concentrations of stabilisers such as albumin; and closer to physiological temperatures, pH and ionic strengths will all tend to increase the stability of the cells. In combination with suitable types and concentrations of surfactant as discussed herein, the skilled worker will thus be able to routinely identify all the necessary conditions to allow good integrity of the cells in any individual case. The method of the invention is preferably carried out with the reagent mixtures of steps i and ii formulated as a single reagent mixture. Evidently, in such a case, steps i) and ii) of the method are conducted simultaneously. This reduces the number of individual reagent mixtures which must be loaded into the instrument and accommodated therein, and furthermore allows a faster assay, since steps i and ii are conducted simultaneously.

Optionally, reaction accelerators such as disclosed in US6818414 may be included in the reagent solution, especially at step i) to increase reaction rate and thus decrease assay time, as long as the included accelerator allows maintaining blood cells in an intact state.

Examples of accelerators, which act specifically on the cholesterol reaction are: (i) membrane intercalators, which increase release of cholesterol from the phospholipid environment of the outer shell of lipoproteins. Examples are amphiphatic alcohols (octanol (see Example 11), hexadecanol; 0.1-100 mM), ceramids (N-hexanoyl-D-erythrosphingosine; 1-1000 μM), diglycerides (1,2- dioctanoyl-s,n-glycerol, hexadecyl glycerol, dipalmitoyl glycerol; 1-1000 μM), and the steroid based antibiotic, fusidic acid;

(ii) triglyceride or lipoprotein lipase (see Example 1 1), which increase release of cholesterol from the inner triglyceride rich core of lipoproteins; (iii) substances that stabilize released cholesterol in the solvent phase, such as serum albumin, polyethylen glycol 6000, cyclodextrins, and dextran; (compounds are available from Sigma-Aldrich, Fluka, and Molecular Probes (Eugene, OR)).

Alternatively, reactions may be accelerated by increasing the reaction temperature. For instance in Example 12 the time needed for the measurement of HDL cholesterol was 180s (3 min) at 37°C. This time (i.e. the time needed to reach the same amount of colored product formed) could be reduced to 105s when the reaction temperature was increased to 42⁰C and further reduced to 75s when the reaction temperature was increased to 47°C.

With regard to the selective reaction of specific class of lipid or specific class of lipoprotein, and the conversion of the reacted lipoprotein to detectable secondary analyte, there are a number of methods which are well known in the art and any of these are suitable for use in the present invention. All of the homogeneous methods described above are suitable and included within the scope of the invention. Further details are supplied below and in the referenced citations.

These methods have previously been unsuitable for use in the method of the invention because it has been assumed that the method must be carried out on separated plasma in the absence of intact blood cells. To be suitable for use in the present invention, these methods need to be adapted so as not to cause cell lysis, or to cause this only at appropriate stages, as described herein. This adaptation can be made by controlling appropriate factors such as tonicity and surfactant type and content as described in greater detail below.

Two methods are currently used routinely for quantification of CH. Both methods are enzymatic and suitable for use in the method of the present invention. In the first method, cholesterol esterase converts cholesterol esters into CH. Cholesterol oxidase then converts CH to choleste-4-ene-3-one and hydrogen peroxide. Finally hydrogen peroxidase uses the hydrogen peroxide formed to convert 4-aminoantipyrin in the presence of phenol to generate a coloured quinoneimine compound. The quinoneimine is monitored by photometry at 500-600 nm wavelength. Other well known colour producing substrates (e.g. TMB which product is blue and monitored at 650 nm) or fluorescent or chemiluminescent substrates may be substituted for the 4-aminoantipyrin/phenol .

The other method substitutes cholesterol dehydrogenase for cholesterol oxidase and determines the amount of CH in the sample on the basis of the amount of the produced NADH or NADPH. With regard to the detection of TG, this is determined routinely in a four step enzymatic reaction, in which lipoprotein lipase hydrolyzes TG to unesterified glycerol and free fatty acids. The glycerol is then phosphorylated by glycerokinase and oxidized by glycerol-3 -phosphate oxidase to di-hydroxy-acetone-phosphate and hydrogen peroxide. In a final colour forming step hydrogen peroxidase uses the hydrogen peroxide formed to convert 4-aminoantipyrin in the presence of phenol into a coloured quinoneimine. The quinoneimine is monitored spectrophotometrically at 500 nm wavelength. By selecting the appropriate substrates and phenols the formed coloured products may be monitored at wavelengths from 450 to 850 nm. Likewise fluorescent or chemiluminescent substrates may be substituted for the 4-aminoantipyrin. Evidently, the final steps in both peroxide-based methods are equivalent and interchangable.

Thus, the enzyme reactions converting the specific plasma lipid component into a detectable chemical product preferably may be performed using the above described enzyme systems.

For CH:

(i) cholesterol esterase + cholesterol oxidase + peroxidase, or (ii) cholesterol esterase + cholesterol dehydrogenase.

For TG:

(iii) lipase + glycerol kinase + glycerol-3 -phosphate oxidase + peroxidase.

As enzymes, commercially available enzymes derived from animals, micro- organisms or plants may be used. Enzymes from specific sources may display selectivity for certain lipoprotein classes, such as lipoprotein lipase and cholesterol esterase from Chromobacterium viscosum or Pseudomonas which react preferentially with the lipoprotein class VLDL. Such enzymes may be used to assay for a specific component, either in isolation or in combination with the other selection methods described herein. The enzymes may be chemically modified so as to change their specificity and stability, e.g conjugation of cholesterol oxidase and cholesterol esterase with PEG in order to make the enzymes less reactive with LDL associated CH (US5807696). Enzymes are typically used at concentrations from 100 - 100,000 U/L.

Measurement of lipid components of specific lipoprotein classes may be performed by several methods that exploit different chemical and physical characteristics of the different lipoprotein classes. In general, these methods rely on the specificity of the enzyme, and/or allow the reaction of the desired component after separating, converting or rendering inactive those other components which might interfere.

Non-ionic, anionic, cationic, and zwitterionic surfactants may be included in order to increase the selectivity of the enzymatic reactions or to increase the rate of reactions. Any suitable surfactant may be used that allows maintaining the intact status of the cells during the course of the reactions. One property of non-ionic surfactants of particular importance for the status of cells, is the hydrophil-lipophil balance (HLB). Surfactants with HLB values below 10 and above 13 are in particular suitable for use in the presence of intact cells. However, surfactants with HLB values between 10 and 13 may also be compatible with intact cells depending on the concentrations used, and depending on the construction and formulation of the assay, such as reaction times and temperatures used, ion strength, pH and types of salts used in the assay mixture, and the presence of stabilizing substances such as serum albumin. Examples of suitable surfactants are polyoxyethylene alkyl ethers (Brij 35 and 78), polyoxyethylene alkyl aryl ethers (Triton X45 and X305, Igepal 210 and 272), polyoxyethylene sorbitan monolaurate monolaurate (Tween 80), polyoxyethylene- cooxypropylene block copolymer (Pluronic F68 and L64), telomere B monoether with polyethylene glycol (Zonyl FSN 100), ethylenediamine alkoxalate block copolymer (Tetronic 1307), 2,4,7,9-tetramethyl-5-decyne-4,7-diol ethoxylate (Surfonyl 465 and 485), polydimethylsiloxane methylethoxylate (Silwet L7600), polyethoxylated oleyl alcohol (Rhodasurf ON-870), polyethoxylated castor oil (Cremophor EL), p-isononylphenoxy-poly(glycidol) (Surfactant 10G), and a polyether sulfonate (Triton X200). Surfactants are used for this purpose typically at concentrations of 0.001-10%, preferably 0.01 to 1%.. The following methods are among those suitable for selectively reacting the particular lipoprotein components as described. In all cases, the referenced published material is incorporated herein by reference.

In homogenous methods to measure the HDL associated CH fraction, the non-HDL particles have been blocked from reaction by different methods and rendered inaccessible to the CH metabolizing enzymes. The most recent development has been highly specific surfactants that selectively dissolve HDL. These include:

(i) In the PEG/cyclodextrin method (US5691159) sulfated alpha-cyclodextrins in the presence Of Mg²⁺ forms soluble complexes with non-HDL which thereby become refractory to breakdown by PEG modified enzymes.

(ii) The polyanion method (US5773304) uses a synthetic polymer together with a polyanion to block non-HDL and make them refractory to solubilization with specific detergents and enzymatic measurement.

(iii) The immunologic method (US6162607) exploits the presence of apolipoprotein B in all non-HDL and its absence in HDL. Antibodies to apoB block non-HDL for reaction with the cholesterol enzymes.

(iv) In the clearance method (US6479249) non-HDL is first consumed in a reaction not generating colour. A specific detergent is then added allowing the enzymes to react with HDL in a reaction generating colour.

(v) The accelerator/detergent method (US6818414) degrades unesterified cholesterol of non-HDL using an accelerator to speed up the reaction, and removes the formed H2O2 in a process not generating colour. In a second step HDL cholesterol is degraded using an HDL-specific detergent in a colour forming process. LDL associated CH has been measured by homogenous methods which did not require a physical separation of non-LDL lipoproteins to measure the LDL CH fraction. Such methods include:

(i) US5888827 describes a method whereby non-LDL is masked by surfactants and cyclodextrins in the presence OfMg²⁺ and becomes refractory to breakdown by PEG modified enzymes.

(ii) US5925534 describes a method using polyanions and surfactants to protect LDL in a sample and allow non-LDL to be enzymatically eliminated whereupon addition of a deprotecting reagent allows the enzymatic determination of LDL associated CH.

(iii) US 60571 18 and US 6194164 describe two different methods ustilizing specific surfactants to selectively eliminate non-LDL associated CH in an enzymatic reaction before determining LDL associated CH.

During recent years homogenous methods for determining VLDL associated CH have been developed. These include:

(i) US6986998 describes a method using albumin and calixarene to block HDL and LDL, respectively, allowing the selective decomposition of VLDL in an enzymatic reaction using VLDL selective enzymes from Chromobacterium viscosum.

(ii) US7208287 describes a method using specific surfactants to selectively decompose VLDL in an enzymatic reaction.

Cholesterol associated with IDL (also called "VLDL remnants" or "remnant-like particles") is commonly determined using ultracentrifugation, high performance liquid chromatography or electrophoresis. Two homogenous methods were recently developed (US7272047 and US 2007/0161068) that use specific surfactants to further the selective enzymatic decomposition of IDL associated cholesterol. measurement of TG of specific lipoprotein classes may be performed by methods that exploit different chemical and physical characteristics of the different lipoprotein classes. These methods are thus analogous to those described above for cholesterol, but utilising enzymatic detection of TG, such as by those methods described herein above.

The detection method used in the assay methods of the present invention is typically photometric, and the indirect product is generally chosen such that it is detectable photometrically (e.g. by its absorbance, fluorescence or chemiluminescence at one or more pre-identified wavelengths). The presence of intact blood cells in such an assay tends to increase the non-specific background signal, especially for absorbance methods, but does not alter the change or rate of change in signal, which is used to assess the concentrations of the analytes in the sample. Since the signal is detectable without removal of the blood cells at any stage, in one preferred embodiment, the blood cells remain intact or substantially intact in the sample throughout the assay method (e.g. for steps i to iv).

A highly preferred method for signal generation is via hydrogen peroxide, which serves as a substrate for the enzymatic oxidation of a colour-producing substance. The oxidizable color producing reagent or reagents that react with formed hydrogen peroxide to produce the detectable chemical product may be any molecule known in the art, the oxidized product of which can be measured by ultraviolet, visual, or infra-red spectroscopy, or by fluorescence or luminescence.

Examples of suitable chromogenic reagents are Trinder reagents, which in the presence OfH₂O₂ react with a coupler to form colored products. Preferred examples of couplers are 4-aminoantipyrin (4AA), 3-methyl-2-benzolinone hydrazone (MBTH), N-methyl-N-phenyl-4-aminoaniline (NCP-04), N-methyl-N-(3- methylphenyl)-4-aminoaniline (NCP-05), N-methyl-N-(3-methoxyphenyl)-4- aminoaniline (NCP-06), N-methyl-N-(3-methoxyphenyl)-3-methoxy-4-aminoaniline (NCP-07). Preferred examples of Trinder reagents are those forming products that can be colorimetrically determined at wavelengths above 600 nm: N-Ethyl-N-(2- hydroxy-3-sulfopropyl)-3,5-dimethyloxyaniline (DAOS), N-Ethyl-N-(2-hydroxy-3- sulfopropyl)-3,5-dimethoxy-4-fluoroaniline (FDAOS), N-(2-hydroxy-3- sulfopropyl)-3,5-dimethoxyaniline (HDAOS), N,N-bis(4-sulfobutyl)-3,5- dimethylaniline (MADB), N-ethyl-N-(2-hydroxy-3-sulfoprpyl)-3,5-dimethylaniline MAOS). Preferred examples that are not Trinder reagents are: 3,3',5,5'-

Tetramethylbenzidine (TMB), N-(3-sulfopropyl)-3,3',5,5^l-tetramethylbenzidine (TMBZ-PS), N,N-bis(2-hydroxy-3-sulfopropyl)tolidine (SAT Blue), N- (carboxymethyl-aminocarbonyl)-4,4-bis(dimethylamino)-biphenyl amine (DA64), 10-(carboxymethylaminocarbonyl)-3,7-bis(dimethylamino)phenotiazine (DA67). The concentration of the chromogen is preferably 0.01-10 g/L, and is limited by solubility.

Examples of suitable fluorescent substrates are dihydrocalceins, dihydroethidium, dihydrofluoresceins, dihydrophenoxazine (Amplex red;10-acetyl-3,7- dihydroxyphenoxazine), and dihydrorhodamines.

Examples of suitable chemiluminescent substrates are luminol (3-aminotriphenylene complexes), Lumigen PS-2 and Lumigen PS-atto.

Although the present inventors have established that the presence of intact blood cells does not prevent the detection of an effective signal in the methods of the invention, where the speed and complexity of the assay method allow, the increased background signal generated by the presence of intact cells may be reduced by elimination of some or all of the intact cells prior to detection of the indirect product. In one embodiment, the present invention therefore provides that the concentration of intact blood cells is reduced after step i but before the detection step iii. In relation to the reduction of intact blood cells, the term "concentration" is used herein to indicate concentration relative to the concentration of the other components (especially the signal-generating secondary analyte or the reaction product or reaction substrate referred to in step (i)). Thus, simple dilution is not intended to be encompassed by such a reduction in concentration, but rather the removal or elimination of whole blood cells without substantial elimination or removal of the other (especially the signal-generating secondary analyte or the reaction product or reaction substrate referred to in step (i)).

There are two key methods by which the concentration of cells may be reduced; by lysis of the cells (fragmented cells block less light), or by their physical removal. Thus, in one preferred method, after step i) of the method, an inhibitor is added to inhibit further reaction (e.g. to inhibit further reaction of the lipid converting enzyme or further generation of the indirect product) and the intact blood cells in the sample are reduced by means of cell lysis. This does not interfere with the assay method because the generation of the indirect product has reached the desired stage, and the inhibitor prevents further generation of detectable products from the material released by cell lysis. A suitable inhibitor of any step in the pathway from the lipid to the indirect product may be used, such as an inhibitor of the lipid converting enzyme. An inhibitor of any other enzyme in the reaction pathway to generate the secondary analyte will, however be effective and where generation of peroxide is a step in the reaction pathway, peroxidise inhibitors are particularly suitable. Cell lysis may be achieved by different means, such as by lowering the osmotic pressure causing the cells to take up water, swell and disrupt, or by using detergents that dissolve the cell membrane. Such methods are known in the art.

Non-ionic, anionic, cationic, and zwitterionic surfactants may be used for the purpose of cell lysis prior to measurement of colored indirect analyte. Any suitable surfactant may be used that does not affect the measurement, typically at concentrations between 0.01-10%, but preferably between 0.1-2%.

Examples of suitable surfactants are the anionic surfactants amine aklylbenzene sulfonate (Ninate 411), sodium dioctylsulfosuccinate (Aerosol OT), sodium N-oleyl- N-methyltaurate (Geropon T-77), sodium olefin sulfonate (Bioterge AS-40), sodium polyoxyethylene lauryl sulphate (Standapol ES-I), and the non-ionic surfactants polyoxyethylene alkyl aryl ether (Triton XlOO and Xl 14) and polyoxyethylene lauryl alcohol (Chemal LA-9). Where cell lysis is conducted, there will evidently be a corresponding release of haemoglobin and other cell products into the reaction medium. In order to gain maximum advantage from the lysis and reduce the background signal as far as possible, it is thus preferred in such cases to select a product (secondary analyte) for which detection is not inhibited by these released products. Secondary analytes detectable at wavelengths above 450 nm, preferably above 500 nm and most preferably 600 nm or higher (e.g. 500 to 1400 nm or 600 to 1200 nm) are thus preferred where cell lysis is used.

As noted above, in the embodiment of the present invention utilising peroxide, inhibition of peroxidase is a suitable method to avoid generating further signal after cell lysis. The common peroxidase inhibitor, Methimazole (2-mercapto- 1 - methylimidazole) is particularly suitable. Examples of other inhibitors of peroxidase reaction are: alphatocopherol, benzhydroxamic acid, diethyldithiocarbamate, dithiothreitol, glutathione, hydroquinone, hydroxylamine, phenyl hydrazine, potassium cyanide, PTU (6-n-propyl-thiouracil), salts of metals (Al²⁺, Hg²⁺, Mn ⁺), sodium azide, Tiron (dihydrobenzene disulfonic acid).

If non-Trinder chromogenic substrates (e.g. DA67) are used, a Trinder coupler (e.g. 4-aminoantipyrine) may be successfully used as a competitive inhibitor (see Table 2 in Example IF below).

Inhibitors of other enzymes used in the conversion of specific lipid to detectable chemical product may also be used, in combination or alone. For the preferred embodiments indicated herein, these include:

Cholesterol oxidase inhibitors: Fenipropimorph, l-fluoro-2,4-dinitrobenzene, N- bromosuccinimide, salts of metals (Ag⁺, Fe²⁺, Hg²⁺), sarcosyl, SDS, trinitrobenzene sulfonate, Triton X-100 (>0.2%).

Cholesterol esterase inhibitors: cetyltriethylammonium bromide, diethyldicarbonate, diisopropyl fluorophosphates, dithiothreitol, 1-hexadecanesulfonyl chloride, 4- hydroxymercury benzoate, N-bromosuccinimide, p-chloromercury benzoate, phenyl methyl sulfonyl fluoride, potassium cyanide, protamine, salts of metals (Ag⁺, Ca²⁺, Cd²⁺, Cu²⁺, Fe²⁺, Fe³⁺, Hg²⁺), Triton X-IOO (>1%).

Glycerol-3-phosphate dehydrogenase inhibitors: ATP, N-ethyl maleimide, p-chloro mercury benzoate, 1,10-phenantroline, salts of metals (Cu²⁺, Mn²⁺, Ni²⁺, Zn²⁺). Lipoprotein lipase. 1 ,1 -bis(aniline)-4.4'-bis(naphtalen)-8,8'-disulfonate, dodecylsulfonyl fluoride, protamine, sodium chloride (>0.5M).

Hepatic lipase inhibitors: cetyltri ethyl ammonium bromide, p-chloromercury benzoate , diethyl 4-nitrophenyl phosphate, EDTA, 1 -hexandecansulfonyl chloride, 4-hydroxymercury benzoate, 2-mercaptoethanol, N-bromosuccinimide, 1,10- phenantroline, protamine, salts of metals (Ag⁺, Ca²⁺, Co²⁺, Cu²⁺, Fe²⁺, Hg²⁺, Mn²⁺, Ni²⁺, Zn²⁺), SDS, Triton XlOO (>1%).

Glycerol-3-phosphate oxidase inhibitors: Acriflavone, atebin, benzoyl formic acid, glyoxylic acid, methylglyoxidde, sodium azide.

The second method by which the intact blood cells in the sample may be reduced is by means of physical separation. This may be by any suitable method including flow past a specific binder for the whole cells, but will most commonly be by filtration. It is notable that filtration to remove cells following reaction step (i) has considerable advantages over filtration of the original blood sample. In particular, the reaction step can be carried out with a small volume of blood sample from a finger-stick. After one or more reagents have been added, however, the volume of the reaction mixture is much greater and separation of the cells can be conducted without the dead-volume being significant. The cells are also more dilute and the filtration thus more effective.

In filtering, the sample is passed through a porous interface (filter, membrane or sieve) of small pore size, in which the cells are trapped. The porous interface may be solid (e.g. sintered glass) or fibrous (e.g. made of cellulose or glass), and the flow may be transversely through or laterally. The sample may be passed through the filter by gravity, centrifugation, capillary forces, pressure or suction. The filtering material may additionally contain reagents (e.g. lectins, antibodies) that capture the blood cells. Other methods use electrostatic attractions. Suitable materials for transverse filtering are given in Table 1. Suitable materials for lateral flow filtering are Hemasep® L (Pall Corp) and LFl and MFl (Whatman).

Alternatively, the intact blood cells may be separated by filtration prior to monitoring formed colored product. The intact blood cells may be separated before or after the enzymatic reactions have been completed, but where this takes place before, then a controlled reaction period is generally necessary. Table 1 lists different filters and their effect on the recovery of colored product and lipids.

Table 1. Filtration recovery of lipids and colored product

In the methods of the present invention including a step of reducing the concentration of intact blood cells after step i but before the detection step iii, the concentration of cells is preferably reduced by at least 50% (e.g. 50 to 90%), preferably by at least 70% (e.g. 70 to 95%), and most preferably by at least 90% (e.g. between 90% and substantially 100%) . As indicated above, the relevant "concentration" is preferably not absolute concentration as might be reduced by dilution, but is rather concentration relative to the sample as a whole, and in particular relative to the concentration of the reaction product or reaction substrate referred to in step (i), or to the secondary analyte.

One of the considerable advantages of the use of whole-blood in the present assay method is the resultant decresase in the assay time. It is thus preferable the at the time from adding the reagent in step i) to the detection of the secondary analyte in step iii) is no more than 7 minutes, preferably no more than 5 minutes, and more preferably in the range 3 to 5 minutes.

It is a further advantage of the present invention that it may be carried out on a minimal volume of blood sample. It is preferred, therefore that the volume of whole blood sample in step i) of the assay method is no greater than 40 μL, preferably no greater than 20 μL and most preferably no greater than 12 μL (e.g. 12 to 25 μL).

In the kits of the present invention, the first reagent mixture and said second reagent mixture are preferably formulated together as a single reagent mixture. This reduces the number of steps, and thus reduces the assay time, and furthermore reduces the reagent storage and handling demands on the assay equipment, allowing the method to be practiced with less sophisticated automatic analysers.

The kit preferably comprises an inhibitor to inhibit further reaction of said specific class of lipid or said specific class of lipoprotein, or to inhibit the generation of further secondary analyte. Suitable inhibitors are disclosed herein are those named above are particularly preferred both individually and in combination.

The kit may additionally contain a lysing agent to cause cell lysis. This is preferably in combination with an inhibitor and suitable lysing agents are again disclosed herein and those mentioned herein (including in the examples attached hereto) are preferred. Surfactants of HLB of 10 to 13 are particularly preferred, as are tonicity adjusters. Reference samples of at least one lipoprotein selected from the group consisting of very low density lipoprotein (VLDL), intermediate density lipoprotein (IDL), low density lipoprotein (LDL), and high density lipoprotein (HDL) may also be incorporated into the kits of the invention.

The invention will now be further illustrated by the following non-limiting examples, and the attached Figures, in which:

Fig.1. illustrates the effect of intact blood cells on the measurement of plasma TG. The upper graph depicts the strong effect of intact cells on the signal. The middle graph depicts the same results normalized to give the same initial absorption level. The intact cells have no or only a very small effect on the development of the curve, i.e. on the conversion of TG to colored product. Thus, intact blood cells do not contribute lipid to the determination of plasma TG. The lower graph shows that lysed cells on the other hand do contribute lipid to the determination of plasma TG.

Fig.2. Absorption profiles of intact and lysed blood cells.

Fig.3. Dose dependent inhibition of hydrogen peroxidase by Methimazole.

Fig 4. illustrates the effect of intact blood cells on the measurement of plasma total CH. The upper graph depicts the strong effect of intact cells on the signal. The middle graph depicts the same results normalized to give the same initial absorption level. The intact cells have no or only a very small effect on the development of the curve, i.e. on the conversion of plasma CH to colored product. Thus, intact blood cells do not contribute lipid to the determination of plasma CH. The lower graph shows that lysed cells on the other hand do contribute lipid to the determination of plasma CH.

Fig.5. illustrates the effect of intact blood cells on the measurement of HDL associated CH. The upper graph depicts the strong effect of intact cells on the signal. The more intact cells are present, the higher the signal. The lower graph depicts the same results normalized to give the same initial absorption level. The inset shows the slope of the curves (i.e. rate of product formation) as a function of cell concentration. The intact cells have a nonsignificant effect on the development of the curve, i.e. on the conversion of HDL associated CH to colored product. Thus, intact blood cells do not contribute lipid to the determination of HDL associated CH.

EXAMPLES

Example 1. Evaluation of the effect of intact blood cells on the measurement of plasma TG.

Blood was drawn from healthy volunteers into EDTA-plasma tubes. Plasma and cells were separated by centrifugation at 200Og for 15 min. All reactions were performed at 37⁰C.

(A)

6 μL of human EDTA plasma with or without 1.5μL of blood cells was added 500 μL of ABX Pentra Triglycerides CP assay reagent. Samples containing blood cells were rapidly hemolysed by the assay reagent, and the released hemoglobin, which absorbs strongly between 400 to 500 nm (Fig. 2), prohibited measurement of plasma TG.

(B)

To overcome this problem, the ABX Pentra Triglyceride CP reagent was reformulated so as to maintain cell integrity. 6 μL of plasma with or without 1.5μL of blood cells or 1.5 μL of blood cells in buffer was added to 500 μL of the reformulated assay reagent, containing lipoprotein lipase (108 kU/L), glycerokinase (505 U/L), glycerol-3 -phosphate oxidase (4150 U/L), peroxidase (495 U/L), ATP (3.14 mmol/L), p-chlorophenol (2.69 mmol/L), 4-aminoantipyrin (0.31 mmol/L), and magnesium chloride (15 mmol/L) in Good's buffer (50 mmol/L), pH 7.4. Color development was measured at 500 nm. The blood cells caused a significant increase in background absorbance (Fig.1 , upper), but no temporal increase in color (Fig.1 , middle). Therefore blood cell associated TG did not participate in the reaction. The uneven progress curve obtained in the presence of intact blood cells is partly a consequence of the high background (>2 absorbance units) and partly a consequence of the Brownian motion of the blood cells. The slight decrease in maximal absorbance seen in the presence of intact blood cells was due to a slow precipitation of the cells, which decreased the number of cells in the pathway of the light with time (Fig.1 , middle, dotted line).

(C) In a direct experiment performed with 5 replicates, 5 μL of blood plasma with or without 5 μL of blood cells was incubated for 5 min at 37⁰C with 500 μL assay reagent containing lipoprotein lipase (107 kU/L), glycerokinase (530 U/L), glycerol- 3-phosphate oxidase (4000 U/L), peroxidase (500 U/L), ATP (3.1 mmol/L), HDAOS (2.5 mmol/L), magnesium chloride (15 mmol/L) and 4-aminoantipyrin (0.3 mmol/L) in Good's buffer (50 mmol/L), pH 7.4. Samples were then centrifuged to sediment blood cells and the supematants were monitored at 600 nm for formation of colored product. The average reading for the five plasma samples was 259 ± 3 mAU, and for plasma + cells it was 259 ± 2 mAU. Blood cell associated TG clearly did not contribute to the measurement of plasma TG.

The non-specific increase in background is due to the light scattering effect of the intact blood cells present in the pathway of the light used to measure the amount of colored product formed. This interference may be reduced by cell lysis prior to measurement, in particular if measurement is performed using wavelengths at 600 nm or higher (Fig. 2)

(D) In another experiment, 5 μL of plasma with or without 5 μL of blood cells or 5 μL of blood cells in buffer were mixed with 500 μL of ABX Pentra Triglycerides CP assay reagent, which causes cell lysis. Formation of colored product was monitored at 600 nm. Cell lysis strongly reduced background and gave smoother progress curves compared to when cells were kept intact, but TG released from the lysed cells contributed to formation of colored product (Fig.1, lower). (E) In yet another experiment, 5 samples of 5 μL of blood plasma + 5 μL of blood cells were mixed with 500 μL of assay reagent. Different amounts of the peroxidase inhibitor, 2-mercapto-l-methylimidazole (Methimazole) were added to the incubation mixtures simultaneously with the initiation of cell lysis by addition of Triton XlOO to a final concentration of 0.1%. The peroxidase inhibitor dose- dependently inhibited the formation of colored product from cell associated TG (Fig.3).

(F) To investigate the ability of different peroxidase inhibitors to quench formation of colored product formed from triglycerides, the inhibitor was added together with a hemolytic concentration of Triton XlOO (0.1% final concentration) to a mixture of whole blood sample and lipoprotein lipase (105 kU/L), glycerokinase (520 U/L), glycerol-3-phosphate oxidase (3900 U/L), peroxidase (500 U/L), ATP (3 mmol/L), magnesium chloride (15 mmol/L), and either HDAOS (2.4 mmol/L) + 4- aminoantipyrine (0.3 mmol/L) or DA67 (10 mmol/L) in Good's buffer, pH 7.4 (50 mmol/L). The absorbance was measured at 630 nm after 2 min incubation with the inhibitor. The results are shown in Table 2.

Table 2. Inhibition of enzymatic conversion of cell-associated triglycerides from lysed cells.

*The non-Trinder peroxidase substrate DA67 was used

In conclusion, intact blood cells do not contribute lipid to the measurement of plasma TG, but the intact cells give rise to high background. Cell lysis lowers the background, but lysed cells contribute lipid to the measurement of plasma TG. Thus, performing the conversion of plasma TG to colored product in the presence of intact blood cells, and then performing cell lysis prior to determination of amount of product formed, in the presence of enzyme inhibitors which block the further conversion of lipid released from lysed cells into colored product, will allow direct determination of plasma lipids in the presence of blood cells, i.e. in whole blood.

Example 2.

Evaluation of the effect of intact blood cells on the measurement of plasma total CH.

(A) 5 μL of human EDTA-plasma or 5μL of human EDTA plasma plus 1.5 μL of blood cells were mixed with 350 μL of Wako HDL CH assay reagent 2 containing cholesterol esterase (4 kU/L), cholesterol oxidase (20 kU/L), and F-DAOS (0.8 mmol/L) in Good's buffer, pH 7.0 (30 mmol/L), supplemented with 100 μL 4- aminoantipyrine (3 mmol/L), 50 μL sodium chloride (1.5 mol/L), and 5 μL peroxidase (160 kU/L). Reactions were performed at 37°C and color development was measured at 650 nm as a function of time. The background absorbance increased dramatically when blood cells were present in the sample (Fig.4, upper). However, the blood cells did not affect significantly the temporal development of color indicating that CH present in the cell membrane was not participating in the measurement (Fig.4, middle).

(B) In a direct experiment performed with 5 replicates, 5 μL of blood plasma with or without 5 μL of blood cells was incubated for 5 min at 37°C with 100 μL Roche HDL-Cholesterol reagent 2 supplemented with 50 μL H-DAOS (25 mmol/L) and 350 μL Good's buffer, pH 7.4 (50 mmol/L). Samples were then centrifuged to sediment blood cells and the supernatants were monitored at 600 nm for formation of colored product. The average reading for the five plasma samples was 430 ± 23 mAU, and for plasma + cells it was 426 ± 21 mAU. Blood cell associated CH clearly did not contribute to the measurement of plasma CH.

(C) In another experiment, 5 μL of plasma with or without 5 μL of blood cells or 5 μL of blood cells in buffer were mixed with 500 μL of ABX Pentra Cholesterol CP assay reagent, which causes cell lysis. Formation of colored product at 37°C was monitored at 600 nm. Cell lysis strongly reduced background and gave smoother progress curves compared to when cells were kept intact, but CH released from the lysed cells contributed to formation of colored product (Fig.4, lower).

(D) To investigate different peroxidise inhibitor's ability to quench enzymatic formation of colored product from cholesterol, the inhibitor was added together with a hemolytic concentration of sodium deoxycholate (0.04% final concentration) to a mixture of sample and cholesterol esterase (1777 U/L), cholesterol oxidase (987 U/L), peroxidase (505 U/L), 4AAP (0.3 mmol/L), and HDAOS (2.5 mmol/L) in Good's buffer pH 7.4. The absorbance was measured at 600 ran after 2 min incubation with the inhibitor. The results are shown in Table 3.

Table 3. Inhibition of enzymatic conversion of cell-associated cholesterol from lysed cells.

In conclusion, intact blood cells do not contribute lipid to the measurement of plasma CH, but the intact cells give rise to high background. Cell lysis lowers the background, but lysed cells contribute lipid to the measurement of plasma CH. Thus, performing the conversion of plasma CH to colored product in the presence of intact blood cells, and then performing cell lysis prior to determination of amount of product formed, in the presence of enzyme inhibitors which block the further conversion of lipid released from lysed cells into colored product, will allow direct determination of plasma lipids in the presence of blood cells, i.e. in whole blood.

(E) In yet another experiment, 5 replicates of 5 μL of plasma with or without 5 μL of blood cells or buffer was mixed with 500 μL of the assay reagent described under 2B above. Enzyme reactions were run for 10 min at 37⁰C and were then terminated and cells lysed by adding 0.1% LDAO (N,N-dimethyldodecylamine N-oxide) and 88 μmol/L methimazole. LDAO causing cell lysis and methimazole inhibiting the peroxidase. Product formation was measured at 630 nm. The average reading for the five plasma samples was 308 ± 14 mAU and for plasma + cells it was 312 ± 6 mAU.

Example 3.

Evaluation of the effect of intact blood cells on the measurement of plasma HDL

CH.

5 μL of human EDTA plasma was mixed with blood cells in different portions to mimic different degrees of cell removal (from 0 to 100% cell removal) and added to 810 μL of Wako HDL-C L-type Reagent 1, containing anti-human apoB antibody, peroxidase (2700 U/L), ascorbate oxidase (2700 U/L), and 4-aminoantipyrin (0.9 mmol/L) in Good's buffer, pH 7.0, and adjusted with sodium chloride to isotonicity to avoid cell lysis. After 4 min incubation at 37⁰C was added 270 μL of Wako HDL- C L-type Reagent 2 containing the cholesterol esterase (4000 U/L), cholesterol oxidase (20,000 U/L), F-DAOS (0.8 mmol/L) in Good's buffer. The presence of increased amounts of blood cells gave rise to an increasing background (Fig 5, upper), but the cells did not affect the time dependent increase in color formation, indicating that cell derived cholesterol did not participate in the measurement (Fig.5, lower).

Experiments 1-3 surprisingly demonstrated that CH and TG present in the blood cell membrane do not contribute to the measurement of CH and TG present in blood plasma when the reaction conditions are such that the intact status of the blood cells are maintained during the experiment. However, intact blood cells do interfere nonspecifically with the measurements, by scattering the light used for the measurement of formed colored product.

Because nonspecific scattering of light is dependent on the size of the light scattering bodies, it may be reduced by decreasing their size, e.g. fragmenting the cells through cell lysis. Fig 2 shows the absorption spectrum between 300 nm and 700 nm of blood plasma in the presence of intact blood cells or lysed blood cells. Cell lysis clearly reduces background substantially and above 600 nm eliminates it almost completely.

However, as shown in Figs 1 and 4, cell lysis renders cell associated CH and TG available for the enzyme systems used to measure plasma CH and TG, leading to an overestimation of these analytes. This problem is solved by including inhibitors of the enzymes converting the lipids to colored product. Fig.3 shows the dose dependent decrease in the formation of colored product by hydrogen peroxidase by the common peroxidase inhibitor, Methimazole (2-mercapto-l-methylimidazole).

Example 4.

Determination of plasma total TG in the presence of blood cells in a point-of-care instrument.

Using the Afmion type point-of-care instrument (Axis-Shield PoC) total TG is determined in blood plasma in the presence of intact blood cells. The TG in the sample is measured using two reagents, Rl and R2 (Table 4).

Using the Sample Device, lOμL of fingerprick blood is aspirated from a patient. The Sample Device is inserted into the Cartridge and placed in the Afinion instrument. Reaction vessel (RV)-I of the Cartridge is filled with 300 μL PBS, RV-2 with 100 μL enzyme/substrate solution (Rl), and RV-3 with 300 μL lysis/inhibition solution (R2) (Table 4). The sample is automatically transferred into RV-I of the Cartridge, which contains 300μL of PBS. 100 μL of diluted sample is transferred from RV-I to RV-2 containing lOOμL Rl . The reaction converting plasma TG to colored product is allowed to proceed for 3 min. Then 100 μL R2 is transferred from RV-3 to RV-2, and the amount of colored product is determined by measuring the absorbance at 630 nm. The hematocrit is determined by transferring 50 μL diluted sample from RV-I to RV-3 and measuring absorbance at 530 nm. The absorbance values obtained for hematocrit and amount of product formed are converted to % and mmol/L, respectively by intrapolating from standard curves constructed using calibrators of known hematocrit and TG concentration.

Table 4

Rl R2

Good's buffer, pH 6.7 50 mmol/L N-bromosuccinimide 30 μmol/L

Magnesium salt 30 mmol/L Triton XlOO 1 %

FDAOS 0.6 mmol/L Methimazole 300 mmol/L

ATP 6 mmol/L

Potassium ferrocyanide 20 μmol/L

4-aminoantipyrine 1 mmol/L

Lipoprotein lipase 108 kU/L

Glycerokinase 1010 U/L

Glycerol phosphate oxidase 8300 U/L

Peroxidase 10000 U/L

Sodium chloride 50 mmol/L

Reagents from Sigma-Aldrich. Lipoprotein lipase from Fluka (Germany) and HDAOS from Probior GmbH (Germany).

Example 5. Determination of plasma total TG in the presence of blood cells in a point-of-care instrument

Determination of total TG in plasma in the presence of intact blood cells is performed with the instrument and reagent 1 of Example 4, but filtration is used instead of cell lysis to reduce the concentration of intact blood cells, and reflectance is measured instead of absorption to quantify formed colored product.

Using the Sample Device, 10 μL of fingerprick blood is aspirated from a patient. The Sample Device is inserted into the Cartridge and placed in the Afinion instrument. The sample is automatically transferred into RV-I of the Cartridge, which contains 200 μL Rl (Table 4). After 3 min 20 μL sample is transferred to RV- 2 containing 100 μL PBS for hematocrit measurement at 850 nm. Another 20 μL sample is transferred from RV-I to RV-3, which contains a filter pad facilitating lateral flow (Hemasep L®) (Pall Corp. USA). The blood cells and the reaction solution containing colored product rapidly separates on the filter with the cells trapped close to the point of application while the solution is wicking laterally away from the cells. The amount of formed colored product is then measured by reflectance at 600 nm from the cell free portion of the filter.

Example 6.

Determination of plasma total CH in the presence of blood cells in a point-of-care instrument.

Using the Afinion type point-of-care instrument (Axis-Shield PoC) total CH is determined in blood plasma in the presence of intact blood cells. The CH in the sample is measured using two reagents, Rl and R2, essentially as described in Example 4, except that Rl is formulated differently (Table 5).

Table 5

Rl R2

Good's buffer, pH 6.7 50 mmol/L N-bromosuccinimide 30 μmol/L

FDAOS 0.6 mmol/L Methimazole 300 mmol/L

Cholesterol oxidase 3000 U/L Triton XlOO 1 %

Cholesterol esterase 3000 U/L

4-aminoantipyπne 1 mmol/L

Peroxidase 6000 U/L

Sodium chloride 90 mmol/L

Example 7

Determination of plasma HDL associated CH in the presence of blood cells in a point-of-care instrument.

Using the Afinion type point-of-care instrument (Axis-Shield PoC) HDL associated CH is determined in blood plasma in the presence of intact blood cells. The HDL associated CH in the sample is measured using three reagents, Rl, R2, and R3 (Table 6). Using the Sample Device, lOμL of fingerprick blood is aspirated from a patient. The Sample Device is inserted into the Cartridge and placed in the Afinion instrument. Reaction vessel (RV)-I of the Cartridge is filled with 300 μL PBS, RV-2 with 100 μL Rl, RV-3 with 150 μL R2, and RV-4 with 250 μL R3. The sample is automatically transferred into RV-I of the Cartridge and diluted in PBS. 50μL of diluted sample is transferred from RV-I to RV-2 and mixed with Rl . After 3 min 50 μL of R2 is transferred from RV-3 to RV-2 and the reaction continued for 3 min. Finally, 50 μL of R3 is transferred from RV-4 to RV-2, whereupon cell lysis and enzyme inhibition occurs. The amount of colored product is determined by measuring the absorbance at 630 nm. The hematocrit is determined by transferring 50 μL of diluted sample from RV-I to RV-4 and measuring absorbance at 530 nm. The absorbance values obtained for hematocrit and amount of product formed are converted to % and mmol/L, respectively by intrapolating from standard curves constructed using calibrators of known hematocrit and CH concentration. Table 6.

Rl R2

Good's buffer, pH 7.0 30 mmol/L Good's buffer, pH 6.7 30 mmol/L Ascorbate oxidase 4000 U/L Cholesterol oxidase 2O kLVL 4-aminoantipyrine 1.3 mmol/L Cholesterol esterase 4000 U/L Peroxidase 3600 U/L F-DAOS 0.8 mmol/L Anti-human β-lipoprotein Sodium chloride 125 mmol/L Antibody 10 g/L Sodium chloride 125 mmol/L R3

N-bromosuccinimide 50 μmol/L Methimazole 500 mmol/L Triton XlOO 2 %

Example 8

Determination of plasma LDL associated CH in the presence of blood cells in a point-of-care instrument.

Using the Afinion type point-of-care instrument (Axis-Shield PoC) LDL associated CH is determined in blood plasma in the presence of intact blood cells. The LDL associated CH in the sample is measured using three reagents, Rl, R2, and R3 (Table 7).

Using the Sample Device, lOμL of fϊngerprick blood is aspirated from a patient. The Sample Device is inserted into the Cartridge and placed in the Afinion instrument. Reaction vessel (RV)-I of the Cartridge is filled with 300 μL PBS, RV-2 with 100 μL Rl, RV-3 with 150 μL R2, and RV-4 with 250 μL R3. The sample is automatically transferred into RV-I of the Cartridge and diluted in PBS. 50μL of diluted sample is transferred from RV-I to RV-2 and mixed with Rl. After 3 min 50 μL of R2 is transferred from RV-3 to RV-2 and the reaction continued for 3 min. Finally, 50 μL of R3 is transferred from RV-4 to RV-2, whereupon cell lysis and enzyme inhibition occurs. The amount of colored product is determined by measuring the absorbance at 630 run. The hematocrit is determined by transferring 50 μL of diluted sample from RV-I to RV-4 and measuring absorbance at 530 nm. The absorbance values obtained for hematocrit and amount of product formed are converted to % and mmol/L, respectively by intrapolating from standard curves constructed using calibrators of known hematocrit and CH concentration.

Table 7.

Rl R2

Good's buffer, pH 7.0 50 mmol/L Good's buffer, pH 7.0 50 mmol/L

F-DAOS 0.4 mmol/L Cholesterol oxidase 4000 U/L

2,6-di-O-methyl-α-cyclodextπn 0.06% Cholesterol esterase 4000 U/L

Sodium chloride 80 mmol/L Peroxidase 6000 U/L

4-aminoantipyπne 3 mmol/L

DDAPS 0.05 %

Sodium chloπde 80 mmol/L

R3

N-bromosuccinimide 60 μmol/L

Methimazole 500 mmol/L

Triton XlOO 2 % Example 9.

Determination of plasma VLDL associated CH in the presence of blood cells in a point-of-care instrument.

The experiment is carried out in an Aflnion point-of-care instrument, using reagents essentially as described in US6986998, but reformulated so as to maintain blood cells in an intact state (Table 8).

Using the Sample Device, lOμL of fingerprick blood is aspirated from a patient. The sample is automatically transferred into reaction vessel 1 (RV-I) of the Cartridge, which contains 190 μL of Rl . First the hematocrit is determined by photometric measurement at 850 nm. Five minutes later 60μL of R2 is transferred from RV-2 to RV-I and the reaction continued for 5 min. Finally, 50 μL of R3 is transferred from RV-3 to RV-I, whereupon cell lysis and enzyme inhibition occurs. The amount of colored product is determined by measuring the absorbance at 630 nm. The absorbance values obtained for hematocrit and amount of product formed are converted to % and mmol/L, respectively by intrapolating from standard curves constructed using calibrators of known hematocrit and VLDL concentration.

Table 8.

Rl R2

Good's buffer, pH 7.0 50 mmol/L Good's buffer, pH 6.7 50 mmol/L F-DAOS 0.4 mmol/L Peroxidase 15000 U/L Ascorbate oxidase 3000 U/L 4-aminoantipyrine 10 mmol/L Calix(8)arene sulfate 50 mmol/L Cholesterol oxidase 2000 U/L Arginine 200 mmol/L Cholesterol esterase 3000 U/L Bovine serum albumin 15 g/L (Chromobacterium viscosum)

R3

N-bromosuccinimide 90 μmol/L Methimazole 500 mmol/L Triton XlOO 2.5 % Example 10 - A prototype assay for triglyceride measurements utilizing whole blood.

A manual prototype assay for the determination of plasma triglyceride concentration utilizing whole blood was constructed for the point-of-care platform Afϊnion. Blood samples were drawn from 10 healthy volunteers, two samples from each person. One sample was centrifuged to separate plasma and cells. The triglyceride concentration of the samples was analyzed using the prototype assay as well as three reference methods: two point-of-care methods, Colestech Lipid Profile (main reference method) and Callegari Triglycerides, and one automatic laboratory method, ABX Pentra Triglycerides CP. The ABX method uses plasma, Cholestech performs in-device filtration of whole blood and thus does the measurement on plasma, and Callegari uses lyzed whole blood.

In the prototype assay 6 μL of whole blood (Prototype wb) or 3μL of plasma

(Prototype pi) was mixed with 306 μL of assay reagent and the mixture incubated at 37C for 10 min. Then 33 μL of a quench solution was added and the formed color immediately measured at 630 nm. As this is a prototype assay not yet fully implemented on the Afinion, the pipette steps are performed manually, and thus affording the prototype test a considerably higher CV compared to the commercial automatic reference methods, Cholestech and ABX (Callegari is an commercial manual method).

The commercial methods were performed according to the protocols supplied by the manufacturers.

The hematocrit was determined by spectrometry (at 540 nm) and using a calibration curve constructed from packed cells. Whole blood values were corrected accordingly. TG assay reagent (306 iiL):

Lipase (chromobacterium viscosum) 108 kU/L

ATP 3.1 mmol/L

Glycerol 3 -phosphate oxidase (Pediococcus sp) 4000 U/L

Glycerokinase (E CoIi) 460 U/L

Horse-radish peroxidise 500 U/L

MgCl₂ 32 mmol/L

DA-67 0.24 mmol/L

PIPES buffer, pH 7.4 50 mmol/L

Sample (6 μL whole blood or 3 uL plasma)

Quench reagent (33 μL): 4-amino antipyrine 0.9 mol/L Triton X-100 1 %

Results are shown in Table 9.

Table 9. Triglycerides

* whole blood with intact cells; ** whole blood with lyzed cells. Example 1 1. A manual prototype assay for total plasma cholesterol in whole blood

A manual prototype assay for the determination of plasma total cholesterol concentration using whole blood was constructed for the point-of-care platform

Afinion. Blood samples were drawn from 10 healthy volunteers, two samples from each person. One sample was centrifuged to separate plasma and cells. The plasma was used for those tests requiring plasma. Whole blood samples were used for tests utilizing whole blood. The cholesterol concentration of the samples was then analyzed using the prototype assay as well as three reference methods: two point-of- care methods, Colestech Lipid Profile (main reference method) and Callegari Cholesterol, and one automatic laboratory method, ABX Pentra Cholesterol CP. The ABX method uses plasma, Cholestech performes in-device filtration of whole blood and thus measures on plasma, and Callegari ues lyzed whole blood.

The experiment was performed as outlined in Experiment 10, but utilizing a different assay reagent.

CH assay reagent: Cholesterol oxidase (rec. microbial in E CoIi) 1000 U/L

Cholesterol esterase {hog pancreas) 600 U/L

Horseradish peroxidase 500 U/L

Lipase (chromobacterium viscosum) 106 kU/L

1-Octanol 58 mmol/L MgCl₂ 32 mmol/L

PIPES buffer, pH 7.4 50 mmol/L

DA-67 0.24 mmol/L

Results are shown in Table 10. Table 10. Total cholesterol

* whole blood with intact cells; ** whole blood with lyzed cells.

Example 12. A manual prototype assay for plasma HDL in whole blood

A manual prototype assay for the determination of plasma HDL concentration using whole blood was constructed for the point-of-care platform Afinion. Blood samples were drawn from 10 healthy volunteers. The samples were centrifuged to separate plasma and cells. The plasma was used for those tests requiring plasma. Whole blood samples were constructed by mixing whole blood and packed cells (from the same donor) 1 :1, to obtain identical hematocrit, 50%, for all samples. The HDL concentration was then analyzed in each sample using the prototype assay as well as three reference methods: two point-of-care instruments, Colestech (main reference method) using capillary blood and Callegari using plasma, and one automatic laboratory instrument, Cobas Mira with WAKO HDL-C M-reagent utilizing plasma. Cholestech performs in-device filtration of the capillary blood sample and thus performs the measurement on plasma. HDL assay reagent 1 TRl) (270 uU:

WAKO HDL reagent 1

NaCl 90 mmol/L

Sample (3μL plasma w/wo 3μL packed cells)

HDL assay reagent 2 (R2) (9OuL): WAKO HDL reagent 2

Quench reagent (54μL):

Methimazole 925 mmol/L

Triton XlOO 0.74%

In the prototype assays the sample was preincubated with Rl for 3 minutes at 37°C followed by addition of R2 and incubation for another 3 minutes at 37°C. Finally Quench reagent was added and the transmission immediately measured at 630nm.

Results are shown in Table 11.

* whole blood with intact cells Example 13.

The LDL levels were computed from the measured values for total cholesterol, triglycerides, and HDL cholesterol by using the Friedewald formula (values in mmol/L):

LDL = Total cholesterol - (HDL + Triglycerides/2.2)

The obtained results are given in Table 12.

Table 12. Computed LDL cholesterol.

The overall bias for the ABX/Wako, Callegari, Prototype wb and Prototype pi methods relative to the Cholestech methods was 5%, 12%, 7% and 6% for, respectively (Table 13).

There was only insignificant difference between Prototype wb and Prototype pi values. In contrast, the Callegari methods, which use lysed whole blood for the total cholesterol and triglyceride measurements but plasma for the HDL-cholesterol measurement, as expected exhibited significantly higher total cholesterol and triglyceride values but not HDL values than all other methods (mean value >3 SD above average of the other mean values). This corroborates our findings reported in Examples 1 -3 that lipids from lysed cells but not lipids from intact cells contribute to the measurement of plasma lipids.

Table 13. Summary of averages of Tables 9-12 and method bias compared to Cholestech.

* Average of TG, CH, and HDL results, relative to results obtained by Cholestech methods . # Computed according to the Friedewald formula. °Wako

Example 14.

The imprecision was investigated for the Prototype wb and pi methods, which currently are only manual, the likewise manual Callegari methods, and the automatic ABX, Wako, and Cholestech methods. Total cholesterol, triglycerides, and HDL were measured by all 5 methods in 2 different samples with 9-10 parallels. As would be expected the average CV was poor for the manual methods (Prototype wb, Prototype pi, Callegari), acceptable for Cholestech and good for ABX/Wako. The poor CV of the manual methods is also reflected in the r- values of these methods (see Tables 9-1 1), which, relative to the Cholestech methods, on average was 0.91 for Prototype wb, 0.92 for Prototype pi, and 0.87 for Callegari, compared to 0.97 for the automatic ABX/Wako methods. Table 14.

Claims

1 ) An enzymatic method for determining the concentration of at least one specific class of lipid or specific class of lipoprotein in the plasma portion of a blood sample comprising plasma and intact blood cells, said method comprising;

1) contacting said sample with a reagent mixture comprising a lipid converting enzyme, wherein said reaction mixture causes the selective reaction of said specific class of lipid or said specific class of lipoprotein whereby to generate a reaction product or consume a reaction substrate;

iii) detecting said indirect product;

iv) relating an amount of said indirect product detected or a rate of formation of said indirect product to the concentration of said specific class of lipid or specific class of lipoprotein in said blood sample;

2) A method as claimed in claim 1 wherein steps i) and ii) are conducted simultaneously by means of at least one reaction mixture.

3) A method as claimed in claim 1 or claim 2 wherein said indirect product is detectable photometrically.

4) A method as claimed in any of claims 1 to 3 wherein, after step i), said intact blood cells in said sample are reduced by at least 50%. 5) A method as claimed in claim 4, wherein after step i), an inhibitor is added to inhibit further reaction of said specific class of lipoprotein and said intact blood cells in said sample are reduced by means of cell lysis.

6) A method as claimed in claim 4 wherein said intact blood cells in said sample are reduced by means of filtration.

7) A method as claimed in any of claims 1 to 6, wherein said specific class of lipoprotein is selected from the group consisting of very low density lipoprotein (VLDL), intermediate density lipoprotein (IDL), low density lipoprotein (LDL), and high density lipoprotein (HDL).

8) A method as claimed in any of claims 1 to 6 wherein said specific class of lipid is triglycerides (TG) or cholesterol (CH).

9) A method as claimed in any of claims 1 to 8 wherein in step i), said reagent mixture comprises a lipid converting enzyme with specificity for said specific class of lipid or said specific class of lipoprotein.

10) A method as claimed in any of claims 1 to 9 wherein in step i), said reagent mixture comprises reagents selectively solubilising said specific class of lipid or the lipids associated with said specific class of lipoprotein.

11) A method as claimed in any of claims 1 to 9 wherein in step i), said reagent mixture comprises reagents inhibiting the reaction of lipoproteins and/or lipids other than said specific class of lipid or said specific class of lipoprotein.

12) A method as claimed in any of claims 1 to 9 wherein in step i), said reagent mixture comprises reagents causing the degradation of lipoproteins and/or lipids other than said specific class of lipid or said specific class of lipoprotein. 13) A kit for use in determining the concentration of at least one specific class of lipid or specific class of lipoprotein in the plasma portion of a blood sample comprising plasma and intact blood cells, said kit comprising;

a) a first reagent mixture formulated to cause the selective reaction of said specific class of lipid or said specific class of lipoprotein whereby to generate a reaction product or consume a reaction substrate;

b) a second reagent mixture formulated to cause conversion of said reaction product or reaction substrate into a detectable indirect product;

c) optionally a reagent for causing cell lysis;

14) A kit as claimed in claim 13 wherein said first reagent mixture and said second reagent mixture are formulated together as a single reagent mixture.

15) A kit as claimed in claim 13 or claim 14 additionally comprising an inhibitor to inhibit further reaction of said specific class of lipid or said specific class of lipoprotein, or further generation of said secondary analyte, and further comprising a lysing agent to cause cell lysis.

16) A kit as claimed in any of claims 13 to 15 further comprising reference samples of at least one lipoprotein selected from the group consisting of very low density lipoprotein (VLDL), intermediate density lipoprotein (IDL), low density lipoprotein (LDL), and high density lipoprotein (HDL).

17) A kit as claimed in any of claims 13 to 16 for use in an automated assay device, preferably a laboratory or 'point-of-care' assay device.