WO1995027208A1

WO1995027208A1 - A method of determining one or more dissolved species in a suspended or emulsified material

Info

Publication number: WO1995027208A1
Application number: PCT/SE1995/000341
Authority: WO
Inventors: Anette Persson; Håkan ROOS; Lennart WAHLSTRÖM
Original assignee: Pharmacia Biosensor Ab
Priority date: 1994-03-31
Filing date: 1995-03-30
Publication date: 1995-10-12
Also published as: EP0753151A1; SE9401102D0; JPH09511065A

Abstract

A method of determining one or more dissolved species in a fluid sample comprises transporting the sample through a flow channel which on a wall thereof has a sensing surface capable of binding the species to form an analyte depleted layer (6) extending from the surface, and determining the extent of binding of the species to the sensing surface. According to the invention, the sample contains suspended or emulsified material and is transported through the flow channel in a laminar flow such that there is formed a flow core (4) of the suspended or emulsified material and a surrounding liquid layer (5) near the flow channel wall, which layer is substantially free from suspended or emulsified material. The sample flow rate is controlled such that the thickness (δ) of the liquid layer (5) near the sensing surface is greater than the thickness (LD) of the analyte depleted layer (6), the binding of the species to the sensing surface thereby being at least substantially independent of the proportion of suspended or emulsified material in the sample.

Description

A METHOD OF DETERMINING ONE OR MORE DISSOLVED
SPECIES IN A SUSPENDED OR EMULSIFIED MATERIAL
The present invention relates to the determination of dissolved species in liquid samples which contain suspended or emulsified matter.

For analysing species dissolved in liquids which also contain non-dissolved matter, such as suspended particle matter or emulsified fluid matter, it is usually necessary to remove the insoluble matter to recover the pure liquid phase before performing the analysis of the desired species or analyte(s). This is particularly the case for assays where the species to be determined, or analyte, is to interact with ligands immobilized to a solid phase, the suspended or emulsified matter impeding the diffusion of the analyte to the solid phase surface and/or interfering directly with analyte/ligand interactions. Common ways of eliminating, for example, particulate matter in a sample are centrifugation, sedimentation or filtration.

An example thereof is the analysis of plasma or serum analytes in blood. Blood is a salt solution, plasma, with cells of various types, blood cells, suspended therein.

Plasma is obtained by separating the blood cells from anticoagulated blood, such as by centrifugation. By instead allowing the blood to clot followed by centrifugation, serum, i.e. fibrin/fibrinogen-free plasma, is obtained.

For several reasons direct analysis of whole blood samples without any separation step would be desired. Not only would the analytical procedure be simplified and less time-consuming thereby, but, perhaps more importantly, would the risk of transmitting infectious or contagious diseases, such as HIV, hepatitis, etc be reduced. Apart from the blood cells influencing the assay, however, the relative blood cell volume, called haematocrit, varies between individuals. Thus, the normal level is within the range of 35-55% with an average for men of about 45% and for women of about 40%. The value obtained in a whole blood assay must therefore be corrected for the haematocrit for

the results to be comparable with established reference or standard preparations.

The above is, of course, also true for any suspension or emulsion where the proportion of suspended or emulsified matter varies between samples.

The object of the present invention is to provide means for the analysis of species dissolved in the liquid phase of suspensions and emulsions, such as e. g. whole blood, which, on the one hand, requires no separation step, and, on the other hand, requires no compensation in the case of variable proportions of suspended or emulsified matter.

In accordance with the present inventive concept this object is achieved by providing ligand or ligands specific for the species to be determined, which ligands are immobilized to a wall of a flow channel, transporting the sample through the flow channel at a laminar flow rate such that there is formed a flow core of the suspended or emulsified matter of the sample and a layer near the flow channel wall which is substantially free from suspended or emulsified matter, and controlling the flow of the sample such that the diffusion of the species to the sensing surface is independent of the amount of suspended or emulsified matter.

The term flow channel is to be interpreted broadly, and the shape and dimensions thereof may vary within wide limits, depending inter alia on the detection principle used to determine the binding of analyte to the flow cell wall. In a simple embodiment, the flow channel is a part of a conduit system for transporting the sample. In other embodiments the flow channel is a specially designed flow cell. For the description of such flow cells, it may, for example, be referred to Ruzicka, J. , Hansen E. H., Flow Injection Analysis; Wiley; New York, 1987; pp 247-256, and Stulik, K., Pacakova, V., Electrochemical Measurements in Flowing Liquids; Ellis Horwood: Chichester, England; 1987.

The necessary dimensions of the flow channel or flow cell

in each particular application is readily determined by the skilled person.

The present inventive concept is based on a phenomenon, called the FÅahreus-Lindqvist effect (Fahraeus R. , Lindqvist T. , Am. J. Physiol. 1931; 96: 562-8), which occurs when a suspension or emulsion, such as anticoagulated blood, is transported through a thin capillary ( < about 0.3 mm) and consists in that the blood cells flow centrally in the capillary with the plasma forming a surrounding sleeve. The effect is due to the fact that when the flow rate above the blood cells differs from that below the blood cells, a lifting force on the blood cells results (cf. an aircraft wing). The lifting force is highest nearest the surface of the capillary where the shear force of the liquid is greatest.

This lifting force tends to force the blood cells towards the centre of the capillary where the flow rate is the same on both sides of the blood cells, resulting in the formation of a blood cell-free plasma layer adjacent to the capillary wall.

When anticoagulated whole blood is transported through a thin layer flow cell having a sensing surface with immobilized ligands or receptors capable of binding an analyte in the blood, the liquid layer adjacent to the sensing surface is therefore plasma which is free from blood cells. Due to the binding of analyte to the sensing surface, there is created a liquid layer extending from the sensing surface which is depleted of analyte and in which the binding of analyte to the immobilized analyte receptors is controlled by the diffusion of analyte to the surface (provided, of course, that the analyte-receptor reaction is not limiting).

If now the above-mentioned cell-free plasma layer is thicker than the analyte depleted layer, the diffusion of analyte to the sensing surface will not be disturbed by the blood cells, and the analytical situation will effectively be that of analysing a plasma sample. As will be explained below, the assay is also independent of the proportion of

blood cells, i.e. no compensation for haematocrit need to be done.

The invention may be applied to any type of heterogeneous assay where an analyte, analyte analogue or other species dissolved in an emulsion or suspension type sample is bound to a ligand or receptor immobilised to a solid phase, and the binding to the solid phase is detected directly or indirectly. As examples of other such emulsion or suspension type samples than anticoagulated blood may be mentioned bacterial suspensions, food emulsions like milk, etc.

The species whose binding to the surface is determined need, of course, not be the primary object of the analysis but may only be a means of indirectly determining a desired component in the suspension or the emulsion. For example, bacterial cells may be assayed for in an indirect analysis by determining remaining antibody against the bacterial cells in an immunological assay.

The detection principle for determining the amount of analyte or other species bound to the sensing surface is not critical to the invention. In a suitable type of detection system the solid phase comprises a sensing structure and a change in a property of the sensing structure is measured as being indicative of binding of the analyte to the immobilised ligand. Among these methods are, for example, mass detecting methods, such as piezoelectric, optical, thermo-optical and surface acoustic wave (SAW) methods, and electrochemical methods, such as potentiometric, conductometric, amperometric and capacitance methods.

Among optical methods may particularly be mentioned those that detect mass surface concentration, such as reflection-optical methods, including both internal and external reflection methods, e. g. ellipsometry and evanescent wave spectroscopy (EWS), the latter including surface plasmon resonance spectroscopy (SPRS), Brewster angle refractometry, critical angle refractometry, frustrated total reflection (FTR), evanescent wave

ellipsometry, scattered total internal reflection (STIR), optical wave guide sensors, evanescent wave based imaging, such as critical angle resolved imaging, Brewster angle resolved imaging, SPR angle resolved imaging, etc. , as well as methods based on evanescent fluorescence (TIRF) and phosphorescence.

Among the optical methods mentioned above, especially SPRS has attracted much attention recently. The phenomenon of SPR is well known. In brief, SPR is observed as a dip in intensity of light reflected at a specific angle from the interface between an optically transparent material, e.g. glass, and a thin metal film, usually silver or gold, and depends on among other factors the refractive index of the medium (e. g. a sample solution) close to the metal surface.

A change of refractive index at the metal surface, such as by the adsorption or binding of material thereto, will cause a corresponding shift in the angle at which SPR occurs. To couple the light to the interface such that SPR arises, two alternative arrangements may be used, either a metallized diffraction grating (Wood's effect), or a metallized glass prism or a prism in optical contact with a metallized glass substrate (Kretschmann effect). For further details on SPR, reference is made to our WO 90/05295. In an SPR-based immunoassay, a ligand may be bound to the metal surface, and the interaction thereof with an analyte dissolved in a fluid sample in contact with the surface is then monitored.

In the following, the invention will be described in more detail, reference being made to the accompanying drawings wherein:
Fig. 1 is a schematic sectional view of a thin layer flow cell;
Fig. 2 is a section along A-A in Fig. 1;
Fig. 3 is a schematic sectional view of whole blood flow in a flow cell of circular cross section;
Fig. 4 is a schematic sectional view of whole blood flow in a flow cell of rectangular cross section;

Fig. 5 is a schematic sectional view of the flow in the longitudinal direction of a flow cell;
Fig. 6 is a diagram showing the results of the determination of IgE in anticoagulated whole blood samples with varied blood cell content as a plot of relative response vs erythrocyte volume fraction (EVF);

Fig. 7 is a diagram showing the results of the determination of IgE concentration in anticoagulated whole blood samples as a plot of relative response vs concentration;
Fig. 8 is a diagram showing the results of the determination of myoglobin in anticoagulated whole blood samples with varied blood cell content as a plot of relative response vs erythrocyte volume fraction (EVF);
Fig. 9 is a diagram showing the results of the determination of myoglobin concentration in anticoagulated whole blood samples as a plot of relative response vs concentration;
Fig. 10 is a diagram showing the results of the determination of protein G in bacterial culture samples as a plot of relative response vs sample dilution;

and
Fig. 11 is a diagram showing the results of the determination of monoclonal antibody against creatine kinase MB in cell culture samples as a plot of relative response vs sample dilution.

An attempt at a more theoretical explanation of the present invention will now be made with particular reference to Figs. 1 to 5.

In heterogeneous assays where the binding of ligand or analyte takes place on a solid phase, such as in the well of a microtiter plate or on a sensor surface (e.g. electrochemistry, SPR, TIRF, SAW etc) the increase of bound analyte on the surface is normally controlled by the diffusion-limited flow of molecules to the surface. This flow may be described as: dR/dt = kMC (1)

where dR/dt is the flow to the surface, kM is the mass transport coefficient and C is the concentration of the analyte.

Consider now the thin layer flow cell illustrated in Fig. 1. This rectangular flow cell has an inlet 1, an outlet 2 and a sensing surface 3 extending along the bottom wall of the flow cell. The flow cell has a width b and a height h, as indicated in Fig. 2. The sensing surface 3, from Xin to Xout, supports immobilized ligands capable of reacting with an analyte in a fluid sample transported through the flow cell. Under normal operating conditions the mass transport coefficient, kM, of such a flow cell is independent of time (Sjolander S. et al., Anal. Chem. 63 (1991): kM(x) - 0.98*(D/h)2/3(f/bx)1/3 (2) where x is the distance from the inlet 1 of the flow cell, D is the diffusion constant of the analyte, h and b are the height and width, respectively, of the flow cell (Fig. 2) and f is the volumetric flow.

A necessary condition for Equation 2 is, of course, that the mass transport coefficient, kM, is much smaller than the heterogeneous association rate constant, i.e. kM kaRmax, where Rmax is the maximum uptake of sensing surface 3.

Assuming that the flow cell in Fig. 1 is the measuring cell in a continuous flow system with a very low dead volume to the measuring cell, ideal square pulses may be injected to the sensing surface 3, or expressed otherwise: C = 0 for 0 > t > tk C = C for 0 < t < tk (3) where tk is the contact time of the analyte with the sensing surface 3.

Since both C and kM thus are independent of time, the analyte uptake level is obtained by integration of Equation 1 over the contact time tk:

tk R(x) = J kMCdt = kM Ctk = 0.98*(D/h)2/3(f/bx)1/3*Ctk (4)
0
As is seen from Equation 4, the analyte uptake level R (x) is independent of the volume of the injected sample.

The only requirement to be satisfied is that the sample volume is sufficiently large for the sample injection to be conducted during the contact time tk at the flow f. From this aspect, the analysis is totally independent of the amount of solid or immiscible matter that may be suspended or emulsified in the sample. In the case of anticoagulated whole blood, for example, the analysis is thus independent of the haematocrit.

Suspended or emulsified matter in the sample may, however, influence the diffusion constant D in Equation 4 and thereby the analyte uptake R (x). is, of course, the diffusion constant normal to the laminar flow (in the z- direction in Fig. 1) that is of interest here. Consider, for example, an anticoagulated whole blood sample, whole blood being a heterogeneous suspension of plasma and blood cells. The diffusion of the analyte takes place in the blood plasma with a diffusion constant equal to that in plasma but the transport is hindered by collisions with solid "particles" in the form of blood cells. Thus, the analyte uptake will be dependent on the haematocrit.

As already mentioned above, the Fahraeus-Lindqvist effect is a phenomenon, occurring when blood is transported in a laminar flow in thin capillaries and is seen as the blood cells being concentrated to a core centrally in the capillary, where the flow rate is the same on both the top and underside of the blood cells, while the plasma forms a blood cell-free liquid sleeve around the core. This means that in assays performed in thin layer flow cells, which typically have dimensions below 100 m, no blood cells will be transported near the sensor surface. The resulting diffusion constant obtained in whole blood will in other words be equal to that in plasma within a layer adjacent to the sensor surface due to the Fahraeus-Lindqvist effect.

This is illustrated in Figs. 3 and 4 which show two types of flow cells. The flow cell of Fig. 3 has a circular cross section, and the sensing surface is provided on the internal side of the cylinder. In Fig. 4, the flow cell has a rectangular cross section, and the sensing surface is provided on one wall of the flow cell. In both Figs. 3 and 4, a central core 4 of blood cells is surrounded by a layer 5 of plasma. The layer 5 has the thickness # (Fig. 3). The sensor surface depletes the plasma layer 5 of analyte, resulting in a depleted plasma layer, sometimes called the diffusion layer, nearest the sensor surface. This depleted layer 6 has the thickness LD (Fig. 3).

If the blood cell-free plasma layer 5 (of thickness #) is thicker than the analyte depleted layer 6 (of thickness LD), it is readily seen that the assay will then be independent of the haematocrit level.

Fig. 5 is a schematic longitudinal section of a thin layer flow cell of the type shown in Fig. 4. The inlet is designated by reference numeral 7, the outlet by numeral 8, and a sensing surface which depletes the sample of analyte is designated by 9. The arrows indicate the parabolic flow profile of the laminar flow. The analyte depleted or diffusion limited layer is indicated by reference numeral 10. Fig. 5 illustrates how this analyte depleted layer 10 of thickness LD increases with the distance from the inlet 7.

The size of the analyte depleted layer may be calculated as (Eddowes M. J., Biosensors, A Practical Approach, Chapter 9, Ed. A.E.G. Cass, Oxford University Press, New York): LD = D/kM (5)
From Equation 5 it is seen that LD increases with the diffusion constant D and decreases with the mass transport coefficient kM. Since D is inversely proportional to the size of the molecule, LD increases with reduced molecular weight, i.e. the smaller the molecular weight is, the more

extensive is the analyte depleted layer. Typical sizes of analyte depleted layers in thin layer flow cells are in the range of 5 to 10 m.

It is thus clear that LD may be increased or reduced by variation of the flow cell dimensions and/or the flow rate. By such variation of LD it may thus readily be insured that LD in a whole blood assay is smaller than the thickness of the blood cell-free layer. It is to be noted in this context, however, that Equation 2 is valid only when the thickness of the analyte depleted layer is less than half the height of the thin layer flow cell. Also, if the assay is not totally mass transport limited, the analyte depleted layer becomes thinner to be zero at total kinetic limitation. It is further to be noted that the tight blood cell core deforms the parabolic flow profile which affects the convection term in the mass transport of molecules. Therefore, Equation 2 does not give a completely adequate description of the mass transport coefficient.

This influence on the flow profile results in an increased uptake of analyte, which increase, however, is moderate.

In summary, the blood cell content of an anticoagulated whole blood sample does not directly influence the uptake of analyte on a flow cell sensing surface but it obstructs the diffusion to the sensing surface. Due to the Fahraeus-Lindqvist effect, a blood cell-free plasma layer is created near the sensing surface and a blood cell dense core where the flow rate is higher.

If this blood cell-free plasma layer is thicker than the analyte depleted layer, the diffusion of the analyte to the sensor surface is not influenced. At least within normal haematocrit levels, assay methods may therefore be constructed that give a very low and insignificant dependence of the haematocrit level despite the above- mentioned deformation of the flow profile due to the dense blood cell core.

The Fahraeus-Lindqvist effect is, however, not restricted to blood cells in anticoagulated whole blood but is generally applicable to suspended or emulsified matter.

For the purposes of the present invention, the size limit of such suspended or emulsified matter depends on the dimensions of the flow cell and sensing area as well as the flow rate. The above presented theoretical considerations therefore apply to flow cell sensor based assays of dissolved analytes in any sample containing suspended or emulsified matter.

In the following Examples, which are intended to illustrate the present invention, by way of example only, the measurements were performed on a commercial SPR-based (Kretschmann effect) biosensor system, BIAcoreTM, and commercial sensing surfaces, Sensor ChipTM CM5, both supplied by Pharmacia Biosensor AB, Uppsala, Sweden. Sensor ChipTM CM5 has a sensing surface in the form of a glass support having a layer of carboxylated dextran covalently bound to the gold film surface. The flow cells consist of upwardly open channels of the cross section 50 m x 500 m and with a length of 2.4 mm, which channels are closed by the sensing surface of Sensor ChipTM CM5 docked against them.

EXAMPLE 1
Whole blood IgE assay with varied haematocrit Preparation of blood samples
Donor EDTA blood was supplemented to 1000 ng/ml with IgE (Pharmacia Diagnostics AB, Uppsala, Sweden), mixed and centrifuged at 2,500 rpm for 10 minutes. Plasma and blood cells were separated from each other and subsequently mixed to provide different mixtures having erythrocyte volume fractions (EVF) of 0,20, 40 and 60%, respectively, to thereby produce blood samples having the same concentration of IgE but a varied cell content covering a normal EVF- range in man.

Preparation of anti-IaE surface
With the Sensor Chip CM5 docked into the BIAcoreTM instrument and a continuous flow of drive buffer, HBS (10 mM Hepes buffer, 0.15 M NaCl, 3.4 mM EDTA, 0.05 % Tween), pH 7.4, maintained at 5 ul/min over the sensing surface, 100 pg/ml of antibody specific for IgE (obtained from

Pharmacia Diagnostics AB, Uppsala, Sweden) in coupling buffer, 10 mM sodium acetate, pH 5.0, were coupled to the sensing surface in accordance with the instrument manufacturer's instructions.

IaE assay
A continuous flow of drive buffer, HBS, was maintained over the sensing surface at 5 l/min. 35 l of each blood sample prepared above was injected at 5 l/min (contact time 420 s). Bound IgE was then detected by the injection of 1.0 mg/ml of anti-IgE antibody as secondary reagent.

Regenerations between blood sample injections were performed with 10 mM glycine-HC1, pH 2.25.

In a similar experiment, the drive buffer flow was changed to 20 l/min, and 35 l of each blood sample were injected at 20 l/min (contact time 105 s) and bound IgE was detected by secondary reagent as above.

The prepared whole blood samples were then centrifuged and the plasma was separated. 35 l of each plasma sample were tested as above at 5 l/min (contact time 420 s) and 20 l/min (contact time 105 s), respectively.

The results (in the form of relative responses normalised to the responses of plasma and plotted vs EVF) are presented in Fig. 6. As appears therefrom, the response at the lower flow rate of 5 l/min was dependent on the EVF and differed significantly between whole blood and plasma.

However, at the higher flow rate of 20 l/min, as expected the response was substantially independent of EVF.

EXAMPLE 2
Whole blood IaE assav with varying IgE concentrations
Samples of well mixed donor EDTA blood were supplemented with IgE to the following concentrations: 0, 100,500, 1000,2000 and 4500 ng/ml. The IgE addition was made from primary standards to obtain the same EVF in all samples. The samples were then thoroughly mixed, and half of each sample was centrifuged to obtain plasma samples.

The whole blood samples and the plasma samples were then analysed for IgE with an anti-IgE immobilised surface as described in Example 1 above at a flow of 10 l/min

followed by injection of anti-IgE as secondary reagent. The results (in the form of relative responses plotted vs IgE concentration) are presented in Fig. 7, indicating good congruence between blood and plasma values.

EXAMPLE 3 Whole blood mvoalobin assav with varied haematocrit Preparation of blood samples
Donor EDTA blood was supplemented to 400 ng/ml with myoglobin (Dako A/S, Copenhagen, Denmark), mixed and centrifuged at 2,500 rpm for 10 minutes. Plasma and blood cells were separated from each other and subsequently mixed to provide different mixtures having EVF's of 0,20, 30, 35,40, 45,50, 60 and 70%, respectively, to thereby produce blood samples having the same concentration of myoglobin but a varied cell content extensively covering the normal EVF-range in man.

Preparation of anti-mvoalobin surface
With the Sensor Chip CM5 docked into the BIAcoreTM instrument and a continuous flow of drive buffer, HBS (10 mM Hepes buffer, 0.15 M NaCl, 3.4 mM EDTA, 0.05 % Tween), pH 7.4, maintained at 5 pl/min over the sensing surface, 50 pg/ml of antibody specific for myoglobin (obtained from Pharmacia Biosensor AB, Uppsala, Sweden) in coupling buffer, 10 mM sodium acetate, pH 5.0, were coupled to the sensing surface in accordance with the instrument manufacturer's instructions.

Mvoalobin assay
A continuous flow of HBS was maintained over the sensing surface at 10 l/min. 12 l of each blood sample prepared above was injected at 10 l/min (contact time 72 s). Bound myoglobin was then detected by the injection of 1.0 mg/ml of anti-myoglobin antibody (Rabbit anti- Myoglobin, Dako A/S, Copenhagen, Denmark) as secondary reagent. Regenerations between blood sample injections were performed with 10 mM glycine-HCI, pH 2.25.

The drive buffer flow was changed to 40 l/min. 30 l of each blood sample were then injected at 40 l/min

(contact time 45 s), and bound myoglobin was detected by secondary reagent as above.

The drive buffer flow was then changed to 2 l/min. 5 l of each blood sample were injected at 2 l/min (contact time 150 s), and bound myoglobin was detected by secondary reagent as above.

The prepared whole blood samples were then centrifuged, the plasma separated and the plasma samples were tested as above at 10 1/min, 40 l/min and 2 l/min, respectively.

The results (in the form of relative responses normalised to the responses of plasma and plotted vs EVF) are presented in Fig. 8. As appears therefrom, the response at the lower flow rate of 2 and 10 l/min was dependent on the EVF. However, at the higher flow rate of 40 l/min, the response was substantially independent of EVF.

EXAMPLE 4 whole blood mvoalobin assay with varvina mvoalobin concentrations
Samples of well mixed donor EDTA blood were supplemented with myoglobin to the following concentrations: 0,20, 40,400, 800 and 1600 ng/ml. The myoglobin addition was made from primary standards to obtain the same EVF in all samples. The samples were then thoroughly mixed, and half of each sample was centrifuged to obtain plasma samples. The whole blood samples and the plasma samples were then analysed for myoglobin with an anti-myoglobin surface as described in Example 3 above at a flow of 10 l/min followed by antimyoglobin as a secondary reagent. The results (in the form of relative response plotted vs myoglobin concentration) are presented in Fig.

9, which indicates good congruence between blood and plasma values.

EXAMPLE 5
Assav for protein G in bacteria culture Preparation of samples
A protein G producing strain of E. coli was fermented, and after the fermentation the bacteria were lysed by

heating to +80 C for 6 minutes to release the protein G product. The culture contained approximately 109 cells/ml.

A sample of the culture was taken for analysis. Part of the sample was centrifuged to remove the cells. The cell- containing and the centrifuged samples were diluted 1:2, 1:4, 1: 8, 1: 16 and 1: 32, respectively, in culture medium.

Preparation of rabbit antibody sensina surface
A rabbit antibody sensing surface was prepared in analogous manner to that described for the preparation of the anti-IgE surface in Example 1.

Protein G assay
The samples were analysed for protein G by injecting them over the rabbit antibody surface in the same way as described in Example 1, the surface being regenerated between samples with 10 mM glycine-HCI, pH 2.5. The results in the form of a plot of relative response (resonance units, RU) vs sample dilution are presented in Fig. 10. As is seen therefrom, there was no difference in the mean response between crude and centrifuged samples of the same concentration showing that the presence of (lysed) cells in the concentration range 107 - 109/ml does not influence the measurements.

EXAMPLE 6
Analvsis of anti-CKMB monoclonals in cell suspension Preparation of culture sample
A hybridoma, Conan MB 2580, producing monoclonal antibodies against creatine kinase MB (CKMB) was cultivated for 4 days in complete culture medium, FDMEM + 10 % fetal calf serum (FCS), to a cell concentration of 1 x 108 cells/ml. The cell suspension was then divided into two tubes. One of the tubes was centrifuged to obtain a cell- free solution of monoclonals and one with cells. Both solutions were then diluted in culture medium (FDMEM without FCS). Half-dilutions were made from 1/1 - 1/32. As a control, the same procedure was performed on complete medium, i.e. FDMEM containing FCS was diluted as above with only FDMEM.

Preparation of RAMFc surface
Rabbit anti-mouse IgG Fc was immobilized to Sensor Chip CM5 in corresponding manner to that described in Example 1.

Monoclonal assay
In corresponding manner to that described in Example 1, the sample dilutions were tested for the presence of monoclonal antibodies (Mabs) by injection to the RAMFc immobilized surface. All dilutions were tested in the order: Cycles 1-6: Cell culture media 1/1 - 1/32 Cycles 7-12: Mab (+ cells) 1/1 - 1/32 Cycles 13-18: Mab (- cells) 1/1 - 1/32 Cycles 19-24: Cell culture media 1/1 - 1/32 Cycles 25-30: Mab (+ cells) 1/1 - 1/32 Cycles 31-36: Mab (- cells) 1/1 - 1/32.

Each cycle comprised injection of 4 l of sample (Mab/cell culture medium), followed by regeneration with 10 l of 10 mM glycine-HCI, pH 1.5.

The results in the form of a plotting of relative response (resonance units, RU) vs sample dilution are presented in Fig. 11. No undesired binding is seen for FDMEM + FCS, and there is specific binding between immobilized antibody and Mab in the sample. The Mab responses are identical for (i) Mab with cells and (ii) Mab without cells, which means that the Mab responses are not influenced by the presence of the cells.

The invention is, of course, not restricted to the embodiments described above and shown in the drawings, but many variations and modifications may be made within the scope of the general inventive concept as stated in the following claims.

Claims

CLAIMS 1. A method of determining one or more dissolved species in a fluid sample, which method comprises transporting the sample trough a flow channel which on a wall thereof has a sensing surface (9) capable of binding said species to form an analyte depleted layer (6; 10) extending from said surface, and determining the extent of binding of said species to the sensing surface, characterized in that the sample contains suspended or emulsified material and is transported through the flow channel in a laminar flow such that there is formed a flow core (4) of the suspended or emulsified material and a surrounding liquid layer (5) near the flow channel wall, which layer is substantially free from suspended or emulsified material, and that the sample flow rate is controlled such that the thickness (#) of said liquid layer (5)

near the sensing surface is greater than the thickness (LD) of said analyte depleted layer (6; 10), the binding of said species to the sensing surface thereby being at least substantially independent of the proportion of suspended or emulsified material in said sample.

2. The method of claim 1, characterized in that said flow channel is a thin layer flow cell.

3. The method of claim 1 or 2, characterized in that the binding of said species to the sensing surface is determined by a method selected from mass-detecting, optical, thermo-optical, surface acoustic wave and electrochemical methods.

4. The method of claim 1,2 or 3, characterized in that said sample is a particle suspension.

5. The method of claim 4, characterized in that said particle suspension is anticoagulated whole blood. <Desc/Clms Page number 18>

6. The method of claim 4, characterized in that said particle suspension is a cell suspension.

7. The method of claim 6, characterized in that said cell suspension is a bacterial cell suspension.

8. The method of claim 7, characterized in that said bacterial cell suspension is one in an indirect method for analysing the bacteria.

9. The method of any one of claims 1 to 8, characterized in that the sample is an emulsion, preferably a food emulsion.

10. The method of any one of claims 1 to 9, characterized in that the method is immunoassay-based.