Membrane assay method

Abstract

Provided herein is technology relating to methods for assaying a cell-containing body sample and particularly, but not exclusively, to treating a sample under conditions to cause cell lysis, preferably by means of a detergent; and subjecting the lysed sample to conditions causing the cleavage of nucleic acid molecules. The technology additionally relates to using nucleic acid cleavage conditions to enhance a membrane assay, a device for carrying out such an assay, and a kit for use in the assay.

Claims

1. An assay device comprising: a chamber for accepting a cell-containing body fluid; and at least one membrane of pore size no greater than 10 μm, wherein said device contains a nuclease and said membrane comprises an immobilized specific binding agent that is specific for a non-nucleic acid analyte of a whole blood sample; and wherein said device further comprises a second binding agent that is specific for said non-nucleic acid analyte and wherein said second binding agent is conjugated to a signal generating moiety.

2. The device as claimed in claim 1, wherein said nuclease is in dried form.

3. The device as claimed in claim 1, wherein said device contains a detergent.

4. The device as claimed in claim 2, wherein said device contains a detergent.

Description

EXAMPLES

Example 1

(1) Whole blood from a healthy donor was centrifuged on the density gradient medium Polymorphprep™ (Axis-Shield PoC AS, Oslo, Norway) as described by the producer. Total white blood cells and red cells were collected in two separate fractions. White cells were counted. White cells were added to whole blood to increase the number of white cells from 7×10.sup.9/L to 30×10.sup.9/L. 12.5 μl of the enriched blood was added to 1 ml of dilution liquid (buffered Na-deoxycholate) containing 1 mM CaCl.sub.2. This solution was divided in two aliquots of 500 μl, one of them were subsequently added 2.5 μl (0.25 U) Micrococcal nuclease. Aliquots of 50 μl of the control solution and the nuclease containing solution were applied to a membrane flow through device (NycoCard Test Device) containing a 0.45 μm pore size nitrocellulose membrane. Aliquots were applied 0 sec, 60 sec and 120 sec after addition of the nuclease. When the blood cell lysate had soaked through the membrane in a flow through manner, 50 μl of a gold conjugated antibody was added followed by 50 μl of washing solution. It should be noted that the membrane bound antibody and the conjugated antibody were directed against different proteins meaning that no sandwich could be formed and that the experiment just shows non specific binding of gold conjugated antibody to the membrane. The red colour of the retained antibody on the membrane was measured using a reflectometer (NycoCard Reader). This instrument expresses the density of the colour as K/S which is proportional to the amount of gold labelled antibody trapped onto the membrane. A white membrane will give a K/S of approximately 0.05 while a very dense, deep red colour will give a K/S of about 5.

(2) K/S after x seconds

(3) TABLE-US-00001 K/S after x seconds Seconds 0 60 120 Control 0.71 0.89 1.24 Nuclease 0.154 0.123 0.136

(4) The table shows for the control a high non-specific background increasing over the incubation period while the background is down to a normal low level (as low as for a plasma-based analysis) even with the first aliquot with less than 30 seconds incubation.

Example 2

(5) In this experiment whole blood was enriched in white cells without the use of density gradient medium centrifugation as was done in example 1. Whole blood from a healthy donor was centrifuged at 2500×g for 15 minutes. The plasma was pipetted off and subsequently the buffycoat and part of the red cell fraction was pipetted off. The buffycoat fraction contained red cells enriched in white cells. Finally pure red cells were taken from the bottom of the tube. The fraction enriched in white cells and the red cell fraction were added to plasma to give the same haematocrit as whole blood and white cells were counted in all three fractions. Haemoglobin was measured spectrophotometrically at 200 times dilution in distilled water at 575 nm using a Shimadzu spectrophotometer.

(6) TABLE-US-00002 O.D..sub.575nm white cells/L Whole blood 0.672 7.5 × 10.sup.9 Enriched in w.c. 0.657  34 × 10.sup.9 Red cells 0.690 0

(7) These three fractions will hereafter be termed C (whole blood), W (blood enriched in white cells) and R (red cell fraction).

(8) 5 μl of C, W and R respectively were gently mixed with 400 μl dilution liquid and 50 μl of each solution applied to a NycoCard test device followed by 50 μl gold conjugated antibody and 50 μl washing solution. As described in example 1, the conjugate antibody and the membrane bound antibody were directed to different proteins. In other words this experiment shows non-specific background only. The lysed dilutions of C, W and R were then shaken vigorously for 10 seconds and 50 μl of each processed in the membrane test device as described above. The conjugate background was measured for both series using the NycoCard Reader.

(9) TABLE-US-00003 K/S K/S shaken sample C 0.136 0.097 W 0.422 0.112 R 0.088 0.094

(10) The table shows that the non-specific background was highly elevated in the case of W and slightly elevated with C compared to the background obtained with a pure red cell fraction (R). Shaking reduced the background of C and W to the same level as for R.

(11) 5 μl of C, W and R were in the next experiment gently mixed with 400 μl of dilution liquid containing 1 mM CaCl.sub.2 and 0.4 U Micrococcus nuclease and processed in the test devices as described above.

(12) TABLE-US-00004 K/S C 0.12 W 0.108 R 0.094

(13) It appears from the table that nuclease digestion of the samples reduced the backgrounds down to “shaking level”. The same advantageous low background was obtained with breaking down DNA either with shearing forces or with enzymatic digestion of DNA. Shaking of the nuclease digested samples did not lower the background further (not shown).

(14) Examples 1 and 2 demonstrated how nuclease digestion of detergent solubilized whole blood reduced the non-specific background in the membrane flow-through test device down to a low level which is the same as the background we obtain with plasma based analysis. The next two examples shows the effect of nuclease treatment when diluted and lysed blood samples were analysed for CRP (C reactive protein) in our automated analyser which is based on the same membrane flow-through system as the manually operated test devices described in examples 1 and 2.

(15) It should be kept in mind that this analyser mixes the small volume of whole blood (1.5 μl) taken from a finger stick with the dilution liquid (200 μl) in a very gentle way. In other words, shearing forces to degrade the DNA released from the lysed white cell nuclei are absent or too small to work efficiently.

Example 3

(16) The blood samples C, W and R described in example 2 were analysed using the automated analyser. The blood was either lysed in the normal dilution liquid or in dilution liquid containing 5 mM CaCl.sub.2 and 0.2 U Micrococcal nuclease.

(17) TABLE-US-00005 Control dilution liquid Dilution liquid + nuclease CRP, mg/L HCT, % CRP, mg/L HCT, % Plasma 0.93 1.37 C 2.49 42.7 1.59 43.6 W 8.8 45.0 2.70 45.0 R 1.72 43.6 1.93 46.4

(18) It appears from the table that the CRP-concentration in the blood sample is about 1 mg/L (plasma based determinations). This is a typical value for a healthy person. Using blood, one must expect a slight elevation of the background due to the red colour of the haemoglobin. This means that the CRP-values for sample R (red cells and no white cells) are the base line or the target value for CRP. Looking at the results obtained with the control dilution liquid, it appears that the CRP result was increased by 45% for the normal whole blood and 512% for the preparation enriched with white cells. Using dilution liquid with nuclease, no CRP increase was observed with whole blood and an increase of 40% was found with the white cell enriched preparation. The conclusion is that CRP was significantly over-estimated when normal dilution liquid was used for dilution of sample W (high white cell counts). It seems that that this sample was slightly over-estimated also when using nuclease. This indicates that higher concentration of nuclease should have been used in this experiment. HCT (haematocrit) determinations show that these were similar for all 3 samples.

Example 4

(19) Three blood samples with low, medium and high plasma concentration of CRP were analysed on the instrument. All three samples had high white cell counts. A normal cell count is about 7×10.sup.9

(20) TABLE-US-00006 Sample CRP, mg/L white cell count 1 1.3 29.6 × 10.sup.9 2 17.2 23.5 × 10.sup.9 3 136.8 26.6 × 10.sup.9

(21) The samples were analysed using normal dilution liquid or dilution liquid containing 5 mM CaCl.sub.2 and 0.5 U Micrococcal nuclease. Based on the conclusions in example 3 the concentration of nuclease was raised from 0.2 (example 3) to 0.5 U in this experiment. The coefficient of variation (CV) was also determined based on 4-6 parallels. CV was not calculated for sample 1 due to the very low value of CRP.

(22) TABLE-US-00007 Control dil. liquid Dilution liq. + nuclease sample CRP CV CRP CV 1 Plasma 1.3 1.5 Blood 23.7 2.6 2 Plasma 17.2 19.5 Blood 31.8 10.2 16.3 1.7 3 Plasma 136.8 137.3 Blood 160 10.1 123.7 3.0

(23) The table shows that the CRP-concentration for all 3 blood samples is over-estimated when the normal dilution liquid was used. The lower the CRP-concentration, the worse was the discrepancy between the determination using blood vs plasma. The determinations using plasma is regarded as the correct value.

(24) Especially severe is the discrepancy with samples containing low CRP-concentrations. Sample 1 was over-estimated 1800%, sample 2 185% and sample 3 117%. This was to be expected if we assume the same fixed non-specific background contribution with all 3 samples.

(25) Using dilution liquid containing nuclease, the situation was quite different. Whole blood determinations were for all 3 samples similar to the plasma values.

(26) Furthermore, the CV of the determinations was significantly lower when nuclease was included in the dilution liquid.

(27) The conclusion was that nuclease mediated DNA-degradation of the lysed blood samples gave a remarkable quality improvement of the automated CRP-assay.

Example 5

(28) Whole blood from a healthy donor was processed as described in Example 2. Whole blood (C), blood enriched in white cells (W) and red cells (R) showed the same haematocrit. White cell counts were 5.9×10.sup.9, 16.9×10.sup.9 and 0×10.sup.9/L, respectively. 25 μl “blood” sample was added to 400 μl buffered detergent containing 2 mM MgCl.sub.2 with or without 1 U nuclease. The blood was mixed very gently with the lysis solution in order to reduce shearing forces to a minimum. 50 μl of this solution was added to a membrane flow-through device (membrane area 9.4 mm.sup.2) and flow time taken. The membrane was coated with anti-CRP antibodies. Subsequently 50 μl gold-conjugated anti-CRP antibody was added followed by 50 μl washing solution. The membrane colour was finally measured using a reflectometer (NycoCard Reader). In this example the red colour on the membrane will represent a true CRP-signal plus a greater or lesser amount of non-specific background signal.

(29) TABLE-US-00008 Sample Flow time (sec) Colour (K/S) C 38 0.81 C + nuclease 22 0.313 W 188 Nd** W + nuclease 23 0.310 R 25 0.27 R + nuclease 23 0.306 Plasma * 24 0.322 * 15 ul plasma was used to give the same plasma load as with the blood samples. **The conjugate did not pass the membrane due to clogging. The colour was dark red.

(30) The table shows that whole blood (C) gave a significant increase in flow time compared to the nuclease treated sample. With sample W the flow time increased dramatically (8 times) and the liquid flow came to a full stop during the application of the gold conjugate. All samples showed the same good flow of about 23 sec when nuclease was added. This was the same flow time as with plasma and with the red cell fraction without nuclease. With respect to signal, whole blood without nuclease would have resulted in a very significant over-estimation of CRP, With W, the flow came to a full stop due to DNA mediated membrane clogging.

(31) Upon nuclease treatment all three samples gave very similar CRP-signals of about 0.31, close to the signal obtained with plasma (0.322) which should be regarded as the target value. This low signal was well above the colour obtained with a membrane coated with an irrelevant antibody (0.093), which represents the background value that would have been obtained with a sample containing zero CRP. This is therefore a low but significant signal of 0.31, and is consistent with the low CRP level of a healthy blood donor.

Example 6

(32) When analysing low concentration blood analytes it is advantageous to process as much blood as possible in order to partly compensate for the low concentration and obtain a readable signal. This experiment was designed to determine the highest possible amount of blood that could be processed through a 0.45 μm nitrocellulose membrane (flow area 9.4 mm.sup.2) upon detergent mediated blood lysis.

(33) Due to the large amount of blood used in parts of this experiment, a high concentration of detergent had to be used to ensure that deficiency of detergent would not be the reason for poor flow.

(34) Blood from a healthy donor (white cell count 6.5×10.sup.9/L) was diluted with the buffered detergent with or without 1 U of nuclease and incubated at room temperature for 30 sec (incubation volume 200 μl). 100 μl of this solution containing from 1.5-25 μl whole blood was added to the flow-through device followed by 50 μl gold conjugated anti-CRP antibody and 50 μl washing solution. Total assay time (start of sample application to end of washing solution) was measured.

(35) TABLE-US-00009 Volume blood processed (μl) Total assay time (sec) 25 Nd*   25 + nuclease 189 12.5 672 12.5 + nuclease 142 6.25 291 6.25 + nuclease 128 3.1 264  3.1 + nuclease 128 1.5 150  1.5 + nuclease 120 *membrane blockage during sample application

(36) The table shows that nuclease digestion of DNA resulted in improved flow at all levels of blood load. Looking at assay time, it appears that 12.5 μl blood with nuclease digestion gave about the same assay time as 1.5 μl blood without nuclease, a factor of about 8. Looking at the results from another angle, if we demand a total assay time for a rapid test to be below 4 minutes, this system tolerates 1.5 μl blood without nuclease and 25 μl with nuclease digestion, a factor of 16.7.

(37) Broadly speaking, nuclease mediated digestion of DNA allows at least 10 times more blood to be processed in this immune-concentration test device.

Example 7—Nuclease Freeze-Drying Procedure

(38) Benzonase™ was prepared at a concentration of 40 U/ml in a buffer containing 12% Trehalose, 0.1% BSA, 1 mM MgC.sub.12, 25 mM Tris pH 7.4. Aliquots of this Benzonaze™ solution was frozen then freeze-dried at −30 degree C. in a manner known per se.

Example 8—Use of a Freeze-Dried Nuclease

(39) Frozen and freeze-dried Benzonase™ beads (as prepared in example 7) were placed into an Afinion™ Analyser CRP cartridge and a blood sample containing elevated white blood cells was processed using the Afinion™ Analyzer. The relative activity of the Benzonase™ was calculated by using the normalized CRP values and expressed as a percentage

(40) Relative nuclease activity in an Afinion™ CRP assay:

(41) A.—Benzonase™ was resuspended at 40 U/ml in the Afinion™ CRP lysis buffer for various periods of time up to 18 hours at 22 degrees C. before analysing the blood sample containing elevated white blood cells.

(42) TABLE-US-00010 Time 5 min 60 min 120 min 18 hours Relative 100% 104% 94% 61% Activity

(43) It is demonstrated that the Benzonase™ appears stable in an aqueous suspension for less than 1 day. A decrease in activity is noted after 120 min.

(44) B—Frozen and freeze-dried Benzonase™ beads, prepared as in Example 7 and placed in the Afinion™ CRP cartridge as described above, were incubated for up to 2 months at 4, 22 or 37 degree C. for various time periods before analysing a blood sample containing elevated white blood cells.

(45) TABLE-US-00011 Temp Day 0 1 month 2 month  4 C. 100 99 96 22 C. 100 100 105 37 C. 102 98 99

(46) Benzonase™ appears stable as a frozen and freeze-dried bead for months.