Turbidimetric analysis on the Hitachi 705 using orosomucoid as a model

An Hitachi 705 programmable analyser was recently purchased for East Birmingham Hospital’s clinical chemistry laboratory. Although the principal use of the instrument was to process traditional spectrophotometric assays, it was considered desirable that the new instrument should be sufficiently versatile to enable turbidimetric analysis to be performed. The work-load for serum orosomucoid estimation was sufficiently high (40/week) to justify an automated approach to save both analytical time and consumable costs when compared to the, then current, use of the Mancini immunodiffusion technique. The advantages of turbidimetric assay of proteins compared with alternative techniques has recently been reviewed [1-1; unfortunately no published methods for turbidimetric analysis using the Hitachi 705 are available and it was unknown whether such techniques were possible using the Hitachi 705. This work was initiated in order to discover ifthe Hitachi 705 could successfully perform turbidimetric analysis of orosomucoid.


Introduction
An Hitachi 705 programmable analyser was recently purchased for East Birmingham Hospital's clinical chemistry laboratory. Although the principal use of the instrument was to process traditional spectrophotometric assays, it was considered desirable that the new instrument should be sufficiently versatile to enable turbidimetric analysis to be performed.
The work-load for serum orosomucoid estimation was sufficiently high (40/week) to justify an automated approach to save both analytical time and consumable costs when compared to the, then current, use of the Mancini immunodiffusion technique.
The advantages of turbidimetric assay of proteins compared with alternative techniques has recently been reviewed [1-1; unfortunately no published methods for turbidimetric analysis using the Hitachi 705 are available and it was unknown whether such techniques were possible using the Hitachi 705.
This work was initiated in order to discover ifthe Hitachi 705 could successfully perform turbidimetric analysis of orosomucoid.

Radial immunodiffusion plates
Conventional radial immunodiffusion analysis of orosomucoid was carried out using gels impregnated with anti-orosomucoid antibody obtained from Hoechst UK Ltd, Salisbury Road, Hounslow, Middlesex, UK. Phosphate buffered saline Dissolve 3"58g disodium phosphate dodecahydrate, 1.56g sodium dihydrogen phosphate dihydrate, 1.8 g sodium chloride, g sodium azide and 40 g of polyethylene glycol (PEG) 6000 in water. Adjust the pH to 6.8 and make up to litre. The solution was filtered through a 0.22 #m millipore filter before use.

Antibody solution
Anti-orosomucoid antisera, purchased from DAKO, Mercia Brocades Ltd, Brocades House, Pyrford Road, West Byfleet, Weybridge, Surrey KT14 6RA, UK, was diluted in phosphate buffered saline and left for 30 min prior to filtration through a 0.22/m millipore filter. Dilution of the antibody in the final * Author to whom correspondence should be addressed. solution was arbitrarily set at in 30. The titre of the neat antisera was 300, i.e. 300#g orosomucoid react with ml antiserum to produce equivalence.

Standardization
The primary standard used was obtained from Hoechst and had an orosomucoid concentration in the range 0.9-1.1 g/1. A pooled, filtered and hepatitis-free serum was ascribed a value of orosomucoid concentration using radial immunodiffusion analysis by reference to the primary standard. This pooled serum then served as a secondary standard and was used throughout this study.

Apparatus
A Pye Unicam 1800 spectrophotometer (Pye Unicam Ltd, York Street, Cambridge CB 2PX, UK) was used for the derivation of optimized conditions of the orosomucoid assay.
The Hitachi 705 used in this study is a programmable discretionary analyser and has recently been described in detail I-2 and 3].
All assays were carried out at 37C.

Manual procedure
The technique adopted for studying the optimization of reaction conditions was a modification of the nephelometric method of Buffone and Lewis [4], but using a turbidimetric end-point mode of assay. Briefly, this consisted of preparing an initial in 10 dilution of serum, control and standards in 9"0 g/1 saline. The diluted sample was then added to both 2.5ml phosphate buffered saline (blank) and 2.5 ml of diluted antibody (test). The absorbance difference at 340nm between the test and blank solutions is proportional to the orosomucoid concentration which was calculated by reference to standards.
Semi-automated procedure using the Hitachi 705 The manual procedure was further modified to enable its adaptation to the Hitachi 705. Unless otherwise stated, all sera were diluted in 10 using a microprocessor-controlled syringe (BCL Dilutrend) by aspirating 50 #1 of serum and dispensing it with 450 #1 saline (9 g/l) into a 2 ml plastic vial. The prediluted sera were placed on the Hitachi sample carousel and the instrument set to aspirate 10/zl of prediluted sample, followed 40s later by the addition of 350 #1 of phosphate buffered saline (reagent 1). A blank absorbance reading was obtained just prior to the addition of 80 #1 diluted antibody (reagent 2) and this was subtracted from a further absorbance reading obtained 4 min later.

Results
Optimization of conditions Using the manual technique described above, a series of antigen/antibody curves were obtained. Figure demonstrates the effect of using different volumes (7, 10, 12, 15, and 20 #l) of a linearly related series of serum standards on absorbance. Conditions were selected so that the point of equivalence (peak of standard curve) occurred at a point well above the highest levels seen in pathological conditions (about 4.0g/l). The adoption of this measure meant excess antibody was always present and that the well-known problems associated with antigen excess [1] were avoided.
The influence of the concentration of pH, PEG 6000, sodium chloride and buffer, on the amount of turbidity produced from a pooled serum sample were investigatedmthe results are shown in figure 2. As a result of this study, the following buffer conditions were selected: (1) Sodium chloride concentration 1.8 g/1 (31 mmol/1).
The speed with which maximum absorbance was achieved after addition of antibody for a serum orosomucoid concentration of 1-5 g/1 is shown in figure 3. The general profile of the curve was similar at other serum orosomucoid concentrations varying from 0.5 g/1 to 3.5 g/1. At 37C, 10min was the minimum time required for turbidity to reach a maximum under the test conditions. Secondly, the Pye Unicam spectrophotometer produces absolute absorbances at 340nm when using the optimized method, whereas the Hitachi uses bichromatic optics requiring measurement of absorbance differences by reference to a secondary wavelength. This facility cannot be disabled and renders the method less sensitive due to the lower absorbance obtained. A typical absorption spectrum is shown in figure 4. A secondary wavelength of 660nm was chosen. These modifications (i.e. a 4 min incubation time and a secondary wavelength of 660nm) were incorporated into the optimized method for analysis of orosomucoid on the Hitachi 705.
The conditions finally adopted for use on the Hitachi 705 are outlined in the Appendix.    figure 5. The graph is virtually a straight line, enabling single-point standardization to be carried out. If significant deviations from linearity occurred, it would be preferable to carry out multipoint standardization; unfortunately, this is not straightforward since the Hitachi 705 software will only accomodate single-point standardization.
There was, however, a less than 2 difference between the results obtained by (1) adopting single-point standardization and assuming linearity of the standard curve up to an orosomucoid concentration of 2g/l; and (2) using multipoint standardization.
This problem can be circumvented by placing a series of standards in the,test sample positions Of the sample rotor and calibrating the Hitachi by single-point standardization. The values obtained for the serial standards in this way can then be plotted on the y co-ordinate against their known values in g/l, either manually or preferably via a microcomputer; the orosomucoid concentration of the test samples are then calculated from the standard curve in the usual way.  In contrast, turbidimetry, using the method described, costs 0.10-0.15/assay and repeat analyses of pathological sera with elevated levels are unnecessary. In addition, results are available within 15 min, whereas the prolonged incubation times associated with radial immunodiffusion ensure results are unavailable for at least 24 h. The precision ofthe turbidimetric method is better than that of the Mancini radial immunodiffusion technique (table 2), while the correlation between the two methods is good ( figure 6).