A research standard for human serum immunoglobulin E.

RESEARCH STANDARD FOR HUMAN SERUM IMMUNOGLOBULIN D 609 of serum containing IgD and use this as a working local standard in each titration. Such a local standard might be formed by fresh serum of suitable IgD content frozen in small aliquots and held at -20°C or colder. This local standard should periodically be carefully assayed against the research standard, using techniques which have been validated statis- tically, and the results should be analysed statis- tically. (5) The relative potency of IgD in a solution of a standard or in any other solution should be expressed as units of activity of IgD per ml of solution. REFERENCES Mancini, G., Carbonara, A. 0. & Heremans, J. F. (1965) Immunochemistry, 2, 235 Rowe, D. S., Anderson, S. G. & Grab, B. (1970) Bull. Wld Hlth Org., 42, 535 A Research Standard for Human Serum Immunoglobulin E by D. S. RowE,l LAURA TACKETr,2 H. BENNICH,3 K. ISHIZAKA,4 S. G. 0. JOHANSSON 5 & S. G. ANDERSON f Immunoglobulin E has recently been recognized as a distinct class of human immunoglobulin. The concentration of IgE in serum is usually measured either by a radio-immunoadsorbent test (Johansson, Bennich & Wide, 1968) or by the radio- active single-radial-diffusion technique (Rowe, 1969). Methods such as these could best be quantified by the inclusion of a biological standard for IgE in each titration run. The activity of IgE as antigen in the solution under test would then be estimated as a potency in relation to the activity of IgE anti- gen in the standard. A research standard for IgE has therefore been prepared, in a way similar to that previously described for research standards for IgG, IgA and IgM (Rowe, Anderson & Grab, 1970), and IgD (Rowe, Anderson & Tackett, 1970). As source material, sera containing high concen- trations of IgE from 91 adult West African donors, bled in Gambia in February 1968, were chosen. The blood was allowed to clot and the serum was separated under sterile conditions. Each serum gave a precipitate in a double-diffusion analysis in agar gel against an antiserum specific for IgE. A total volume of 1400 ml of serum was obtained. It was 1 WHO International Reference Centre for Immuno- globulins, Lausanne, Switzerland. 'Statistical Services Section, National Institute for Medical Research, Mill Hill, London N.W.7, England. 'The Wallenberg Laboratory, Uppsala, Sweden. 'The Johns Hopkins University School of Medicine, Baltimore, Md. 21212, USA. 'Blodcentralen Akademiska Sjukhuset, Uppsala, Sweden. Division of Biological Standards, National Institute for Medical Research, Mill Hill, London N.W. 7, England. transferred to the WHO International Reference Centre for Immunoglobulins at Lausanne, Switzer- land, for further testing, and then to the Division of Biological Standards, National Institute for Medi- cal Research, London, England, for freeze-drying. When in the laboratory, the serum was stored at -20°C; it was transported by air in insulated con- tainers with solid carbon dioxide. On 13 August 1968 the samples were thawed, pooled, and passed through a glass fibre pre-filter (designated by Millipore as type AP25) in order to clarify the material. On 14 August 1968 the material was distributed into ampoules and freeze-dried. (During the filling, 24 ampoules were weighed. The mean wet weight of contents was 1.013 g with a range from 1.011 g to 1.014 g.) The ampoules were further dried over phosphorus pentoxide, filled with pure dry nitrogen and sealed by fusion of glass. The ampoules were coded 68/341 and stored in the dark at - 20°C. The dry weight of the contents of each of 6 of the ampoules was determined. The mean was 92.84 mg and the range 92.64 mg-93.18 mg. The water content was determined on 6 ampoules. The mean content was 0.66% and the range was 0.38 %-1.34%. The oxygen content of the gas in each of 3 ampoules was esti- mated using the mass spectrometer. The average was was 0.56% and the range was 0.1 %-0.8%. Stability ofIgE in the research standard An estimate was made of the probable stability of the IgE present in 68/341. An accelerated degra- dation test was carried out by holding ampoules at 2586E -200C, +40C, and +37°C for a period of 1 year. The ampoule contents were reconstituted by the addition of 1 ml of water and the amount of IgE as antigen in each solution assayed by the radio- active single-radial-diffusion method by laboratories 1 and 3, and by the radio-immunoadsorbent test by laboratory 2. These two techniques differ in principle. Radio-immunoadsorption is an inhibition test, whereas radioactive single radial diffusion is based on precipitation. There is evidence that the former may be the more sensitive to the destruction of antigenic determinants of IgE (Stanworth et al., 1970). Laboratories 1 and 2 each used the same antiserum, which had been prepared by immunizing a sheep with the Fc fragment of E-myeloma protein ND; laboratory 3 used an antiserum prepared by immunizing rabbits with E-myeloma protein PS. All antisera were adsorbed with serum proteins, other than IgE, in soluble form, and were demonstrated to be specific for IgE by gel-diffusion analysis and other tests. The results were assessed by a statis- tical comparison of dose-response curves of the materials held at the different temperatures, using standard methods previously described (Rowe, Anderson & Grab, 1970) (see accompanying table). The responses were transformed if necessary and were plotted against log dose to obtain a linear relationship based on a series of at least 3 dilutions of each material. Laboratories 1 and 3 carried out 3 assays using a separate ampoule of each material for each assay. Laboratory 2 carried out 6 assays, 2 assays on each of 3 ampoules of each material. When the results of all the laboratories were considered together there was no evidence of signi- ficant change of potency of IgE in material stored at +4°C or + 37°C compared with IgE in material stored at - 200C. However, the confidence limits were wide for the 370C material and some indi- vidual results showed significant loss of potency, e.g., the +4°C material as tested by laboratory 3 and the +370C material as tested by laboratory 2. On the basis of these results 68/341 was considered to be acceptable as a reference reagent for quanti- tative purposes. However, as a further check on the stability of IgE in this material, accelerated degradation tests should be repeated after longer periods of storage. The pooled serum from which 68/341 had been freeze-dried was also held in frozen form at -20°C for 1 year, sealed in all-glass ampoules in approxi- mately 1-ml volumes. When this material was compared with 68/341 held at -20°C no significant difference in potency of IgE was detected (see accom- panying table). This finding suggests that IgE may be stable in serum frozen at - 20°C and that freeze- drying did not damage the antigen. The IgE content of the research standard It is recognized that current practice is to express immunoglobulin concentrations in terms such as mg/ml; however, no definitive value for the IgE content of 68/341 in such terms is given here. The concentration of IgE in preparations such as 68/341 might be estimated by immunochemical comparisons with representative solutions of isolated IgE of known concentrations. However, a variety of factors may affect such estimates. These probably include the nature of the isolated immunoglobulin and differences between various antisera and various quantitative immunochemical methods. The attempt ACCELERATED DEGRADATION OF IgE IN 68/341 AND COMPARISON OF FROZEN MATERIALa Laboratory Test b Freeze-dried Freeze-dried Frozen ___________ I ______ ~~+4'C +37C - 200C I RSRD 93.3 (74.7 -117.5) 97.1 (64.3 -142.67) 98.8 (83.7 -116.6) 2 RIST 100.5 (92.2 -109.6) 86.2 (78.1 - 95.3) 100.9 (73.7 -138.0) 3 RSRD 97.4 (95.9 - 98.8) 100.3 (92.3 -109.1) not done Mean log potency f 97.02 (88.47-106.41) 94.36 (71.25-124.96) 99.25 (90.94-108.31) a Results are expressed as mean percentage potencies (with 95 % confidence limits) relative to freeze-dried material stored at -20'C. b RSRD = radioactive single radial diffusion; RIST = radio-immunoadsorbent test. 610 NOTES RESEARCH STANDARD FOR HUMAN SERUM IMMUNOGLOBUTLIN E 611 to calibrate preparation 67/86 by weight of immuno- globulin has illustrated these difficulties." Unitage of the research standard 68/341 The material 68/341 has been established in the United Kingdom as the British Research Standard for Human Serum Immunoglobulin E and an arbitrary unitage has been assigned to it such that 1 unit of activity of IgE is the activity present in 0.009284 mg of the freeze-dried powder. Since the mean weight of contents of each ampoule is 92.84 mg each contains on average 10 000 units of activity of IgE. Availability of the research standard The research standard for IgE is now available to investigators wishing to use it in estimating concentrations of IgE in their own laboratories. Workers in North and South America should request 68/341 from Dr J. L. Fahey, National Cancer Institute, Immunoglobulin Reference Centre, Bethesda, Md. 20014, USA. Workers in the United Kingdom should request 68/341 from The Director, Division of Biological Standards, National Insti- tute of Medical Research, Mill Hill, London N.W.7, England. Workers elsewhere should request 68/341 from The Director, WHO International Reference Centre for Immunoglobulins, 21 rue du Bugnon, 1011 Lausanne, Switzerland. Recommendations for the use ofthe research standard (1) On receipt the research standard should be stored in the dark at - 20°C. (2) 1 ampoule of the research standard may con- veniently be reconstituted by the addition of 1.0 ml of distilled water. The powder should dissolve readily on standing for 1 hour at room temperature to give a slightly turbid solution. An appropriate series of dilutions of this material should be pre- ' Rowe, D. S., Anderson, S. G. & Grab, B. Paper in preparation. pared and used on the same day that the material is reconstituted. (3) The total volume of the solution of the pre- paration made by the addition of 1.0 ml of water will exceed 1.0 ml. It has been calculated that the average volume of the solution will be 1.07 ml. This solution will therefore contain 9346 units of IgE in 1.00 ml. (4) Since the research standard is not available in sufficient quantity for general inclusion in every titration run it is suggested that each investigator should prepare for himself a stable preparation of serum containing IgE and use this as a working local standard in each titration. Such a local standard might be formed by fresh serum of suitable IgE content frozen in small aliquots and held at - 20°C or colder. This local standard should perio- dically be carefully compared with the research standard, using techniques which have been vali- dated statistically, and the results of the comparison should be analysed statistically. (5) The relative potency of IgE in a standard or any other solution should be expressed as units of activity of IgE per ml of solution. ACKNOWLEDGEMENTS We are indebted to Dr I. A. McGregor and Dr R. J. M. Wilson, Medical Research Council Laboratories, Fajara, The Gambia, West Africa, to their staff and to the blood donors for their invaluable assistance in making suitable serum available to us. REFERENCES Johansson, S. G. O., Bennich, H. & Wide, L. (1968) Immunology, 14, 265 Rowe, D. S. (1969) Bull. Wld Hlth Org., 40, 613 Rowe, D. S., Anderson, S. G. & Grab, B. (1970) Bull. Wld Hith Org., 42, 535 Rowe, D. S., Anderson, S. G. & Tackett, L. (1970) Bull. Wld Hlth Org., 43, 607 Stanworth, D. R., Housley, J., Bennich, H. & Johansson, S. G. 0. (1970) Immunochemistry, 7, 321