Portal de Periódicos da CAPES

The story of IgND

2005; Elsevier BV; Volume: 115; Issue: 3 Linguagem: Inglês

10.1016/j.jaci.2004.11.027

1097-6825

S. G. O. Johansson,

Glycosylation and Glycoproteins Research

University studies are not a way to support a young family, at least not in Sweden! Thus as soon as I was considered suitably qualified, albeit as a laboratory technician, I jumped at the opportunity to start work. In the fall of 1960, I joined the budding Section of the Blood Transfusion Service at the University Hospital in Uppsala. But routine laboratory work was not really exciting enough, and therefore when my boss, Claes F. Högman, MD, and his colleague Johan Killander, MD, started a research group in the highly topical field of immunochemistry-immunology, I was pleased to be invited to join. My first task was to establish a technique for quantitation of immunoglobulins. From the blood serology aspect, it was of interest to be able to characterize anti-Rh antibodies into the two, at that time, well-known immunoglobulin classes, then referred to as γG and γM, but in addition, we wanted to know more about the structure and function of the immunoglobulins. For the biochemical studies, another medical student, Hans Bennich, was recruited. Hans was then working on a research project on the separation of serum proteins at the Institute of Biochemistry under the famous Professor Arne Tiselius, one of the Nobel Laureates in this field at Uppsala University. Hans and I started to work with the immunoglobulins project around 1962, a close collaboration that would continue for more than 20 years. It soon became clear that Hans, in his work to isolate an immunoglobulin, needed sera with high concentrations. It was known that in a rather uncommon disease, multiple myeloma, affecting the immunoglobulin-producing plasma cells, a very high and homogenous increase in the cell product could be secreted. In electrophoresis this product was seen as a narrow and usually strongly stained band referred to as an M-component. I set up immunoelectrophoresis and started to identify each M-component detected in the routine screening of sera using electrophoresis. The M-components, being either IgG, IgM, or IgA, were purified, and using reduction-alkylation and digestion with proteolytic enzymes, such as papain and pepsin, Hans prepared many fragments of varying size and carbohydrate content, fragments like Fab (fragment antigen binding), Fc (fragment crystallizing; this was actually true only for some IgGs), or light and heavy chains. To be able to characterize the immunologic properties of these fragments, I developed immunoassays using antisera that we raised in rabbits. Both my clinical immunology activities and Hans' biochemistry work developed nicely, and by 1964, we had developed considerable technical skills and had accumulated large amounts of information on the complex structures of immunoglobulins. We were less good at producing publications, although I did report on our experiences of immunoglobulin quantitation at the Annual Meeting of the Swedish Physicians in 1963. However, some news about our activities must have spread through the grapevine because in January of 1966, a young scientist, Malcolm “Mac” Turner, came from Birmingham (UK) to work with Hans. Mac was a nice person and also a very useful link to immunochemistry science in the UK. In early 1965, I found a serum with an M-component that I could not identify as either IgG, IgM, or IgA. This was very exciting, especially in the light of information claiming that a D. S. Rowe had, during a stay at Johns Hopkins in Baltimore, found an M-component that fulfilled all the criteria for representing a new class of immunoglobulins. Following the alphabetical order, this had been termed IgD. I quickly managed to get in touch with Hans, who was then out of town for a few weeks. He also became excited, and over the phone, he instructed his laboratory technician to start, as a first step, a simple gel filtration, separating serum proteins by size. However, the unusual protein peak had the same position as IgG, and when I analyzed the fractions, I could indeed confirm that the M-component was IgG, which, for unknown reasons, was unwilling to react in serum. This experience was a bit disappointing! In June 1965, I found a second M-component that I could not identify (Fig 1). This time, to spare Hans the work and me the shame, I did the gel filtration separation myself at my hospital laboratory and analyzed the fractions for IgG, IgM, and IgA. This serum was different; the main protein peak indicated a molecular size similar to that of IgA, but in no fraction could I find the M-component. Could it be one of the subclasses of IgA, which was discussed among immunologists at that time? Could it be an IgA with hidden class-specific antigenic determinants? Maybe it was an IgD? Through Mac Turner, we contacted David Rowe, now back in Birmingham, UK, to get some anti-IgD. Or could it, we wondered, like David Rowe's M-component, be a hitherto unknown product of the uncontrolled growing plasma cells of a patient with myeloma? With the experiences of the odd IgG some months earlier fresh in our memory, Hans at first was not too excited about my new finding. However, after some weeks, I managed, despite the almost holy Swedish summer vacations, to convince him that this case was even more interesting than the previous one. Thus Hans started to look at the serum and came back excited, both with interesting information and lots of fractions for me to test for immunoglobulin content. Biochemically, the M-component differed from the other immunoglobulins and was referred to as IgX in our laboratory, adopting the Swedish custom to use “X” to designate an unknown. A 3.5S fraction, obtained after papain digestion, presumably an FcX, was used to immunize rabbits. It should carry any isotype antigen, and with an anti-IgX, we should be able to find a normal counterpart to the IgX M-component. We had a lot of experience of digesting IgG, and we knew the typical physicochemical properties of the 3.5S Fcγ fragment very well. IgX must be similar, or at least that was what we thought. The anti-FcX gave a very nice, heavy precipitate with the IgX M-component in gel diffusion and was used in a number of gel-diffusion and hemagglutination techniques to test sera from healthy blood donors and sera from patients sent to me for quantitation of IgG, IgM, and IgA as part of the clinical workup. However, in no case did we see the faintest precipitin line. The idea then came up that the production of IgX could be genetically restricted, and therefore I asked the patient if we could get blood samples from his family members. He was most cooperative and brought his brothers and a highly pregnant daughter. When their sera were tested, a strong beautiful precipitin line was obtained with the sample from the daughter. Great! Everything was crystal clear: IgX was indeed a new immunoglobulin class, and it was linked to pregnancy. The discovery was celebrated. There was just one small annoying fact: the precipitate was very strong and seemed to fuse with the positive control serum, which was the purified IgX M-component. Maybe we should repeat the test, but should we use a fresh serum sample? In the meantime the daughter had given birth to a healthy child, and she was still on the ward. Thus I went to see her. Sitting at her bedside, probably looking very “nondoctorish,” I convinced her to give us a new sample. Some 30 years later, she confided that my visit had scared her! But that sample was negative. Maybe the nonserious advice to “celebrate every successful experiment” should be more generally adhered to. There we were, stuck in a corner with an immunoglobulin with a unique structure and unique antigenic properties but without a normal counterpart.1Johansson S.G.O. Bennich H.H. Immunological studies of an atypical (myeloma) immunoglobulin.Immunology. 1967; 13: 381-394PubMed Google Scholar So back to basics! The papain digestion was repeated, but all 5 or so major fractions with sizes from 3.5S to 7S were used for immunization. Leif Wide, assistant professor at the Department of Clinical Chemistry, was consulted about the possibility of adapting his very sensitive techniques for human chorionic gonadotropin and other hormones already used by us for IgG and IgM to measure IgX. These projects took most of 1966. During that time, we obtained several anti-IgD antisera, none of which reacted with the IgX M-component. At the Swedish Physicians Meeting in November 1966, I reported on the unique IgX M-component. There was no real excitement. One world-famous Swedish clinical chemist told us that odd M-components were common, he in fact had several, collected over some years, in his freezer, and they were nothing to be excited about. Interestingly, later, using anti-IgD from David Rowe, he found that 4 of these were IgD myelomas. We had now worked with the IgX M-component for 18 months, and we still had not found any counterpart in normal serum. A month later, in January 1967, everything changed. The radioimmunoassay used (Fig 2) on the basis of the antiserum to the 7S fragment of IgX was inhibited in a dose-response way by sera of blood donors. We suddenly had all the evidence needed for definition of a new immunoglobulin class: a unique biochemical structure carrying unique antigens generally present in the population. According to the convention at that time, it was provisionally named using the initials of the patient from whom it originated, IgND. Actually, a couple of years ago, I heard a plenary speaker at an international congress tell the audience that “ND” was the abbreviation of “not detected”! Among the very first blood donor sera tested for IgND to represent a reference population was a remarkable outlier with an IgND level about 30 times the mean of the group. In all the studies I had performed on serum levels of IgG, IgM, and IgA, neither in normal nor in pathologic sera had I ever seen such an outlier. I contacted the donor, a young girl. She insisted on being healthy but did admit that she had experienced some breathing problems a couple of weeks earlier, when she had spent her Christmas holiday with her parents. They had a dog. I asked a friend at the outpatient allergy clinic to have a look at her, and he confirmed dog allergy–induced asthma. In my freezers I had many sera from patients with asthma. I had been quantitating serum IgG levels for a clinician who treated asthma with gammaglobulin injections. I quantitated IgND in about 50 of these sera and asked him, without knowledge of my results, to classify the patients as being allergic or not allergic. Serum IgND was increased in about 60% of the sera from patients with allergic asthma.2Johansson S.G.O. Raised levels of a new immunoglobulin class (IgND) in asthma.Lancet. 1967; 2: 951-953Abstract PubMed Google Scholar We were a bit concerned about these findings. As we understood it, it was definitely proved that the serum activity capable of mediating an allergic reaction, referred to as a “reagin,” was linked to IgA.3Vaerman J.P. Epstein W. Fudenberg H. Ishizaka K. Direct demonstration of reagin activity in purified γ1A-globulin.Nature. 1964; 203: 1046Crossref PubMed Scopus (10) Google Scholar Ishizaka et al4Ishizaka K. Ishizaka T. Hathorn E.M. Blocking of Prausnitz-Küstner sensitization with reagin by “A chain” of human γ1A-globulin.Immunochemistry. 1964; 1: 197-207Crossref PubMed Scopus (3) Google Scholar had provided the final evidence by showing that isolated heavy chains of IgA could block the Prausnitz-Küstner (PK) test and thus did have the structure critical for a reagin molecule. Because IgND seemed to be linked to allergy, we, together with Leif Wide, developed a radioimmunoassay for IgND antibodies to common allergens. The antiglobulin principle, used daily at the blood center for tracing anti-Rh antibodies in the sera of pregnant women, was adopted. The proteins of commercial allergen extracts were covalently coupled to cyanogen bromide–activated cellulose particles. The test, called the RAST, gave results that correlated well with the clinical allergy diagnosis.5Wide L. Bennich H. Johansson S.G.O. Diagnosis of allergy by an in-vitro test for allergen antibodies.Lancet. 1967; 2: 1105-1107Abstract PubMed Google Scholar However, in this work I had the first encounter with supposedly standardized allergen extracts, and it was rather shocking; they varied 100- to 10,000-fold in allergen content measured as IgND-binding capacity. The study of IgND in asthma, the report on the RAST, and findings of a dose-response blocking of the PK test with purified IgND, performed by Dr Denis R. Stanworth in Birmingham,6Stanworth D.R. Humphrey J.H. Bennich H. Johansson S.G.O. Specific inhibition of the Prausnitz-Küstner reaction by an atypical human myeloma protein.Lancet. 1967; 2: 330-332Abstract PubMed Google Scholar were all published in Lancet in the fall of 1967. However, in March 1967, we had contacted the Ishizaka group and informed them about IgND and invited them to compare reagents. In recent articles they had reported that they had an antiserum to something they called γE. It was said to be able to block the factor causing erythema to ragweed and, more interestingly, the reagin activity to ragweed allergen E in PK testing. The Ishizakas were excited and sent us some anti-γE, and we sent them an anti-IgND and some purified IgND. This was the start of several years of scientific collaboration and personal friendship. Together we approached the World Health Organization International Immunoglobulin Reference Centre. Both groups sent them what we had available of antisera, and Hans and I also sent purified IgND. Unfortunately, because of the very low concentration in serum, it had not been possible to purify γE. In February 1968, Hans, Kimi, and I, together with a couple of collaborators and experts, spent a week at the World Health Organization Centre in Lausanne. The conclusion of this workshop was that the unique biochemical and antigenic properties of IgND, supported by the unique antigenic properties of γE, represented a new immunoglobulin. The very first report was written,7Bennich H.H. Ishizaka K. Johansson S.G.O. Rowe D.S. Stanworth D.R. Terry W.D. Immunoglobulin E, a new class of human immunoglobulin.Bull World Health Organ. 1968; 38: 151-152PubMed Google Scholar announcing the presence of the new immunoglobulin class, IgE, taking the fifth and, as it turned out, final alphabetical position. The discovery of IgE has had a very positive effect on the clinical and scientific aspects of allergy, and the local and international collaborations that resulted in its formal recognition are in the greatest scientific tradition. It has also had a very positive personal effect, and I have been greatly honored—for example, some 20 international allergy societies, including the American Academy of Allergy, Asthma and Immunology and the American College of Allergy, Asthma and Immunology, have kindly nominated me as an Honorary Fellow.