Originally prepared for Bright Sparcs by ASAP on behalf of the

Australian Academy of Science Biographical Memoirs of Deceased Fellows

Pehr Victor Edman 1916-1977

Pehr Victor Edman was born in Stockholm, Sweden, in April 1916 and died in Munich, FRG, in March 1977. He was born into a lawyer's family and received his schooling in Stockholm. In 1935 he began medical studies at the Karolinska Institute and graduated with his primary medical qualifications in 1938. He became interested in research and, following graduation, continued to work at the Karolinska Institute, largely in the laboratory of Professor Eric Jorpes. He appears to have systematically taught himself organic chemistry at this time by extensive reading. During the war years his research was interrupted by a long period of service in the medical corps of the Swedish Army. He was awarded the degree of Doctor of Medicine in 1946. The subject of his thesis was the purification and analysis of angiotensin from bovine blood. His earlier published studies concerned heparin and secretin, which were interests of his mentor Jorpes.

At this juncture Edman began to take the independent research direction which he followed almost uninterruptedly for the rest of his career. He accepted a grant to work for a year in the Northrop-Kunitz laboratory at the Princeton branch of the Rockefeller Institute of Medical Research. Swedish medical research had been isolated during the war and he was anxious to learn of the progress made in the United States. Moreover, his work on angiotensin had made him realize that simple compositional analysis would not be helpful in providing a basis for understanding the biological function of peptides or proteins. The realization that proteins were not colloids but that each had a definite molecular weight and a specific structure was beginning to emerge, especially as a result of the work of the Uppsala group. Edman knew that the order of aminoacids linked by peptide bonds was an essential part of the unique makeup of any given protein. At Princeton he began experiments to try to find a way to chemically decode the aminoacid sequence of proteins.

In the early years of Edman's attempts in his area two general procedures were being used to attack the sequence problem. Various reagents had been found useful in labelling the amino terminal (or first) aminoacid through its reactive amino group and allowing identification as a derivative. One of these reagents, luorodinitrobenzene (FDNB), which gave the linitrophenol (DNP) derivative of the aminoterminus, was used by Sanger in his epochal work on the structure of insulin. By using the FDNB reaction with sets of overlapping peptides derived from partial cleavage of insulin, Sanger, by 1956, was able to deduce a unique structure for the insulin molecule. This was the first primary structure of a protein to be decoded, but despite the undoubted importance of the feat it was clear that the method was too cumbersome to have wide application.

Another reagent used for aminoterminal determination was phenylisocyanate (PITC), introduced for this purpose by Abderhalden and Brockmann in 1930. As with FDNB the hydrolysis to release the aminoterminal amioacid derivative destroyed many of the other peptide bonds, leaving the remaining protein useless for analysis. In Princeton, Edman realsed that if phenylisothiocyanate (PITC) were used the nucleophilic sulphur would weaken he adjacent peptide bond, raising the possibility of finding conditions for its hydrolysis that did not cleave the remainder of the molecule. This remaining peptide could then be subjected to a second reaction with PITC and the second aminoacid determined, and so on theoretically to the carboxyterminal end of the molecule. Whether Edman thought of this solution completely independently, whether some unrelated paper uncovered in his wide reading drew PITC to his attention, or whether some colleague in Princeton or Stockholm suggested its use is not known. In view of the success which the reaction ultimately achieved, the latter seems unlikely in the absence of any claims or reminiscences to this effect. In his review of aminoacid sequencing methods written in 1969 Edman is at pains to stress that the PITC reaction was not at all derived from the earlier PIC reaction, as they had different mechanisms of action. However, in view of the superficial similarities between the reagents and the similar uses to which they were put in protein chemistry, this seems a little strained. By the time Edman returned to Sweden in 1947 he had performed enough experiments to know that the idea was practicable and could form the basis of a protein sequencing technique.

Edman took up an associate professorship at Lund and continued to work almost exclusively on protein degradation. The derivative resulting from the coupling of PITC with an aminoacid, the 3-phenyl-2-thiohydantoin (PTH) aminoacid, proved to be a stable compound in almost all cases. Edman synthesized the PTH derivatives of all the amino acids found in proteins and developed chromatographic systems to identify and quantify them; conditions for the coupling of PITC to the amino terminus and the cleavage of the PTH derivative which worked smoothly for all peptide bonds were found. After two years work Edman was able to publish the chemical details of a method capable, in theory, of solving the problem of primary structure of proteins and of providing the essential information about innumerable proteins essential for further advances in protein biochemistry. The characteristic ultraviolet absorption spectra of the PTH-aminoacids made them particularly suited to quantitative studies. It permitted useful measurements of the subunit structure and molecular weight of proteins and, in conjunction with column chromatography, it was an alternative to the ninhydrin reaction for aminoacid analysis. However, these possibilities could not be fully exploited until the recent developments in high performance liquid chromatography. The method became widely known and was given the eponym 'Edman degradation' by Kai Linderstrom-Lang of the Carlsberg Laboratories.

In the early 1950s Mr "Jack" Holt, a well-known Victorian racehorse trainer, died and his will provided that the income from his estate, to be held in trust, was to be used for medical research at St. Vincent's Hospital in Melbourne. By 1956 the hospital authorities had decided to establish a separate research institution in the hospital rather than disburse the funds as research grants to existing hospital units. The decision had also been taken to develop a non-clinical basic science area, preferably biochemistry. The School of Medical Research, as it was originally known, had its own governing board and for practical purposes functioned independently of the hospital administration.

Edman applied for the position of Director of Research. He was clearly the outstanding candidate and moreover his interests coincided with the preference for biochemistry as the focus of the School. In 1957 Edman accepted he offer of the position of first Director of Research at St Vincent's School of Medical Research in Melbourne. His reasons for this move, which were said to have been a mixture of general dissatisfaction with scientific resources in Lund and the impending breakdown of his marriage, must have been strong and not primarily directed to career improvement. He had just accomplished an outstanding piece of individual biochemical research which would have made him welcome in many leading centres in the northern hemisphere. In Australia, in Melbourne, he would be largely isolated from these centres; the School was a new institution, with no traditions, no established workers, nor any support staff; and although situated in a teaching hospital of the University of Melbourne, it had itself no academic or university affiliation. Moreover, Australia at that time was not known for generous government research funding. Despite this formidable array of disincentives Edman decided to move alone to Melbourne to continue his work, initially without trained help and without close colleagues.

In Australia Edman completed a few small projects on other aspects of protein structure hat he had begun in Sweden, but otherwise he worked almost entirely on the phenylisothiocyanate (PITC) degradation. This work fell into three phases: improvements in the conditions for the degradation, largely focussed on the elimination of side reactions; 'automation' of the reaction sequence; and application of the degradation to various sequence problems. The latter phase overlapped the other two and usually involved the interests of visiting scientists who had come to Edman's laboratory to learn or use his technique.

By 1960-61 the three-stage degradation reaction had been essentially perfected. Its universal application and repetitive nature suggested to G.S. Begg, Edman's Australian technical assistant, that it would be suitable for automation. Edman realized that the no. of existing proteins (about ten million) made manual sequencing an impossible task and was quickly converted to the idea of automation. The close control of reaction conditions possible with automation also gave promise of higher and more constant repetitive yield than were possible manually. High repetitive yields are crucial to repetitive processes, whether synthetic or degradative. Geoffrey Begg had been one of the early technical staff employed by Edman after his arrival in Melbourne. He had no formal qualifications but from a combination of courses at technical college and self-instruction had achieved a remarkable expertise in practical chemistry and glassblowing, mechanical engineering, and electronics. The sequenator project provided a perfect opportunity to use these multiple talents which complemented Edman's academic and theoretical knowledge. Edman and Begg worked as a team on the sequence automation project, with no sustained input from other workers. It was typical of Edman's thorough approach to all tasks that he became a sufficiently adept tool-maker in this period to do much of the fitting and turning himself.

The basis of what was to become the protein sequenator was developed to a prototype stage in a period of a few weeks in the autumn of 1961 - the glass cup spinning on its cylindrical axis, addition of reagents via a catheter, reactions in a thin liquid film on the wall of the spinning cup, and extractions by solvent moving upwards over the film into a groove. Within two years Edman and Begg had built, in their own workshop, a machine capable of reliably carrying out the reactions of the degradation. They had found new conditions and reagents suitable for the physical conditions of the spinning cup; for example, the open cup in general required less volatile chemicals and the narrow delivery and effluent tubes demanded special attention to the surface tension and rheological properties of the solvents. Edman's wide knowledge of classical organic chemistry enabled quick progress in converting the manual reaction to its automated form. In 1964 Edman reported his preliminary findings to a meeting in Scotland. In 1967 in the first issue of the European Journal of Biochemistry with Begg as coauthor he published his definitive paper demonstrating an unbroken automated determination of the aminoterminal sixty aminoacids of humpback whale myoglobin at the rate of one residue per hour. The extent of this advance can be gauged from the knowledge that at that time the most extensive manual degradation encompassed about fifteen residues at a rate of one per day. Many laboratories could not establish the manual degradation at all, owing to a failure to appreciate the importance of pure reagents in eliminating side reactions. During the next few years Edman's aim was to improve the repetitive yield obtained from the machine; an increase from the 98% of the 1967 paper to 99% was calculated to double the length of determinable sequence. The protein sequenator in Melbourne remained unique until late in 1969 when the Beckman Instrument Company in the United States put on the market a commercial version based on Edman's design. Edman played no part in the commercialization of his machine. The Board of the School discussed the possibilities of patenting the sequenator but soon accepted Edman's strong view that he should publish fully without patent protection. Edman was elected a Fellow of the Australian Academy of Scienec in 1968 and a Fellow of the Royal Society of London in 1974. Edman became an Australian citizen in the mid sixties.

In 1972 Edman resigned from St Vincent's School of Medical Research and became Director of Protein Chemistry I at the Max Planck Institute for Biochemistry at Martinsreid near Munich. In 1968 he had remarried; his second wife Agnes Henschen had come from Stockholm and this had given him a reason to think of a return to Europe. In the ten years since his arrival in 1957 the School had remained small, and attempts to raise support for expansion on the basis of the success of the sequenator project proved not very successful. Now, as many years before in Lund, he believed that the importance of his work was not properly recognized and that he would continue to have inadequate resources in Melbourne. A move to the new laboratories of the Max Planck Institute seemed to provide an answer to his needs. Edman set up his laboratory in Munich along the lines of that in Melbourne and with the same aim of increasing the efficiency of the degradation. In addition, with his aid, Agnes Henschen began to make substantial progress in her studies of fibrinogen structure. Sadly, Edman developed a cerebral tumour and died after a short period of coma in 1977.

Edman played little if any role in broader scientific administration or politics in Australia. Although his School had no formal academic affiliation, there is no evidence that he would not have been accepted in these arenas. Some efforts to arrange a personal appointment at the University of Melbourne came to nothing. Thus he remained something of an enigma in the scientific community. He was slow to publish, with approximately one and a half papers per year during his Australian period, which made difficulties for those wishing to implement the method. If his impact on the Australian scene was limited, it was paradoxically the result of his single-minded pursuit of the sequence degradation. Such work, despite Edman's reputation, was not very attractive to students and he never built up a tradition of a flow of graduate students. Once the initial work on the manual or especially the automated reaction was complete, the details would easily have been completed by others in his or other laboratories. One cannot help thinking that his impact would have been so much greater had he seen himself able to move strongly into new areas of protein structure and function. Biological research often requires the appreciation of the importance of an approximate result for advancement.

Nevertheless the Fellowships of the Australian Academy of Science and the Royal Society of London indicate in how much esteem his work was held internationally, and this judgement has been supported by later events. Technical advances in related fields, especially in liquid chromatography and sensitive ultra violet detectors, have led to the development of low-level or microsequencing techniques which still employ the reactions described by Edman forty years ago and which play an indispensible role in gene isolation and molecular cloning.

Edman's reputation as a reclusive person, often difficult to dealwith, did not arise from those who worked with him in the laboratory. He was reserved by Australian standards, but courteous and always helpful, often humourous, and took pleasure in organising social occasions both at his home and outdoors in the country. To those who came to know him in the laboratory two aspects probably had a lasting influence. In those days he was a rare example in the hospitals and the world of Australian medical research of someone who devoted himself fulltime to nonclinical research. This served as an example, to those who came across him, of the possibility of such a career. At a time when biochemistry in Australia was largely concerned with the intricacies of metabolic pathways, an area where the great discoveries had already been made, Edman understood and stressed the importance of the information-containing macromolecules. The double helical structure of DNA had been proposed by Watson and Crick only a few years before Edman's arrival in Australia. The possibility of obtaining a corresponding understanding of the more complex structures of proteins which Edman's work opened up inspired his colleagues with the belief that they were in a position to participate directly in a new era of biological investigation.

F.J. Morgan works in the Department of Biochemistry, La Trobe University.

This memoir was originally published in Historical Records of Australian Science, vol. 8, no. 2, 1990