Lives of the fellows

Keith William Taylor

b.19 April 1929 d.2 February 2012
MB BChir Cantab(1955) PhD(1960) MRCP(1987) FRCP(1994)

Keith Taylor was professor of biochemistry at the University of London. A man of substantial intellect, enormous energy and boundless enthusiasm, he was a pioneer in our understanding of the causes of type one diabetes. He himself was diagnosed with this form of diabetes as a student at Cambridge and the condition remained with him into ripe old age, both as a burden to endure and as a source of intellectual inspiration.

Born in Coxhoe, Durham, he was the son of Frederick William Taylor, a headmaster, and Edith Mary Taylor, a housewife. He was educated at Adams Grammar School in Newport, Shropshire, and went on to Cambridge University and then King’s College Hospital Medical School in London. He qualified MB BChir in 1955. After house jobs at King’s and the Brooke Hospital, he carried out research in Cambridge and at Harvard. He then went to King’s College Hospital as a clinical assistant. From 1968 to 1975 he was a reader at the University of Sussex, and professor of biochemistry in Sydney from 1975 to 1979. He was then appointed professor of biochemistry at the London Hospital, and finally honorary consultant in diabetic medicine at King’s College Hospital. He won several international prizes, including the major award of Diabetes UK, the Banting lecture.

Early in his research career he started working with Philip Randle [Munk’s Roll, Vol.XII, web] in Cambridge, taking advantage of their ability to measure insulin in the blood. There they demonstrated the impact of short-chain fatty acids on insulin secretion, which in time led to the characterisation of the Randle cycle. Later, after Keith was recruited to King’s College by R D Lawrence [Munk’s Roll, Vol.VI, p.275], then the preeminent diabetes centre, he wrote Diabetes and its management (Oxford, Blackwell Scientific, 1973), a major textbook of diabetes, with his two clinical colleagues, Wilfrid Oakley [Munk’s Roll, Vol.XI, p.424] and David Pyke [Munk’s Roll, Vol.XI, p.461]. This is still available and regularly updated.

Taylor and Pyke also studied co-twins of patients with diabetes, on the assumption that the co-twin would have pre-diabetes. It became apparent that the co-twin did not always show changes in insulin secretion or glucose tolerance, indeed that there was, in some, no sign that they would develop diabetes. By implication, the disease could not be strongly inherited and other factors must operate. In 1970, in a paper in the British Medical Journal, they reported that ‘the aetiology of diabetes depends on several factors and an inherited element is only one of the components in its causation’ Br Med J, 1970 Dec 12;4[5736]:649-51). It was already apparent that these non-diabetic twins tended to be the co-twins of the young-onset diabetes cases now called type one diabetes, the type of diabetes that Keith himself had. Keith’s ability to think laterally, by juxtaposing his state-of-the-art biochemical techniques with knowledge of clinical medicine, threw him in with Bob Gamble. Bob was an epidemiologist and together they noted the seasonal nature of the presentation of type one diabetes and an association with antibodies to Coxsackie B4 viruses. Keith then showed how viruses, even the virally induced cytokine interferon, can damage insulin-secreting cells.

Things were now moving fast in London, where a small group of which he was a member, known locally as ‘the mafia’, began to publish papers thick and fast on the origins of type one diabetes, including the genetic human leukocyte antigen association, on twins and families and on islet cell autoantibodies. Everything pointed towards a virus causing the disease in genetically susceptible individuals. But evidence from Andrew Cudworth [Munk’s Roll, Vol.VII, p.129], one of the ‘mafia’, that the onset of type one diabetes was a slow process over many years seemed to confound the viral hypothesis, which relied on seasonality of diagnosis. Years later it was found that the appearance of these autoantibodies could itself be seasonal and that viral particles could be detected in the pancreatic islets of newly-diagnosed patients. So, the man who had initiated the issue 40 years earlier was there again to lead the pursuit of viruses as a cause of type one diabetes. Keith arranged for us to write a book on the subject, which he co-edited, and which was published in 2012 (Diabetes and viruses New York, London, Springer) and to establish an international group, Viruses in Diabetes International Study Group (VIDIS). When VIDIS arranged an inaugural symposium at a major international meeting, Keith was to have given the key-note lecture, but sadly he died earlier that year, shortly after completing his editorial work. His final paper appeared posthumously (‘Virus antibody survey in different European populations indicates risk association between coxsackie virus B1 and type 1 diabetes.’ Diabetes. 2014 Feb;63[2]:655-62).

It is always difficult to capture a man in an avalanche of important discoveries, prizes and awards, and Keith was no exception. He had no airs and graces, was astonishingly modest and unassuming, and offered no sense of the effortless superiority that characterised some of his colleagues. His family, students and colleagues will, above all, remember his unalloyed enthusiasm and his zest for knowledge and argument. This small man with a big heart leaves an enormous legacy. Keith Taylor was an inspiration.

He was survived by his wife, Margaret Catherine Taylor, whom he had married in 1957, his two sons, daughter and two grandchildren.

R D G Leslie

[BMJ 2012 344 4032 – accessed 26 February 2015]

(Volume XII, page web)

<< Back to List