Unravelling the Double Helix

by Gareth Williams

  On the autopsy table a few hours later, the hallmarks of the serial killer could be seen all over the murder scene. Healthy lung feels light and floats on water; it crackles faintly when pressed between the fingertips, like the popping of miniaturised bubble-wrap, because the alveoli are still partly inflated from the patient’s dying breaths. The lower lobe of the patient’s right lung looked quite different – dark and solid, like liver. The pathologist removed the lungs, took a fresh scalpel and cut out a chunk of tissue from deep inside the affected lobe. He dropped it into a sterile glass pot which was taken immediately to Dr J. Bell Ferguson, Medical Officer for Smethwick. Together with dozens of other samples of sputum, blood and lung from pneumonia patients, the specimen was packed into an ice-filled box and put on the fast train to London, addressed to a laboratory with a grand-sounding name near Covent Garden.

  So far, all this may seem irrelevant to the plot, but this is how real science happens. On a bench in this laboratory, specimens from pneumonia patients in Smethwick threw up a result that many researchers would have ignored because it did not make sense. Even after the finding was confirmed, most experts refused to believe it because it broke too many rules – and undermined the foundations on which some notably successful careers had been built.

  It was only many years later, when its full significance was finally appreciated, that this discovery was seen to be as momentous as Brown’s discovery of the nucleus, or Mendel’s Studies of Plant Hybridisation, or Thomas Hunt Morgan’s chromosomal maps of mutations. This was the first step on the long and winding road to understanding what DNA really does.

  Hunting in pairs

  ‘Pneumonia’ is an infection in the lungs; ‘lobar’ means that at least one of the lobes, the major anatomical territories of the lungs, has been invaded. Pneumonia runs a variable course. In the frail and elderly, it can appear almost humanitarian: ‘the old man’s friend’, bearing him away gently to a better place. By contrast, lobar pneumonia was a brute that slaughtered tens of thousands of healthy people in the prime of life.

  It took years to hunt down the killer. In 1881, Louis Pasteur found a round bacterium (coccus) in the saliva of a boy dying from rabies. This had nothing to do with rabies (which is caused by a virus), but the same bacterium was soon isolated from patients with lobar pneumonia and found to induce fatal pneumonia when injected into rabbits – hence the name ‘pneumococcus’. High magnification showed that pneumococci occurred in pairs, often surrounded by a transparent capsule that looked like a clear halo when the bacteria were stranded in a drop of India ink on the microscope slide.

  Curiously, not all pneumococci were murderers. A single virulent pneumococcus would kill a mouse in hours, while millions of non-virulent ones could be injected with no ill effects. It turned out that virulence hinged on the capsule: strains of pneumococci with a well-developed capsule were lethal, while non-capsulated strains could not gain a foothold. The capsule is a form of stealth technology which enables pneumococci to sneak past the white blood cells which constantly patrol the bloodstream and tissues. The white cells normally gobble up invading bacteria (‘phagocytosis’), but are unable to kill bacteria that are protected by a capsule.

  At first sight, the pneumococcus seemed a promising target for serum therapy of the sort that Simon Flexner had used against meningitis. Antibodies were readily raised by injecting dead pneumococci into horses and they worked like a dream in laboratory mice; unfortunately, they often failed in patients with lobar pneumonia. The reason that some pneumococci sidestepped antibody therapy was discovered by Friedrich Neufeld (known universally as Fred), a quiet and methodical German bacteriologist who worked in Berlin with Robert Koch, the world-famous microbe-hunter and Nobel laureate. Neufeld first showed that phagocytosis was the crucial defence in aborting an infection. Capsulated pneumococci resisted being phagocytosed, but their protection could be wiped out by antibodies against the capsule; the antibodies coated the capsule and pulled the bacteria into bite-sized clumps that the phagocytic cells devoured as usual. These antibodies were astonishingly potent: a tiny dose could save a mouse injected with enough pneumococci to kill 1,000 million unprotected mice.

  Next, Neufeld worked out why immune serum sometimes failed. Horses injected with dead pneumococci always produced effective antibodies, but these did not always protect against live pneumococci isolated from a different source. Neufeld proposed that each pneumococcus must carry a particular ‘antigen’ (the specific molecule against which the antibodies are tailor-made) and that the antibodies made against one isolate of pneumococci would only kill others that carried the same antigen. By testing hundreds of isolates of pneumococci and their ability to make life-saving antibodies in mice, Neufeld deduced that there must be three distinct kinds of pneumococcus antigen, which he called Types I, II and III. Pneumococci of one type raised antibodies which killed all pneumococci of that type, but had no effect on the other two. Only the correct antibody caused pneumococci to clump under the microscope; this ‘agglutination’ reaction became a quick diagnostic test for ‘typing’ an unknown pneumococcus. It followed that immune serum would be useless in treating lobar pneumonia if the antibodies had been raised against pneumococci of a different type from the one that was attacking the patient.

  All this made Neufeld famous. George Bernard Shaw grafted his work into the plot of The Doctor’s Dilemma (1906) as a miracle cure for tuberculosis.* And in 1910, Simon Flexner wrote to ask for samples of the three types of pneumococcus and their corresponding antibodies. Neufeld was happy to oblige, and shipped the materials to Alphonse Dochez, a young American bacteriologist at the Rockefeller. Over the next couple of years, Dochez worked his way through all the pneumococci that had come through the door of Rufus Cole’s pneumonia ward. He found several new strains which were distinct from Neufeld’s Types I, II and III. Continuing Neufeld’s numbering, he called this ragbag group ‘Type IV’ (later known familiarly as ‘the American scrapheap’).

  Neufeld’s research brought some benefit to the bedside; the pneumococcus that had struck down the patient could now be typed and then hit with the correct antiserum. Immune therapy began to enjoy greater success, but still fell short of a miracle cure.

  Robert Koch died of a heart attack in 1910, shortly after attending a conference in Landau. His name was added formally to his Institute in 1912, to celebrate the thirtieth anniversary of his discovery of the tuberculosis bacillus. In 1917, Fred Neufeld succeeded his mentor and hero as Director of the Robert Koch Institute for Infectious Diseases. In that role, Neufeld kept his head down while pushing ahead with the science. He was Protestant, politically inert and lived quietly at home with his mother. When the National Socialist German Workers’ Party was born in 1920, it seemed to pose no threat.

  In the summer of 1927, Neufeld celebrated his ten years as director with a trip to London. He went to visit a reclusive Englishman who never appeared at conferences and rarely published, but had recently made curious, almost heretical claims about the types of pneumococcus.

  Grim and almost sordid

  Dudley House in Endell Street, near Covent Garden, had seen better days. Its lower two floors were now a post office, above which sat the Pathological Laboratory of the Ministry of Health. Despite its grand title, the lab was described as ‘grim and almost sordid’, with ‘very little up to date equipment even for that period’. The government employees who worked there spent their time monitoring outbreaks of infection – essential but dull work which could have been designed to stifle creativity.

  A man who was perfectly at ease here was Frederick Griffith, now in his late forties. Born in Cheshire, Griffith followed his elder brother Stanley into medical school in Liverpool and began his medical career by working with him on the mammoth Royal Commission on Tuberculosis (1901–11). Stanley then headed off to Cambridge and established himself as an expert on bovine tuberculosis. Fred’s career was more mundane. He became ‘a civil servant and proud of it’, as a local government bacte
riologist in London.

  Fred Griffith specialised in mapping outbreaks of infections such as pneumonia, scarlet fever and tuberculosis. This kept him shackled to the laboratory bench, identifying bacteria in samples sent from across the country. He was variously described as ‘immensely hard-working and meticulously careful’, ‘modest and retiring’, ‘scrupulously honest’ and ‘a lovable personality to those few who got to know him’. That last tribute said it all. Griffith was a loner, uncomfortable outside his small circle of associates.

  Most scientists go to conferences, as a break from routine and a chance to catch up with friends, enemies and whatever is new in research. Griffith had a pathological hatred of meetings and avoided them like the plague. During his forty-year career, he presented just one paper, when the International Microbiological Congress descended on London in 1936. His colleagues had to push him into the taxi that took him to the meeting, where he made his lecture so dreadful that he would never be invited again. Griffith also starved himself of the oxygen of publicity by being ‘reluctant to write papers’, but his few publications were highly regarded by Fred Neufeld and the pneumonia team at the Rockefeller.

  Apart from big brother Stanley, Griffith’s closest friend was William Scott, his colleague in the Pathological Laboratory. Scott was a few years younger but had seen more of the world: Munich, Cambridge, the Straits Settlements and Paris, where he acquired perfect French and a wife to match. Scott and Griffith met in 1914 while tracing meningitis outbreaks near London and found that their personalities meshed. For the rest of their lives, they worked together in the one-room lab at Dudley House, doing all their own bench work and keeping the team cheerful with their ‘dry wit’. At night, Scott returned to his family and mansion in Dulwich, and Griffith to his bachelor flat in Eccleston Square in Pimlico, which he shared with his favourite niece, a housekeeper and an Irish terrier called Bobby.

  Few photos survive of Fred Griffith. A formal portrait shows a balding, intense man who chose not to smile for the camera. Two others capture him off-guard: tanned and relaxed, on one of his regular skiing holidays in the Alps, and sitting with his dog on the Downs above the futuristic house that he had built on the South Coast near Brighton (Figure 11.1).

  Figure 11.1 Fred Griffith with his dog, Bobby.

  Taking the rough with the smooth

  The routine workload in the lab over the post office left Griffith and Scott little time for research. Somehow, though, Griffith managed to squeeze in a long series of peculiar experiments. What he found made no sense at first – but it lit a slow-burning fuse that, twenty-five years later, detonated the biggest bang of twentieth-century biology.

  Griffith’s curiosity was first aroused by odd results in the samples sent in by Dr J. Bell Ferguson, medical officer in Smethwick. In 1920, most cases of lobar pneumonia had been caused by Type II pneumococci, but these steadily lost ground to Type IV, which predominated in 1922. While looking for an explanation, Griffith asked Ferguson to send serial sputum samples for up to fifteen days after cases reached hospital – and was surprised to find that some patients harboured two or three other types of pneumococci, which appeared several days after the one initially identified.

  Where had these new types come from? Griffith thought it very unlikely that a single patient would pick up ‘multiple infections with different and unalterable varieties of pneumococci’. Instead, he suggested that the types that appeared later were all derived from the one which first attacked the patient. This was revolutionary, because it contradicted Neufeld’s dogma, now twenty years old, that the type of a pneumococcus was a permanent fixture.

  In early 1927, Griffith began experiments to test his unlikely hypothesis. He had already isolated distinct strains of pneumococcus which would make it easy to tell if one type had changed into another. These were two variants – which he named ‘S’ and ‘R’ – of each type (I, II, III and IV). S were all capsulated and lethal, and were named for the ‘Smooth’ appearance of the shiny colonies they formed on an agar jelly plate. The R variants of each type were non-capsulated and harmless, and were named ‘Rough’ after the granular surface of their colonies. The difference between S (virulent) and R (harmless) was not subtle. A mouse would be killed by a single S pneumococcus of any type injected under its skin, but was unharmed by 500 million of its R cousins.

  Griffith began by injecting mice separately with either live R (harmless), or S (lethal) which had been killed by heating. As expected, all these mice survived. Then he injected other mice with a mixture of live R and dead S, which should also have been harmless. Unexpectedly, some of these mice died of septicaemia, just as if they had been given live S – which is exactly what he found in the blood that he sucked out of their hearts after death. Somehow, the dead S had induced the harmless R to grow a capsule and become killers. Initially, he mixed live R and dead S pneumococci of a single type, e.g. Type II. Not surprisingly, the live S that he isolated from the mice that died were of the same type, in this case Type II.

  Griffith then tweaked the experiment to see whether he could change one type into another. He injected eight mice with a live R + dead S mixture, but now the live R were Type II and the dead S were Type I. Six of the eight mice were unharmed but the other two died of septicaemia – and paradoxically achieved immortality, because the live S pneumococci that killed them were Type I (Figure 11.2). These ‘transformed’ pneumococci were the real thing – capsulated and lethal. From beyond the grave, the dead Type I S had passed on both its licence to kill and its antigenic identity to the harmless R.

  If they could be believed, these results had wide implications. They smashed the assumption that the type of a pneumococcus was immutable and transmitted unchanged to all its progeny. This could explain why serum therapy sometimes failed, even with the right antibodies: the pneumococcus could wriggle out of the line of fire by turning into another type against which the serum would be impotent.

  And if type was no more than a flag of convenience, what was now to be made of the work which had made the reputation of Fred Neufeld, Director of the Robert Koch Institute?

  Figure 11.2 Griffith’s experiments on the transformation of pneumococci. Unexpectedly, mice died when given a mixture of harmless live R Type II and heat-killed S Type I – because their blood was full of live S Type I. An unknown substance from dead S had made the live R lethal and changed their type from II to I.

  Shock of the new

  Neufeld’s visit to Dudley House in early summer 1927 was unforgettable. He had intended to quiz Griffith about his earlier paper on the multiple types of pneumococci which had appeared in pneumonia patients in Smethwick, but Griffith showed him the unfinished manuscript in which he claimed that he had made pneumococci change their type. By then, Griffith had repeated the experiments many times and confirmed the finding. Immediately on returning to Berlin, Neufeld put everything else aside and, with Griffith’s blessing, began running those baffling experiments again to see if Griffith really had destroyed one of the laws of bacteriology.

  Fred Griffith submitted his paper to the mid-list Journal of Hygiene in late August 1927, and it appeared in January 1928 under the unsensational title, ‘The significance of pneumococcal type’. It ran to over 22,000 words, over twenty times as long as the paper which Watson and Crick published in Nature twenty-five years later. The experiment that should have made Griffith famous fills just two pages. Dead S pneumococci of any type could ‘transform’ – i.e. pass on their virulence and type to – any type of live R. For example, live S Type IV were recovered from the hearts of mice that died after being injected with live R Type I mixed with dead S Type IV. Accidental contamination with live S was ruled out; the dead S (cooked at 60 °C for three hours) really were dead and even massive doses could not hurt a mouse when injected on their own. Interestingly, S that were heated to 100 °C were unable to transform R, indicating that whatever mediated transformation was destroyed by the higher temperature.

  To explain his find
ings, Griffith speculated that every pneumococcus has a ‘vestigial’ capacity to make all types of capsule, and that dead S might somehow kick-start the production of its own type of capsule. He did not mention the possibility that transformation might represent a genetic change – even though his paper on the Smethwick samples, published six years earlier, had made the bold claim that Type I could turn into Type IV by ‘mutation of type characters’. And so, despite the powerful new evidence staring him in the face, Fred Griffith missed the real ‘significance of pneumococcal type’, and the biggest trick of his career.

  Predictably, the first reaction to his paper was ‘scepticism and disbelief’, but English bacteriologists kept quiet because Griffith was regarded as a good chap, even if a bit odd. Nobody was interested enough to repeat his experiments to see where he had gone wrong. However, Fred Neufeld did – and he proved that the reclusive Englishman in his run-down lab over the post office had got it right. Neufeld’s paper was published in March 1928, just a few weeks after Griffith’s, but in the Zeitschrift für Immunitätforschung which was largely invisible outside Germany.

  Nothing happened for another year and then another paper appeared, this time in the Journal of Experimental Medicine, America’s top medical research journal (editor, Dr Simon Flexner). The author was Hobart Reimann at the Peking Union Medical College, the Rockefeller’s overseas research base in Beijing. Reimann’s conclusion left no room for doubt. By ‘applying the methods of Griffith’, he had confirmed that ‘bacteria of the R form . . . may even transform into S forms of a different type’.

  There was no reaction from Griffith. He had lost interest in pneumococci and had moved on to investigate the dreaded ‘puerperal fever’ which killed mothers and their newborn babies. In a farewell review article on the pneumococcus, he barely mentioned his revolutionary discovery, and he told a colleague that it was ‘now up to the chemists’ to follow up pneumococcal transformation.