In 1923, Frederick Griffith, a medical officer in the British Ministry of Health, demonstrated that pneumococcal types (I, II, III, or IV) could exist in either R or S forms. R and S designated the attenuated and virulent strains of certain bacteria, respectively: non-virulent colonies (R) appeared rough and irregular under the microscope, while the colonies of virulent strains (S) were smooth and shiny. Griffith showed that R and S strains could be converted from one to the other in living organisms as well as in the laboratory. Griffith's discovery initially meshed nicely with Avery's understanding of virulence: the smooth, virulent colonies were encapsulated, whereas the non-virulent strains had lost their capsules. Avery's concept of type specificity was not therefore directly challenged, for reversion of R to S form always led to the production of a specific soluble substance identical with the S type from which the R form had originated.
In their work with pneumonia patients at the Hospital, Avery's group had concluded that pneumococcal types I and II usually died out during convalescence and were often replaced by type IV. However, Griffith understood that the R and S forms were associated with the properties of non-virulence and virulence respectively. As early as 1922, he arrived at the alternative theory that the virulence of types I and II were attenuated during convalescence, and that this change was accompanied by the mutation of type characters, which transformed and degraded the pneumococci into those of the heterogeneous and less virulent type IV. In this way, Griffith believed, he had shown the affinities, rather than differences, between the various pneumococcal types.
In 1928, Griffith published a further demonstration of transformation of one pneumococcal type to another in vivo (within a living organism), which jolted the field of pneumococcal immunology. He had injected into mice the living cells of the R form of Type I pneumococcus mixed with heat-killed cells of the S form of Type II pneumococcus. The mice developed pneumonia infections and died, and from their blood Griffith isolated colonies of the S form of Type II. Not only did the R form convert to S, but the pneumococcal type was transformed as well, from I to II. In the bacteriology of the 1920s, the conversion of the R to the S form could be regarded as an adaptation to the environment. However, the transformation of Type I to Type II was the equivalent of the transformation of one species into another, a phenomenon never before observed.
Avery was initially skeptical of Griffith's findings and for some time refused to accept the validity of his claims, believing that they were the result of inadequate experimental controls. Avery's research on therapeutic sera led him to conclude that pneumococcal types were fixed and that specific therapeutic agents could thus be developed to combat the various types. A transformation from type to type in vivo presented a disturbing clinical picture, as well as a challenge to the theoretical formulations of contemporary bacteriology. Soon, however, new evidence came to light in favor of Griffith's claims, as his results were duplicated by Fred Neufeld in Germany and at the Rockefeller Institute Hospital by Michael H. Dawson and Richard H. P. Sia. Dawson and Sia confirmed Griffith's findings in 1929 and developed a method of achieving transformation in vitro (outside the body). When Dawson left the laboratory in 1930, James Lionel Alloway took the pursuit one step further. Alloway broke open the S-form bacteria to set the contents free, then passed the culture through so fine a filter that the shells, together with any unbroken cells, were removed. When this extract, in cell-free form, was added to a growing culture of the R form, transformation took place. From the cell-free extract, a material called the "transforming principle," seemed to direct the transformation of one type of pneumococcus into another. It could be precipitated, out of solution, as a solid substance with alcohol.
It was with some reluctance that Avery eventually accepted that pneumococci could be made at will to undergo transmissible hereditary changes in immunological specificity. However, once he accepted the new phenomenon, he immediately visualized its far-reaching implications not only for bacteriology and genetics, but also for general biology and medicine. Alloway left the lab in 1932 and the following year published the lab's last paper on the subject of transformation before the landmark 1944 article. Experiments on the transforming principle were conducted by other members of Avery's group, notably Colin M. MacLeod, who joined the Rockefeller Institute in the summer of 1934. At this time Avery was diagnosed with Grave's disease--a debilitating condition caused by an excessive secretion of thyroid hormones that can lead to protrusion of the eyes, nervous excitability, and a rapid heartbeat--and did not fully recover for several years. Although he remained an interested observer and advisor in MacLeod's work, it was several years before he once again actively participated in research. Between 1934 and 1937, MacLeod was able to derive a rough strain from encapsulated type II pneumococcus that proved very susceptible to transformation and remarkably stable in the R form. This strain was used for most of the subsequent research on transformation. He was also able to improve the reliability of the test for transforming activity and began new approaches to fractioning the extracts. However, in 1937, believing that he had reached an impasse in research on bacterial transformation, MacLeod suspended work on the subject until the fall of 1940, when he and Avery resumed the search for the identity of the substance that caused transformation.
Although dedicated to learning the secrets of the transforming principle, this research was not the sole focus of the laboratory. As part of the Rockefeller Hospital, the members of the laboratory always had clinical responsibilities, and these influenced the group's other research efforts. In the mid-1930s, the laboratory began investigating the sulfonamide drugs which had been recently shown to be highly effective against pneumonia. In the late 1930s, members of the lab began to investigate the antibiotic effects of certain soil bacilli.