Some machines, like the pH meter discussed in last month’s issue, we primarily remember for their unit of measurement—that wondrous shorthand that is the culmination of arduous theory and experimentation. The history of science is peppered with such units brought on by great minds and new technologies, and it is often that their theoretical or their practical pioneers receive the honor of having these units named after them. Celsius, Ampere, Dalton, and Svedberg are but a few such names etched into the brains of practically any biomedical bench scientist. Yet while their names live on in text and in lab notebooks, there remains an encyclopedia of names that are defined less by a new “how” to measure than by a related and incessant doubt, “How do I know I’m measuring what I want to measure?”

Figure 1, Craig, L.C. (1944) J. Biol. Chem. 155, 519-534

The problem of separating and isolating specific substances from a mixture has been a central headache for chemistry since its modern start on Lavoisier’s lab bench. For later nineteenth and twentieth century chemists inclined to think of biological molecules, this headache often turned into a migraine. For if an organism is composed of thousands of different nucleic acids, lipids, proteins and peptides, sugars, minerals, salts, and soluble ions, how is it even possible to conceive of isolating a pure and single type of molecule from such an extraordinarily complex mixture? (This seems one of those rare problems whose difficulty isn’t easier to grasp with hindsight.)

More pressing concerns during WWII, however, propelled one chemist to make such a seemingly impossible task for small molecules a reality. In 1943, Rockefeller chemist Lyman C. Craig published a single-author paper outlining a method for the separation of complex chemical mixtures¹. If an unknown mixture of interest were mixed in two known immiscible solvents, compounds in the mixture could be purified on the basis of how well they partitioned into one solvent or the other upon separating. If this mixing then separating were repeated sequentially, one could observe the unique distribution of each pure compound among the fractions, even if the compounds were highly related. Craig applied these techniques—later called Countercurrent Distribution (CCD)—to mixtures of the anti-malarial drug quinicrine (atabrine), which was then in use by the US Army in the Pacific. Using CCD, Craig was able to isolate microgram amounts of quinicrine from blood and urine samples of treated patients, and was, as a result, able to inform clinicians of the pharmacological profile of the drug for the first time.

That, of course, was only the beginning. From the late 1940s through the 1960s, Craig applied CCD techniques to purify and to analyze many other useful compounds, from the antibiotics gramicidin, tyrocidine, bacitracin, and various penicillins, to fatty and bile acids, to insulin and other hormones. Two of his famed CCD machines with which much of this work was done are on display in the museum in Caspary Hall. The original, a stainless steel cylinder with counter-rotating drums (Accession no. ADDHERE), was built by Craig and his technician Otto Post, and allowed for twenty mixing and extraction cycles in a single run². Later models (Accession no. ADDHERE) relied on intricate glass separation funnels that could be rocked such that a mixture could proceed through up to 1000 separation cycles in a single run! Simple in theory, these machines were quite elaborate and technically impressive, and what is most striking is their apparent dynamism: they were meant to move. To see them operate must’ve induced a combination of awe and excitement as such wizardry, it is little wonder that the Craig lab on the sixth floor of Flexner historically had no shortage of interested postdocs and students.

In the end, you might be wondering why this technology didn’t survive. Around the time Craig published his paper on what is essentially an extraction scheme from two liquid phases, a British pair of scientists proposed doing something similar though slightly different by immobilizing one liquid phase on a gel³. Dubbed “partition chromatography” by their inventors A. J. P. Martin and R. L. M. Synge, this advance marked the beginning of modern chromatographic methods of separation (affinity, thin layer, high-pressure, etc.) and ultimately proved easier and more effective than Craig’s labor-intensive technique. Martin and Synge would go on to win the 1952 Nobel Prize in Chemistry for their invention, an honor that, at first glance, makes Craig’s 1963 Lasker Award appear prosaic. But this is not the case. As Stanford Moore wrote in Craig’s biography for the National Academy of Sciences, “Craig always kept in mind the principle that methods are a means to an end and not an end in themselves.” For Craig, proper purification was just a starting step for further analyses, a theme evident in his bibliography of 300 or so papers. Whether it was an important chemical structure or a pharmacokinetic study of drugs or hormones, Craig’s energy remained focused on biological problems that could be addressed with the methods of chemistry. And when no methods were available, he was fearless in designing newer, and seemingly magical, means of separation.

References
1) Craig, LC (1943) J. Biol. Chem. 150:33-45.
2) Craig, LC (1944) J. Biol. Chem. 155, 519-34.
3) Martin, AJP and RLM Synge (1941) Biochem. J. 25(1-2):91-121.