A Serendipitous Contamination

All cells receive messages via hormones, neurotransmitter molecules and growth factors. These molecules bind to protein receptors in the cell membrane and relay their information in the form of "second messengers" to the cell's interior. It has long been known that cells contain a vitamin-like substance called myoinositol. Our discovery 35 years ago that cell-surface receptor activation leads to increased metabolic turnover of an inositol-containing phospholipid, phosphatidylinositol (PI), was serendipitous. Had there not been phospholipid-derived contaminants in an "RNA" fraction I was studying, Mabel Hokin and I probably would never have come across this important phenomenon. It turns out that the "PI effect" involves the release of components of inositol phospholipids that serve as second messengers. The story begins in August 1949, when I arrived at H.A. Krebs' laboratory at the University of Sheffield, U.K., to work towards a biochemistry Ph.D. As was often his style, Krebs suggested that I formulate my own problem and work on it with little supervision. I was a rather independent fellow, far too independent given my limited experience in biochemistry, and I happily accepted his proposal. We therefore had little contact with each other, apart from brief exchanges in the hall before he could dart into a nearby lab. But his comments were always supportive, and when it came time for me to submit a manuscript, he would go over the draft meticulously with me. The first time he did this, he wrote only my name under the title and turned to me with a smile. At that time, great advances were being made in our understanding of metabolic pathways. There was much less interest in physiological aspects of biochemistry. My medical background put me in a good position to examine a problem that ultimately led to the discovery of the PI effect. I was interested in working out an in vitro system to study the mechanism of protein synthesis, and after failing with a stomach mucosa model, I turned to pancreas slices. I had considerable success with this system, and was able to show—apparently for the first time—both net synthesis and agonist-stimulated secretion of a protein (amylase) in vitro. This problem occupied the greater part of my thesis research. Some intriguing observations by T. Caspersson and J. Brachet in the 1940s showing a parallelism between protein synthetic rate and the RNA content of a variety of cells suggested that protein synthesis involved RNA. Several months before I left England, two Russians, M.A. Guberniev and L.I. Il'ina, reported that when secretion in salivary glands, stomach and pancreas was stimulated by injecting into dogs congeners of the important neurotransmitter, acetylcholine, there were increases of up to tenfold in the incorporation of 32P into "RNA." I confirmed their observations in pancreas slices, but found only about a twofold stimulation of 32P incorporation. Before completing my doctoral studies, I began to suspect that the stimulation of 32P incorporation into "RNA" might be due to contamination with phospholipid hydrolytic products as a result of the alkaline hydrolytic step used for isolating ribonucleotides of RNA. My suspicions were aroused by my detecting radioactivity with a Geiger counter in areas of the chromatogram not coinciding with the U.V.-absorbing ribonucleotide spots. These thoughts occurred to me only days before Mabel Hokin and I were to set sail for Halifax, Nova Scotia, in March 1952. I had time to incubate pancreas slices with 32P, but I did not have time to remove contaminating 32P from the phospholipids and to count their radioactivity. So I carried onto the ship a large rack of ethanol-ether extracts of the lipids, and when we arrived for our postdoctoral studies in J.H. Quastel's laboratory at McGill University, we completed the experiment. The lipid fractions from cholinergically stimulated tissue showed enormous increases in 32P-specific radioactivity, indicating increased phospholipid turnover. This work led to the publication of our first paper on the subject, in the Journal of Biological Chemistry in 1953. When methods became available a few years later for separating individual phospholipids from small samples of tissue, we identified the lipid showing increased phosphate turnover as PI. The same year our first paper appeared Watson and Crick published their Nature paper on the double helix structure of DNA. It is interesting how rapidly the nucleic acid story unfolded, whereas the PI story had to wait 30 more years. There were many reasons for this. Until the 1960s, virtually nothing was known about calcium's central role in mediating physiological responses. Bob Michell presented his revolutionary hypothesis of a link between receptor-activated PI turnover and cellular calcium mobilization in 1975. The PI problem couldn't really begin to be solved until around 1980, however, when an intracellular pool of "trigger calcium" was demonstrated by Irene Schulz and techniques were developed to allow highly impermeable putative second messenger inositol phosphates to enter cells. Clearly the PI effect was a discovery well before its time. By the time I left the field in the mid-'60s, we had published 48 papers on the PI effect or related subjects. My research interests gradually shifted to studying the membrane sodium-potassium pump, after we found a large PI effect in the salt gland of the albatross, which misled us into thinking the effect had something to do with sodium-potassium transport in that organ. The PI response in the salt gland eventually turned out to be related to calcium mobilization, as elsewhere. However, kinetic studies in this organ led to our formulation of the original version of the PI cycle. This has stood the test of time, but the current area of excitement—cell membrane-bound phosphatidylinositol mono- and bisphosphate and the latter's breakdown product, second messenger inositol 1,4,5-trisphosphate (IP3) and its metabolites—has now been incorporated into the cycle. Many laboratories, in particular those of Michael Berridge, Robin Irvine and Philip Majerus, have shown that IP3 and two related inositol phosphates transmit information from cell surface receptors to the interior. There they activate a variety of cellular responses, including mobilization of calcium from intracellular stores. Diacylglycerol was in our original PI cycle, but its second messenger function was not known until recently when Y. Nishizuka and co-workers demonstrated that it activates an important enzyme (protein kinase C) that phosphorylates certain target proteins and thus regulates additional cellular processes. Recently I returned to this field, studying the neurotransmitter-induced formation of a compound that closely resembles IP3. I doubt whether many investigators who started a field 35 years earlier often have the good fortune to do exciting things in the same area 35 years later. Such a happy circumstance is an excellent prescription for keeping scientifically young. Hokin is chairman of the Department of Pharmacology at the University of Wisconsin.

A Serendipitous Contamination

Interested in reading more?

Become a Member of