Abstract
In physical science the first essential step in the direction of learning any subject is to find principles of numerical reckoning and practicable methods for measuring some quality connected with it. I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced to the state of Science, whatever the matter may be. Sir William Thomson (Lord Kelvin), Popular Lectures and Addresses, vol. 1, 2nd ed. “Electrical Units of Measurement”, a lecture delivered at the Institution of Civil Engineers, 3 May 1883; London: Macmillan and Co., 1891, pp.80–81.
Ed.: a donkey, a fool
planned, high-quality research projects that address important questions and use rigorous experimental approaches to establish and validate important findings. Indeed, when experimental approaches or data appeared to be based on assumptions rather than facts, Lou would sometimes quote the saying, ‘don’t assume, it makes an ass 1 of you and me’. Louis Sokoloff had a very high and lasting impact on those who worked in his laboratory as well as those with whom he interacted at scientific meetings and other occasions. Lou died on 30 July 2015 in Washington, DC at age 93, but his spirit continues to live and grow within scientists who were influenced by his training and collaborations, by discussions with him, and by his publications. Lou’s legacy includes his unique achievements that focused on basic research and transformed neuroscience and also led to clinical applications that benefitted many patients.
Louis Sokoloff was born on 14 October 1921, in Philadelphia, PA, and earned his BA and MD degrees from University of Pennsylvania in 1943 and 1946, respectively. His internship was at Philadelphia General Hospital from 1946 to 1947, and in 1947, Lou and Betty Kaiser were married. In 1948, Lou embarked on his scientific career as a fellow in the laboratory of Seymour Kety where he became involved in studies of the cerebral circulation. These studies initiated his life-long quest to understand relationships between changes in rates of blood flow and glucose supply and demand with functional activity of brain cells. Early studies used the N2O method to measure blood flow to the brain as a
whole, and, when coupled with determination of arteriovenous differences for glucose and oxygen, global metabolic rates could be determined under various conditions. The limitations of this method were recognized after studies demonstrated that decrements in flow and metabolism were readily detected but increases were small or negligible. Kety developed the principles of exchange of inert gas between blood and tissues and derived an equation that led to an approach to measure local rates of blood flow. During the span from 1953 to 1955, the team developed the use of a radioactive gas, [131I]trifluoroiodomethane, along with quantitative autoradiography to measure and visualize simultaneously local rates of blood flow throughout the entire brain. Stimulation of blood flow in visual cortex by presenting light flashes to the retina provided the first example of imaging functional activity in brain of awake subjects. Development of quantitative autoradiography by use of calibrated radioactive standards that were included with brain samples during exposure to the X-ray film was also a major, but generally unrecognized, technical advance that was the basis for subsequent quantitative autoradiographic studies using 14C- and 3H-labeled tracers. Blood flow assays were subsequently improved by using a non-volatile compound, [14C]antipyrine, followed by the more freely diffusible [14C]iodoantipyrine. Blood flow assays must be short to prevent back flux of tracer to blood, and therefore, they are often used to examine functional responses to brief stimuli or at intervals during extended stimulation.
In the mid-1950’s, Lou was also interested in measuring local metabolic rates, with the rationale that metabolism would be more directly linked to changes in energy demand arising from shifts in cellular activity. He, therefore, worked on devising a model using [14C]glucose but recognized that experimental periods must be short to minimize loss of labeled products causing quantification of product accumulation to be vulnerable to various errors; the project was abandoned. Shortly thereafter, he learned about studies by others showing that 2-deoxyglucose (DG) was a substrate for hexokinase and the DG-6-P was not further metabolized via glycolysis; he kept this in mind for the future. In 1968, Lou spent a sabbatical year in France and developed expertise in enzyme kinetics. Upon his return to NIH, he developed a metabolic model for competition of [14C]DG with glucose for transport and metabolism. Initial studies were carried out in the early 1970s, and the detailed description of the method, published in 1977, became a citation classic. Studies using auditory, visual, and motor system activation clearly showed that metabolic activation could be visualized and quantified; subsequent technical advances included digitizing the autoradiographs and color coding metabolic rates. The DG method was quickly extended for use in humans by developing the use of [18F]fluorodeoxyglucose (FDG) and positron emission tomography. These methods stimulated the field of brain imaging and led to an explosion of studies in animals and humans.
Development of the DG method carefully took into account the possibility that glucose-6-phosphatase activity could convert DG-6-P back to DG, which would then be lost from brain and cause underestimation of glucose utilization rate. Biochemical assays showed that the enzyme activity in brain was very low and varying the duration of the DG assay period between 30 and 60 min showed no change in calculated rate that would fall with increased duration of the experimental period if there were significant phosphatase activity. Nevertheless, reports from other laboratories generated a prolonged controversy that took more than a decade to resolve. Because Lou Sokoloff took personal responsibility for the validity of all of the studies that used the DG method, his laboratory replicated the findings and then elucidated the sources and consequences of reported high phosphatase activity. The bottom line is that the claims of high phosphatase activity were artifacts arising from incorrect assumptions in metabolic modeling of precursor–product relationships and methodological flaws arising from impurities in isolated glucose fraction, unrecognized loss of glucose due to chemical reactions in the test tube during its purification, and quantitative copurification of the major contaminant. These detailed analyses also established the impact of tissue heterogeneity (i.e., regions of interest comprised of gray and white matter) on calculated values and revealed that DG-6-P is further metabolized to other phosphorylated compounds, including DG-1-P, that is acid labile and is converted back to DG in the test tube during the routine perchloric acid extraction procedure. Loss of DG-1-P from the apparent product pool contributed to artifacts of precursor–product modeling and may underlie the slow loss of label products in tissue (after the specified 45–60 min experimental period) if it enters intracellular acidic compartments. Use of an ethanol-based extraction procedure protected DG-1-P and enabled determination of the relationships between the steady state brain: plasma distribution spaces for glucose, DG, and methylglucose. Methylglucose could then be used to determine local glucose concentrations in brain so that the correct value for the lumped constant (the correction factor for differences in transport and phosphorylation of DG and glucose that is sensitive to hypoglycemic conditions) could be used for glucose utilization rate calculation. The ethanol extraction procedure was also essential for revealing the presence of quite high glycogen levels in brain; glycogen content is too low in acid extracts and is sensitive to animal handling. Glucose-6-phosphatase activity was also assayed in cultured brain cells and demonstrated to be negligible. To sum up, the phosphatase issue was a diversion from the progress of the laboratory, but careful, detailed studies clearly proved that it had no meaningful impact on the DG method when carried out as designed. In addition, these studies opened up new and unexpected areas of investigation, as is often the case in research.
The history of Lou Sokoloff’s life, work, and details of his interactions with colleagues are available in Lou’s autobiography published in The History of Neuroscience in Autobiography, volume 1, edited by Larry R. Squire, Society for Neuroscience, 1996, pp. 454–497 (free download at http://www.sfn.org/About/History-of-Neuroscience/Autobiographical-Chapters). Videos can be viewed at http://www.sfn.org/About/History-of-Neuroscience/Autobiographical-Videos-of-Prominent-Neuroscientists/Louis-Sokoloff or at http://wn.com/louis_sokoloff, and also at http://iscbfm.org/Meetings/Hightlights/Prof--Sokoloff-s-video.aspx.
Brain mapping studies using the blood flow, [14C]DG and [18F]FDG methods were widely acclaimed, and Lou received a number of honors, including the F. O. Schmitt award in 1980, election to the National Academy of Sciences, USA (1980), Albert Lasker Clinical Medical Research Award (1981), the Karl Lashley Award (American Philosophical Society, 1987), and in 1988, Seymour Kety and Louis Sokoloff received the first National Academy of Sciences award in the Neurosciences. Lou served as Editor-in-Chief of the Journal of Neurochemistry (1974–1977), president of the American Society for Neurochemistry (1977–1979), president of the Association for Research in Nervous and Mental Disease (1983), and founding member and first president (1981–1983) of the International Society for Cerebral Blood Flow and Metabolism. The history of the Society was reviewed by Paulson, Kanno, Reivich, and Sokoloff in Journal of Cerebral Blood Flow & Metabolism (2012) 32, 1099–1106.
Lou and Betty Sokoloff loved French cuisine and wine, and they were generous hosts to many scientists who visited and worked in his laboratory. The international community has lost a friend, colleague, and first-rate scientist, but his impact on brain mapping and imaging functional activity via quantitative changes in blood flow and glucose utilization lives on. We can honor Lou’s achievements and his memory by applying his scientific values to our work: detailed knowledge of the past literature, choice of important problems, careful experimental design, thoughtful analysis of data, and interpretation of the results from an integrative perspective. Luck can provide a nourishing environment and essential interactions, but excellence does not occur by chance. Curiosity, perseverance, attention to detail, and dedication to discovery are critical elements that led to development of quantitative imaging of blood flow and metabolism and brain mapping. We miss a founding father of our discipline but live with his spirit.