|
Jacob Robbins
National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, USA Redundancy in biological systems becomes more and more evident as scientific data accumulate. It can involve individual nucleotides or amino acids in DNA or proteins, as well as biological processes at various levels of complexity. In an organism, redundancy can have survival value. From a historical perspective, the failure to recognize that redundancy is possible can interrupt progress in scientific investigation. This happened in work on the cellular uptake of thyroid hormone. When early experiments failed to show a specific entry mechanism for thyroxine into cells, it was concluded that this relatively nonpolar hormone molecule could easily traverse the limiting membrane of the cell by diffusion. The same was assumed for triiodothyronine, and for many years the scientific community remained comfortable with this conclusion, but in the 1970s new experiments by Rao and colleagues destroyed this state of complacency. The other major center where this revolutionary advance took place was in Georg Hennemann's laboratory. Rao's group, using suspensions of freshly isolated rat hepatocytes, was able to demonstrate energy dependent uptake of T3 but not T4. Hennemann and his colleagues began their work with primary monolayer cultures of hepatocytes and they embarked on a long series of experiments which set the standard in the field. They first had to recognize that there was more than one component accounting for the accumulation of hormone. One was nonsaturable up to micromolar levels and could represent nonspecific adherence to the cell surface as well as passive diffusion into the cell interior. When this was subtracted from the total uptake it became possible to define a saturable, relatively high affinity component (its Km was 61nM) that was stereospecific and dependent on cellular ATP, glucose, Na+-K+ ATPase, and extracellular sodium. This led to similar studies in several other laboratories, including my own, using a variety of experimental systems and various types of cells. Although not everyone reached the same conclusion - for example, some described facilitated transport that was not energy dependent - the consensus that has developed is that most if not all types of cells possess a specific entry mechanism for T3, that cells differ in whether they have a similar mechanism for T4 and, if so, whether the two hormones share the same transport system, that even for T3 there may be more than one specific transport mechanism. My own view is that even a single cell type often has more than one way to accumulate thyroid hormone in what might be called a specific manner. The reviews listed below document the literature on this subject. It is significant that essentially all of the major reviews that one can find emanate from Georg Hennemann or his closest collaborators, Roelof Docter and Eric Krenning. In this article I will touch briefly on the following questions: 1. Does thyroid hormone (TH) cross the plasma membrane against a concentration gradient? 2. How does TH get across? 3. Are there receptors for TH on the plasma membrane? 4. Where does TH go once inside the cell? 5. Does interference with a specific entry route affect TH metabolism or action? 6. Are multiple entry routes a feature of the well known thyroid homeostasis? It can readily be seen that many of these questions are interrelated, which does not make for a very orderly discussion. The first question asks whether thyroid hormone transport meets the criterion for active transport. We know that thyroid hormone in the extracellular milieu is mostly bound to carrier proteins, but the free hormone hypothesis, which is now well established, tells us that it is the free hormone concentration outside the cell that is the important parameter. We also know that the hormone inside the cell is bound to proteins and perhaps to membranes. Oppenheimer and colleagues have compared the free T3 concentrations in rat liver cytosol and plasma in vivo, and in GH1 pituitary cell cytosol and culture medium in vitro. They found that the intracellular concentration exceeded the extracellular by 2-3 fold. In my laboratory we studied this question in myeloblast and neuroblastoma cells by comparing nuclear T3 binding in the intact cell and in isolated nuclei. The affinity constant (Ka) for the nuclear receptor in intact cells, calculated on the basis of free T3 concentration in the medium, exceeded the Ka in isolated nuclei by 3 to 6 fold. This suggests that the cytosolic free T3 is higher than in the medium. D-T3, which is not specifically transported into the cells, did not show a difference in Ka. Thus at least one entry path for T3 appears to be by active transport - accumulation against a concentration gradient. Furthermore, any simple diffusion through the plasma membrane should favor exit of T3 rather than entry if the concentration is higher inside the cell. It may, however, be possible for facilitated passive transport into the cell, which depends on a transport protein, to coexist with an active transport pathway. The second and third questions concern the mechanism by which the hormone gains entry into the cell. Here again there may be more than one specific entry process to consider. One that has been clearly shown by direct observation is receptor mediated endocytosis. Using rhodamine labeled T3 and video- enhanced fluorescence microscopy, Cheng and colleagues demonstrated on the plasma membrane of mouse fibroblasts that the hormone is collected, together with unrelated proteins, in clathrin coated pits and then enters the cell in coated vesicles. Endocytosis inhibitors, such as dansylcadaverine, were later shown to interfere with T3 entry into several other cell types. Presumably the process is initiated by the binding of hormone to a membrane protein and there may well be several such proteins with different functions once entering the cell. They would include, for example, the presumptive p29 subunit of type II deiodinase which Leonard and colleagues have shown to bind T4 and enter astrocytes by endocytosis. Also included is the fibroblast LDL receptor which internalizes T4 that is bound to the apolipoprotein B of plasma LDL, as shown in my laboratory by Benvenga. In Georg Hennemann's laboratory, there has been an intensive, ongoing search for a different kind of TH receptor in the cell membrane, a carrier protein responsible for active transport or, perhaps, facilitated diffusion. This would be an integral membrane protein that transposes the hormone from one side of the membrane to the other - a so called transporter such as exists for other small molecules. The work of Lakshmanan in my laboratory showing that system L amino acids, especially phenylalanine, inhibit thyroid hormone accumulation by neuronal cells, may indicate a relationship between the hormone and the amino acid transporters. Hennemann's group prepared monoclonal antibodies to a 52kDa membrane protein in hepatocytes that inhibited TH uptake and its intracellular deiodination. A very recent achievement in their search is the induction of sodium dependent T4 and T3 transport into frog oocytes by injection of rat liver mRNA. The phenylalanine transporter, which is sodium independent, has not yet been cloned. I expect that this work will lead eventually to the characterization of one or more thyroid hormone transporters and to open up a field with important biological and medical possibilities. I have already alluded to the fourth question which is concerned with the destination of the internalized hormone and the possibility that different modes of entry might lead to differences in targeting to intracellular locations. One that has been studied in greatest detail, by Leonard and colleagues, is the p29 protein in astrocytes. T4 binding accelerates endocytosis of the protein and redirects it to endosomes instead of lysosomes. Eventually p29 is recycled to the plasma membrane, presumably no longer carrying the hormone. The T4 undergoes 5' deiodination, mainly by the type II deiodinase as in other cells of the CNS. This deiodination, which produces the more active hormone, T3, proceeds at a faster rate when T4 is diminished as a result of inactivation or degradation of the deiodinase when the cell is exposed to T4. This could be the way that cells in the nervous system regulate the intracellular production of T3 from T4, even while they accumulate T3 from plasma by another entry pathway. Whether T4 or T3 transported by any of the entry mechanisms equilibrates with the cytosolic hormone pool or is sequestered in a cytoplasmic organelle remains an open question. Another destination for T3 in the cell is the nucleus where it binds to receptors that activate transcription of DNA, ultimately leading to synthesis of the proteins that are responsible for many or most of the actions attributed to thyroid hormone. What is the route that plasma T3 takes to reach the nucleus? We don't know this in much detail, but we can say that active or specific T3 entry through the plasma membrane is an important first requirement. This is based on a number of experiments done in my own and several other laboratories in which various types of cells were allowed to accumulate T3 in the presence or absence of inhibitors of energy metabolism, endocytosis, or sodium transport. Nuclei were then isolated and it was shown that nuclear T3 uptake was reduced by 25 to 85%, and was often more affected than total cell uptake. It is not possible to decipher from these data which of the several specific transport pathways is responsible for nuclear uptake. Perhaps all of them contribute. In neuronal cells, T3 derived from accumulated T4 must also contribute. In any case, it seems that specific transport of either T3 or T4 across the plasma membrane increases the efficiency of hormone uptake by nuclear receptors. This discussion of nuclear uptake provides an answer to the fifth question which concerns the role of specific cell entry pathways in hormone action. The still controversial area of thyroid hormone actions not involving nuclear receptors could also depend on specific cell entry pathways provided that these actions take place inside the cell and not on the surface. Several laboratories have identified intracellular enzymes that take up thyroid hormone from the extracellular medium, but their role in the hormone's effects has not been clarified. Even though some or all of these may prove to be significant in hormone action, it is of interest to consider the work of Weisiger with hepatocytes. He has put forth the argument that low affinity binding to a reservoir of intracellular proteins, rather than trapping a ligand, is actually a mechanism for stabilizing and regulating the delivery system that enables the ligand to reach its ultimate destination in a timely manner. This resembles the passive role of the carrier proteins in the circulation and may well be important for thyroid hormone passage through the cytoplasm. Intracellular metabolism of thyroid hormone, in addition to providing T3 from T4, is important for the disposal of hormone from the body. It includes conjugation reactions as well as deiodination. Hennemann and colleagues have shown by the use of membrane transport inhibitors that these reactions also are dependent in large part on the passage of hormone into the cell interior through specific entry pathways. Finally I return to the redundancy issue and its role in thyroid physiology. A central feature of the thyroid system is that it is designed to assure a steady supply of active thyroid hormone to its sites of action. This is manifested at many levels, from the response of the pituitary thyroid axis in nutritional iodine deficiency, to the effect of insufficient T4 secretion on the rate at which T3 is produced from T4 in the brain. Does the presence of multiple entry pathways for thyroid hormone into cells contribute to thyroid homeostasis? Probably it does help to have more than one way to penetrate the cell membrane. Nevertheless, Hennemann has described a patient with a rare type of thyroid hormone resistance in whom the only apparent defect was a decrease in the entry of T4 into the rapidly equilibrating pool in the liver, accounting for a decrease in T3 production. Assuming that only one active entry pathway was affected, this indicates that the other pathways could not meet the hormonal needs of the patient. If this is the case, we would like to know which of the several entry pathways is so important for thyroid homeostasis. Of even greater interest is the possibility that some of these pathways could have specific functional roles. This may be the result of their delivering the hormone to particular sites within the cell, or their being subject to particular regulatory mechanisms. The outcome could be a unique hormonal action or a particular effect on the hormone's metabolism. In the foregoing discussion there are suggestions that such phenomena may indeed occur. Continued investigation in this direction is almost certain to be rewarding. |
