THE IODIDE SPACE IN RABBIT BRAIN Thesis by Nawal Abd El-Hay Ahmed In Partial Fulfillment of the Requirements For the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 19158 (Submitted May 21, 19158) 11 To my husband and my daughter iii ACKNOWLEOOMENTS It has been a profitable experience to have been a graduate student in the Caltech Biology Division. I am grateful to my advisor, Dr. Van Harreveld, for his in valuable help during the course of my graduate work and unselfish ccintribution of time, for his understanding, wisdom and kindness. I am proud to have been his student. I also take this opportunity to thank Mrs. Ruth Estey for her assistance, Plotience and excellence in the typing of the drafts of this thesis. Special thanks to Drs. Justine Garvey and D. H. Campbell for the use of their counter during the course of. this work. iv ABSTRACT In the present investigation labeled iodide was used to investi gate the interrelationship between brain, blood and cerebrospinal fluid, to examine active transport across the blood-brain- and the blood cerebrospinal fluid barriers, and to estimate the extracellular sp:l.ce of the brain. The iodide sp:i.ce in the brain and the iodide concentration in cerebrospinal fluid after intravenous administration of radioactive iodide are determined by the following mechanisms. Iodide p:i.sses into the cerebrospinal fluid but active transport in the choroid plexus moves most of the iodide back again into the plasma, keeping the concentration at a very law value. An extracellular fluid is formed at the blood-brain barrier possibly in a similar way. The iodide concentration of this fluid is unknown but is probably higher than that in the cerebrospinal fluid. Diffusion of iodide across the brain-cerebrospinal fluid barrier transports this ion from the brain into the cerebrospinal fluid which is constantly renewed •\=:link action". The iodide sp:i.ce was found to be 2.4% four to five hours after the intravenous administration of l3lI, the iodide content of the cerebro spinal fluid vra.s 1.2% of that of the TCA serum filtrate. The iodide sp:l.ce increased to 10.6% in prep:l.rations in which in addition to l31r unlabeled iodide (to a serum concentration of 25 to 50 nM) vra.s adminis tered to saturate the active transport processes in the choroid plexuses and blood-brain barrier. The iodide activity of the cerebrospinal fluid v in these experiments increased to 29.3% of that in the TCA serum fil trate. In experiments in which the inhibitor of iodide transport, perchlorate (8 mM), was injected intravenously with the l31:r-, the iodide space was 8.2% and the iodide concentration in the cerebro spinal fluid 26.4%. These experiments demonstrate the effect of saturation and inhibition of active transport on the iodide space. They show furthermore that the depression of the active transport did not raise the iodide concentration in the cerebrospinal fluid to the plasma concentration. The relatively low (l/3 of that in the serum TCA filtrate) iodide concentration in the cerebrospinal fluid under these circumstances was ascribed to a differential permeability of the blood-cerebrospinal fluid barrier for iodide and chloride. The sink action can be eliminated by perfusion of the ventricles with an artificial cerebrospinal fluid containing iodide. Ventriculo cisternal perfusion with l31:r- alone resulted in an iodide space of 7 .2% after 4.5 hours. An iodide space of 10.2% was determined by a canbined intravenous administration and yentricular perfusion with an artificial cerebrospinal fluid containing the same concentration of l31:r as present in the plasma. When in similar experiments perchlorate was administered intravenously, the iodide sp:i.ce rose to 16.8%. The iodide sp:i.ce determined by simultaneous intravenous injection and ventricular perfusion with both labeled and unlabeled iodide, in a 20.8%. concentration sufficient to saturate the active transport, was In the latter instances the sink action is eliminated and also active transport is inhibited or saturated. It was postulated that under vi these conditions the iodide concentration in plasma and brain extra cellular fluid are approximately the same. The use of the iodide space as a measure of the brain extracellular space was discussed. 1 INTRODUCTION Determination of extracellular spa.ce. In most tissues the extracellular s{Bce can be determined by the intravenous administration of a compound which does not readily penetrate the cell membrane but {Bases quickly through the capillary wall. After allowing time for equilibration of the compound in plasma and extracellular material, the extracellular space is estimated by determining the concentrations of the compound in both serum and tissue. The most common compounds used for this purpose are inulin, sucrose, thiocyanate and radioactive tracers such as thiocyanate "Illd iodideo The extracellular space is computed in the following way, assuming that the electrolyte composition in the extracellular fluid and the plasma are the same and that the marker does not penetrate into the intracellular compartment. = concentration of the compound in tissue Extracellular S{Bce x 100 concentration of the com~po und in plasma 24 A Na s{Bce of 11% was found in this way in rabbit muscle (Hahn, Hevesy and Rebbe, 1939). With inulin an extracellular S{BCe of 9.6% was estimated in rat muscle by Creese, d'Silva and Hashish (1955), of 10.0% by Law and Phelps (1966). The naturally present chloride and sodium have in the {Bst been used to determine the extracellular S{Bce on the (erroneous) assumption that these ions are present exclusively in the extracellular com{Brtment. Since a certain percentage of these ions are present in the cellular elements such values are too high. 2 Wilde (1945) determined a chloride sp:i.ce of 17% for rat muscles in this way. The sulfate sp:i.ce determined with labeled so was found to 4 be I€J to 80% of the chloride space (Walser, Seldin and Grollman, 1954). The inulin sp:i.ce was 80% of the chloride sp:i.ce after 1 to 2 hours of equilibration (Cotlove, 1954). The application of these methods to the central nervous system is difficult and uncertain because of the presence of a barrier between blood and brain (the blood-brain barrier) which prevents the quick penetration of the marker compounds into the nervous tissue. Goldmann (1909) showed that the intravenous infusion of the dye, trypan blue, did not result in staining of the central nervous tissue. However, when thi~ dye was introduced directly into the ventricles of the brain (i.e., into the cerebrospinal fluid system), the tissue was easily stained. Later it was found that many other compounds are not readily transported from the blood into the brain tissue. This supported the idea of the presence of a barrier between blood and brain. The blood-brain barrier isolates the brain and prevents rapid clia.pges in the surrounding of the nerve cells. Although penetration of many compounds through the blood- brain barrier is slow, it occurs at a measurable rate. It was believed, therefore, that if sufficient time is allowed, an equality of marker compound concentrations in plasma and extracellular material of the brain could be achieved, making it possible to estimate the extracellular sp;i.ce in brain tissue. Based on this concept Woodbury (1958) determined a so space 4 of 3.9% in rat brain by using labeled sulfate. After 4 hours, this 3 value was reached, which remained at the same level for 16 hours but then increased. He also found an inulin space of about 4 to 5°/o after 16 hours e~uilibration. Barlow, Domek, Goldberg and Roth (1961) de termined a 35so space between 2 to 5.5% in cats after maintaining a 4 steady level of labeled sulfate in the plasma for 4 to 8 hours. Morrison (1959) found an inulin space of less than 5% in dogs and rats after 3 hours. Reed and Woodbury (1963) estimated a 6% iodide space in rats 4 hours after intravenous injection. Streicher (1961) deter mined a thiocyanate space of 4 to 17% in rat brain after 1/2 to 2 hours. The larger values were roughly correlated with a higher thiocyanate concentration in the plasma. The results of these investigations, in which the marker compounds were administered intravenously led to the acceptance of a small (3 to 5%) extracellular space in central nervous tissue. Composition of cerebrospinal fluid. The cerebrospinal fluid was at one time considered as a plasma ultrafiltrate (Holmes and Tower, 1955). The solvents in plasma and cerebrospinal fluid would be distributed in accordance with a Gibbs- Donnan equilibrium. The physical requirements of a typical Gibbs-Donnan system are, first, that a membrane, separating two electrolyte solutions, is per meable for all ion species but one, and secondly, that all water move ments across the membrane are prevented by a hydrostatic pressure difference between the compartments. Such a system can be scheroatized 4 as in Fig. lA in "Which it is assumed that a membrane separates the two compartments I and II, and that the diffusable ions are Na+ and Cl-. The indiffusable ion is indicated by An-. It is assumed that the two compartments are of equal and constant volume and that at the + start the Na concentration is the same in both compartments, and is equal .to the Cl- concentration in II and the An- concentration in I. The osmoconcentrations are thus the same in the two compartments. In Fig. l.B the concentrations of these ions are indicated by the heights of the columns in the initial state. Chloride ions will now start to diffuse across the membrane from II to I, and since equality of the positive and negative charges has to be maintained in each compartment, Na+ ions have to accompany the Cl- + ions. Thus the Na concentration becomes higher in I than in II. Then Na+, accompanied by Cl- , starts to diffuse back from compartment I into II. After some time the diffusion of NaCl from I into II equals that from II into I, and a Donnan equilibrium has been reached. The concen trations of the ions under these conditions are shown in Fig. lC. In compartment I, the Na+ concentration has increased, the An- concentra- tion is unchanged, and it now contains Cl-. In compartment II, the Na+ and Cl- concentrations are equal but decreased. The Gibbs-Donnan equilibrium demands that the products of the concentrations of the diffusable positive and negative ions, on the two sides of the membrane, are equal, i.e.