Neurochem. Int. Vol. 22, No. 3, pp. 293-300, 1993 0197-0186/93 $6.00+0.00 Printed in Great Britain.All rights reserved Copyright © 1993PergamonPress Ltd FREE CHOLINE AND CHOLINE METABOLITES IN RAT BRAIN AND BODY FLUIDS: SENSITIVE DETERMINATION AND IMPLICATIONS FOR CHOLINE SUPPLY TO THE BRAIN JOCHEN KLEIN,* RONALD GONZALEZ, t ANDREA KOPPEN and KONRAD LOFFELHOLZ Department of Pharmacology, University of Mainz, Obere Zahlbacher Str. 67, W-6500 Mainz, Germany (Received 25 June 1992 ; accepted 14 August 1992) Al~traet--In the central nervous system, choline is an essential precursor of choline-containing phos- pholipids in neurons and glial cells and of acetylcholine in cholinergic neurons. In order to study choline transport and metabolism in the brain, we developed a comprehensive methodical procedure for the analysis of choline and its major metabolites which involves a separation step, selective hydrolysis and subsequent determination of free choline by HPLC and electrochemical detection. In the present paper, we report the levels of choline, acetylcholine, phosphocholine, glycerophosphocholine and choline-containing phospholipids in brain tissue, cerebrospinal fluid and blood plasma of the untreated rat. The levels of free choline in blood plasma (11.4 #M), CSF (6.7 #M) and brain intracellular space (64.0 pM) were sufficiently similar to be compatible with an exchange of choline between these compartments. In contrast, the intracellular levels of glycerophosphocholine (1.15 mM) and phosphocholine (0.59 mM) in the brain were considerably higher than their CSF concentrations of 2.83 and 1.70/~M, respectively. In blood plasma, glycerophosphocholine was present in a concentration of 4.58 #M while phosphocholine levels were very low or absent (< 0.1 #M). The levels of phosphatidylcholine and lyso-phosphatidylcholine were high in blood plasma (1267 and 268 #M) but very low in cerebrospinal fluid (< 10 pM). We concluded that the transport of free choline is the only likely mechanism which contributes to the supply of choline to the brain under physiological conditions. Choline is an important nutrient in man (Zeisel, 1981 ; Zeisel et al., 1991) and serves as a precursor of both acetylcholine and major choline-containing mem- brane phospholipids such as phosphatidylcholine and sphingomyelin. The current interest in the choline metabolism of the central nervous system stems largely from the fact that important diseases such as the senile dementia of the Alzheimer type are associated with cellular degeneration and central cholinergic dys- functions (Bartus et al., 1982; Coyle et al., 1983). Therefore, the analysis of choline metabolism in the central nervous system is of major importance due to the close interrelationship of phospholipid metab- *To whom correspondence should be addressed. tRecipient of a DAAD fellowship; on leave from the Dept. of Pharmacology, School of Medicine, Universidad de Costa Rica, San Jose, Costa Rica. ~Abbreviations: CSF, cerebrospinal fluid; GPCh, glycero- phosphocholine; (lyso-) PtdCh, (lyso-) phosphatidyl- choline. olism, choline availability and the synthesis of acetyl- choline (Blusztajn and Wurtman, 1983 ; Tucek, 1988 ; Klein et al., 1992). Therapeutic approaches for senile dementia utilizing choline or one of its precur- sors have been tested or are under investigation (Hollander et al., 1986; Kumar and Calache, 1991). As the brain synthesizes choline at very low rates (Bremer and Greenberg, 1961; Crews et al., 1981), the supply of choline to the brain must predominantly occur via transport from the arterial blood. Several studies reported on the transport of free choline from arterial blood into the brain (Diamond, 1971 ; Dross and Kewitz, 1972 ; Freeman et al., 1975 ; Cornford et al., 1978). In addition, some authors claimed that a transport of lipid-bound choline at the blood-brain barrier may contribute to the maintenance of brain choline homeostasis (Illingworth and Portman, 1972 ; Mann and Bennett, 1979). The physiological rel- evance of the transfer of lipid-bound choline was, however, disputed (Pardridge et al., 1979). To our knowledge, no data have been reported on the poss- 293