- Email: [email protected]
Free amino acids and related substances in human glial tumours and in fetal brain: comparison with normal adult brain
Brain Research, 117 (1976) 105-113
105
© Elsevier/North-Holland Biomedical Press, Amsterdam - Printed in The Netherlands
F R E E A M I N O ACIDS A N D R E L A T E D SUBSTANCES IN H U M A N G L I A L T U M O U R S A N D IN F E T A L BRAIN: C O M P A R I S O N W I T H N O R M A L A D U L T BRAIN
JEANNE-MARIE LEFAUCONN1ER, CHRISTIANE PORTEMER and FERNANDE CHATAGNER Laboratoire de Chimie Biologique, 96 Boulevard Raspail, 75006-Paris (France)
(Accepted March 30th, 1976)
SUMMARY An ion exchange automatic chromatographic analysis of the free amino acid concentrations of 18 human glial tumours and of 4 human fetal brains was carried out and the concentrations were compared to those of 13 biopsy specimens of normal adult brain. In addition, the concentrations of the amino acids of the glial tumours were compared to those of 7 intracerebral metastases of various origin. The chromatograms of several tumour specimens showed an unidentified peak overlapping proline. As far as the amino acid concentrations are concerned they varied depending upon the origin of the sample. The concentrations of most amino acids were higher in fetal brain than in adult brain with the exception of aspartic acid, glutamic acid, glutamine, cystathionine and GABA. Two peptides: glutathione and homocarnosine were absent in fetal brain and were present in adult brain. In glial tumours, homocarnosine and some amino acids, namely aspartic acid, glutamic acid and GABA, showed lower concentrations than in normal brain. Some amino acids were in the same concentration as in normal brain: taurine, phosphoethanolamine, glutamine and cystathionine. Most of the others were in higher concentrations than in normal brain, mainly proline. The results suggest that the concentrations of 5 compounds: taurine, proline, cystathionine, GABA and homocarnosine, taken as a whole, provide information on the origin of the sample.
INTRODUCTION The amino acid concentrations of normal adult human brain and of fetal human brain have already been reportedl,10,13,~5,16,18. In both cases the analysis reflects the
106 amino acid composition of an heterogeneous population of neurons and glial cells. It seemed therefore worthwhile to determine the amino acid content o f glial tumours, c o m p o s e d mainly o f cells o f glial origin. To our knowledge, however, only partial information is available in this connection 11. The work reported here is a comparative study o f amino acid concentrations determined in 18 glial tumours, 4 fetal brains and 13 normal adult brain biopsy specimens. In addition, we c o m p a r e d the amino acid concentrations o f the glial tumours with those of 7 metastases to determine whether the alterations, if any, found in glial turnouts were characteristic o f gliomas or due to malignancy. MATERIAL AND METHODS N o r m a l brain biopsy specimens Biopsy specimens of normal brain tissue were obtained at c r a n i o t o m y f r o m 13 adults undergoing neurosurgical operations for treatment o f vascular malformations or for ablation o f intracranial tumours. Table I lists the indications for operation and the site at which tissue was removed for biopsy. In every case normal tissue resection was necessary to gain access to a deep-seated lesion. The operation was undertaken under general anaesthesia (Thiopental, Phenoperidine and Pancuronium), with artificial respiration. These biopsy specimens were put into dry ice within less than 10 rain after removal from the brain. Tumour tissue specimens These were obtained from patients operated on under the conditions described above. The specimens consisted of: (I) T u m o u r s o f glial origin: 13 glioblastoma
TABLE
1
Sourees of biopsy specimens of normal brain tissue and corresponding cystathionine concentrations Patient
Age
Sex
Indication .[or operation
Site of biopsy
Cystathionine ( Imloles/g)
l 2 3
45 23 32
M M F
Temporal pole Frontal pole Frontal pole
0.59 0.61 0.32
4 5 6 7 8 9 10 11 12 13
33 22 47 47 53 69 55 59 26 63
M M F M F M M M M M
Middle Cerebral Art. Aneurysm Carotid Aneurysm Ant. Communicating Art. Aneurysm Arterio-venous fistula Meningioma Meningioma Meningioma Meningioma Meningioma Metastasis Glioma Glioma Acoustic nerve neurinoma
Frontal lobe Frontal pole Temporal lobe Temporal lobe Frontal pole Frontal lobe Parietal lobe Temporal lobe Temporal lobe Cerebellum
0.76 0.25 4.79 3.05 0.25 2.35 1.20 3.55 1.05 1.74
107 TABLE 1I
Amino acid content o f normal adult human brain tissue, fetal brain tissue, glioblastomas, and cerebral metastases The results are expressed in/~moles/g of wet weight and represent means 4- S.E.M. n -- number of samples. Student's t test was used to calculate the statistical differences. Fetal brain and glioblastomas are compared with adult brain. Metastases are compared with glioblastomas. Abbreviations: Ph.ethanolamine - phosphoethanolamine; Ox. Glutathione = oxidized glutathione; a-aminobut, acid ~-aminobutyric acid.
Taurine Ph.ethanolamine Aspartic acid Threonine Serine Glutamic acid Glutamine Proline Ox.Glutathione Glycine Alanine a aminobut, acid Valine (4 cystine) Methionine Cystathionine lsoleucine Leucine Tyrosine Phenylalanine GABA Ornithine Lysine Histidine Homocarnosine Arginine Total a. acids
Adult brain (n 13)
Fetal brain (n = 4)
Glioblastomas (n 13)
Metastases (n 7)
1.03 0.66 1.52 0.29 0.48 6.84 3.49 0.04 0.23 1.03 0.95 0.03 0.21 0.04 1.57 0.07 0.16 0.10 0.09 0.94 0.02 0.20 0.10 0.17 0.10 20.7
3.14 i 0.42*** 2.60 4- 0.25*** 0.65 4- 0.12"* 0.91 4- 0.11"** 1.10 ± 0.13"** 2.91 4- 0.17"** 1.06 4- 0.13"** 0.31 4- 0.10"* 0"* 1.24 ± 0.17 1.69 4- 0.17'* 0.06 ± 0.01 ** 0.50 4- 0.09** 0.07 4- 0.02 0.05 4- 0.03 0.24 4- 0.04*** 0.55 4- 0.07*** 0.19 4- 0.02 0.21 4- 0.03* 0.46 ± 0.12" 0.08 4- 0.01 0.43 4- 0.09* 0.17 ± 0.03 0 *** 0.32 ± 0.06*** 20.1 ± 1.8
1.20 0.60 0.76 0.73 0.76 4.22 3.77 0.45 0.36 3.09 1.92 0.06 0.37 0.07 1.65 0.12 0.27 0.14 0.15 0.54 0.11 0.46 0.15 0.06 0.17 22.3
3.08 1.66 1.20 0.44 0.75 4.15 1.10 0.85 0.10 2.44 1.71 0.06 0.56 0.07 0.15 0.21 0.52 0.21 0.25 0.07 0.07 0.50 0.18 0.01 0.38 20.1
± 0.09 4- 0.1 l 4- 0.14 4- 0.05 4- 0.04 4- 0.32 4- 0.28 3_ 0.03 4- 0.05 4- 0.14 4- 0.11 -I- 0.01 4- 0.04 4- 0.01 4- 0.40 4- 0.02 4- 0.02 4- 0.03 4- 0.02 4- 0.02 4- 0.01 ± 0.05 4- 0.03 ± 0.03 ± 0.03 ± 1.0
4- 0.10 -I- 0.13 4- 0.18" 4- 0.14" ± 0.15 4- 0.92* q- 0.56 4- 0.09*** 4- 0.11 ± 0.57** 4- 0.36* ± 0.01 ** ± 0.05* 4- 0.02 4- 0.34 4- 0.02* ± 0.04 ± 0.02 4- 0.02 4- 0.13" 4- 0.03 4- 0.10" ± 0.03 4- 0.03** ± 0.07 ± 2.4
4- 0.87* 4- 0.23*** 4- 0.27 ~ 0.10 4- 0.12 4- 0.26 4- 0.17" 4- 0.12"** 4- 0.08 3_ 0.25 4- 0.26 i 0.02 ~: 0.12 4- 0.03 ± 0.06** £ 0.09 4- 0.17 4- 0.05* ± 0.08 4- 0.03* 4- 0.02 4- 0.13 4- 0.03 ~ 0.01 ** 4- 0.17 5:1.8
* P < 0.05. ** P < 0.01. *** P < 0.001.
m u l t i f o r m e o f t h e c e r e b r a l h e m i s p h e r e s in a d u l t s , 2 g l i o b l a s t o m a m u l t i f o r m e o f t h e p o s t e r i o r f o s s a in c h i l d r e n , 1 r e c u r r e n t g l i o b l a s t o m a , 1 a s t r o c y t o m a , a n d 1 e p e n d y m o m a . (2) M e t a s t a s e s : 2 f r o m t h e s k i n ( m e l a n o m a ) , 1 f r o m t h e l u n g , 1 f r o m t h e b r e a s t , 3 f r o m u n k n o w n origin.
Fetal brain T h e f e t u s e s w e r e o b t a i n e d a t legal a b o r t i o n . T h e c r o w n - r u m p f e t u s e s w a s f r o m 120 m m t o 140 m m .
length of the
108
Isolation and assay of brain amino acids Brain samples were prepared according to O'Neal and Koeppe 14, then treated with ether to remove lipids contaminants and stored a t - - 2 0 °C until chromatographed, Chromatographic analysis was carried out with the single column Technicon Auto Analyser, using lithium buffers as described by Vega and Nunn 23 with a slight modification for the first buffer (pH 2.75 instead of 3.01). The results given are the mean of two analyses made on each sample. RESULTS
Description of the chromatograms The amino acids studied are listed in Table II in the same order as the corresponding peaks on the chromatograms. The general appearance of these chromatograms was very similar to that reported by Perry et al. 15, with the following differences: (1) The peak of oxidized glutathione preceded that of proline. (2) Cystine and valine were often inadequately resolved from each other; when they were resolved the amount of cystine was negligible compared with that of valine, whereas when they were mixed the ratio of the surface of the peak at 440 and 570 nm was very close to that observed for valine alone (this ratio is: 0.231 for valine and 0.512 for cystine). For this reason cystine was shown in Table Il as valine ( + cystine), but the concentrations were calculated as if it were pure valine. In the normal adult brain chromatograms certain peaks remained unidentified. Some of these were constant, in particular a small one at the beginning, but most were inconstant. In the fetal brain chromatograms all the measurable peaks were identified. In the tumour tissue several peaks not seen on the chromatograms of normal brain tissue were visible but none constantly. In 8 of the 13 glioblastomas and in 2 of the 7 metastases a large peak in the position of proline, but with a ratio of the surface of the peak at 440 and 570 nm varying from 0.30 to 0.80, was noted. Chromatography of the samples with the large unidentified peak was repeated several months later. In each case this peak was found to have been replaced by a much lower peak of which the ratio of the surface at 440 and 570 nm was about 6.0 and therefore characteristic of proline. This unidentified peak may consequently result from the superposition of proline and another labile substance as yet unknown but which could be identified on fresh samples. The concentrations given for proline in Table II are the means of those obtained at the first analysis when proline was correctly separated and at the second analysis when proline was initially mixed with the unidentified substance. Normal brain tissue The mean values for the amino acids concentrations in normal brain biopsy specimens are given in the first column of Table II. The amino acids with the highest concentrations were glutamic acid, glutamine, aspartic acid, cystathionine, taurine, glycine, alanine, G A B A and phosphoethanolamine. No regional differences were detected in these biopsy specimens with the exception of taurine. Taurine was present
109 in concentrations varying from 0.65 to 1.48/~moles/g in the cerebral biopsy specimens and in a concentration of 3.16 /zmoles/g in the cerebellar specimen. The value of taurine for the cerebellar specimen was not included in the calculation of the mean. Wide individual variations were found for certain amino acids. For example, the concentration of glycine ranged from 0.39 to 2.12/~moles/g, that of alanine from 0.56 to 2.02 /zmoles/g and that of cystathionine from 0.22 to 4.79 #moles/g. This variability for alanine and glycine differs of the results of Perry et al. ~6. We also found in the case of these 2 amino acids, of aspartic acid, and of GABA, concentrations higher than those reported by Perry et al. TM for biopsy specimens. This may be due to the interval (1-10 rain) between removal and freezing of the samples. Lajtha and Toth 8, analysing the amino acid concentrations of mouse and guinea pig brains found higher alanine and GABA concentrations in brains kept for 20 rain at 20 °C than in brains from heads frozen immediately after decapitation.
Fetal brain tissue
The concentrations of free amino acids found in the brains of 4 fetuses (Table II, column 2) agree as regards the relative proportions of the different amino acids with those reported by Okumura et al. la and A'Zary et al. 1. We observed that most amino acids--chiefly taurine, phosphoethanolamine and proline - - were higher in fetal than in adult brains, while others (aspartic acid, glutamic acid, glutamine, cystathionine and GABA) were lower in fetal than in adult brains. The most striking differences were in the 2 peptides homocarnosine and glutathione which were absent in fetal brains while they were present in adult brains, and in cystathionine which was present in very low amount in fetal brains but reached very high levels in adult brains. Among the amino acids which decreased during cerebral development, the greatest relative decline was shown by proline, which diminished more than tenfold.
Glial tumours
The amino acid content of 13 glioblastoma multiforme in adult subjects is presented in Table II (column 3) and compared with that of normal brain. The amino acids which were present in smaller amount in the glioblastomas, namely aspartic acid, glutamic acid, GABA, and homocarnosine, were those which are thought to be for the major part synthesised in the brain. Some amino acids, taurine, phosphoethanolamine, glutamine and cystathionine, were present in the same concentrations in malignant tissue and in normal tissue. Among the numerous amino acids which were present in greater amounts, proline showed the greatest relative increase (tenfold). In addition to the 13 glioblastoma multiforme we analysed the other glial tumours listed under material and methods. The main features of these other tumours were: a very low level of GABA (0.06/zmoles/g and of cystathionine (0.09/~moles/g) in the recurrent glioblastoma, a very low level of GABA (0 and 0.06/~moles/g) and the absence of homocarnosine in the 2 glioblastoma multiforme of the posterior fossa in children and an absence of GABA, homocarnosine and cystathionine in the ependymoma.
110
IJ m O L E S / i
i
r ,:.:, :.:.:
TAURINE
J
PROLINE
[]
NORMAL ADULT
[]
FETAL B R A I N
[]
GLIOBLASTOMAS
[]
METASTASES
BRAIN
CYSTATHIONIN E
GABA
HO MOCA RNOS I N E
rl = 1 3
R =4
n
-- 13
1"1 = 7
Fig. 1. Concentrations of taurine, proline, cystathionine, GABA, and homocarnosine in normal adult human brain, fetal brain, glioblastomas and metastases. Each bar represents the mean for n samples. The vertical lines represent the S.E.M. The results are expressed in Hmoles/g wet wt. Metastases
This group is not homogeneous, owing to the various origin of the tumours composing it. They have however a common characteristic: they originate from turnouts outside the nervous system. When compared to glial tumours, metastases showed much higher concentrations of taurine, phosphoethanolamine and proline and very low concentrations of cystathionine, G A B A and homocarnosine, this latter dipeptide being present only in trace amounts. Possibly the small amounts of normal tissue contaminating the samples were responsible for the presence of G A B A and homocarnosine. DISCUSSION The concentrations of most of the amino acids studied varied according to the origin of the sample. This applied especially to taurine, proline, cystathionine, GABA and homocarnosine (Fig. 1). Our finding of lower concentrations of taurine in adult brain than in fetal brain
Ill agrees with other findings in most of the species hitherto studied 7A2,13,21,24. The concentrations of taurine in glial tumours were not significantly different from those in normal adult brain, but in metastases they were much higher. The concentrations of proline were low in all the specimens of normal brain analysed. The ratio of proline in fetal brain and gliomas to that in normal adult brain was about 10:1. In metastases the corresponding ratio was about 20:1. This suggests that a low level of proline is a characteristic feature of normal adult human brain. Cystathionine was found in very low concentrations in fetal brain, a result which agrees with that of Volpe and Laster 24 in the monkey brain and of Okumura et al. 13 in the human fetal brain. As was first reported by Tallan et al. 22 and later by others 2,4,13,16,19,27, cystathionine was found in high amount in normal adult brain (1.57 ~_ 0.40 /~moles/g) and we observed wide variations from sample to sample (from 0.22 to 4.79 /~moles/g). Such a variability has already been observed and 4 explanations have been proposed for it: (1) individual variations, (2) regional differences, (3) a variable amount of grey and white matter in the samples and (4) pathological alterations in apparently normal brain in the vicinity of intracerebral tumours. The first hypothesis has been suggested by Perry et al. 16 in the man and demonstrated in the monkey by Sturman et al. e°. The second hypothesis is based upon observations made in human brains obtained at necropsy 1'5 and in monkey brains 2°. The third explanation was first proposed by Shimizu et al. 19 who found higher concentrations of cystathionine in the white matter than in the grey matter of 3 human brains obtained at necropsy. Perry et al. 16, in contrast, reported a content of cystathionine in grey matter obtained by biopsy higher than that of specimens of corpus callosum obtained at necropsy and suggested that the difference of concentration of cystathionine in the white and in the grey matter which has been observed on autopsy specimens might have been due to the post-mortem activity of cystathionase in the grey matter. However, differences in the concentration of cystathionine in the white matter and in the grey matter noted by Sturman et al. 20 in monkeys could not in their experiments have been due to post-mortem changes. The fourth hypothesis, in other words pathological alterations in the brains of patient with brain tumours, was put forward by Okumura et al. 13. They observed high values of cystathionine in tissue surrounding tumours and low values in the brain of a patient suffering of'genuine epilepsy' and also in the brain of an individual accidently deceased. This hypothesis was supported by the results of Promyslov and Sokovnina 17, who found much higher concentrations of cystathionine in tissue taken far from the tumour site in rabbits with glial turnouts, than in the brains of control animals. These hypotheses, however, are not mutually exclusive and our results are compatible with them all. Indeed, we found the lowest values for cystathionine in the temporal and frontal poles and the highest in the body of the temporal lobe. This might result from regional differences or from the fact that the frontal and the temporal pole resected at operation contained more grey than white matter. As for the possible alterations in cystathionine content due to the presence of turnout tissue, it was noticeable that the only specimens which were certainly normal were those from individuals with aneurysms; they were also those with the lowest cystathionine concentrations (Table l).
112 The cystathionine content in malignant tissue varied markedly with the histological nature of the tumour. In tumours of glial origin the average value for cystathionine was high. This suggests that this amino acid is synthesized in glial cells. However, high levels of this amino acid have also been observed in neuroblastomas ~ and some hepatic tumours a,a,zS, so that one may wonder whether a high concentration of cystathionine is not simply an indication of malignancy rather than specific for gliomas. Our findings of very small amounts of cystathionine in metastases of diverse (non-glial) origin suggest that the high concentrations noted in glioblastomas may well be characteristic of malignant glial cell metabolism. As Gjessing 5 suggested in the case of neuroblastomas, the high concentrations of cystathionine found could be explained by low cystathionase activity resulting of the vitamin B6 deficiency of tumour tissue. This hypothesis could be checked by determination of the concentration of pyridoxal phosphate and by measures of the activity of cystathionase. It is noteworthy that, in glioblastomas as in normal tissue, the cystathionine concentration varied from sample to sample, the lowest amount being found in the recurrent glioma and in the turnouts in which the concentration of total amino acids was low. The highest concentrations of G A B A and homocarnosine were found in normal adult brain. Like Okumura et al. 13 we found GABA in fetal brain, but we could not detect any homocarnosine. In glial tumours G A B A and homocarnosine were present in appreciable amounts. Wolleman and Devenyi 26, however, failed to detect G A B A in any of 25 brain tumours. The presence of GABA and homocarnosine in glial tumours suggests that they may be, at least partly, associated with glial cell metabolism. To sum up, the presence of characteristic amounts of different amino acids in different types of tissue suggests that a tissue could be identified from its amino acid content. Yet, identification of the origin of a sample may not be easy on individual chromatograms especially in the case of glioblastomas, some of which giving chromatograms similar to those of metastases. ACKNOWLEDGEMENTS We are grateful to the neurosurgeons: Drs Chai, Cophigon, Houdart, Lebesnerais, Rey, Tavernier, Thurel and Visot and to the other members of the staff of the operating rooms for their assistance in obtaining tumour specimens and normal brain biopsy specimens. We are indebted to Drs Bou6 and Hauw and to Prs Gautray and Lanvin for providing the fetal specimens. This study was supported by contract 75-1-197-6 of the lnstitut National de la Sant6 et de la Recherche M6dicale.
REFERENCES 1 A'Zary, E., Saifer, A., Schneck, L. and Volk, B. W., The free amino acids of foetal brain in TaySachs disease, J. Neurochem., 18 (1971) 1369-1370. 2 Brenton, D. P., Cusworth, D. C. and Gaull, G. E., Homocystinuria. Biochemical studies of tissues including a comparison with cystatbioninuria, Pediatrics, 35 (1965) 50 56.
113 3 Geiser, C. F., Baez, A., Schindler, A. M. and Shih, V. E., Epithelial hepatoblastoma associated with congenital hemihypertrophy and cystathioninuria: presentation of a case, Pediatrics, 46 (1970) 66 73. 4 Gerritsen, T. and Waisman, H. A., Homocystinuria: absence ofcystathionine in the brain, Science, 145 (1964) 588. 5 Gjessing, L. R., Studies of functional neural tumors, 11I. Cystathionine in the tumor tissue, Scand. J. clin. Lab. Invest., 15 (1963) 479482. 6 Helson, L., Peterson, R. H. F. and Schwartz, M. K., Cystathionine excess in children with hepatic cancer, Cancer Res., 33 (1973) 1570-1573. 7 Lajtha, A. and Toth, J., Perinatal changes in the free amino acid pool of the brain in mice, Brain Research, 55 (1973) 238 241. 8 Lajtha, A. and Toth, J., Post-mortem changes in the cerebral free amino acid pool, Brain Research, 76 (1974) 546-551. 9 Levi, G. and Morisi, G., Free amino acids and related compounds in chick brain during development, Brain Research, 26 ( 1971 ) 131-140. 10 Mardens, Y. and van Sande, M., Identification and quantitation of ninhydrine positive substances in the human central nervous system, Automation in Analyt. Chem., European Technicon Symposium, 1967, p. 239. 11 Molnar, G., A simple thin-layer chromatographic method for the study of amino acids in cerebral tumors using frozen tissue slices, Clin. chim. Acta, 27 (1970) 535-541. 12 Oja, S. S., Uusitalo, A. J., Vahvelainen, M. L. and Piha, R. S., Changes in cerebral and hepatic amino acids in the rat and guinea pig during development, Brain Research, 11 (1968) 655-661. 13 Okumura, N., Otsuki, S. and Kayemana, A., Studies on free amino acids in human brain, J. Biochem. (Tokyo), 47 (1960) 315-320. ! 4 0 ' N e a l , R. M. and Koeppe, R. E., Precursors in vivo of glutamate, aspartate and their derivatives of rat brain, J. Neurochem., 13 (1966) 835-847. 15 Perry, T. L., Berry, K., Hansen, S., Diamond, S. and Mok, C., Regional distribution of amino acids in human brain obtained at autopsy, J. Neurochem., 18 (1971) 513-519. 16 Perry, T. L., Hansen, S., Berry, K., Mok, C. and Lesk, D., Free amino acids and related compounds in biopsies of human brain, J. Neurochem., 18 (1971) 521-528. 17 Promyslov, M. Sh. and Sokovnina, Ya. M., Content of L-Cystathionine in brain tissue in some nervous system diseases, Vop. Med. Khim., 19 (1973) 66 68. 18 Robinson, N. and Williams, C. B.,Amino acids in human brain, Clin. chim. Acta, 12 (1965) 311-317. 19 Shimizu, H., Kakimoto, Y., and Sano, I., A method of determination of cystathionine and its distribution in human brain, J. Nearochem., 13 (1966) 65-73. 20 Sturman, J. A., Rassin, D. K. and Gaull, G. E., Relation of three enzymes of transsulphuration to the concentration of cystathionine in various regions of monkey brain, J. Neurochem., 17 (1970) 1117-1119. 21 Sturman, J. A. and Gaull, G. E., Taurine in the brain and liver of the developing human and monkey, J. Neurochem., 25 (1975) 831-835. 22 Tallan,H. H., Moore, S. and Stein, W. H., L-Cystationine in human brain, J. biol. Chem., 230 (1958) 707-713. 23 Vega, A. and Nunn, P. B., A lithium buffer system for single-column amino acid analysis, Anal. Biochem., 32 (1969) 446453. 24 Volpe, J. J. and Laster, L., Trans-sulphuration in primate brain: regional distribution of cystathionine synthase, cystathionine, and taurine in the brain of the Rhesus monkey at various stages of development, J. Neurochem., 17 (1970) 425437. 25 Vofate, P. A. and Wadman, S.K., Cystathioninuria in hepatoblastoma, Clin. chim. Acta, 22 (1968) 373-378. 26 Wolleman, M. and D6v6nyi, T., The ~, aminobutyric acid content and glutamate decarboxylase activity of brain tumors, J. Neurochem., 10 (1963) 83 88. 27 Yamamoto, Y., Mori, A. and Jinnai, D., Amino acids in the brain. Alterations of amino acids in rabbit brain caused by repetitive convulsive fits and comparison of amino acid contents in epileptic and non epileptic human brain, J. Biochem. (Tokyo), 49 (1961) 368-372.
105
© Elsevier/North-Holland Biomedical Press, Amsterdam - Printed in The Netherlands
F R E E A M I N O ACIDS A N D R E L A T E D SUBSTANCES IN H U M A N G L I A L T U M O U R S A N D IN F E T A L BRAIN: C O M P A R I S O N W I T H N O R M A L A D U L T BRAIN
JEANNE-MARIE LEFAUCONN1ER, CHRISTIANE PORTEMER and FERNANDE CHATAGNER Laboratoire de Chimie Biologique, 96 Boulevard Raspail, 75006-Paris (France)
(Accepted March 30th, 1976)
SUMMARY An ion exchange automatic chromatographic analysis of the free amino acid concentrations of 18 human glial tumours and of 4 human fetal brains was carried out and the concentrations were compared to those of 13 biopsy specimens of normal adult brain. In addition, the concentrations of the amino acids of the glial tumours were compared to those of 7 intracerebral metastases of various origin. The chromatograms of several tumour specimens showed an unidentified peak overlapping proline. As far as the amino acid concentrations are concerned they varied depending upon the origin of the sample. The concentrations of most amino acids were higher in fetal brain than in adult brain with the exception of aspartic acid, glutamic acid, glutamine, cystathionine and GABA. Two peptides: glutathione and homocarnosine were absent in fetal brain and were present in adult brain. In glial tumours, homocarnosine and some amino acids, namely aspartic acid, glutamic acid and GABA, showed lower concentrations than in normal brain. Some amino acids were in the same concentration as in normal brain: taurine, phosphoethanolamine, glutamine and cystathionine. Most of the others were in higher concentrations than in normal brain, mainly proline. The results suggest that the concentrations of 5 compounds: taurine, proline, cystathionine, GABA and homocarnosine, taken as a whole, provide information on the origin of the sample.
INTRODUCTION The amino acid concentrations of normal adult human brain and of fetal human brain have already been reportedl,10,13,~5,16,18. In both cases the analysis reflects the
106 amino acid composition of an heterogeneous population of neurons and glial cells. It seemed therefore worthwhile to determine the amino acid content o f glial tumours, c o m p o s e d mainly o f cells o f glial origin. To our knowledge, however, only partial information is available in this connection 11. The work reported here is a comparative study o f amino acid concentrations determined in 18 glial tumours, 4 fetal brains and 13 normal adult brain biopsy specimens. In addition, we c o m p a r e d the amino acid concentrations o f the glial tumours with those of 7 metastases to determine whether the alterations, if any, found in glial turnouts were characteristic o f gliomas or due to malignancy. MATERIAL AND METHODS N o r m a l brain biopsy specimens Biopsy specimens of normal brain tissue were obtained at c r a n i o t o m y f r o m 13 adults undergoing neurosurgical operations for treatment o f vascular malformations or for ablation o f intracranial tumours. Table I lists the indications for operation and the site at which tissue was removed for biopsy. In every case normal tissue resection was necessary to gain access to a deep-seated lesion. The operation was undertaken under general anaesthesia (Thiopental, Phenoperidine and Pancuronium), with artificial respiration. These biopsy specimens were put into dry ice within less than 10 rain after removal from the brain. Tumour tissue specimens These were obtained from patients operated on under the conditions described above. The specimens consisted of: (I) T u m o u r s o f glial origin: 13 glioblastoma
TABLE
1
Sourees of biopsy specimens of normal brain tissue and corresponding cystathionine concentrations Patient
Age
Sex
Indication .[or operation
Site of biopsy
Cystathionine ( Imloles/g)
l 2 3
45 23 32
M M F
Temporal pole Frontal pole Frontal pole
0.59 0.61 0.32
4 5 6 7 8 9 10 11 12 13
33 22 47 47 53 69 55 59 26 63
M M F M F M M M M M
Middle Cerebral Art. Aneurysm Carotid Aneurysm Ant. Communicating Art. Aneurysm Arterio-venous fistula Meningioma Meningioma Meningioma Meningioma Meningioma Metastasis Glioma Glioma Acoustic nerve neurinoma
Frontal lobe Frontal pole Temporal lobe Temporal lobe Frontal pole Frontal lobe Parietal lobe Temporal lobe Temporal lobe Cerebellum
0.76 0.25 4.79 3.05 0.25 2.35 1.20 3.55 1.05 1.74
107 TABLE 1I
Amino acid content o f normal adult human brain tissue, fetal brain tissue, glioblastomas, and cerebral metastases The results are expressed in/~moles/g of wet weight and represent means 4- S.E.M. n -- number of samples. Student's t test was used to calculate the statistical differences. Fetal brain and glioblastomas are compared with adult brain. Metastases are compared with glioblastomas. Abbreviations: Ph.ethanolamine - phosphoethanolamine; Ox. Glutathione = oxidized glutathione; a-aminobut, acid ~-aminobutyric acid.
Taurine Ph.ethanolamine Aspartic acid Threonine Serine Glutamic acid Glutamine Proline Ox.Glutathione Glycine Alanine a aminobut, acid Valine (4 cystine) Methionine Cystathionine lsoleucine Leucine Tyrosine Phenylalanine GABA Ornithine Lysine Histidine Homocarnosine Arginine Total a. acids
Adult brain (n 13)
Fetal brain (n = 4)
Glioblastomas (n 13)
Metastases (n 7)
1.03 0.66 1.52 0.29 0.48 6.84 3.49 0.04 0.23 1.03 0.95 0.03 0.21 0.04 1.57 0.07 0.16 0.10 0.09 0.94 0.02 0.20 0.10 0.17 0.10 20.7
3.14 i 0.42*** 2.60 4- 0.25*** 0.65 4- 0.12"* 0.91 4- 0.11"** 1.10 ± 0.13"** 2.91 4- 0.17"** 1.06 4- 0.13"** 0.31 4- 0.10"* 0"* 1.24 ± 0.17 1.69 4- 0.17'* 0.06 ± 0.01 ** 0.50 4- 0.09** 0.07 4- 0.02 0.05 4- 0.03 0.24 4- 0.04*** 0.55 4- 0.07*** 0.19 4- 0.02 0.21 4- 0.03* 0.46 ± 0.12" 0.08 4- 0.01 0.43 4- 0.09* 0.17 ± 0.03 0 *** 0.32 ± 0.06*** 20.1 ± 1.8
1.20 0.60 0.76 0.73 0.76 4.22 3.77 0.45 0.36 3.09 1.92 0.06 0.37 0.07 1.65 0.12 0.27 0.14 0.15 0.54 0.11 0.46 0.15 0.06 0.17 22.3
3.08 1.66 1.20 0.44 0.75 4.15 1.10 0.85 0.10 2.44 1.71 0.06 0.56 0.07 0.15 0.21 0.52 0.21 0.25 0.07 0.07 0.50 0.18 0.01 0.38 20.1
± 0.09 4- 0.1 l 4- 0.14 4- 0.05 4- 0.04 4- 0.32 4- 0.28 3_ 0.03 4- 0.05 4- 0.14 4- 0.11 -I- 0.01 4- 0.04 4- 0.01 4- 0.40 4- 0.02 4- 0.02 4- 0.03 4- 0.02 4- 0.02 4- 0.01 ± 0.05 4- 0.03 ± 0.03 ± 0.03 ± 1.0
4- 0.10 -I- 0.13 4- 0.18" 4- 0.14" ± 0.15 4- 0.92* q- 0.56 4- 0.09*** 4- 0.11 ± 0.57** 4- 0.36* ± 0.01 ** ± 0.05* 4- 0.02 4- 0.34 4- 0.02* ± 0.04 ± 0.02 4- 0.02 4- 0.13" 4- 0.03 4- 0.10" ± 0.03 4- 0.03** ± 0.07 ± 2.4
4- 0.87* 4- 0.23*** 4- 0.27 ~ 0.10 4- 0.12 4- 0.26 4- 0.17" 4- 0.12"** 4- 0.08 3_ 0.25 4- 0.26 i 0.02 ~: 0.12 4- 0.03 ± 0.06** £ 0.09 4- 0.17 4- 0.05* ± 0.08 4- 0.03* 4- 0.02 4- 0.13 4- 0.03 ~ 0.01 ** 4- 0.17 5:1.8
* P < 0.05. ** P < 0.01. *** P < 0.001.
m u l t i f o r m e o f t h e c e r e b r a l h e m i s p h e r e s in a d u l t s , 2 g l i o b l a s t o m a m u l t i f o r m e o f t h e p o s t e r i o r f o s s a in c h i l d r e n , 1 r e c u r r e n t g l i o b l a s t o m a , 1 a s t r o c y t o m a , a n d 1 e p e n d y m o m a . (2) M e t a s t a s e s : 2 f r o m t h e s k i n ( m e l a n o m a ) , 1 f r o m t h e l u n g , 1 f r o m t h e b r e a s t , 3 f r o m u n k n o w n origin.
Fetal brain T h e f e t u s e s w e r e o b t a i n e d a t legal a b o r t i o n . T h e c r o w n - r u m p f e t u s e s w a s f r o m 120 m m t o 140 m m .
length of the
108
Isolation and assay of brain amino acids Brain samples were prepared according to O'Neal and Koeppe 14, then treated with ether to remove lipids contaminants and stored a t - - 2 0 °C until chromatographed, Chromatographic analysis was carried out with the single column Technicon Auto Analyser, using lithium buffers as described by Vega and Nunn 23 with a slight modification for the first buffer (pH 2.75 instead of 3.01). The results given are the mean of two analyses made on each sample. RESULTS
Description of the chromatograms The amino acids studied are listed in Table II in the same order as the corresponding peaks on the chromatograms. The general appearance of these chromatograms was very similar to that reported by Perry et al. 15, with the following differences: (1) The peak of oxidized glutathione preceded that of proline. (2) Cystine and valine were often inadequately resolved from each other; when they were resolved the amount of cystine was negligible compared with that of valine, whereas when they were mixed the ratio of the surface of the peak at 440 and 570 nm was very close to that observed for valine alone (this ratio is: 0.231 for valine and 0.512 for cystine). For this reason cystine was shown in Table Il as valine ( + cystine), but the concentrations were calculated as if it were pure valine. In the normal adult brain chromatograms certain peaks remained unidentified. Some of these were constant, in particular a small one at the beginning, but most were inconstant. In the fetal brain chromatograms all the measurable peaks were identified. In the tumour tissue several peaks not seen on the chromatograms of normal brain tissue were visible but none constantly. In 8 of the 13 glioblastomas and in 2 of the 7 metastases a large peak in the position of proline, but with a ratio of the surface of the peak at 440 and 570 nm varying from 0.30 to 0.80, was noted. Chromatography of the samples with the large unidentified peak was repeated several months later. In each case this peak was found to have been replaced by a much lower peak of which the ratio of the surface at 440 and 570 nm was about 6.0 and therefore characteristic of proline. This unidentified peak may consequently result from the superposition of proline and another labile substance as yet unknown but which could be identified on fresh samples. The concentrations given for proline in Table II are the means of those obtained at the first analysis when proline was correctly separated and at the second analysis when proline was initially mixed with the unidentified substance. Normal brain tissue The mean values for the amino acids concentrations in normal brain biopsy specimens are given in the first column of Table II. The amino acids with the highest concentrations were glutamic acid, glutamine, aspartic acid, cystathionine, taurine, glycine, alanine, G A B A and phosphoethanolamine. No regional differences were detected in these biopsy specimens with the exception of taurine. Taurine was present
109 in concentrations varying from 0.65 to 1.48/~moles/g in the cerebral biopsy specimens and in a concentration of 3.16 /zmoles/g in the cerebellar specimen. The value of taurine for the cerebellar specimen was not included in the calculation of the mean. Wide individual variations were found for certain amino acids. For example, the concentration of glycine ranged from 0.39 to 2.12/~moles/g, that of alanine from 0.56 to 2.02 /zmoles/g and that of cystathionine from 0.22 to 4.79 #moles/g. This variability for alanine and glycine differs of the results of Perry et al. ~6. We also found in the case of these 2 amino acids, of aspartic acid, and of GABA, concentrations higher than those reported by Perry et al. TM for biopsy specimens. This may be due to the interval (1-10 rain) between removal and freezing of the samples. Lajtha and Toth 8, analysing the amino acid concentrations of mouse and guinea pig brains found higher alanine and GABA concentrations in brains kept for 20 rain at 20 °C than in brains from heads frozen immediately after decapitation.
Fetal brain tissue
The concentrations of free amino acids found in the brains of 4 fetuses (Table II, column 2) agree as regards the relative proportions of the different amino acids with those reported by Okumura et al. la and A'Zary et al. 1. We observed that most amino acids--chiefly taurine, phosphoethanolamine and proline - - were higher in fetal than in adult brains, while others (aspartic acid, glutamic acid, glutamine, cystathionine and GABA) were lower in fetal than in adult brains. The most striking differences were in the 2 peptides homocarnosine and glutathione which were absent in fetal brains while they were present in adult brains, and in cystathionine which was present in very low amount in fetal brains but reached very high levels in adult brains. Among the amino acids which decreased during cerebral development, the greatest relative decline was shown by proline, which diminished more than tenfold.
Glial tumours
The amino acid content of 13 glioblastoma multiforme in adult subjects is presented in Table II (column 3) and compared with that of normal brain. The amino acids which were present in smaller amount in the glioblastomas, namely aspartic acid, glutamic acid, GABA, and homocarnosine, were those which are thought to be for the major part synthesised in the brain. Some amino acids, taurine, phosphoethanolamine, glutamine and cystathionine, were present in the same concentrations in malignant tissue and in normal tissue. Among the numerous amino acids which were present in greater amounts, proline showed the greatest relative increase (tenfold). In addition to the 13 glioblastoma multiforme we analysed the other glial tumours listed under material and methods. The main features of these other tumours were: a very low level of GABA (0.06/zmoles/g and of cystathionine (0.09/~moles/g) in the recurrent glioblastoma, a very low level of GABA (0 and 0.06/~moles/g) and the absence of homocarnosine in the 2 glioblastoma multiforme of the posterior fossa in children and an absence of GABA, homocarnosine and cystathionine in the ependymoma.
110
IJ m O L E S / i
i
r ,:.:, :.:.:
TAURINE
J
PROLINE
[]
NORMAL ADULT
[]
FETAL B R A I N
[]
GLIOBLASTOMAS
[]
METASTASES
BRAIN
CYSTATHIONIN E
GABA
HO MOCA RNOS I N E
rl = 1 3
R =4
n
-- 13
1"1 = 7
Fig. 1. Concentrations of taurine, proline, cystathionine, GABA, and homocarnosine in normal adult human brain, fetal brain, glioblastomas and metastases. Each bar represents the mean for n samples. The vertical lines represent the S.E.M. The results are expressed in Hmoles/g wet wt. Metastases
This group is not homogeneous, owing to the various origin of the tumours composing it. They have however a common characteristic: they originate from turnouts outside the nervous system. When compared to glial tumours, metastases showed much higher concentrations of taurine, phosphoethanolamine and proline and very low concentrations of cystathionine, G A B A and homocarnosine, this latter dipeptide being present only in trace amounts. Possibly the small amounts of normal tissue contaminating the samples were responsible for the presence of G A B A and homocarnosine. DISCUSSION The concentrations of most of the amino acids studied varied according to the origin of the sample. This applied especially to taurine, proline, cystathionine, GABA and homocarnosine (Fig. 1). Our finding of lower concentrations of taurine in adult brain than in fetal brain
Ill agrees with other findings in most of the species hitherto studied 7A2,13,21,24. The concentrations of taurine in glial tumours were not significantly different from those in normal adult brain, but in metastases they were much higher. The concentrations of proline were low in all the specimens of normal brain analysed. The ratio of proline in fetal brain and gliomas to that in normal adult brain was about 10:1. In metastases the corresponding ratio was about 20:1. This suggests that a low level of proline is a characteristic feature of normal adult human brain. Cystathionine was found in very low concentrations in fetal brain, a result which agrees with that of Volpe and Laster 24 in the monkey brain and of Okumura et al. 13 in the human fetal brain. As was first reported by Tallan et al. 22 and later by others 2,4,13,16,19,27, cystathionine was found in high amount in normal adult brain (1.57 ~_ 0.40 /~moles/g) and we observed wide variations from sample to sample (from 0.22 to 4.79 /~moles/g). Such a variability has already been observed and 4 explanations have been proposed for it: (1) individual variations, (2) regional differences, (3) a variable amount of grey and white matter in the samples and (4) pathological alterations in apparently normal brain in the vicinity of intracerebral tumours. The first hypothesis has been suggested by Perry et al. 16 in the man and demonstrated in the monkey by Sturman et al. e°. The second hypothesis is based upon observations made in human brains obtained at necropsy 1'5 and in monkey brains 2°. The third explanation was first proposed by Shimizu et al. 19 who found higher concentrations of cystathionine in the white matter than in the grey matter of 3 human brains obtained at necropsy. Perry et al. 16, in contrast, reported a content of cystathionine in grey matter obtained by biopsy higher than that of specimens of corpus callosum obtained at necropsy and suggested that the difference of concentration of cystathionine in the white and in the grey matter which has been observed on autopsy specimens might have been due to the post-mortem activity of cystathionase in the grey matter. However, differences in the concentration of cystathionine in the white matter and in the grey matter noted by Sturman et al. 20 in monkeys could not in their experiments have been due to post-mortem changes. The fourth hypothesis, in other words pathological alterations in the brains of patient with brain tumours, was put forward by Okumura et al. 13. They observed high values of cystathionine in tissue surrounding tumours and low values in the brain of a patient suffering of'genuine epilepsy' and also in the brain of an individual accidently deceased. This hypothesis was supported by the results of Promyslov and Sokovnina 17, who found much higher concentrations of cystathionine in tissue taken far from the tumour site in rabbits with glial turnouts, than in the brains of control animals. These hypotheses, however, are not mutually exclusive and our results are compatible with them all. Indeed, we found the lowest values for cystathionine in the temporal and frontal poles and the highest in the body of the temporal lobe. This might result from regional differences or from the fact that the frontal and the temporal pole resected at operation contained more grey than white matter. As for the possible alterations in cystathionine content due to the presence of turnout tissue, it was noticeable that the only specimens which were certainly normal were those from individuals with aneurysms; they were also those with the lowest cystathionine concentrations (Table l).
112 The cystathionine content in malignant tissue varied markedly with the histological nature of the tumour. In tumours of glial origin the average value for cystathionine was high. This suggests that this amino acid is synthesized in glial cells. However, high levels of this amino acid have also been observed in neuroblastomas ~ and some hepatic tumours a,a,zS, so that one may wonder whether a high concentration of cystathionine is not simply an indication of malignancy rather than specific for gliomas. Our findings of very small amounts of cystathionine in metastases of diverse (non-glial) origin suggest that the high concentrations noted in glioblastomas may well be characteristic of malignant glial cell metabolism. As Gjessing 5 suggested in the case of neuroblastomas, the high concentrations of cystathionine found could be explained by low cystathionase activity resulting of the vitamin B6 deficiency of tumour tissue. This hypothesis could be checked by determination of the concentration of pyridoxal phosphate and by measures of the activity of cystathionase. It is noteworthy that, in glioblastomas as in normal tissue, the cystathionine concentration varied from sample to sample, the lowest amount being found in the recurrent glioma and in the turnouts in which the concentration of total amino acids was low. The highest concentrations of G A B A and homocarnosine were found in normal adult brain. Like Okumura et al. 13 we found GABA in fetal brain, but we could not detect any homocarnosine. In glial tumours G A B A and homocarnosine were present in appreciable amounts. Wolleman and Devenyi 26, however, failed to detect G A B A in any of 25 brain tumours. The presence of GABA and homocarnosine in glial tumours suggests that they may be, at least partly, associated with glial cell metabolism. To sum up, the presence of characteristic amounts of different amino acids in different types of tissue suggests that a tissue could be identified from its amino acid content. Yet, identification of the origin of a sample may not be easy on individual chromatograms especially in the case of glioblastomas, some of which giving chromatograms similar to those of metastases. ACKNOWLEDGEMENTS We are grateful to the neurosurgeons: Drs Chai, Cophigon, Houdart, Lebesnerais, Rey, Tavernier, Thurel and Visot and to the other members of the staff of the operating rooms for their assistance in obtaining tumour specimens and normal brain biopsy specimens. We are indebted to Drs Bou6 and Hauw and to Prs Gautray and Lanvin for providing the fetal specimens. This study was supported by contract 75-1-197-6 of the lnstitut National de la Sant6 et de la Recherche M6dicale.
REFERENCES 1 A'Zary, E., Saifer, A., Schneck, L. and Volk, B. W., The free amino acids of foetal brain in TaySachs disease, J. Neurochem., 18 (1971) 1369-1370. 2 Brenton, D. P., Cusworth, D. C. and Gaull, G. E., Homocystinuria. Biochemical studies of tissues including a comparison with cystatbioninuria, Pediatrics, 35 (1965) 50 56.
113 3 Geiser, C. F., Baez, A., Schindler, A. M. and Shih, V. E., Epithelial hepatoblastoma associated with congenital hemihypertrophy and cystathioninuria: presentation of a case, Pediatrics, 46 (1970) 66 73. 4 Gerritsen, T. and Waisman, H. A., Homocystinuria: absence ofcystathionine in the brain, Science, 145 (1964) 588. 5 Gjessing, L. R., Studies of functional neural tumors, 11I. Cystathionine in the tumor tissue, Scand. J. clin. Lab. Invest., 15 (1963) 479482. 6 Helson, L., Peterson, R. H. F. and Schwartz, M. K., Cystathionine excess in children with hepatic cancer, Cancer Res., 33 (1973) 1570-1573. 7 Lajtha, A. and Toth, J., Perinatal changes in the free amino acid pool of the brain in mice, Brain Research, 55 (1973) 238 241. 8 Lajtha, A. and Toth, J., Post-mortem changes in the cerebral free amino acid pool, Brain Research, 76 (1974) 546-551. 9 Levi, G. and Morisi, G., Free amino acids and related compounds in chick brain during development, Brain Research, 26 ( 1971 ) 131-140. 10 Mardens, Y. and van Sande, M., Identification and quantitation of ninhydrine positive substances in the human central nervous system, Automation in Analyt. Chem., European Technicon Symposium, 1967, p. 239. 11 Molnar, G., A simple thin-layer chromatographic method for the study of amino acids in cerebral tumors using frozen tissue slices, Clin. chim. Acta, 27 (1970) 535-541. 12 Oja, S. S., Uusitalo, A. J., Vahvelainen, M. L. and Piha, R. S., Changes in cerebral and hepatic amino acids in the rat and guinea pig during development, Brain Research, 11 (1968) 655-661. 13 Okumura, N., Otsuki, S. and Kayemana, A., Studies on free amino acids in human brain, J. Biochem. (Tokyo), 47 (1960) 315-320. ! 4 0 ' N e a l , R. M. and Koeppe, R. E., Precursors in vivo of glutamate, aspartate and their derivatives of rat brain, J. Neurochem., 13 (1966) 835-847. 15 Perry, T. L., Berry, K., Hansen, S., Diamond, S. and Mok, C., Regional distribution of amino acids in human brain obtained at autopsy, J. Neurochem., 18 (1971) 513-519. 16 Perry, T. L., Hansen, S., Berry, K., Mok, C. and Lesk, D., Free amino acids and related compounds in biopsies of human brain, J. Neurochem., 18 (1971) 521-528. 17 Promyslov, M. Sh. and Sokovnina, Ya. M., Content of L-Cystathionine in brain tissue in some nervous system diseases, Vop. Med. Khim., 19 (1973) 66 68. 18 Robinson, N. and Williams, C. B.,Amino acids in human brain, Clin. chim. Acta, 12 (1965) 311-317. 19 Shimizu, H., Kakimoto, Y., and Sano, I., A method of determination of cystathionine and its distribution in human brain, J. Nearochem., 13 (1966) 65-73. 20 Sturman, J. A., Rassin, D. K. and Gaull, G. E., Relation of three enzymes of transsulphuration to the concentration of cystathionine in various regions of monkey brain, J. Neurochem., 17 (1970) 1117-1119. 21 Sturman, J. A. and Gaull, G. E., Taurine in the brain and liver of the developing human and monkey, J. Neurochem., 25 (1975) 831-835. 22 Tallan,H. H., Moore, S. and Stein, W. H., L-Cystationine in human brain, J. biol. Chem., 230 (1958) 707-713. 23 Vega, A. and Nunn, P. B., A lithium buffer system for single-column amino acid analysis, Anal. Biochem., 32 (1969) 446453. 24 Volpe, J. J. and Laster, L., Trans-sulphuration in primate brain: regional distribution of cystathionine synthase, cystathionine, and taurine in the brain of the Rhesus monkey at various stages of development, J. Neurochem., 17 (1970) 425437. 25 Vofate, P. A. and Wadman, S.K., Cystathioninuria in hepatoblastoma, Clin. chim. Acta, 22 (1968) 373-378. 26 Wolleman, M. and D6v6nyi, T., The ~, aminobutyric acid content and glutamate decarboxylase activity of brain tumors, J. Neurochem., 10 (1963) 83 88. 27 Yamamoto, Y., Mori, A. and Jinnai, D., Amino acids in the brain. Alterations of amino acids in rabbit brain caused by repetitive convulsive fits and comparison of amino acid contents in epileptic and non epileptic human brain, J. Biochem. (Tokyo), 49 (1961) 368-372.