Polyamine biosynthesis in primary tumors of human central nervous system: Review of current knowledge

Progress m Neuroblology Vol 25, pp. 289 to 295, 1985 Printed m Great Britain All rights reserved

0301-0082/85/$0 00 +0 50 Copyright © 1986 Pergamon Press Ltd

POLYAMINE BIOSYNTHESIS IN PRIMARY HUMAN CENTRAL NERVOUS SYSTEM: CURRENT KNOWLEDGE

TUMORS OF REVIEW OF

GIUSEPPE SCALABRINO*, MARIA ELENA FERIOLI* a n d GIOVANNI LUCCARELLIt

*lnstztute of General Pathology and C N.R. Centre for Research m Cell Pathology, Umverstty of Mdan t Neurologlcal Instttute " C Besta', 20133-Mdan, Italy (Recewed 15 June 1985)

Contents 1. 2. 3 4.

Introduction Levels of polyammes in CSF and m erythrocytes of patients with CNS tumors Levels of polyammes and polyamine biosynthetic decarboxylases m pnmary CNS tumors General conclusions References

289 290 292 293 293

1. Introduction

The polyamines putrescine, spermidine and spermine, which are ubiquitous organic cations of low molecular weight in all living organisms, are distributed in the different areas of mammalian central nervous system (CNS) (Kremzner, 1970; Russell et al., 1970; Shaskan and Snyder, 1973; Shaw and Pateman, 1973; Harik and Snyder, 1974; Seiler and Lamberty, 1975; Seiler and Schmidt-Glenewinkel, 1975; Halliday and Shaw, 1976). The levels of polyamines vary markedly between brain regions (Kremzner et al., 1970; Snyder et al., 1973). Areas with considerable white matter contain higher levels of spermidine, although this correlation with white matter is by no means perfect (Shimizu et al., 1964; Kremzner, 1973; Snyder et ai., 1973). These polyamines are known to be synthesized in human nervous tissues, because the presence of the four enzymes of the biosynthetic pathway of polyamines, i.e. L-ornithine decarboxylase (EC 4.1.1.17) (ODC), S-adenosyl-L-methionine decarboxylase (EC 4.1.1.50) (AMD), spermidine synthase (EC 2.5.1.16) and spermine synthase (EC 2.5.1.--) in mammalian CNS is now well documented (Kremzner et al., 1970; Shaskan et al., 1973; Raina et al., 1976). These polyamines can diffuse from human nervous tissues into cerebrospinal fluid (CSF) (for review see Shaw, 1979a). The structures for the three main polyamines and the scheme for polyamine biosynthesis in eukaryotic ceils are shown in Fig. 1. It has been demonstrated that interconversion of polyamines can also take place in mammalian CNS (Halliday and Shaw, 1976; Sturman et al., 1976; Seiler and Bolkenius, 1985). Although the physiological function of these amines is still not well understood at the molecular level, an abundant literature suggests that the concentrations of polyamines inside the eukaryotic cells are highly regulated and that polyamines play essential roles in cellular growth (whether normal or neoplastic) and differentiation (for review see Scalabrino and Ferioli, 1981, 1982). Like other mammalian cells, nerve cells, both normal and neoplastic, actively take up polyamines and metabolize them, as has been demonstrated by in vivo studies with radiolabeled polyamines (Russell et al., 1970; Shaskan and Snyder, 1973; Shaw, 1979b; Antrup and Seiler, 1980) and by in vitro studies with cultured cells (Chen and Liu, 1981; Chen and Rinehart, 1981). For details on the biochemical functions of polyamines and on their biosynthesis or catabolism in mammalian CNS, the reader is referred to other more ample recent reviews (Shaw, 1979a; Scalabrino and Ferioli, 1981, 1982). This review does not attempt to Address correspondence and reprint requests to: Dr. G Scalabnno, IsUtuto di Patologia Generale, Via Mangiagalli, 31, 20133-Milan, Italy. 289 J P N 25/4--A

290

G SCAI.ABRINO et a/

NH2

COOH I H2N-(CH2)3-CH-NH2

C02~CRVITdIVE

ORNITHINE

DECARBOXYLASE (ODC)

H2N-(CH2)4-NH2

OH OH

PUTRESCINE

S-ADENOSYL-L-METHIONINE (SAM)

SP~R~ZDZV~S~rHASE

~

N

OECARBOXYLATEO SAM (O-SAM)

~C02

/

SAM DECARB2XYLASE

(A,','O)

METHYLTHIOADENOSINE (MTA) CH3-S-Ad

H2N-(CH2)3-NH-(CH2)4-NH2

SPERMIDINE

SPERMINE SYNTHASE , , . ~ - ' - D- SAM

~

".---,~MTA

H2N-(CH2)3-NH-(CH2)4-NH-(CH2)3-NH2

SPERMINE

FIG 1. Scheme of the pathway of the bmsynthesls of the three mare polyammes m mammahan cells

elucidate the role of polyamines in human CNS tumors, but only to provide a brief survey of the literature on the polyamine content and metabolism in human CNS tumors and to briefly discuss the possible usefulness of polyamines and their biosynthetic enzymes in clinical oncology of CNS primary tumors. The topic of polyamines and the tumors of the human CNS is obviously part of the much broader topic of polyamines and human tumors as a whole (for review see Scalabrino and Ferioli, 1982), with many similarities and a few specific aspects. Generally speaking there are two different, although somehow complementary, ways to approach this problem. The first is the determination of the levels of the three main polyamines, putrescine, spermidine and spermine in CSF of treated or untreated patients with primary CNS tumors. To this regard during the last decade there have been a large number of articles (Marten, 1977, 1978, 1981a,b; Seidenfeld and Marten, 1979; Marten and Seldenfeld, 1981) and many of these may be considered relevant to this presentation. The second is the determination of the levels of the polyamines and/or of the polyamine biosynthetic decarboxylases, namely ODC and AMD, in tumor tissues obtained surgically from untreated patients with primary CNS tumors (Harik and Sutton, 1979; Scalabrino et al., 1982). Recently, a third approach has been tried, consisting of the determination of the polyamine levels in erythrocytes of patients with primary CNS tumors (Moulinoux et al., 1984). 2. Levels of Polyamines in CSF and in Erythrocytes of Patients with CNS Tumors

Since the human CNS is relatively isolated from the rest of the body and is immersed in CSF, enhanced production of polyamines in the nervous tissues frequently causes and is, therefore, mirrored by similar changes in the polyamine levels in CSF. However, this is a general statement and not without exceptions, since, as will be described later, there is no such correlation for many types of human CNS tumors (Marten, 1977; Seldenfeld and Marten, 1979). Furthermore, increases in the CSF polyamine levels, albeit to a lesser degree than in patients with CNS tumors, have been found in patients with a variety of non-neoplastic CNS diseases, such as infectious diseases, degenerative diseases and vascular diseases (Marten et al., 1976, 1979). Increased polyamines have also been found

POLYAMINF~ AND H U M A N CNS TUMORS

291

in blood and urine of patients with illnesses, whether neoplastic or not, in organs other than CNS, in some instances, even with diseases not characterized by cell proliferation (for review see Scalabrino and Ferioli, 1982). Therefore, it has become quite clear that overproduction of one or more polyamines is not strictly specific for tumor cells. Moreover, increased CSF or other body fluid polyamine levels might be due either to rapid growth of cells or to necrosis of cells, both of which are accompanied by a high leakage of polyamines. In normal human CSF the levels of putrescine and spermidine have been determined, while spermine is not quantifiable (Marton, 1978; Marton et al., 1979). Despite these and other limitations, polyamines have been widely considered to be helpful markers, alone or together with others, for the diagnosis of human brain tumors and for the evaluation of the degree of tumor malignancy, of tumor recurrence or of effectiveness of the therapy (Marton, 1977, 1978, 1981a,b; Seidenfeld and Marton, 1979; Marton and Seidenfeld, 1981). High levels of putrescme and spermidine have been found in the CSF of untreated patients with different types of malignant brain tumors, such as glioblastoma multiforme (grade IV astrocytoma), anaplastic astrocytoma (grade III astrocytoma), medulloblastoma, ependymoma and pituitary tumors (Marton, 1978; Marton et al., 1974, 1976, 1979; Fulton et al., 1980, 1982; Moulinoux et al., 1984). Interestingly enough, there is good correlation between patient status and CSF levels of putrescine and sometimes of spermidine for patients with medulloblastoma, although some few false-negative results have been reported (Marton, 1981a,b; Marton et al., 1979, 1981; Moulinoux et al., 1984). More importantly, in patients with this type of cancer, an elevation in CSF putrescine level may be the earliest indicator of tumor recurrence (Marton, 1981a,b; Marton et al., 1979, 1981; Phuphanich et al., 1985). The correlation between clinical status of the cancer patients and the CSF levels of polyamines (especially putrescine) was not always so good for patients with glioblastoma multiforme or with anaplastic astrocytoma, particularly when these CNS tumors were localized within the cerebral hemispheres (Fulton et al., 1980). It has recently been reported that patients with CNS tumors adjacent to the ventricular system or the subarachnoid space, i.e. to the CSF pathways, had significantly elevated CSF polyamine levels, correlated well with the degree of malignancy and with the size of the tumor (Fulton et al., 1980; Marton, 1981a,b; Pierangeli et al., 1981). While polyamines might, therefore, reflect the cell growth activity of brain tumor, their use diagnostically is limited when the compounds cannot reach the CSF. Studies of bram diffusion and capillary permeability of putrescine substantiate this (Pierangeli et al., 1981). Furthermore, the CSF levels of putrescine and spermidine frequently declined in patients with medulloblastomas or astrocytomas during favorable responses to the therapy followed by a clinical improvement and, in some instances, these levels become very close to those seen in reference patients with various non-neoplastic CNS diseases (Marton, 1978; Marton et al., 1976). Seidenfeld and Marton (1979) have collated the results of the studies of putative CNS tumor markers and have calculated the relative sensitivity and specificity of each marker and combinations thereof. These authors have concluded that desmosterol and polyamine levels in CSF can be of predictive value in monitoring recurrence and therapy progress. Nevertheless, attempts to correlate the CSF polyamine levels with the degree of malignancy of the different primary brain tumors and to the tumor size have failed to fully materialize, which is the major reason why determination of the CSF polyamine levels has never become a useful routine procedure for diagnosis and monitoring of primary human brain tumors. Interestingly enough, in the last few years abnormally high concentrations of polyamines (noticeably spermidine and spermine) have been significantly frequently observed in erythrocytes from patients with different types of neoplasia (for review see Scalabrino and Ferioli, 1982). Recently, this has been demonstrated to be true also for human primary CNS tumors (Moulinoux et al., 1984). In more detail, the spermidine levels of the erythrocytes of patients suffering from glioblastoma are significantly higher than those of healthy humans (Moulinoux et al., 1984).

292

G

St ALABRINO ('I al

3. Levels of Polyamines and Polyamine Biosynthetic Decarboxylases in Primary CNS Tumors Much more promising are the few studies in which polyamme contents of human brain neoplastic tissues from untreated patients have been assayed. All the data available, although limited in number for certain types of tumors, agree that CNS tumors generally have higher putrescine concentrations than those in normal human brain (Kremzner et a l , 1970; Harik and Sutton, 1979, Hank and Marton. 1981). Furthermore, the extent of this elevation in putrescine has been found to be a reliable indicator of the degree of malignancy (by conventional histopathologlcal criteria) of brain tumors of the astrocytoma group (Harik and Sutton, 1979). All the data available about polyamme contents m CSF of paUents with CNS tumors or in brain neoplastic tissues concur that the level of putrescine (i.e. the product of the reaction catalyzed by ornithme decarboxylase) is the most reliable and the only one which may be of some usefulness in clinical oncology of patients with primary CNS tumors On the basis of this, labeled putrescine has recently been proposed as a valuable tracer for location of brain tumors by positron emission tomography (Volkow et aL, 1983). Because of all the doubts about the significance of polyamine determinations in CSF from cancer patients and because of the futile question as to which polyamine it would be best to measure in each of the various types of human CNS neoplasla, we decided to measure the activities of the two polyamine biosynthetic decarboxylases, namely ornithme decarboxylase (ODC) and adenosylmethlonine decarboxylase (AMD), in different types of primary human CNS tumor t~ssues surgically exc~sed from untreated patients, and to correlate the levels with the degree of malignancy by conventional histopathoioglcal criteria (Scalabrino et al., 1982) In experimental oncology research with Morns rat hepatomas with vastly different growth rates (Williams-Ashman et al., 1972), and with chemical carcinogenesis in mouse skin (O'Brien et al., 1976; Boutwell et al., 1979), It has been found that the degree of ODC enhancement can be a useful biochemical indicator of neoplastic growth. Whenever possible we studied primary CNS tumors with well defined growth rates and grades of malignancy. The number of patients with each type of CNS tumor was large enough for the results to be evaluated by statable statistical tests. Our major finding was a correlation between the activities of polyamme b~osynthet~c decarboxylases (particularly ODC) and the growth rate and degree of malignancy of the neoplasm for most of the primary CNS tumors we tested. The ODC activity m the astrocytoma group significantly and progressively increased from mfratentorial pllocyt~c astrocytoma (grade I) to supratentorial glioblastoma multiforme (grade IV) through supratentorial astrocytoma of grades II and III (Scalabrmo et al., 1982). It is welt known that the growth rate ~s much slower for low-grade, well differentiated astrocytomas than for the high-grade, malignant gliomas, including glioblastoma multlforme (Steel, 1980). More importantly, medulloblastoma, a highly malignant brain tumor, had the highest level of ODC activity among all the primary CNS tumors of whatever histogenetic origin we tested (Scalabrino et al., 1982). The activity of AMD in the same types of brain tumors significantly and progresswely increased from the infratentorial pilocytic grade I astrocytoma to the supratentorial grade III astrocytoma, so that it, as well as ODC, paralleled quite well the degree of histopathological malignancy (Scalabrlno et al., 1982). However, there was no further increase in AMD activity from supratentorial grade III astrocytoma to glioblastoma multiforme (grade IV). Therefore, the increase in AMD activity in the astrocytoma group tumors was not entirely s~milar to that in ODC actwity. The surprising finding was. however, that in medutloblastomas there is a clear dichotomy between the levels of the two polyamine biosynthetic decarboxylases, with the highest ODC levels and the lowest AMD levels of any of the human brain tumors of neuroeplthehal tissue we have tested (Scalabrino et al., 1982). We also wish to emphasize our results for various types of human meningiomas. When, as suggested by Rubinstein (1972), we divided the meningiomas into atypical and typical forms on the basis of the presence or absence of mitotic figures,

POLYAMINESAND HUMAN CNS TUMORS

293

regardless of histologic variants, we found that ODC levels in the atypical forms were significantly higher than those in the typical forms, without corresponding significant differences in the AMD levels (Scalabrino et al., 1982). It should be recalled that it has been claimed that the incidence of mitotic figures in tumors of meningeal tissues is an important criterion for assessment of the likelihood of recurrence and a rough prognosticator for rapid growth (Crompton and Gautier-Smith, 1970; Harik and Sutton, 1979). Furthermore, a good correlation has recently been reported (Mondovi et al., 1983) between the degree of CNS tumor malignancy and levels of activity of diamino oxidase, which is involved in polyamine catabolism (Scalabrino and Ferioli, 1981), for astrocytoma tumors but not for medulloblastomas, which have surprisingly little enzyme activity (Mondovi et al., 1983). 4. General Conclusions

When we compare the usefulness of polyamine contents in CSF and CNS tumor specimens with that of activity of the polyamine biosynthetic decarboxylases in CNS tumors after their surgical removal, for diagnosis and, even more importantly, prognosis in the clinical oncology of CNS primary tumors, we feel that: (1) measurements of CSF polyamine levels can be useful for short-time evaluation of a specific course of therapy and for detection of remission or relapse of the neoplastic disease, but are of little or no use for evaluating the degree of malignancy of the tumor, particularly when used without other CNS tumor markers and (2) the activities of the polyamine biosynthetic decarboxylases, especially ODC, arc by far better indicators of the degree of malignancy of the tumor, but are of no use for evaluating the effectiveness of the therapy, because, obviously, repeated samples can not be obtained. Therefore, the two polyamine biosynthetic decarboxylases, especially ODC, arc important tools to use for study not only of the CNS ontogeny (Anderson and Schanberg, 1972; Schmidt and Cantoni, 1973; Shaskan et al., 1973; Sturman and GauU, 1974; Gilad and Kopin, 1979; Slotkin, 1979; Laitinen et al., 1982; Ruel et al., 1984) but also of human CNS tumors. It is important to mention here that mammalian CNS shows a peculiar ontogenic pattern of the activities of the two polyamine biosynthetic decarboxylases, because only ODC is very active during embryonal life and only AMD is very active during adult life (Anderson and Schanberg, 1972, 1975; Schmidt and Cantoni, 1973; Shaskan et al., 1973; Grillo et al., 1983). From the point of view of the levels of the polyamine biosynthetic decarboxylase activities, medulloblastoma, which shows a very high level of ODC activity and a very low level of AMD activity, is confirmed to be a typical embryonic tumor of CNS. The pivotal role of ODC in studies of the growth of and differentiation processes in the CNS is further stressed by the recent finding that specific attenuation of the levels of ODC activity in neuroblastoma cells is temporally related to the differentiation of this cell line when cAMP analogs are added to the culture medium (Chen et al., 1982). We hope this review will stimulate further studies of the levels, functions and properties of polyamine biosynthetic decarboxylases in primary human CNS tumors. References ANDERSON, T. R. and SCHANBERG,S. M (1972) Ormthine decarboxylase activity in developing rat bram. J. Neurochem. 19, 1471-1481. ANDERSON,T. R. and SCHANBERG,S. M. (1975) Effect of tyroxine and corttsol on brain ornithme decarboxylase activity and swimming behavior in developing rat. Bioehem. Pharmac. 24, 495-501. ANTRUP, H and SELLER,N. (1980) On the turnover of polyamines sl~rmidinc and spermine in mouse brain and other organs. Neurochem. Res 5, 123-143. BOUTWELL, R K., O'BRmN, T. G., VER~U~,A. K , WEEK~, R. G., DE YOUNG, L. M., ASnENDEL, C. L. and ASTRUP, E. G. (1979) The induction of ornithine decarboxylase activity and its control m mouse skin epidermis. Adv. Enzyme Reg. 17, 89-112. CHEN K. Y and LIU, A. Y. C. (1981) Differences in polyamine metabohsm of the undifferentiated and differentiated neuroblastoma cells. FEBS Left. 134, 71-74. CrmN, K Y. and RI~mI~A~T, C. A. JR. (1981) Difference m putrescine transport in undifferentiated versus differenttated mouse NB-15 neuroblastoma cells. Biochem. biophys. Res. Commun 101, 243-249.

294

G

S t` ALABRINOet a/

CHEN, K Y, PRESEPE, V , PARKEN. N and LIU, A Y O 11982) Changes of ornlthlne decarboxylase activity

and polyamlne content upon dlfferentmtlon of mouse NB-15 neuroblastoma cells J cell Phi',*lol 110. 285-290 CROMPTON, M R and GAUTIER-SMITH, P C 11970) The prediction of recurrence m menlnglomas J ~eurol Neurosurg Psvchtat 33, 81~87 FULTON, D S, LEVIS, V A , LUSlCH, W. P, WILSON, C B and MAgTON, L J. 11980) Cerebrosplnal fluid polyamlnes m patients with ghoblastoma multlforme and anaplasUc astrocytoma Cancer Res. 40, 3293- 3296 FULTON, D A , MARTON, L J , LUBICH, W P and WILSON, C B (1982) Polyamlne levels m CFS from patients with pituitary tumors or nonneoplastlc pituitary disease Archs NeuroL 39, 47-48 GILAD. G M. and KOPIN, i J 11979) Neurochemlcal aspects of neuronal ontogenests m the developing rat cerebellum' Changes in neurotransmltter and polyamme synthesizing enzymes J Neurochem 33, 1195-1204 GRILLO, M A , FOSSA,T and DIANZANI, U (1983) Argmase, ormthlne decarboxylase and S-adenosylmethlonme decarboxylase in chicken brain and retina Int J. Btochem 15, 1081-1084. HALLIDAY, C A and SHAW, G G 11976) The distribution and metabolism of putresctne, spermldme and spermlne injected Into the cerebral ventricles of rabbits J Neurochem 26, 1199-1205 HARIK, S I and MARTON, L J (1981) CSF putresclne levels The enzymatlc-lsotoplc method vs liquid chromatography Archs Neurol, 38, 91-94 HAR1K. S I and SNYDER, S H (1974) Putresclne regional distribution in the nervous system of the rat and the cat Brain Res 66, 328 331 HARIK, S I and SUTTON. C H (1979) Putresclne as a biochemical marker of malignant brain tumors Cancer Res 39, 5010-5015 KREMZNER, L T (1970) Metabolism of polyamlnes in the nervous system Fedn Proc 29, 1583-1588 KREMZNER, L T (1973) Polyamme metabolism in normal and neoplastic neural tissue In Polyammes in Normal and Neoplastic Growth, pp. 27-40 Ed D H. RUSSELL Raven Press, New York. KREMZNER, L. T , BARRETT,R E and TERRANO, M. J (1970) Polyamlne metabohsm in the central and peripheral nervous system Ann N Y Acad Scl 171, 735-748 LAITINEN, S I., LAITINEN, P H , HIETALA, O A , PAJUNEN,A E. I. and PIHA, R S (1982) Developmental changes in mouse brain polyammes metabolism Neurochem. Res 7, 1477-1485 MARTON, L J (1977) Polyammes and brain tumors Sam Cancer Inst. Monogr. 46, 127-131 MARTON, L J 11978) Potential of cerebrosplnal fluid polyamme determinations in the diagnosis and therapeutic monitoring of brain tumors. Adv Polyamine Res 2, 257-263 MARTON, L J (1981a) Polyammes Relation to brain tumor therapy and monitoring Cancer Treat Rep 65 (Suppl 2), 107-108 MARTON, L J (1981b) CSF polyamlnes. Potential as brain tumor markers Archs Neurol 38, 73 74 MAgTOS, L J and SE1DENFELD, J (1981) Approaches to the study of polyammes as cancer markers In: Polyam,nes m Biology and Medicine, pp 337-348 Eds D R MORRIS and L J MARTON Marcel Dekker, Inc, New York MARTON, L J , HEBY, O and WILSON, C B (1974) Increased polyamme concentrations m the cerebrospinal fluid of patients with brain tumors Int J Cancer 14, 731-735 MARTON, L J , HEBY, O , LEVIS, V A , LUmCH, W P , CRAFTS, D. C and WILSON, C B (1976) The relationship of polyamines In cerebrospmal fluid to the presence of central nervous system tumors Cancer Res 36, 973-977 MARTON, L J , EDWARDS, M. S, LEVIS. V A , LUBICH, W P. and Wlt,Sos, C B 11979) Predictive value of cerebrosplnal fluid polyamlnes m medulloblastoma. Cancer Res. 39, 993-997 MARTON, L. J , EDWARDS, M. S, LEVIS, V. A , LuBICt-I, W. P. and WILSOS, C B (1981) CSF polyamInes: A new and important means of monitoring patients with medulloblastoma. Cancer 47, 757-760. MONDOVt, B, RICCIO, P , RIccIo, A and MARCOZZl, G S (1983) Amine oxldase activity in malignant human brain tumors. Adv Polyamme Res 4, 183-191. MOULINOUX, J P, QUEMENER,V , LE CALVE, M , CHATEL, M and DARCEL, F 11984) Polyammes in human brain tumors. J Neuro-Oncol 2, 153-158. O'BRIEN, T G (1976) The Induction of ormthme decarboxylase as early, possibly obligatory, event in mouse skin cancerogenesls Cancer Res 36, 2644-2653 PHUPHANICH, S. LEVIS. V A., LUBICH, W P and MARTON, L J (1985) An update on the use ofCSF polyamlne levels to monitor patients with recurrent medulloblastoma Neurology 35 (Supp/ 1), 112-113 PIERANGEL1, E , LEVIS, V. A , SE1DENFELD, J. and MARTON, L J. (1981) Putrescine diffusion In cat brain and capillary permeablhty in rat brain Relation to CSF putrescine levels in brain tumor patients Fur J Cancer 17, 143-147 RAINA, A , PAJULA, R L and ELORANTA,T 11976) A rapid assay method for spermldme and spername synthases. Distrlbutmn of polyamine-syntheSlzlng enzymes and methlonine adenosyltransferase in rat tissue FEBS Lett 67, 252-255 RUBINSTE1N, L J (I 972) Tumors of the central nervous system. In. Atlas of Tumor Pathology, 2rid series, Fascicle 6. pp 169-190 Armed Forces Institute of Pathology, Washington DC RUEL, J , CHI~NARD, C., COULOMBE, P and DUSSAULT, J. H. (1984) Thyroid hormones modulate ornlthme decarboxylase in the immature rat cerebellum. Can. J. Physm/. Pharmac. 62, 1279-1283 RUSSELL, D. H., MEDINA, V. J. and SNYDER, S. H (1970) The dynamics of synthesis and degradation of polyammes in normal and regenerating rat hver and brain. J btol Chem. 245, 6732-6738 SCALABRINO, G and FERIOLI, M. E (1981) Polyamines in mammalian tumors Part I Adv. Cancer Res 35, 151-268.

SCALABRINO, G and FERIOLL M E 11982) Polyammes in mammalian tumors. Part II. Adv Cancer Res 36, 1-102 SCALABRINO, G , MODENA, D , FERIOL1, M E., PUERARI, M. and LUCCARELLI, G. (1982) Degree of mahgnancy in human primary central nervous system tumors' Ornithme doearboxylase levels as better indicators than adenosylmethlonine decarboxylase levels J Sam Cancer Inst. 68, 751-754

POLYAMINF~ AND HUMAN CNS TUMORS

295

SCHMIDT, G. L. and CANTONI, G. L. (1973) Adenosylmethionine decarboxylase m developing rat brain. J Neurochem. 20, 1373-1385 SEIDEN~LD, J. and MARTON, L. J. (1979) Biochemical markers of central nervous system tumors measured m cerebrospinal fired and their potential use m diagnosis and patient management: A review J. Nam. Cancer Inst. 63) 919-931. SELLER, N. and BOLKENIUS, F. N. 0985) Polyamme reutihzation and turnover in brain. Neurochem. Res. 10, 529-544. SELLER,N. and LAMBERTY,U. (1975) Interrelations between polyamlnes and nucielc acids: Changes of polyamme and nucleic acid concentrations in the developing rat brain. J Neurochem. 24, 5--13. SEILER,N. and SCHMIDT-GLENEWINKEL,T. (1975) Regional distribuuon of putrescme, spermldme and spermme in relation to the distribution of RNA and DNA in the rat nervous system J. Neurochem 24, 791-795 SHASKAN, E, G. and SNYDER, S. H (1973) Polyamine turnover in different regions of rat brain J. Neurochem. 20, 1453-1460 SHASKAN, E. G., HARASZTi, J. H. and SNYOER, S H. (1973) Polyamines: Developmental alterations in regional disposition and metabolism m rat brain. J. Neurochem. 20, 1443-1452. SHAW, G. G. (1979a) The polyamlnes m the central nervous system. Bzochem. Pharmac. 28, 1-6. SHAW, G. G (1979b) The synthesis and turnover of spermidlne and spermme m mouse brain. Neurochem. Res. 4, 269-275. SHAw, G G. and PATEMAN, A. J. (1973) The regional distribution of the polyammes spermidine and spermine in brain J. Neurochem. 20, 1225-1230. SmMIZU, H., KAKIMOTO,Y. and SANO, I. (1964) The determination and distribution of polyamines in mammalian nervous system. J. Pharmac. exp. Ther. 143, 199-204. SLOTKIN, T A. (1979) Ornlthine decarboxylase as a tool m developmental neurobtology. Life Scz. 24, 1623-1630. Sr,,VOER, S. H , SHASKAN, E. G. and HARIK, S. I. (1973) Polyamine disposition m the central nervous system In' Polyam~nes m Normal and Neoplastic Growth, pp. 199-213. Ed D. H. RUSSELL. Raven Press, New York. STEEL, G. G 0980) Growth kinetics of brain turnouts In. Bra,n Tumors, pp. 10-20 eds. D G. T THOMASand D. I. GRAHAM. Butterworths, London. STURMAN, J A and GAULL, G. E. (1974) Polyamme biosynthesis in human fetal liver and brain. Pediat. Res 8, 231-237. STURMAN, J. A , INGOGL1A, N. A. and LINDQUIST, T D. (1976) Interconversion of putrescine, spermldine and spermme m goldfish and rat retina Life Sci 19, 719-724 VOLKOW, N., GOLDMAN, ,% S., FLAMM, E. S) CRAVIOTO, H., WOLF, A. P. and BRODIE, J. D. 0983) Labeled putrescine as a probe m brain tumors. Science 221,673-675. WILLIAMS-ASHMAN,n . G., COPPOC, G. L. and WEaER, G. (1972) Imbalance m orntthine metabolism m bepatomas of different growth rates as expressed tn formation of putrescme, spermidme and spermine Cancer Res 32, 1924-1932.