Polyamines in mammalian ageing: An oncological problem, too? A review

Mechanisms of Ageing and Development, 26 (1984) 149-164

149

Elsevier Scientific Publishers Ireland Ltd.

POLYAMINES IN MAMMALIAN AGEING: AN ONCOLOGICAL PROBLEM, TOO? A REVIEW

GIUSEPPE SCALABRINO* and MARIA E. FERIOL1 Institute of General Pathology and CNR Centre for Research in Cell Pathology, University of Milan, Via Mangiagalli 31,20133 Milano (Italy)

(Received October 14th, 1983) (Revision received February 14th, 1984) SUMMARY This review surveys the literature about changes in polyamine contents and levels of activity of the enzymes involved in the polyamine biosynthetic pathway in organs of ageing mammals. The literature about changes in the polyamine levels in physiological fluids in healthy ageing humans is also reviewed. Generally speaking, decreases in the levels of the main polyamines (noticeably putrescine and spermidine) are observed in different mammaliar~ organs with ageing. The polyamine levels in serum and in urine of healthy human beings are also age-related, declining progressively with increasing age. Some major enzymes (i.e., ornithine decarboxylase (EC 4.1.1.17) and S-adenosyl-Lmethionine decarboxylase (EC 4.1.1.50)) involved in the polyamine biosynthetic pathway show similar trends. Hormonal induction of omithine decarboxylase activity is strongly reduced in organs of aged animals, as it is in neoplastic organs. There is also some evidence for an age-related decrease in the level of ornithine decarboxylase and its inducibility in mammalian cells cultured in vitro. Some in vitro effects of spermidine and spermine on aged structures or systems are briefly summarized. There is no evidence yet that this generally reduced capacity of mammalian aged organs for polyamine biosynthesis is one of the factors responsible for the well known high incidence of some neoplasias in elderly humans. In view of the typical stimulatory effects of the tumour promoters on polyamine biosynthesis in target tissues and the effects of senescence on the same metabolic pathway, it can be excluded that the ageing process has a tumour promoting effect by itself. However, although the exact mechanism responsible for the increased occurrence of some tumors during mammalian senescence is still obscure, there are enough experimental data (both in humans and in animals) to indicate that the reduced polyamine biosynthetic capacity of aged mammals can account for the slower course of some tumors in elderly patients. *To whom correspondence and reprint requests should be addressed. 0047-6374/84/$03.00 Printed and Published in Ireland.

© 1984 ElsevierScientific Publishers Ireland Ltd.

150 K e y words: Ageing; Polyamines; Oncology; Ornithine decarboxylase Ut navem, ut aedificium idem destruit facillime qui construxit, sic hominem eadem optime quae conglutinavit natura dissolvit; iam omnis conglutinatio recens aegre, invererata facile divellitur. (Cicero, De Senectute, XX, 72) As the person who has built a ship or a house, likewise takes it down with the greatest ease, so the same nature which glued the human machine, takes it asunder most skilfully. Moreover, every fastening o f glue, when it is fresh, is with difficulty torn asunder, when old, very easily. (Ocero, On Old Age, XX, 72)

PREAMBLE The three main natural polyamines, putrescine, spermidine and spermine (for formulae, see Fig. 1), are ubiquitous in living entities and are crucial for all types of cellular growth (both restricted and unrestricted) and in the cell duplication process of prokaryotic and eukaryotic cells (for reviews, see refs. 1-3). In keeping with this, polyamines have recently been def'med, in a happy metaphor, as "a growth industry" [4]. During the past fifteen years, extensive literature has accumulated on the close, although non-specific, connections between the polyamines and cancer, in both animals and humans (for reviews, see refs. 1 and 2). These simple but fundamental concepts are firmly established and known to even those scientists who are not well acquainted with the topic of regulation of cellular growth. If we borrow the terminology of the physicists, who divide forces into positive and negative, all the different developmental processes for mammals that follow in sequence during the life span can be reasonably divided into a "positive" phase, from conception to adulthood, and a "negative" phase, senescence, which ends in death. This is an oversimplified and schematic division because both of these phases have some common features. They have predetermined timetables, albeit variable for different organs and for different species, and they are both genetically programmed [5-7]. It is known that the phase of positive development is regulated by coordination of expression and repression of different genes, while the phase of negative development is regulated essentially by progressive instability of genes or even their silence, or by the progressive accumulation of genetic errors in information-containing molecules. Since mammalian organisms are not able to renew totally and indefinitely all vital cell structures, senescence is predictable, pre-set in the genome [ 5 - 7 ] . Therefore, it seems to us appropriate to consider the pattern and the metabolism of polyamines in mammalian ageing not only as a problem specific to the biochemistry of gerontology, but also as something included in and complementary to the much broader area of the biochemistry of development of mammalian cells and organs. Ageing and cancer are two phenomena related to each other in some aspects, although the relationship between these two processes is not yet clear [8-15]. The main biochemical similarities and the main biochemical differences between senescence and neoplasia

151 have been listed and analyzed in the reviews of Pitot [8,9] and of Hocman [10]. Among the many points of similarity between ageing and cancer, suffice it to mention that both ageing and the incidence of cancer increase with increasing duration of life-span and, what is much more relevant, that both these biologic processes appear to be genetically programmed, although for cancer we have data to support this at present only for some mammalian neoplasias. Interestingly enough, in recent studies on the cellular transforming genes, namely the so-called oncogenes, a number of genes in normal eukaryotic cells with potential oncogenic activity have been identified [ 16-18]. A genetic basis for high susceptibility to chemical carcinogens is also indicated by the finding that high inducibility of the cytochrome P-450 enzyme aryl hydrocarbon hydroxylase with carcinogens is associated with increased occurrence of bronchogenic carcinoma [19]. This review has different aims. First, we survey the results of all studies available, both in rive and in vitro, in which polyamine content and/or metabolism in mammalian ageing has been investigated. Secondly, we discuss whether or not the data about polyamine metabolism in aged mammalian organs shed further light on the connections between senescence and cancer, although most of our conclusions on this topic can only be speculative. Lastly, our review also fills the lacunae in the above-mentioned reviews [8-15] and in the others available on the biochemical changes in mammalian senescence [20-28] that do not mention the polyamines. BASICCONCEPTSIN THE BIOCHEMISTRYAND BIOSYNTHESISOF POLYAMINES Polyamines are strongly cationic compounds synthesized in virtually all animals, plant and bacterial cells and present as such or as their metabolic conjugates in physiological fluids [29-34]. The major polyamines in mammals are spermidine and spermine and their precursor, putrescine. The polyamine pattern varies markedly from one species to another, and in a given species it differs greatly in various tissues and also depends on the growth conditions, growth rate, age, etc., of the tissue [1,2,29-34]. Characteristic patterns of biosynthesis and accumulation of polyamines are maintained during growth and differentation in various types of cells and tissues [1,2,29-34]. High concentrations of polyamines, however, can also be found in non-growing tissues, such as pancreas, prostate gland and lactating mammary gland, that are active in protein synthesis [34,35]. Because of their structural characteristics, polyamines are able to interact with nucleic acids, proteins and phospholipids through ionic forces and hydrogen or hydrophobic bonds. Although the importance of polyamines in many biological processes, such as cell proliferation, DNA replication, RNA synthesis, translation, transport and membrane reactions, has been shown, many aspects of polyamine action at the molecular level and their relations to chemical structure are still poorly understood [29-34]. The most important functions of natural polyamines in intermediary metabolism are related to the metabolism of nucleic acids. As strong bases, polyamines have strong affinity for the nucleic acids and through specific binding they stabilize the nucleic acids against denaturation and digestion by nucleases. They also have some regulatory functions in

152 the synthesis of nucleic acids and proteins, they enhance the synthesis of RNA by DNAdependent RNA polymerase, stabilize and maintain the association of the ribosomal subunits and promote the attachment of ribosomes to endoplasmic reticulum membranes [36,37]. Furthermore, polyamines stimulate the methylation of tRNA, increase the aminoacylation of tRNA and enhance the binding of aminoacyl tRNA to ribosomes [36,37]. Polyamines induce in vitro the interconversion of tRNA from the inactive to the active configuration [38]. The various species of ribonucleic acids (mRNA, rRNA and tRNA) are essential constituents of the protein synthesis machinery. Therefore, any conformational changes in these molecules induced by interaction with intracellular polyamines could be crucial in the regulation and in the correctness of the translation process. Furthermore, it has been demonstrated with cell-free systems from bacterial and mammalian cells that the three main polyamines increase the overall rate of polypeptide synthesis and partially replace the Mg2. required for this process [36,37]. Almost every step in polypeptide synthesis can be stimulated in vitro by the addition of an appropriate amount of putrescine, spermidine or spermine [36-38]. Therefore, it is quite clear that the two roles, structural and metabolic, of polyamines inside the cells are connected. Likewise, it is clear why, with time, the polyamines have come to be considered ever more important as growth stimulating factors for both microorganisms and animal ceils. Although the rise in polyamine content in growing cells often precedes the increase in DNA and RNA content, it is still an open question whether or not these increases are causally linked or, conversely, simply temporally coordinated each another. However, the idea that the mere measurements of the concentrations of one or more of the polyamines in the tissues were enough to unravel the physiological significance of these substances in living organisms, was quickly recognized as an oversimplification. As a consequence of this, much effort has been spent on studies of the enzymes of the polyamine biosynthesis pathway, to describe enzyme changes in various experimental models, some connected, some not connected, with cell growth. Four enzymes are known to be involved in the biosynthesis of polyamines in eukaryotic cells: two decarboxylases and two synthases, also called aminopropyltransferases. The scheme for polyamine biosynthesis in mammalian cells is shown in Fig. 1. Many microorganisms and higher plants are also able to produce putrescine from agmatine by decarboxylation of arginine, but all mammalian ceils produce putrescine only by enzymic decarboxylation of ornithine [36]. Ornithine decarboxylase (EC 4.1.1.17) (ODC) is the rate-limiting enzyme in polyamine biosynthesis in mammalian cells [39-42]. Generally speaking, it is present in very small amounts in quiescent cells, and its activity can be increased manyfold within a few hours of exposure to such different stimuli as hormones, drugs, and growth factors, or during tissue regeneration [1,2,29-34]. Putrescine is converted to spermidine by addition of an aminopropyl group derived from enzymic decarboxylation of S-adenosylL-methionine by S-adenosyl-L-methionine decarboxylase (EC 4.1.1.50) (SAMD). Like ODC activity, SAMD activity is regulated by many hormones and other growthpromoting stimuli [1,2,29-34]. The transfer of the aminopropyl group from decarboxylated S-adenosyl-L-methionine to putrescine is catalysed by spermidine synthase

153 NHa(CHa)3 CH (NH a) COOH ..~ CH3CONH(CHa)3CHO---.,..C.ABA L-ornithine CH3CONH (CHa)4 NH~"~ N-acetyl-y-zmmobutyraldehyde ~,~ ( ~ 4m°n°-N- acetylputrescine

CH3CONH(CH2)3NH(CH2)4NHa I1'- acetyispermidint ~

C02a~cetyl_Co k ~ ~...~'NH a(CH2)4NH~ -.........~ ~ putrescin, ~ ( , ~ /~

2

m

a

*

h

,

0

CH3-S-Ad 5'- methyl- th,oadenosina Ad CH3-S-(CH2)2CH(NH2)COOH SO_adenosyl-L- meth,o....

,

,

0

a

f

CH:~CONH(CHa)3NH (CHa)4NH(CHa)3NHa

/ acetyl- CoA

NHa(CHal3NH(CH214NH (CH;~13NHa sparmine

CHI-S-Ad 5"-me*hy[- thioadenosine

Ne-(Z-carboxyethy[)-spermidine ; splrmic acid ~) (~) (~) (~ (~) (~) (~)

L-ornilhine dlcarboxyiase (E[ /,1117) S-zdanosyl- L-m/lhionine d/carboxylasa( [[ /, 11 50) spermidma synthasl ([C 251.16) spermine synthasa spermidine/spermine N~-acetyltransferasa polyamine oxidase acatyl-roA- 1./,o diaminobufane N-zcalyl*ransferase

Fig. 1. Scheme of the pathways of the biosynthesis,catabolism and interconversionof the three main polyaminesin mammaliancells. (EC 2.5.1.16). Another aminopropyl group is added to spermidine to form spermine, and this reaction is catalysed by a second aminopropyltransferase termed spermine synthase [32]. The other product of both the aminopropyltransferase reactions is 5'-methylthioadenosine, which is produced in amounts stoichiometric with spermidine or spermine, but it is very rapidly split by a phosphorylase to produce adenine and 5'-methylthioribose-1 -phosphate [32,34]. Although the spermidine synthase and spermine synthase reactions are effectively irreversible, it has been demonstrated that polyamine interconversion is a physiological event which takes place by the successive action of two enzymes, spermidine-N 1-acetyltransferase and polyamine oxidase [43,44]. The first and probably rate-limiting step of the polyamine interconversion pathway is acetylation in the NX-position by a cytosol enzyme, ~e. the Nl-acetyltransferase. The Nl-acetylpolyamines produced by this reaction are in tum substrates for the cytoplasmic polyamine oxidase, which cleaves the polyamine at the internal nitrogen to yield N-acetylpropionaldehyde and putrescine or spermidine depending on the substrata [43,44]. Acetylated derivatives of the chief polyamines are found normally in blood and urine in small amounts [1,2]. The scheme of polyamine interconversion is also presented in Fig. 1. Three of the enzymes described above, ODC, SAMD, and spermidine NLacetyltransferase, have very rapid rates of turnover [ 1,29-34]. ODC has been shown to have a very

154 short half-life and together with the other two enzymes appear to be the key regulator of putrescine and polyamine content in mammalian tissues [1,2,29-34]. The short halflives of these three enzymes permit the rapid change in their activity in response to stimuli that has been observed in many situations [1,2,29-34]. The two polyamine synthases of mammalian tissues have considerably higher activities than ODC and SAMD and their half-lives are much longer [1,2,29-34]. It is therefore unlikely that these two synthases are part of the control mechanism for polyamine synthesis and their levels. For additional details on the biochemical functions of polyamines and on their biosynthesis or catabolism, the reader is referred to other more ample recent reviews [1,2,29-34,39-45]. CHANGES IN POLYAMINECONTENTOF ORGANS AND CELLS DURING AGEING As mentioned above, there is a large body of evidence that the polyamines play an essential role in cell growth and differentiation. Consequently, since it is well known that protein synthesis and metabolism decrease in mammalian organisms with advancing age, it is reasonable to expect the polyamine content to be decreased in the tissues of aged animals compared with those of young animals. Markedly decreased spermidine contents have been shown in several different organs, such as liver, thymus, kidney, spleen, brain and skeletal muscle of aged rats [46-48]. The decrease is generally proportional to the increase in age [46-48]. On the other hand, the spermine content of the chief organs of aged rats is quite erratic, since it is even increased in some organs (liver, kidneys, spleen, thymus), unchanged in other organs and decreased in brain [46,47,49]. The putrescine content of the liver of aged rats decreases with increasing age of the animal [50,51]. For the polyamine content in the prostate gland of aged rats, see the next section, in which polyamine contents are correlated with the levels of the activities of the two polyamine biosynthetic decarboxylases. Erythrocytes are a good model for studying the cellular age-dependent changes, since red blood cells of different ages are present in the blood at any moment. Studies with normal human erythrocytes separated from blood taken from healthy subjects have demonstrated that there are markedly decreasing concentrations of all three main polyamines in aged erythrocytes compared with the young ones (reticulocyte rich) [52,53]. Thus, the results of studies with whole aged organs (with or without the capacity of cell renewal) and with circulating cells, such as blood cells, agree at least in showing a decrease in putrescine and spermidine levels with increasing age. For the polyamine contents of blood cells of children with progeria, see the paragraph devoted to the polyamine levels in the physiological fluids of aged human subjects. CHANGES IN LEVELS OF THE ENZYMES OF THE POLYAMINE BIOSYNTHETIC PATHWAY DURING AGEING From the few data available on the polyamine content of some organs of aged rats, one can once again see that the main enzymes involved in polyamine biosynthetic and

155 catabolic pathways show similar trends, with their activities decreasing with increasing age of the animal. This has been demonstrated to be true for ODC, SAMD, spermidine synthase and diamine oxidase in some organs, such as thymus, spleen, liver and small intestine of rats of various ages [54,55]. ODC is also decreased in the heart, lung and cerebral cortex of aged rats [48]. This decrease in ODC activity is apparent in both the nuclear fractions and the soluble fractions of all the above organs, except for the nuclear fraction of the lung, in which it surprisingly increased with ageing [48]. Being an androgen-responsive organ, the rat prostate has been widely used as an excellent experimental model for investigating not only hormone-induced tissue growth, but also various possible physiological functions of the main natural polyamines. The ventral lobe of the rat prostate is one of the richest sources among rat tissues of the major natural polyamines and of their biosynthetic enzymes. However, the tissue model based on rat ventral prostate has the disadvantage that this part of the gland also secretes polyamines into seminal fluid. In contrast with rat ventral prostate, considerably smaller amounts of polyamines are found in the dorsolateral lobe of the gland. In the ventral part of the prostate gland of old rats there are decreases in ODC activity [56,57] and in SAMD activity [57]. The ODC activity of the dorsolateral part of the prostate gland of the rat is too low to be quantified with sufficient precision and consequently it is not possible to evaluate the changes due to the ageing of the animal [57]. As for SAMD activity, in the dorsolateral and in the ventral part of the rat prostate there are age-related decreases [57]. These decreases in the activities of both decarboxylases are not attributable to age-related changes in their properties, such as the stability of the enzymes, their K m values, their requirement for effectors (pyridoxal phosphate and putrescine), the changes in their half-lives [57]. The age-related decreases in ODC and SAMD activities in the prostates of aged rats are also not due to the presence of inhibitors or inactivators of these enzymes [57]. Another study by the same authors [58] demonstrated that the rate of inactivation of ODC or SAMD activity in the'ventral prostate of old rats was comparable to that in the ventral prostate of young rats. Further, based upon immunological titration of SAMD, Shain e t al. [59] demonstrated that the quantitative relationship between the enzymatic activity of SAMD and the antigenic mass of SAMD in prostate is the same for young mature rats and for aged rats. This strongly argues for a quantitative decrease in the prostatic content of active and inactive SAMD protein being responsible for the age-related decrease in prostatic SAMD activity [59]. All these results strongly support the idea that the age-related decreases in prostatic ODC and SAMD activities in the rat may be at least partially the result of an alteration in the regulation of prostatic gene activity by androgens. In fact, Shain and Moss [57] have identified three types of age-related decrease in polyamine decarboxylase activities in rat prostate in relation to the response to treatment with exogenous testosterone: (1) a decrease fully reversible by exogenous testosterone; (2) a decrease partially reversible by exogenous testosterone; and (3) a decrease not reversible by exogenous testosterone. Surprisingly enough, these enzymatic decreases in ~,. prostates of old rats are not closely correlated with the changes in the polyamine contents of the gland. In fact, whereas the putrescine content in the ventral part of the prostate gland of aged rats was lower than in young ones,

156 spermidine and spermine contents of the same part of the organ of aged rats were actually increased [60]. The small increase in total polyamine amount observed in the ventral prostate of old rats is mainly due to a progressive rise in spermine and, to a lesser degree, to the lesser rise in spermidine with increasing age [61 ]. On the other hand, in the dorsolateral part of prostates of old rats the putrescine level was unchanged, while both spermidine and spermine contents were diminished [60]. Therefore, it seems that the two anabolic polyamine decarboxylases are not the main regulators of the polyamine content of the prostate in aged rats. INDUCTION OF THE ACTIVITIES OF THE POLYAMINE BIOSYNTHETICDECARBOXYLASES IN ORGANS OF AGED ANIMALS

I N VIVO

As has been above mentioned, ODC and SAMD activities can be easily induced in the target organs by many different stimuli. It has been demonstrated that induction in the livers of old rats requires a longer time and is quantitatively less than in livers of young animals. This was observed in regenerating rat liver, in which ODC induction begins later in aged rats than in young adult rats [62]. Furthermore, in young rats the second peak of ODC activity in the regenerating liver remnant occurred almost immediately after the first peak, whereas in older animals there was a delay of several hours before the second peak [63,64]. As for SAMD activity in rat liver regeneration, in old rats there was no detectable increase in SAMD activity at any time after partial hepatectomy [65,66]. Amazingly enough, the time course of ODC induction by hormones in the livers of old rodents seems to differ between species and in relation to the hormone used. In the rat, ODC induction after dexamethasone administration is quantitatively and temporally similar in young and in old animals [55], whereas growth hormone, only when administered at a low dose, causes a greater induction of hepatic ODC activity in weanling rats than in adult rats [67]. In the mouse the peak of ODC induction after growth hormone administration is reached later in old than in young animals [68]. In the ventral prostate of aged rats a given dose of testosterone c a u l s , over the same time interval, a significant increase in ODC activity but not in SAMD activity [57]. Notably, the level of ODC activity in aged ventral prostate after testosterone treatment was only about half the level in the ventral prostates of untreated young rats [57]. Additonaliy, testosterone causes an induction of dorsolateral prostate SAMD activity [57]. Endocrine regulation of the activities of the polyamine biosynthetic decarboxylases in the prostate gland of aged rats is also demonstrated by changes in enzyme activities after orchiectomy. In young rats and in old ones, orchiectomy rapidly (within two days) causes marked decreases of both ODC and SAMD activities in both parts of the prostate [57]. The effect of orchiectomy on the prostatic polyamine content of aged rats requires much more time to become apparent than the effect on the decarboxylase activities, since decreased contents of some of the main polyamines are evident only by one week after the operation [60].

157 CHANGES IN ORNITHINE DECARBOXYLASE LEVELS IN MAMMALIAN CULTUREDCELLS WITH AGE OF THE CULTURE Tissue (cell) culture techniques have been used extensively to study the ageing process at the cellular level. It is convenient to consider two sorts of experimental approach to ageing at the cellular level. First, the limited lifespan of some cultured human cell lines in vitro has led to their extensive employment as a model system for studying human cellular ageing. However, there is some doubt about whether or not changes in diploid cell populations after serial passage in vitro accurately reflect cellular ageing in vivo [69]. An alternative cell system consists of cultures of cells taken from donors of different ages. The results of only one study of polyamine metabolism in mammalian cultured cells ageing in vitro are available thus far. This study shows that the onset of the in vitro cellular senescence of human embryonic lung fibroblasts is associated with a greatly reduced capacity of the older cells to increase ODC activity in response to the stimulus of fresh culture medium [70]. Furthermore, this decreased response of ODC activity is accompanied by a decreased growth rate of, and some morphologic changes in, the cells, indicating that the onset of cellular senescence can be recognized from many different parameters. Interestingly enough, increasing intracellular putrescine content does not increase the total number of human fibroblast doublings and does not delay the onset of the terminal phase of senescence after 50 -+ 10 doublings [which is the normal number of cultured human fibroblast doublings (the so-called "Hayflick limit") and therefore a major indicator of ageing of the culture], although the increased intracellular putrescine content markedly lengthens the survival time of the cultured cells without division capacity, namely the post-senescence phase time [70]. MISCELLANEOUS IN STRUCTURES

VITRO

EFFECTS OF POLYAMINES ON AGED SYSTEMS OR

The effects of spermidine and sperrnine on the in vitro incorporation of labelled leucine into microsomal protein from livers of rats of different ages have been examined [71]. When liver microsomes obtained from young rats are incubated with liver cell sap preparations obtained from old rats, spermidine stimulates the incorporation of labelled leucine into protein, whereas spermine has just the opposite effect. On the other hand, when liver microsomes obtained from old rats are incubated with liver cell sap preparations obtained from young rats, only spermine has a stimulatory effect on the incorporation of labeled leucine into protein [71]. The observation that spermine is active as stimulator only with "old" microsomes and, conversely, that spermidine has the same effect only with "young" microsomes, indicates specific regulatory roles of these polyamines in the various periods of life, at least at this level. The effects of spermidine and spermine on in vitro phosphorylation and acetylation of individual histones of the cerebral cortex of rats of different ages have been studied [72].

158 The stimulatory effect of these two polyamines on the phosphorylation of individual histones in the young animals decreased progressively with increasing age and no effect on any histone could be seen in the old rats [72]. The same is true for the enhancement of acetylation of the nucleosomal histones by spermidine and spermine, sin~ it aim decreased with increasing age of the rats [72]. Therefore, spermidine and spermine can act as modulators of the expression of specific genes, since it is well known that phosphorylation and acetylation of histones enhance the template activity of chromatin and DNA. Moreover, these results, together with those showing a general age-related modification in structure and function of chromatin [73], are consistent with the idea that alterations of content and of the biochemical functions of spermidine and spermine in ageing may be responsible, at least in part, for some age-dependent changes in the expression of genes. Interestingly enough, spermine, but not spermidine or putrescine, has a restorative effect on the oxidative and phosphorylative capacities of heat.aged isolated mitochondria of rat liver [74]. CHANGES IN POLYAMINE LEVELS IN THE PHYSIOLOGICAL FLUIDS OF AGED HUMAN SUBJECTS The data for healthy old subjects are very scarce and conflict with each other. Some authors found no age-related differences in polyamine levels in blood [75], urine [76] or plasma [77] of old normal human beings compared with young or younger adults. There is one report [78] of a decrease in the average level of urinary putrescine excretion in old humans, albeit limited to females. Other authors have reported a decrease in the urinary excretion of spermine in elderly male subjects only [79]. However, in a more accurate and more complete study, it was shown that the levels of all three major polyamines in both serum and urine were age-dependent [80]. The polyamine levels in both these physiological fluids are highest at birth, declining progressively with increasing age [80]. Moreover, the rate of this decrease diminishes as the age of the subject increases "[801. In the one case of progeria studied so far from the polyamine point of view, no marked derangements in polyamine levels of the physiological fluids were seen [52], but the putrescine content was decreased in the erythrocytes, while the spermidine and spermine contents were increased in the leukocytes of the child with this disease of premature ageing [52]. C O N C L U S I O N S A N D SPECULATIONS

Although the topic chosen for this review is very specific, it must already be apparent to the reader (1) that potyamine functions are not limited to the regulation of growth processes (see the above hints about polyamines and mitochondrial function and polyamines and reproductive physiology) and (2) that the changes in polyamine metabolism

159 in mammalian senescence are probably connected with or responsible for other agerelated disorders characteristic of geriatric pathology as well as of cancer. From the results reported in this review, three types of conclusions can be drawn. The first concerns the changes in levels and metabolism of polyamines during mammalian ageing; the second concerns comparing polyamine metabolism and content in ageing and cancer; the third concerns the possibility of relating ageing to cancer through the polyamines. For the biochemistry of senescence, it can be easily concluded that the biosynthesis of polyamines, and therefore their levels, and the inducibility of the related biosynthetic enzymes are generally markedly diminished in aged mammalian tissues, both replicating and non-replicating. This is the second decrease in the rate of polyamine biosynthesis observed in eukaryotic ceils during life, the first occurring after embryonic life [51,8 I]. However, the kind of approach used in the in vivo investigations of the changes in polyamine content and/or metabolism during ageing suffers from some drawbacks, such as (1) it is merely quantitative, (2) it does not answer the question as to whether the observed changes are primary or secondary to the ageing process, (3) it does not completely differentiate the age of the whole animal from the age of the tissues or organs taken from the animal. Nonetheless, it is worth noting that there is quite good agreement between the results obtained in in vivo studies and those obtained in in vitro study, since both show progressive decreases in ODC activity with increasing age. Therefore, it is conceivable to consider that the decrease in polyamine biosynthesis and content in aged cells or organs is a reliable additional parameter for evaluating the degree of senescence of mammalian cells or tissues. Recently, the uncommon D-aspartate isomer has been proposed as a new indicator for the age of white matter in mammalian brain, since it accumulates with age [82]. When we try to compare polyamine metabolism in ageing and cancer, we can make the following statements. (1) One fundamental difference is that polyamine biosynthesis and content are generally markedly enhanced in organs undergoing carcinogenesis [ 1,2] and in fully developed tumours, whereas they are generally markedly diminished in aged organs and tissues. Such a difference is easy to understand on the basis of the rate of cell division and growth, which is generally increased in tumours and decreased throughout ageing. (2) One important similarity is that the induction of ODC activity in both aged organs or tissues (see before) and in tumours [83] is generally strongly reduced. These characteristics can therefore be added to those listed by Pitot [8,9] in his tables for comparison of the processes of ageing and neoplasia. Unfortunately, however, we deplore the large disparity that exists between the number of papers on polyamines in ageing and the number of papers on polyamines in tumours. This disparity precludes making other comparisons of polyamine metabolism in ageing and cancer. For instance, we have no information about the presence of qualitative enzymic changes, such as different forms of ODC or ODC isozymes, in aged tissues or organs, although this has been demonstrated in neoplastic tissues [ 1]. Lastly, when we attempt to relate ageing and cancer by way of the polyamines, some

160 speculations are possible. It is well established that polyamines are key intracellular factors, although by no means the only ones available, for permitting and favouring both regulated and unregulated cell growth. If this is indeed so, one could reasonably expect a low incidence of neoplasias during senescence on the basis of the decreased capacity of the aged tissues and cells to synthesize polyamines. Instead, the variations in age distribution of the appearance of different types of human cancer clearly demonstrate ti~at the incidence of neoplasias of epithelial origin increases greatly with age, while the incidence of neoplasias of the reticuloendothelial system, the central nervous system and connective tissue increases only slightly with age [84]. What the exact mechanisms of this large increase in incidence of some neoplasias with increasing age of the individuals are, remains obscure and can only be speculated upon. We can exclude any direct genetic mechanism as a cause of this phenomenon, although such a genetic mechanism has been demonstrated for most childhood neoplasms. Experimental studies demonstrate that the high incidence of neoplasia in ageing does not seem to be due to the ageing process itself, but rather to a cumulative effect of prolonged exposure to carcinogens throughout the lifetime of the individual, combined with the waning immune response observed in old healthy mammals [11,85,86]. To date, however, there is no uniform opinion about the nature of the interrelation between senescence and the high incidence of spontaneous tumours. A recent careful study carried out with humans strongly suggests that the increase in tumour incidence with advancing age of the host is due to a "tumour susceptibility factor", which is common to most tissues and increases with advancing age [12]. Other experimental observations provide evidence for certain differences in the agedependent metabolism of carcinogens as well as in the extent of the interaction of the carcinogen metabolites with DNA of various tissues [87]. Whatever causes the high incidence of neoplasias in ageing, the characteristics of polyamine metabolism during senescence can not reasonably be considered to be factors favouring the incidence of neoplasia in this period of life, since the capacity to synthesize polyamines and to induce their biosynthetic enzymes, especially the two decarboxylases, declines progressively with age, while neoplastic transformation by many different carcinogenic stimuli requires the capacity to strongly enhance the activities of these enzymes involved in the polyamine biosynthetic pathway [1,2]. Rather, the large differences in the rates of polyamine biosynthesis between aged organs and young ones together with other factors may help explain why the same neoplasias have quite different courses in elderly patients than in young or in adult patients. Accordingly, the reduced polyamine biosynthetic capacity of tissues of aged organisms may well be one factor responsible for the generally slow course of some neoplasias, notably chronic leukaemias and carcinomas of the small intestine, in elderly patients. In other words, it is tempting to speculate that the difference between aged and young organs in the rates of polyamine biosynthesis is roughly maintained in corresponding tumours, i.e. between tumours arising in tissues of aged organisms and tumours arising in tissues of young organisms, leading to a reduced availability of one type of key growth factor, the polyamines, inside the neoplastic cells. Although no study has yet been made of polyamine synthesis in tumours of aged patients

161

or animals, our hypothesis is supported by the f'mding that both human and experimental turnouts with slow growth rates have lower levels of ODC activity and of putrescine content than tumours, either human or experimental, with high growth rates (for reviews, see refs. 1 and 2). Finally, three points should be discussed briefly. First, the connection, if any, between ageing and the incidence of benign tumours in man has not yet been studied. Therefore, there is just not enough information available to indicate even tentatively the role of changes in polyamine metabolism in aged organs in this problem. Second, in this review we did not discuss thoroughly how modifications of polyamine metabolism during ageing might be linked with the various theories (perhaps it would be better to call them speculations) on the mechanisms of ageing. This was deliberate, because of our special research interest at present (which is ontology rather than gerontology) and because we chose to analyse (as expressed in the title) the connections between polyamines and cancer in ageing rather than between polyamines and senescence tout court. We wish to stress in this review the appealing idea that the lesser capacity for polyamine synthesis in aged organs agrees better with the concept that defective gene expression rather than modification of gene structure is responsible for development of ageing (see the paragraph on miscellaneous in vitro effects of polyamines on aged systems or structures). Third, there is a widespread tendency to assign a tumour-promoting effect to the ageing process [14]. However, this idea is not supported by information about polyamine metabolism. It is well known, in fact, that the tumour promoters are also powerful inductors of ODC activity and polyamine synthesis in their target tissues [ 1]. This does not seem to be the case for ageing. ACKNOWLEDGEMENT

We wish to thank Prof. E. Ciaranfi (Milan) for the helpful discussions. We apologize to those investigators whose work we have inadvertently not cited. REFERENCES 1 G. Scalabrino and M.E. Ferioli, Polyamines in mammalian tumors. Part 1. Adv. Cancer Res., 35 (1981) 151. 2 G. Scalabrino and M.E. Ferioli, Polyamines in mammalian tumors. Part II. Adv. Cancer Res., 36 (1982) 1. 3 M.H. Goyns, The role of polyamines in animal cell physiology. J. Theor. BioL, 97 (1982) 577. 4 S.S. Cohen, The polyamines as a growth industry. Fed. Prec., 4I (1982) 3061. 5 S. Goldstein, The biology of aging. N. Engl. J. Med., 285 (1971) 1120. 6 L. Hayflick, Current theories of biological aging. Fed. Prec., 34 (1975) 9. 7 L. Hayflick, The cell biology of human aging. N. Engl. £ Med., 295 (1976) 302. 8 H.C. Pitot, Carcinogenesis and aging - Two related phenomena? Am. J. ParhoL, 87 (1977) 444. 9 H.C. Pitot, Interactions in the natural history of aging and carcinogenesis. Fed. Prec., 37 (1978) 2841. 10 G. Hocman, Biochemistry of ageing and cancer. Int. J. Biochem., 13 (1981) 659. 11 R.A. Gatti and R.A. Good, Aging, immunity, and malignancy. Geriatrics, 25 (1970) 158.

162 12 D. Dix, P. Cohen and J. Flannery, On the role of aging in cancer incidence. J. Theor. Biol., 83 (1980) 163. 13 V.N. Anisimov and V.S. Turusov, Modifying effect of aging on chemical carcinogenesis. A review. Mech. Ageing Dev., 15 (1981) 399. 14 V.N. Anisimov, Carcinogenesis and aging. Adv. CancerRes., 40 (1983) 365. 15 F.F. Holmes (ed.), Ageing and Cancer, [Rec. Res. Cancer Res., Vol. 87], Springer-Verlag, Berlin, 1983. 16 G.M. Cooper, Cellular transforming genes. Science, 218 (1982) 801. 17 T.G. Krontiris, The emerging genetics of human cancer. N. Engl. J. Med., 309 (1983) 404. 18 R.V. Gilden and N.R. Rice, Oncogenes. Carcinogenesis, 4 (1983) 791. 19 H.V. Gelboin, Carcinogens, drugs and cyto'chromes P-450. N. Engl. J. Med., 309 (1983) 105. 20 G. Hocman, Biochemistry of ageing. Int. J. Biochem., I0 (1979) 867. 21 G. Hocrnan, Biochemistry of ageing. If. Int. J. Biochem., 12 (1980) 515. 22 J. Bjorksten, Biochemistry of aging. In H.E. Spiegel (ed.), Contemporary Theories and Techniques [Clinical Biochemistry, Vol. 2], Academic Press, New York, 1982, pp. 209-253. 23 R.N. Butler (ed.), Symposium on "Overview of th'e biology of aging". Fed. Prec., 38 (1979) 1955-1971. 24 J.R. Florini (ed.), Symposium on "Molecular biology and aging". Fed. Prec., 38 (1979) 19721992. 25 L. Hayflick (ed.), Symposium on "Cell biology of aging". Fed. Prec., 38 (1979) 1847-1872. 26 M. Rothstein, BiochemicaIApproaches to Aging, Academic Press, New York, 1982. 27 G.M. Martin, Cellular aging - Clonal senescence. Am. J. Pathot, 89 (1977) 484. 28 G.M. Martin, Cellular aging - Postreplicative cells. Am. J. Pathot, 89 (1977) 513. 29 J.M. Gaugas (ed.), Polyamines in Biomedical Research, John Wiley, New York, 1980. 30 D.R. Morris and L.J. Marten, (eds.), Polyamines in Biology and Medicine, Marcel Dekker, New York, 1981. 31 S.S. Cohen(ed.), Symposium on "Metabolism and functions of polyaminesin oukaryotic systems". Fed. Prec., 41 (1982) 3061-3088. 32 A.E. Pegg and P.P. McCann, Polyamine metabolism and function. Am. Z PhysioL, 243 (1982) C212. 33 G. Quash and A.M. Roch, Les polyamines chez les eucaryotes: les m6canismes hom6ostatiques intracellulaires et s~riques. Ann. BioL Clin., 37 (1979) 317. 34 H.G. Williams-Ashman and Z.N. Canellakis, Polyamines in mammalian biology and medicine Persp. Biol. Med., 22 (1979) 421. 35 T. Oka, J.W. Perry, T. Takemoto, T. Sakai, N. Terada and H. Inoue, The multiple regulatory roles of polyamines in the hormonal induction of mammary gland development. Adv. Polyamine Res., 3 (1981) 309. 36 C.W. Tabor and H. Tabor, 1,4-Diaminobutane (putrescine), spermidine and spermine. Annu. Rev. Biochem., 45 (1976) 285. 37 A. Raina and J. J~nne, Physiology of the natural polyamines putrescine, spermidine and spermine. Med. Biol., 53 (1975) 121. 38 T.T. Sakai and S.S. Cohen, Effects of polyamines on the structure and reactivity of tRNA. Prog. Nucl. Acid Res. Mol. Biol., 17(1976) 15. 39 D.H. Russell, Ornithine decarboxylase: a key regulatory protein. Med. Biol., 59 (1981) 286. 40 D.H. Russell, Ornithine decarboxylase may be a multifunetional protein. Adu. Enzyme Regui., 21 (1983) 201. 41 E.S. Canellakis, D. Viceps-Madore, D.A. Kyriakidis and J.S. HeUer, The regulation and function of ornithine decarboxylase and of the polyamines. Curt. Top. Cell. Regul., 15 (1979) 155. 42 E.S. Canellakis, D.A. Kyriakidis, J.S. Heller and J.W. Pawlak, The complexity of regulation of ornithine decarboxylase. Med, Biol., 59 (1981) 279. 43 A.E. Pegg, H. Hibasami, I. Matsui and D.R. Bethell, Formation and intereonversion of putreseine and spermidine in mammalian ceils. Adv. Enzyme ReguL, 19 (1981) 427. 44 N. Seiler, F.N. Bolkenius and O.M. Rennert, Int~reonversion, catabolism and elimination of polyamines. Med. Biol., 59 (1981) 334.

163 45 O. Heby, Role of polyamines in the control of cell proliferation and differentiation. Differentiation, 19 (1981) 1. 46 J. J~inne, A. Raina and M. Slimes, Spermidine and spermine in rat tissues at different ages. Acta Physiol. Scand., 62 (1964) 352. 47 M.E. Ferioli and R. Comolli, Changes of liver and kidney polyamine levels during ageing. Exp. Gerontol,, ]0 (1975) 13. 48 R. Das and M.S. Kanungo, Activity and modulation of ornithine decarboxylase and concentrations of polyamines in various tissues of rats as a function of age. Exp. Gerontol., ] 7 (1982) 95. 49 E.G. Shaskan, Brain regional spermidine and spermine levels in relation to RNA and DNA in aging rat brain. J. Neurochem., 28 (1977) 509. 50 J. Jfinne, Studies on the biosynthetic pathway of polyamines in rat liver. Acta Physiol. Scand., Suppl., 300 (1967) 1. 51 J. J~nne, E. H61tt[i and S.K. Guha, Polyamines in mammalian liver during growth and development. In H. Popper and F. Schaffner (eds.), Progress in Liver Disease, Grune and Stratton, New York, 1976, pp. 100-124. 52 K.D. Cooper, J.B. Shukla and O.M. Rennert, Polyamine distribution in cellular compartments of blood and in aging erythrocytes. Clin. Chim. Acta, 73 (1976) 71. 53 O.M. Rennert and J.B. Shukla, Polyamines in health and disease. Adv. Polyamine Res., 2 (1978) 195. 54 J. J~inne and E. H61tt~i, Putrescine metabolizing enzyme activities in some rat tissues during postnatal development. Acta Chem. Scand., 2 7 (1973) 2399. 55 M.E. Ferioli, G. Ceruti and R. Comolli, Changes in rat liver ornithine decaxboxylase activity during ageing and effect of stimulation by dexamethasone. Exp. Gerontol., 11 (1976) 153. 56 E.E.K. Takyi and G.H. Thomas, Ornithine decarboxylase activity during neonatal development of the rat ventral prostate. Biochim. Biophys. Acta, 497(1977) 652. 57 S.A. Shain and A.L. Moss, Aging in the AXC rat: differential effects of chronic testosterone treatment on restoration of diminished prostate L-ornithine decarboxylase and S-adenosyl-L-methionine decarboxylase activities. Endocrinology, 109 (1981) 1184. 58 S.A. Shain and A.L. Moss, Aging in the AXC rat: equivalence of the rates of inactivation of L-ornithine decarboxylase and S-adenosyl-L-methionine decarboxylase in prostate of young and aged rats. Endocrinology, 109 (1981) 1192. 59 S.A. Shaln, J.K. Hilliard and C. de Leon, Aging in the AXC rat: diminished prostatic S-adenosylL-methionine decarboxylase (AMDC) activity appears to reflect altered expression of the AMDC gene. Endocrinology, 113 (1983) 1292. 60 S.A. Shaln and C.M. Lancaster, Aging in the AXC rat: prostatic polyamine content is not correlated with prostatic L-ornithine decarboxylase and S-adenosyl-L-methionine decarboxylase activity. Mech. Ageing Dev., 20 (1982) 33. 61 E.E.K. Takyi, D.J.M. Fuller, L.J. Donaldson and G.H. Thomas, Deoxyribonucleic acid and polyamine synthesis in rat ventral prostate. Effects of age of the intact rat and androgen stimulation of the castrated rat with testosterone, 5~-dihydrotestosterone and 5a-androstane-3fl,17fl-diol. Bio. chem. J., 162 (1977) 87. 62 E. Cavia and T.E. Webb, Modified induction of ornithine decarboxylase by factors which affect liver regeneration. Biochim. Biophy~ Acra, 262 (1972) 546. 63 E. H61tt/i and J. J/inne, Ornithine decarboxylase activity and the accumulation of putrescine at early stages of liver regeneration. FEBS Lett., 23 (1972) 117. 64 A. Ralna, J. JS.nne, P. Hannonen, E. H61tt~i and J. Ahonen, Polyamine-synthesizing enzymes in regenerating liver and in experimental granuloma. In D.H. Russell (ed.), Polyamines in Normal and Neoplastic Growth, Raven Press, New York, 1973, p. 167. 65 D.H. Russell, Discussion: Putrescine and spermidine biosynthesis in growth and development. Ann. N.Y. Acad. Sci., 17] (1970) 772. 66 D.H. Russell and J.B. Lombardini, Polyamines: (1) Enhanced S-adenosyl-L-methionine decarboxylase in rapid growth systems, and (2) The relationships between polyamine concentrations and RNA accumulation. Biochim. Biophys. Acta, 240 (1971) 273.

164 67 D.H. Russell, S.H. Snyder and V.J. Medina, Growth hormone induction of omithine decarboxylase in rat liver. Endocrinology, 86 (1970) 1414. 68 S. Jacobus and D. Gershon, Age-related changes in inducible mouse liver enzymes: ornithine decarboxylase and tyrosine aminotransferase. Mech. Ageing Dev., 12 (1980) 311. 69 R.R. Kohn, Aging and cell division. Science, 188 (1975) 203. 70 P.E. Duffy and L.T. Kremzner, Omithine decarboxylase activity and polyamines in relation to aging of human fibroblasts. Exp. Cell Res., 108 (1977) 435. 71 R. ComoUi, Polyarnine effects on 14Cqeucine transfer to microsomal protein in a rat liver cell free system during ageing. Exp. Gerontol., 8 (1973) 307. 72 R. Das and M.S. Kanungo, Effects of polyamines on in vitro phosphorylation and acetylation of histories of the cerebral cortex of rats of various ages. Biochem. Biophys. Res. Commun., 90 (1979) 708. 73 J. Gaubatz, M. Ellis and R. Chalkley, The structural organization of mouse chromatin as a function of age. Fed. Proc., 38 (1979) 1973. 74 J.E. Phillips and R.R.J. Chaffer, Restorative effects of spermine on oxidative phosphorylation and respiration in heat-aged mitochondria. Biochem. Biophys. Res. Commun., 108 (1982) 174. 75 O.M. Rennert, J. Frias and J.B. Shukla, Polyamine metabolism in cystic fibrosis. Tex. Rep. Biol. Med., 34 (1976) 187. 76 A. Lipton, L. Sheehan, R. Mortel and H.A. Harvey, Urinary polyamine levels in patients with localized malignancy. Cancer, 38 (1976) 1344. 77 P. Chaisiri, M.E. Harper, R.W. Blarney, W.B. Peeling and K. Griffiths, Plasma spermidine concentrations in patients with turnouts of the breast or prostate or testis. Clin. Chim. Acta, 104 (1980) 367. 78 T.P. Waalkes, C.W. Gehrke, D.C. Tormey, R.W. Zumwalt, J.N. Hueser, K.C. Kuo, D.B. Lakings, D.L. Ahmann and C.G. Moertel, Urinary excretion of polyamines by patients with advanced malignancy. Cancer Chemother. Rep., 39 (1975) 1103. 79 S. Beninati, M. Piacentini, A. Spinedi and F. Autuoti, Urinary polyamine excretion in man: I. Influence of sex and age. Biomedicine, 33 (1980) 140. 80 D. Rudman, M.H. Kutner, R.K. Chawla, M.G. Goldsmith, R.D. Blackston and R. Bain, Serum and urine polyamines in normal and short children. J. Clin. Invest., 64 (1979) 1661. 81 J.A. Sturman and G.E. Gaull, Polyamine biosynthesis in human fetal liver and brain. Pediatr. Res., 8 (1974) 231. 82 E.H. Man, M.E. Sanhouse, J. Burg and G.H. Fisher, Accumulation of D-aspartic acid with age in the human brain. Science, 220 (1983) 1407. 83 G. Scalabrino, H. P6s6, E. H61ttii, P. Hannonen, A. KaUio and J. J~'nne, Synthesis and accumulation of polyamines in rat liver during chemical carcinogenesis, lnt. Z Cancer, 21 (1978) 239. 84 R, Doll, An epidemiological perspective of the biology of cancer. Cancer Res., 38 (1978) 3573. 85 R. Peto, F.J.C. Roe, P.N. Lee, L. Levy and J. Clack, Cancer and ageing in mice and men. Br. J. Cancer, 32 (1975) 411. 86 S. Rockwell, Effects of host age on the transplantation, growth, and radiation response of EMT6 tumors. CancerRes., 41 (1981) 527. 87 A.J. Likhachev, H. Ohshima, V.N. Anisimov, A.I. Ovsyannikov, S.Y. Revskoy, L.K. Keefer and E.J. Reist, Carcinogenesis and aging. II. Modifying effect of aging on metabolism o.f methyl(acetoxymethyl)nitrosmalne and its interaction with DNA of various tissues in rats. Carcinogenesis, 4 (1983) 967.