Polyamine Metabolism in Mammalian Cells

A. Ornithine Decarboxylase. ODC is a pyridoxal phosphate-dependent enzyme that is found only at very low levels in resting cells. Cellular levels of ODC can be significantly increased, sometimes by a 100-fold or more, by stimulation with growth factors, hormones, drugs, and other regulatory molecules; however, ODC still represents only a minor portion of the total cellular protein (e.g., from 0.01 % of the cytosolic protein in androgen-stimulated mouse kidneys to 0.00012% in thioacetamide-stimulated rat liver). A macromolecular inhibitor of ODC, termed antizyme, has been isolated in a number of cells and may be another regulatory factor controlling ODC activity as well.

The use of the competitive inhibitor α-methylornithine conclusively demonstrated that formation of putrescine by ODC was essential for DNA replication and, consequently, mammalian cell growth. Later synthesis of the more potent inhibitor α-difluorometh- ylornithine (DFMO; eflornithine, Ornidyl®) alllowed extensive work to explore the relationship between polyamine depletion caused by ODC inhibition, its consequent effects on cell replication, and, in some cases, cell differentiation.

DFMO (an enzyme-activated, mechanism-based, irreversible inhibitor of ODC) and some other substrate and product-related inhibitors, which have somewhat improved biochemical properties, have been widely used as biochemical and pharmacological tools. In fact, DFMO has been extensively explored as therapy for several types of cancer and protozoan infections. A number of studies have shown that treatment of cells in culture with DFMO significantly decreases intracellular putrescine and spermidine with little or no effect on spermine, whereas an enormous increase occurs in the overall levels of decarboxylated AdoMet. Other more potent inhibitors such as (2R, 5R)-δ-methyl-α-acetylenic putrescine had greater effects on spermine levels but did not achieve complete depletion as was the case with spermidine.

B. S-Adenosylmethionine Decarboxylase. Mammalian AdoMetDC is the enzyme that provides decarboxylated AdoMet as an aminopropyl donor in the formation of spermidine from putrescine. Once it has been decarboxylated, AdoMet is committed to polyamine production, as no other reactions utilizing decarboxylated AdoMet at any physiologically significant rate are known. Therefore, the production of decarboxylated AdoMet is kept low and constitutes the rate-limiting factor in spermidine formation.

Mammalian AdoMetDC is activated by putrescine and repressed by spermidine, linking the supply of decarboxylated AdoMet to the need for spermidine and the availability of the other substrate (putrescine) for spermidine synthesis. AdoMetDC has an enzyme- bound pyruvate as cofactor and, analogous to ODC, is present in cells only at very low levels (e.g., equal to 0.015% of the soluble protein in ventral prostate and to 0.0007% in liver). Similarly, AdoMetDC is also regulated by a number of external growth factors and stimuli.

Mammalian AdoMetDC is a dimer of two pairs of subunits with molecular weights of 30,621 and 7681. These subunits are formed by the cleavage of a proenzyme chain (M_r about 38,000) in a reaction that forms the pyruvate prosthetic group at the amino-terminal end of the larger subunit. The processing-cleavage step may be an autocatalytic reaction of the proenzyme chain and is accelerated in the presence of putrescine. Because this step is essential for the production of the active sites of the enzyme, it provides an attractive target for the future design of therapeutically useful inhibitors of AdoMetDC.

Methylglyoxal bis(guanylhydrazone) (MGBG) is a potent inhibitor of AdoMetDC and was used in early experiments to prevent formation of decarboxylated AdoMet and, thus, synthesis of spermidine. Although such experiments suggested the importance of polyamines for cell growth, they were not conclusive because MGBG is nonspecific and has a variety of effects on mitochondrial and other physiological functions unrelated to effects of inhibition of polyamine biosynthesis.

MGBG can be considered as a structural analogue of spermidine and is taken up by cells via a specific polyamine transport mechanism. Thus, the reversal of the effects of MGBG by spermidine may be as much due to interference of its cellular uptake or binding as to its replacement of intracellular polyamines.

Even though other inhibitors of AdoMetDC have been utilized, only recently have newer specific and more potent compounds become available, such as 5' - deoxy - 5' [N- methyl - N- (aminooxyethyl)] - amino - adenosine, 5'-deoxy-5'[N-methyl-N-(3-hydrazino- propyl)]aminoadenosine, and S-(5'-deoxy-5'-adeno- syl)-methyl-thioethylhydroxylamine. These irreversible inhibitors of AdoMetDC apparently bind to the active site of the enzyme and form a covalent bond with the enzymatic pyruvate prosthetic group. An enzyme-activated inhibitor of AdoMetDC has also been described—5'-{[(Z)-4-amino-2-butenyl]methyl- amino}-5 '-deoxyadenosine.

These inhibitors produce profound but expected changes in intracellular polyamine levels and consequently inhibit cell growth. For example, they reduce spermidine and spermine as well as decarboxylated AdoMet and 5'-methyl-thioadenosine and cause an enormous increase of putrescine in L1210 cells. Following depletion of spermidine and spermine, cell growth is arrested in spite of the large concentration of available putrescine; addition of either one depletes polyamine fully restored normal cell growth. Utilization of DFMO and an AdoMetDC inhibitor together completely blocks the increase of all three amines, demonstrating the potential utility of inhibition of AdoMetDC as a target for drug design.

C. Aminopropyltransferase. Spermidine synthase is responsible for the transfer of the aminopropyl group from decarboxylated AdoMet to putrescine. A second enzyme, spermine synthase, carries out a similar reaction and adds another aminopropyl group to spermidine. Although the mechanisms of action of these two aminopropyl transferases are analogous, both spermidine synthase and spermine synthase are specific and discrete enzymes, each having its own unique substrate.

The cellular amounts of both of these aminopropyl transferases are normally much higher than either ODC or AdoMetDC and are apparently regulated by the levels of their substrates (e.g., decarboxylated AdoMet). However, apparently the disposition of available decarboxylated AdoMet toward spermidine or spermine is probably determined by the relative amounts of the two synthases, and marked changes in spermidine synthase activity have been observed in response to hormones, tissue regeneration, and cell growth factors.

The multisubstrate analogue S-adenosyl-l,8-diam- ino-3-thioctane (AdoDato), designed as an inhibitor of spermidine synthase, strongly inhibits this enzyme in mammalian cells. Another potent inhibitor of the same enzyme, cyclohexylamine, is competitive with respect to putrescine with a K_i of about 0.2 μM.

Although AdoDato is an effective in situ inhibitor of spermidine synthase in mammalian cells, its overall utility is limited because of the large increase in decarboxylated AdoMet resulting from the compensating rise in AdoMetDC after the inhibition of spermidine synthase. This decarboxylated AdoMet, in turn, can be used for the synthesis of spermine via spermine synthase, and thus the total polyamine pools are not significantly altered. Cyclohexylamine, however, does decrease overall polyamine levels in murine tumors.

Design and synthesis of specific spermine synthesis inhibitors have been of recent interest because of the relative ineffectiveness of ODC inhibitors in decreasing spermine levels and because the physiological significance of spermine synthase, present in mammalian cells but not found in many microorganisms, is not clearly understood at present.

The first inhibitor shown to appreciably block spermine synthase in vitro was S-methyl-5'-methyl- thioadenosine. Another compound, S-adenosyl-1,12- diamino-3-thio-9-azadodecane (AdoDatad), structurally related to AdoDato by addition of an aminopropyl moiety, was designed and found to be a potent, specific, multisubstrate analogue inhibitor of spermine synthase. This enzyme step is also blocked effectively by n-butyl-1,3-diaminopropane. However, neither AdoDatad nor n-butyl-1,3-diaminopropane had any effect on cell proliferation, although spermine was significantly depleted, suggesting either no requirement for spermine or that the increased amounts of spermidine found could compensate for the lack of spermine.

D. Polyamine Transport. Besides having the ability to synthesize polyamines, cells also possess a specific membrane transport system for the uptake of exogenous polyamines. Although its biochemical mechanism is not well understood, this uptake system is regulated by the intracellular polyamine content. Thus, when polyamines are depleted by use of specific inhibitors, the transport system responds to increased uptake and vice versa.

Of particular import is the fact that the transport system reduced the therapeutic effectiveness of polyamine biosynthetic inhibitors such as DFMO because extracellular polyamines (e.g., from diet, intestinal microorganisms, cell turnover) can be utilized via the transport system of polyamine depleted cells. For example, DFMO is significantly more active against a polyamine transport mutant L1210 cell line in vivo than against the parent cell line. DFMO is also more effective against tumors in rodents when intestinal polyamine oxidase is inhibited, thus preventing formation of polyamines from their N¹-acetyl derivatives.

Many polyamine analogues are actually substrates for the transport system and thus act as competitive inhibitors in relation to the natural polyamines. In fact, a number of carrier systems likely have some overlap in specificities. One system is Na⁺-dependent and can be regulated by compounds that change Na⁺ flux. Characterization of such systems should ultimately allow the design and synthesis of specific inhibitors to regulate entry and efflux of the polyamines.

MGBG toxicity is quite obviously related to its transport and ultimate accumulation within the cell. Such high levels of MGBG cause inactivation of mitochondria and finally inhibition of macromolecular biosynthesis. Although MGBG accumulation is via the polyamine transport system, it does not seem to mimic the natural polyamines and other polyamine analogues in down-regulating the polyamine transport system. As the system continues to transport MGBG into cells that already contain very high levels of the drug, MGBG’s lack of repression of such transport may be a factor in its significant cytotoxicity.

Various N-alkyl polyamine analogues have been synthesized and are taken up by the polyamine transport system. For example, N¹N⁸bis(ethyl)-spermidine reduced ODC levels and intracellular putrescine, spermidine, and also spermine in L1210 cells, whereas N¹,N¹²-bis(ethyl)spermine was even more active in that it also reduced AdoMetDC levels and caused almost complete depletion of all the polyamines. Such effects are produced in part because these analogues are recognized by the polyamine regulatory systems for ODC and AdoMetDC biosynthesis. In addition, efflux of intracellular polyamines increases by some yet undefined mechanism, which normally regulates intracellular polyamine levels.

The bis(ethyl)polyamines are also potent inducers of SAT, and enzyme increases of several 100-fold are seen in cells within hours after exposure to the analogues. Such induced increases in SAT may also affect the rapid excretion of intracellular polyamines because such acetylation facilitates conversion of nonexcreted spermine to spermidine and putrescine, which are both excreted from the cell.