The natural polyamines are ubiquitous multifunctional organic cations which play important roles in regulating cellular proliferation and survival. Here we present a novel approach to investigating polyamine functions by using optical isomers of MeSpd (α-methylspermidine) and Me2Spm (α,ω-bismethylspermine), metabolically stable functional mimetics of natural polyamines. We studied the ability of MeSpd and Me2Spm to alter the normal polyamine regulation pathways at the level of polyamine uptake and the major control mechanisms known to affect the key polyamine metabolic enzymes. These include: (i) ODC (ornithine decarboxylase), which catalyses the rate-limiting step of polyamine synthesis; (ii) ODC antizyme, an inhibitor of ODC and polyamine uptake; (iii) SSAT (spermidine/spermine N1-acetyltransferase), the major polyamine catabolic enzyme; and (iv) AdoMetDC (S-adenosyl-L-methionine decarboxylase), which is required for the conversion of putrescine into spermidine, and spermidine into spermine. We show that the stereoisomers differ in their cellular uptake and ability to downregulate ODC and AdoMetDC, and to induce SSAT. These effects are mediated by the ability of the enantiomers to induce +1 ribosomal frameshifting on ODC antizyme mRNA, to suppress the translation of AdoMetDC uORF (upstream open reading frame) and to regulate the alternative splicing of SSAT pre-mRNA. The unique effects of chiral polyamine analogues on polyamine metabolism may offer novel possibilities for studying the physiological functions, control mechanisms, and targets of the natural polyamines, as well as advance therapeutic drug development in cancer and other human health-related issues.
- alternative splicing
- optical isomer
- upstream open reading frame
Polyamines are achiral organic polycationic compounds that play a critical role as regulators of important cellular processes such as proliferation, differentiation and modulation of ion channels . As expected for a molecule with multiple cellular roles, polyamine levels are tightly controlled through a series of feedback circuits to maintain appropriate intracellular concentrations. The key regulatory enzymes in polyamine biosynthesis are ODC (ornithine decarboxylase) and AdoMetDC (S-adenosyl-L-methionine decarboxylase). During catabolism of polyamines, SSAT (spermidine/spermine N1-acetyltransferase) is the rate-controlling enzyme of PAO (polyamine oxidase)-mediated back-conversion of spermine and spermidine into spermidine and putrescine respectively. On the basis of the observations that increased polyamine levels are correlated with, and can promote, neoplastic growth, it has been proposed that reducing or depleting intracellular polyamines is an appropriate therapy for cancer and other diseases involving undesired cellular proliferation. Towards this end, inhibitors of ODC and AdoMetDC [2–4] have been identified, and polyamine analogues have been designed with the intent of affecting multiple targets in the polyamine pathway. Several of these compounds alone or in combination are currently in clinical trials [5–8] as potential anti-cancer agents.
Owing to their positive charge at physiological pH, a major fraction of polyamines is electrostatically bound to anionic cellular sites such as RNA, DNA and phospholipids. Thus, polyamines are involved in regulation of many levels of gene expression, e.g. by influencing chromatin condensation, DNA stability and structure, and RNA processing and translation. One very specific target of polyamine-mediated regulation is the programmed frameshifting of eukaryotic ODC antizyme [9,10] and TY1 transposon . Antizyme is a small regulatory protein which binds to and inhibits ODC and targets it to the proteasome for degradation. In addition, antizyme negatively regulates the polyamine uptake system [12,13]. The synthesis of antizyme is controlled by cellular polyamine concentration at the level of translation, whereby polyamines greatly increase the efficiency of +1 ribosomal frameshifting, allowing the production of functional, full-length protein, thus completing an autoregulatory circuit controlling polyamine synthesis and transport.
The feedback regulation of AdoMetDC by spermidine and spermine involves control of translational initiation. The AdoMetDC 5′ leader sequence harbours an uORF (upstream open reading frame), which codes for a small regulatory peptide with the sequence MAGDIS . When polyamine levels are low, ribosomes pause after uORF termination but are able to continue to the AdoMetDC reading frame. By contrast, elevated polyamine levels stabilize the ribosomal pause in the vicinity of uORF termination codon, and inhibit the completion of uORF peptide synthesis, reducing translation of the AdoMetDC ORF.
We recently discovered a novel regulation mechanism for the expression of SSAT, where polyamines modulate the alternative splicing of SSAT pre-mRNA . Low polyamine levels result in accumulation of an unproductive splice variant (SSAT-X), which is rapidly degraded by the protein synthesis-dependent mRNA surveillance pathway known as NMD (nonsense-mediated mRNA decay). By contrast, high polyamine levels favour generation of the productive mRNA variant (SSAT), resulting in synthesis of active SSAT enzyme protein, leading to decreased polyamine levels.
Although complex feedback mechanisms have been found for all key metabolic enzymes controlling cellular polyamine levels, the underlying regulatory mechanisms of these specific processes are not yet understood. Thus new specific tools are needed to manipulate the polyamine metabolism. α-Methylated polyamine analogues are functional mimetics for natural polyamines both in vivo and in vitro [16–18]. MeSpd (α-methylspermidine) and Me2Spm (α,ω-bismethylspermine) are not cytotoxic and fulfil many cellular functions of the natural polyamines and support proliferation. Because they are metabolically much more stable than their natural counterparts and replace the natural polyamines by enhancing polyamine efflux and inhibiting uptake, the α-methylated analogues are convenient tools for polyamine research. Unlike the natural polyamines, the α-methylated analogues are chiral. Since chirality plays an important role in Nature by controlling the binding and metabolic transformation of the compounds in living systems, we rationalized that the stereoisomers of α-methylated polyamine analogues may possess divergent biological properties. Indeed, the investigation of these novel chiral analogues led to the discovery of the hidden stereospecificity of several enzymes of polyamine metabolism, such as PAO , SMO (spermine oxidase) and deoxyhypusine synthase .
In the present paper, we investigated the differences between the stereoisomers of MeSpd and Me2Spm in their cellular uptake and their ability to regulate the key enzymes of polyamine metabolism. Our results revealed significant differences between the effects of stereoisomers on the major known polyamine feedback regulatory mechanisms which control intracellular polyamine levels.
HEK-293 (human epithelial kidney-293) cells and a human prostate carcinoma cell line, DU145, were obtained from A.T.C.C. Nontransgenic mouse primary fetal fibroblasts were prepared as previously described in . The optical isomers of MeSpd and Me2Spm were synthesized as described in . DFMO (difluoromethylornithine) was obtained from ILEX Oncology. [14C]-labelled spermine tetrahydrochloride (specific activity 112 mCi/mmol), spermidine trihydrochloride (specific activity 113 mCi/mmol) and carboxyl-S-adenosyl-L-methionine (specific activity 54 mCi/mmol) were from GE Healthcare.
HEK-293 cells were cultured in DMEM (Dulbecco's modified Eagle's medium; Sigma) plus 10% FBS (fetal bovine serum) at 37 °C and 5% CO2. Primary fetal fibroblasts and DU145 cells were cultured in DMEM supplemented with 10% heat-inactivated FBS (Gibco) and 50 μg/ml gentamycin (Gibco), and incubated in a humidified atmosphere at 37 °C, 10% CO2. When natural polyamines were used, 1 mM aminoguanidine was included to prevent the oxidation of Spd (spermidine) and Spm (spermine) by serum amine oxidases. An analogue concentration of 100 μM was used to effectively replace the natural polyamines with the analogues. The cells were harvested by trypsinization, washed with PBS, pelleted and stored at −70 °C. The cell number was measured electronically with a Coulter Counter, model Z1. Cell pellets were lysed in buffer [20 mM Tris/HCl, pH 7.4, 1 mM EDTA, 0.1% Triton X-100, 1 mM dithiothreitol and protease inhibitor cocktail Complete EDTA-free (Roche)] and incubated for 20 min on ice. For AdoMetDC activity assay, the cell samples were lysed in buffer containing 50 mM potassium phosphate buffer, pH 7.2, 0.1 mM EDTA, 0.1% Triton X-100 and 1 mM dithiothreitol. Samples for polyamine measurement were taken and the rest of the lysate was centrifuged at 12000 g for 20 min at 4 °C. The supernatant fraction was used for enzymatic assays of SSAT, ODC and AdoMetDC. Total protein concentration was measured using Coomassie Brilliant Blue (Bio-Rad) with dilutions of BSA (Sigma) as standards.
Enzyme activities and polyamines
Intracellular polyamines, polyamine analogues and nucleosides were measured by HPLC according to the method published in . The amount of DNA was measured from pellets of polyamine samples according to the method of Giles and Myers . ODC and SSAT activities were measured as described previously in [25,26]. AdoMetDC activity was measured by a similar method to ODC activity, using reaction buffer consisting of 50 mM potassium phosphate buffer, pH 7.4, 1.25 mM dithiothreitol, 3 mM putrescine, 0.2 mM SAM (S-adenosyl-L-methionine) and 25 μCi/ml [14C-carboxyl]S-adenosyl-L-methionine.
Alternative splicing of SSAT
Primary mouse fetal fibroblasts were plated on 6-well plates and grown overnight. Then, fresh medium containing 10 μg/ml CHX (cycloheximide) was added. After 1 h, polyamine analogues (100 μM) were added, and the cells were further incubated for 7 h. RNA extraction, DNase treatment, cDNA synthesis and quantitative RT–PCR (reverse transcription–PCR) were carried out as previously described in .
Uptake competition experiments
DU145 cells were plated on to 6-well plates at a density of 0.3×106 cells/well. After overnight incubation, pre-warmed serum-free medium supplemented with 10 μM [14C]Spd or [14C] Spm (specific activity 25 mCi/mmol) and competing polyamine analogues (1, 10 or 100 μM) was added. After 10 min incubation, the medium was aspirated, the plates washed twice with ice-cold PBS and the cells lysed in 500 μl of 0.5 M NaOH. The lysate (400 μl) was mixed with 3 ml of Optiphase HiSafe scintillation cocktail (PerkinElmer) and counted using a liquid scintillation counter (1450 Microbeta PLUS, Wallac). Total protein concentration was measured in the lysate using Coomassie Brilliant Blue.
Antizyme frameshifting in HEK-293 cells and in vitro
The frameshifting assay was essentially carried out as described in [27,28]. In brief, HEK-293 cells were plated at 7×103 cells/well in 48 wells of a 96-well plate. After 48 h culture in 2.5 mM DFMO, the cells were transfected with 50 ng/well of p2luc reporter plasmid  and 0.2 μl of Lipofectamine™ 2000. The reporter plasmid contains antizyme frameshifting sequences needed for polyamine induction between the Renilla and firefly luciferase genes. The firefly luciferase ORF lacks an initiation codon and can only be expressed as a fusion protein with Renilla luciferase if translational frameshifting occurs. After 12 h, polyamine analogues, 1 mM aminoguanidine and 2.5 mM DFMO were added. The cells were lysed 24 h later and relative light units were measured following an autoinjection of the Dual Luciferase Reporter Assay reagents using a Veritas Luminometer (Turner Biosystems). in vitro frameshifting assays were carried out according to manufacturer's specifications using the TNT Quick Coupled Transcription/Translation system (Promega) in a final volume of 10 μl supplemented with various concentrations of polyamine analogues. After incubating for 1 h at 30 °C, reactions were diluted 10-fold in Passive Lysis Buffer supplied with the Dual Luciferase Reporter Assay (Promega) reagents. Frameshifting efficiency was calculated as a ratio of the firefly to Renilla luciferase activities standardized to an in-frame control.
AdoMetDC uORF translation in vitro
Plasmids containing an AdoMetDC uORF–luciferase fusion construct, or a similar construct where the uORF initiation codon AUG was mutated to AUA, were a gift from Dr David Morris (Department of Biochemistry, University of Washington, Seattle, U.S.A.) . The plasmids were produced in Escherichia coli, purified with QIAFilter Maxi Kit (Qiagen) and linearized with EcoRI. Capped mRNAs were synthesized using a RiboMAX Large Scale RNA production system (T7) in vitro transcription kit (Promega) and ribo m7G cap analogue (Promega) according to the manufacturer's instructions. After purification, in vitro translation reactions (25 μl) were carried out using a Flexi rabbit reticulocyte lysate (Promega) and added polyamine (100 μM Spm and Me2Spm, 250 μM Spd and MeSpd). The endogenous Spd and Spm levels in the lysate were 440 and 40 μM respectively. Final Mg2+ and K+ concentrations in the reactions were 3.2 and 75 mM respectively. Luminescence was measured as quadruplicates with Victor2 Multilabel Counter (Wallac) from 5 μl of the lysate mixed with 100 μl of luciferase assay reagent (Promega).
Values are means±S.D. One-way ANOVA with Tuckey's post-hoc test was used for multiple comparisons with the aid of GraphPad Prism 4.03 software (GraphPad Software). *, ** and *** refer to P values of <0.05, <0.01 and <0.001 respectively.
Polyamine concentrations, ODC, AdoMetDC and SSAT activities
To examine the effect of the stereoisomers of MeSpd and Me2Spm on intracellular polyamine concentrations and on the enzyme activities of ODC, SSAT and AdoMetDC, DU145 prostate cancer cells were treated with 100 μM of the analogues for 72 h, with or without 5 mM DFMO, an irreversible ODC inhibitor. As shown in Tables 1 and 2, S-MeSpd and S,S-Me2Spm were the most effective stereoisomers to inhibit ODC activity. In fact, S,S-Me2Spm downregulated ODC activity even better than DFMO. In contrast, the analogues only slightly induced SSAT activity, S- and S,S-isomers being the most potent ones. All the stereoisomers accumulated in the cells to concentrations near that of the natural polyamines, and decreased the intracellular levels of natural polyamines, especially when used in combination with DFMO. S-MeSpd and S,S-Me2Spm accumulated at somewhat higher intracellular concentrations than the other stereoisomers, and most effectively depleted the natural polyamines. When different analogue concentrations (1, 10 or 100 μM) were tested, it was found that the higher the analogue concentration, the more efficient the reduction of the natural polyamines (results not shown). As we reported previously , all analogues supported cell growth during DFMO-induced polyamine depletion for 3 days (results not shown). Of Me2Spm stereoisomers, only S,S-Me2Spm was metabolized to MeSpd (Table 2), and both MeSpd enantiomers were metabolized to MeSpm (α-methylspermine) by spermine synthase (Table 1). These MeSpm enantiomers can be then further converted into MeSpd and Spd by SMO, with S-MeSpm being preferred to R-MeSpm (results not shown).
Table 3 depicts AdoMetDC activity and the levels of SAM and deSAM (decarboxylated SAM) in DU145 cells treated with DFMO and the stereoisomers or natural polyamines for 24, 48 and 72 h. As expected, DFMO caused an induction of the enzyme activity, and led to accumulation of deSAM, and the treatment with natural polyamines or their analogues suppressed the enzyme activity and reduced the amount of deSAM. Based on both activity and nucleoside data, S-MeSpd suppressed the enzyme activity more than R-MeSpm, and R,R-Me2Spm was the best suppressor of AdoMetDC activity among the Me2Spm stereoisomers. Table 3 also shows that the methylated analogues were not as efficient AdoMetDC suppressors as their corresponding natural polyamines.
Uptake of the stereoisomers
The ability of the stereoisomers to compete for uptake with the natural polyamines was investigated in DU145 cells by 10 min competition assays with [14C]-labelled Spd or Spm (Figure 1). In such a short time, the antizyme does not downregulate the polyamine uptake. S-MeSpd and S,S-Me2Spm were clearly the least effective competitors. Similar results were obtained with mouse primary fetal fibroblasts (results not shown). Since we found differences in the ability of the stereoisomers to inhibit ODC activity, we hypothesized that the stereoisomers might differently affect the synthesis of the functional antizyme by inducing ribosomal frameshifting on antizyme mRNA. Therefore, antizyme-dependent uptake of the stereoisomers was studied by blocking antizyme expression by pre-incubating DU145 cells with a translation inhibitor, CHX. After 1 h, the analogues were added for a further 2 h, and the intracellular concentrations of the analogues were measured by HPLC. As shown in Figure 2, without CHX treatment there were no differences in the accumulation of the distinct stereoisomers of MeSpd or Me2Spm. By contrast, when CHX was added, S- and S,S-isomers accumulated at higher concentrations than the others, suggesting that these stereoisomers more effectively induce antizyme expression.
ODC antizyme frameshifting
To investigate in more detail whether S- and S,S-isomers induce antizyme frameshifting more efficiently than the other isomers, HEK-293 cells were transfected with a dual luciferase reporter plasmid containing the antizyme sequences required for polyamine-induced frameshifting between the Renilla and firefly luciferase genes . The firefly luciferase ORF lacks an initiation codon and can only be expressed as a fusion protein with Renilla luciferase if translational frameshifting occurs. Frameshifting efficiency was calculated as the ratio of the firefly to Renilla luciferase activities standardized to an in-frame control. Figure 3(a) indicates that S-MeSpd and S,S-Me2Spm were more potent than the other stereoisomers at inducing antizyme frameshifting. To rule out the possibility that the better frameshifting efficiency was caused by higher intracellular analogue concentrations, frameshifting was also tested in rabbit reticulocyte lysate. As depicted in Figure 3(b), differences existed between the stereoisomers, S-MeSpd and S,S-Me2Spm again being more efficient than the others.
Alternative splicing of SSAT
Since the stereoisomers differently affected SSAT activity, we next tested whether there were differences between the stereoisomers in their ability to regulate the alternative splicing of SSAT pre-mRNA. Mouse primary fetal fibroblasts were first exposed to the protein synthesis inhibitor CHX in order to block the NMD pathway and rapid degradation of the unproductive splice variant. After 1 h, polyamine analogues were added, and incubation was continued for 7 h. Figure 4 shows the amount of the unproductive splice variant (SSAT-X) relative to total SSAT mRNA (SSAT plus SSAT-X) as measured by quantitative RT–PCR. The treatment with CHX led to marked accumulation of the unproductive variant due to inhibition of NMD, and S,S-Me2Spm and S-MeSpd were the most effective isomers in decreasing the generation of the unproductive variant of SSAT.
Translational regulation of AdoMetDC in vitro
The expression of AdoMetDC is regulated mainly through translation of its uORF, where polyamines suppress the translation of downstream AdoMetDC protein . Since our cell culture data showed differences in the AdoMetDC activity and deSAM level between the stereoisomers, we further investigated whether they differently affected the translation of AdoMetDC uORF. We translated the uORF–luciferase fusion construct in rabbit reticulocyte lysate with or without added natural polyamines or the analogues. As expected, all polyamines suppressed the translation of the fusion construct (Figure 5). However, marked differences existed between the tested compounds, the natural polyamines being better suppressors than the analogues. In line with our cell culture data we found that S-MeSpd and R,R-Me2Spm were significantly more efficient suppressors of AdoMetDC uORF translation than the other stereoisomers.
Polyamines participate in several important cellular processes such as cell cycle regulation, differentiation, apoptosis, protein synthesis and modulation of ion channels. Some of those processes require covalent binding of polyamines to proteins, of which a well-known example is the hypusination of eIF5A (eukaryotic translation initiation factor 5A) protein . Other polyamine effects are mediated through their ionic interactions with cellular anionic sites. The difference between polyamines and inorganic cations such as Mg2+ is that the charges are distributed along a conformationally flexible carbon backbone, allowing them to interact with various targets, especially the polynucleotides RNA and DNA. Through these interactions, the polyamines influence the expression of many important genes, among which are the genes involved in their own metabolism.
The results from the present study revealed significant differences between the different stereoisomers of MeSpd and Me2Spm in their ability to regulate several components of polyamine homoeostasis, such as uptake and the expression of the rate-limiting biosynthetic and catabolic enzymes, ODC, AdoMetDC and SSAT. Earlier studies have indicated that polyamines and their close structural analogues induce +1 ribosomal frameshifting in ODC antizyme mRNA [9,11,27,32]. Our results from uptake and frameshifting studies show that S- and S,S-isomers were more effective than the other stereoisomers of MeSpd and Me2Spm at inducing antizyme frameshifting, and the results from in vitro frameshifting are in line with the cell culture studies. Furthermore, S- and S,S-stereoisomers were the most efficient in downregulating ODC activity. In fact, S,S-Me2Spm alone was even more potent in decreasing ODC activity than the inhibitor DFMO.
Since antizyme is known not only to promote ODC degradation but also to inhibit polyamine uptake [12,13,33], we examined the uptake and intracellular accumulation of these analogues. In 10 min uptake experiments, where the analogues apparently did not affect the amount of antizyme, S-MeSpd and S,S-Me2Spm were found to be the least efficient inhibitors of the uptake of Spd and Spm respectively. However, unexpectedly, after a longer culture time, S- and S,S-stereoisomers accumulated at slightly higher intracellular concentrations than R- and R,R-isomers. When the concentrations of the added analogue as well as its metabolic products were added together (Tables 1 and 2), the differences between the stereoisomers were smaller, but still present. The reason why S- and S,S-isomers induced antizyme frameshifting, but still accumulated at higher concentration than the other isomers, is not understood. The analogues may divergently modulate some yet unidentified cellular regulatory systems.
The mechanisms of polyamine-mediated regulation of the key enzymes of polyamine metabolism are currently not fully understood. It is possible that polyamines directly interact with specific site(s) of target mRNA(s). This view is supported by the work of Higashi et al. , who recently found that Spd binds to and induces a structural change in a bulged-out region of double-stranded RNA, and that the effect is different than that induced by Mg2+. However, it has also been suggested that polyamines may modulate frameshifting through interactions with the translational machinery, such as tRNAs or the ribosome, and further detailed mechanistic studies are needed to clarify this point.
The interpretation of our cell culture results is complicated by the fact that the intracellular concentrations of S-MeSpd and S,S-Me2Spm were somewhat higher than those of the other isomers. However, this observation must be interpreted cautiously, as the intracellular localization and the level of free analogue cannot readily be determined. To overcome these technical problems, we used in vitro systems to verify the results from cell culture experiments, which indicated that differences existed between the stereoisomers. Our view that the regulation is due to the stereospecificity is also supported by our recent findings that the stereoisomers of α-methylpolyamines significantly differ in their abilities to condense DNA and protect it from hydrogen peroxide-induced oxidative stress in in vitro systems where no endogenous polyamines are present (T.J. Thomas and M.T. Hyvönen, unpublished work).
Earlier publications indicate that some polyamine metabolic enzymes are stereospecific for their substrates, such as spermidine synthase  and ODC . The use of novel chiral α-methylated polyamines led to the discovery of a hidden stereospecificity for PAO , SMO and deoxyhypusine synthase . In addition, others have reported that deoxyhypusine hydroxylase [37,38], and similarly spermine synthase , show stereocontrol for unsaturated spermidine analogues.
A vast number of polyamine analogues, of which most are achiral, have been synthesized for studying the physiological roles of polyamines and developing drugs for the treatment of cancer and parasitic diseases. Among these compounds are N-alkylated, C-alkylated, unsaturated, fluorinated and amino-oxy analogues . The increasing knowledge about polyamine metabolism and their physiological roles has opened up possibilities to develop potent antimetabolites such as the cytotoxic N-alkylated analogue DENSPM (N1N11-diethylnorspermine), which is currently undergoing trials for the chemotherapy of colon carcinoma. One key issue in the drug development is acceptable host toxicity, since total polyamine depletion is difficult to achieve and also may be highly toxic to the host. An important goal is to find differences in polyamine metabolism and cellular targets between the host and parasite, or normal and malignant tissue, which can be used to selectively target the drugs. Functional polyamine mimetics with diverse effects on polyamine metabolism may offer the means to selectively regulate key metabolic pathways. Here we show that the optical isomers of α-methylated polyamine analogues are promising examples of this approach, by providing evidence that their metabolism and biological effects can be modulated by relatively minor changes such as the configuration of the chiral centre. For example, S-MeSpd supports hypusine synthesis in DU145 cells during prolonged polyamine depletion with DFMO, whereas R-MeSpd does not . Thus, the combined use of DFMO and R-MeSpd may offer a practical means to achieve an eIF5A-dependent cytostatic response in vivo. In further analogue development studies, the size, position and chemical nature of substituents can be altered to achieve additional novel biological or metabolic properties for research and/or therapeutic use.
Mervi Hyvönen and Christine Anderson performed the experiments. Nikolay Grigorenko, Alex Khomutov and Jouko Vepsäläinen synthesized the stereoisomers. Mervi Hyvönen, Michael Howard, Tuomo Keinänen, Juhani Jänne and Leena Alhonen planned the experiments. Mervi Hyvönen, Michael Howard and Tuomo Keinänen wrote the manuscript.
This work was supported by the Academy of Finland [grant numbers 124185 and 128702] and the Russian Foundation for Basic Research [grant number 08-04-91777]. A portion of this work was funded by the Muscular Dystrophy Association and the National Institutes of Health.
The authors thank Dr David Morris for providing AdoMetDC uORF–luciferase fusion constructs, and Mrs Anne Karppinen, Mrs Arja Korhonen and Mrs Tuula Reponen (University of Kuopio, Finland) for skilful technical assistance.
Abbreviations: AdoMetDC, S-adenosyl-L-methionine decarboxylase; CHX, cycloheximide; DFMO, difluoromethylornithine; DMEM, Dulbecco's modified Eagle's medium; eIF5A, eukaryotic translation initiation factor 5A; FBS, fetal bovine serum; HEK-293 cell, human epithelial kidney-293 cell; MeSpd, α-methylspermidine; MeSpm, α-methylspermine; Me2Spm, α,ω-bismethylspermine; NMD, nonsense-mediated mRNA decay; ODC, ornithine decarboxylase; PAO, polyamine oxidase; RT–PCR, reverse transcription–PCR; SAM, S-adenosyl-L-methionine; deSAM, decarboxylated SAM; SMO, spermine oxidase; Spd, spermidine; Spm, spermine; SSAT, spermidine/spermine N1-acetyltransferase; uORF, upstream open reading frame
- © The Authors Journal compilation © 2009 Biochemical Society