The polyamines putrescine, spermidine, and spermine are widely distributed polycationic compounds essential for cellular functions. Intracellular polyamine pools are tightly regulated by a complex regulatory mechanism involving de novo biosynthesis, catabolism, and transport across the plasma membrane. In mammals, both the production of polyamines and their uptake from the extracellular space are controlled by a set of proteins named antizymes and antizyme inhibitors. Dysregulation of polyamine levels has been implicated in a variety of human pathologies, especially cancer. Additionally, decreases in the intracellular and circulating polyamine levels during aging have been reported. The differences in the polyamine content existing among tissues are mainly due to the endogenous polyamine metabolism. In addition, a part of the tissue polyamines has its origin in the diet or their production by the intestinal microbiome. Emerging evidence has suggested that exogenous polyamines (either orally administrated or synthetized by the gut microbiota) are able to induce longevity in mice, and that spermidine supplementation exerts cardioprotective effects in animal models. Furthermore, the administration of either spermidine or spermine has been shown to be effective for improving glucose homeostasis and insulin sensitivity and reducing adiposity and hepatic fat accumulation in diet-induced obesity mouse models. 

Numerous experiments have shown that due to their polycationic nature, polyamines can readily bind to negatively charged biomolecules including DNA, RNA, proteins, and phospholipids, modulating in many cases the function of these macromolecules (12, 13). When polyamine metabolism was pharmacologically or genetically altered, many relevant biochemical, and cellular processes resulted affected, including translation, transcription, signal transduction, cell proliferation, and differentiation, apoptosis, or cell stress response (14–19). However, the molecular mechanisms by which polyamines exert their biological effects are only partially understood (20). In humans, only two genetic diseases affecting enzymes of the biosynthetic pathway of polyamines have been described so far (21, 22). However, several lines of evidence have implicated a dysregulation of the polyamine system in hyperproliferative and neurodegenerative diseases (23–27). The alteration of polyamine metabolism in several types of cancer has sustained the interest in using the polyamine pathway as a target for anticancer therapy (28, 29). In this regard, it is clear that the limitation of the cellular polyamine content has a potential interest in cancer chemoprevention or in the treatment of other pathologies such as certain infectious diseases and type 1 diabetes (28, 30, 31). On the other hand, experiments conducted in different cellular and animal models have revealed that the exogenous administration of spermidine or other natural polyamines may exert beneficial effects by affecting processes such as cellular stress, chronic inflammation, or dysregulated lipid or glucose metabolism (32– 34).

 In this review, we will discuss different aspects related with the homeostasis of polyamines in mammalian tissues, including the relevance of the polyamines generated by the gut microbiota, how the endogenous levels of polyamines can be affected by the administration of exogenous polyamines, and the impact that these treatments may exert on the evolution of the biochemical changes associated with aging or obesity-related metabolic diseases. POLYAMINE FUNCTIONS Polyamines are essential for life. In fact, the inactivation of the genes that control the biosynthesis of putrescine or spermidine are embryo-lethal in mice (35, 36). This is not surprising due to the multiple essential cellular actions that have being ascribed to polyamines. Thus, owing to their general interaction with nucleic acids, they can affect many processes in which DNA, RNA, or proteins participate as substrates (13, 37, 38). Particularly interesting is the relationship between polyamines and reactive oxygen species (ROS). Although polyamine catabolism may generate potentially toxic products such as H2O2 and polyamine derived aldehydes associated to pathophysiological consequences (39, 40), numerous in vitro and in vivo experiments have suggested that spermine and spermidine may act as scavengers of ROS, and then protecting DNA from oxidative damage (41–43). 

The diamine putrescine is synthesized by ornithine decarboxylase (ODC), a key enzyme that presents a complex regulation (57). This diamine can be converted into spermidine by the addition of an aminopropyl moiety donated by decarboxylated S-adenosyl methionine (dcAdoMet) through the action of spermidine synthase (5, 58). The tetra-amine spermine is synthesized by another aminopropyl transferase (spermine synthase) through the incorporation of another aminopropyl group to the amino butyl end of spermidine (5, 59). The second key-rate enzyme in the synthesis of spermidine and spermine is S-adenosyl methionine decarboxylase (AdoMetDC, AMD1), which produces dcAdoMet used as aminopropyl donor for the formation of both polyamines (60). Other biochemical route that participates in the regulation of the intracellular levels of polyamines is the so-called back-conversion pathway of polyamines (Figure 2). Spermidine/spermine N1 acetyltransferase (SSAT) is a cytosolic enzyme that plays a key role in reducing intracellular polyamine levels, through the acetylation of spermine and spermidine to produce in first instance acetyl-spermine or acetyl-spermidine (61, 62), which are either excreted from the cell or oxidized to spermidine or putrescine, respectively, by an acetyl polyamine oxidase (PAOX) (63). PAOX is a FAD-dependent enzyme located in peroxisomes that generates, apart from the corresponding polyamine, the potentially toxic byproducts 3-acetamidopropanal and H2O2 (40).

The polyamine levels in the intestinal lumen change over time after a meal, from millimolar levels immediately after the ingestion, to much lower values in the fasting period (96). Luminal polyamines can be taken up mainly by the small intestine. Experiments using radiolabeled polyamines administered intragastrically to rats showed that polyamines can be absorbed and distributed to different tissues (88). This distribution was not uniform, since polyamines were accumulated preferentially in those tissues stimulated to proliferate (88). The first barrier for the uptake of polyamines from the intestinal lumen is formed by the epithelial cells of the intestinal mucosa. In isolated enterocytes, as in other mammalian cells in culture, polyamines can be taken up by different specific polyamine transporters, as already commented in the polyamine transport section. The in vivo polyamine uptake by the intestinal cells is more complex due to the existence of different polyamine transporters in the apical and basolateral membranes, as shown by studies using brush-border and basolateral membrane vesicles of the enterocyte (97). According with experimental data, luminal polyamines could be taken by enterocytes by transport across the apical membrane and extruded across the basolateral membrane by low affinity transporters to the systemic circulation (96). It was also hypothesized that the majority of luminal polyamines could be passively absorbed via the paracellular route (96). Whereas, most of spermidine and spermine taken up by the intestinal cells are not metabolized in these cells, a variable proportion of putrescine is transformed into other compounds including spermidine, γ-aminobutyric acid (GABA) and succinate (88, 98). 

In the small intestine of rats, putrescine can be transformed into succinate acting as a source of instant energy (99). The absorption of polyamines appears to be rapid, since experiments using an ex vivo rat model revealed that values about 70% of the 14C-polyamines administered to the jejunal lumen were found in the portal vein, after 10 min of polyamine administration (100). Most of the studies on luminal polyamine uptake and their distribution through the body have been based on the acute administration of a low dose of labeled polyamines to rats. Recently, as described in other section, many studies have reported beneficial effects of a prolonged oral administration of either spermidine or spermine to rodents (101–104). However, in most studies tissue polyamine levels were not reported. In mouse models, prolonged administration of polyamine-rich diets have been seen to increase blood levels of spermidine and/or spermine (56, 105, 106). In aged mice spermidine levels significantly increased in blood (107) and liver (101) after supplementation of the drinking water with 3 mM spermidine for 6 months. In line with this, a 28-day oral supplementation of adult mice with 50 mg/kg of spermidine resulted in a significant increase of spermidine in whole blood and heart (but not in brain) of females, but not in males (106). In humans it has been shown that a prolonged intake (2 months) of polyamine-rich products such as natto (fermented soy) produces a significant rise in the levels of spermine (but not spermidine) in blood (56). 

More recently, the results of a clinical trial using spermidine supplements in older human subjects have been reported, showing no differences in blood polyamine levels between controls and spermidine-supplemented individuals at 3 months of follow-up (106)s. A part of the absorbed luminal polyamines remains in the intestinal cells. This is not surprising since the intestinal tissue is one of the most rapidly proliferating tissues, and in general the polyamine uptake of intestinal cells is associated with the proliferative stage. Dietary polyamines might be important for the development of the digestive tract, and also for the maintenance of the adult digestive tract (108). Moreover, polyamine reservoirs are not only responsible for maintaining the rapid turnover and high proliferation rates of the intestinal epithelial cells but also for enhancing the integrity of the intestinal barrier. Thus, polyamines are able to stimulate the production of intercellular junction proteins, such as occludin, zonula occludens 1, and E-cadherin, which are essentials to regulate the paracellular permeability and reinforcing epithelial barrier function (109). Furthermore, intestinal polyamine pools are necessary for the postnatal development of the gastrointestinal tract. These data were confirmed in a study using pup rats, where the polyamine administration was able to induce the production of mucus and secretory IgA in the small intestine, while rats fed with a polyamine-deficient diet developed intestinal mucosal hypoplasia (110).

 In addition, the oral administration of polyamines to suckling rats accelerated the maturation process affecting intestine, liver and pancreas. Interestingly, the precocious maturation of the intestine only was achieved when spermine was given orally and not when other routes of administration were used, suggesting that this process requires the interaction of the polyamine with the luminal side of the mucosa (111). On the other hand, although it has been postulated that the polyamines present in the human milk, could contribute to prevent food allergic processes in neonates, more studies appear to be necessary (112). As commented above, antizymes and antizyme inhibitors play a relevant role in cellular polyamine uptake. However, the influence of these proteins on intestinal polyamine absorption is uncertain. Both ODC and AZ1 mRNA transcripts are abundantly expressed in small intestine (113). Treatment of the intestinal epithelial cell line IEC-6 with 10µM of spermidine or spermine induced the synthesis of the antizyme protein negatively affecting polyamine uptake (114). In this cell line AZ1 is induced in the absence of amino acids, through a mechanism involving mTORC1 (115). Other studies using a human colon adenocarcinoma cell line (Caco-2) revealed that amino acid supplementation also modulates antizyme induction (116), whereas amino acid restriction increased spermidine uptake by Caco-2 cells, by an antizyme-independent mechanism (117). The levels of AZ mRNA have been analyzed in cells isolated from jejunal crypt-villus axis. Whereas, AZ mRNA levels were high in cells from the small intestinal crypts, the message fell to near undetectable levels in cells of the villus tip (118).