1-aminocyclopropane-1-carboxylic acid (ACC) in plants: more than just the precursor of ethylene!
Bram Van de Poel1,2 and Dominique Van Der Straeten2,*
1Department of Cell Biology and Molecular Genetics, University of Maryland, College Park, MD, USA
2Laboratory of Functional Plant Biology, Department of Physiology, Ghent University, Ghent, Belgium
Edited by: Domenico De Martinis, ENEA Italian National Agency for New Technologies, Energy and Sustainable Economic Development, Italy
Reviewed by: Brad Binder, University of Tennessee-Knoxville, USA; Chi-Kuang Wen, Chinese Academy of Sciences, China
*Correspondence: Dominique Van Der Straeten, Laboratory of Functional Plant Biology, Department of Physiology, Ghent University, K.L. Ledeganckstraat 35, 9000 Ghent, Belgium e-mail: email@example.com
This article was submitted to Plant Physiology, a section of the journal Frontiers in Plant Science.
Author information ►Article notes ►Copyright and License information ►
Received 2014 Oct 10; Accepted 2014 Oct 28.
Copyright © 2014 Van de Poel and Van Der Straeten.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
This article has been cited by other articles in PMC.
Ethylene is a simple two carbon atom molecule with profound effects on plants. There are quite a few review papers covering all aspects of ethylene biology in plants, including its biosynthesis, signaling and physiology. This is merely a logical consequence of the fascinating and pleiotropic nature of this gaseous plant hormone. Its biochemical precursor, 1-aminocyclopropane-1-carboxylic acid (ACC) is also a fairly simple molecule, but perhaps its role in plant biology is seriously underestimated. This triangularly shaped amino acid has many more features than just being the precursor of the lead-role player ethylene. For example, ACC can be conjugated to three different derivatives, but their biological role remains vague. ACC can also be metabolized by bacteria using ACC-deaminase, favoring plant growth and lowering stress susceptibility. ACC is also subjected to a sophisticated transport mechanism to ensure local and long-distance ethylene responses. Last but not least, there are now a few exciting studies where ACC has been reported to function as a signal itself, independently from ethylene. This review puts ACC in the spotlight, not to give it the lead-role, but to create a picture of the stunning co-production of the hormone and its precursor.
Keywords: 1-aminocyclopropane-1-carboxylic acid (ACC), ethylene, conjugation, deaminase, transport, signaling
THE DISCOVERY OF ACC
The discovery of ethylene as a plant growth regulator can be attributed to the work of the Russian scientist Neljubov (1901). He reported that dark-grown pea seedlings showed a reduced hypocotyl growth in combination with an exaggerated hypocotyl bending when exposed to illumination gas (Neljubov, 1901). Neljubov (1901) could pinpoint ethylene gas as the active component that caused dark-grown pea seedlings to bend, by flowing the illumination gas over several filters prior to exposing the seedlings. This typical ethylene response of dark-grown seedlings was later defined as the triple response: (1) shortening of the hypocotyl and roots, (2) radial swelling of the hypocotyl, and (3) the exaggeration of the apical hook (Knight et al., 1910). In 1934, conclusive evidence that ethylene is a natural product from plants, was presented by the English scientist (Gane, 1934). It took another 30 years before the primary steps of the ethylene biosynthesis pathway were elucidated (see Figure 1). Lieberman and Mapson (1964) first reported that ethylene could be produced from the amino acid methionine, taking advantage of the high rates of ethylene production from apples for their experimental work (Figure 1). 13 years later, Adams and Yang (1977) made tremendous progress in understanding the biosynthesis pathway of ethylene, when they discovered that S-adenosyl-L-methionine (SAM) was an intermediate between methionine and ethylene. Yang and co-workers also showed that 5′-methylthioadenosine (MTA) was formed as a by-product from SAM and that MTA could be recycled back to methionine (Murr and Yang, 1975). The elaboration of the different reaction steps of the methionine cycle in plants, now often referred to as the Yang-cycle, was mainly inspired by the biochemical similarities between the plant pathway and the methionine salvage cycle which was already known for prokaryotes, yeast, and mammalians. An-up-to-date overview of the methionine and SAM metabolism in plants is given by Sauter et al. (2013). The major discovery that made the methionine cycle in plants unique from all other organisms, was the characterization of 1-aminocyclopropane-1-carboxylic acid (ACC) as the intermediate between SAM and ethylene (Adams and Yang, 1979). Adams and Yang (1979) were able to identify ACC as the precursor for ethylene by feeding experiments on apple tissue, using radio-labeled methionine. Upon incubation of apple disks, they observed a shift from ethylene production in air, toward an unknown compound that was retained in the tissue when treated with nitrogen (lack of oxygen inhibits oxidation of ACC toward ethylene). By using a pH-dependent ion mobility assay, they could characterize this unknown component as an amino acid. Subsequently, the component was identified as ACC, using co-migration of synthetic ACC for both paper-chromatography and paper-electrophoresis (Adams and Yang, 1979). They further showed that the conversion of radioactively labeled methionine toward ethylene decreased when unlabeled ACC was supplemented, yet the conversion of labeled ACC to ethylene was almost not affected when unlabeled methionine was supplemented, suggesting that externally supplied ACC is in fact used to produce ethylene. Additional evidence for ACC being the intermediate precursor between SAM and ethylene was obtained by treating apple tissue with [S]-trans-2-amino-4(2′-aminoethoxy)trans-3-butenoic acid, also known as AVG (2-amino-ethoxy-vinylglycine), a pyridoxal-5′-phosphate (PLP or vitamin B6) dependent enzyme inhibitor, which was later known to inhibit the enzymatic conversion of SAM toward ACC. The identification of ACC as the precursor of ethylene was a major breakthrough in the understanding of the ethylene biosynthesis pathway in plants, and was part of the foundation for many new discoveries in the field of ethylene biology.
Structural scheme of ethylene biosynthesis and 1-aminocyclopropane-1-carboxylic acid (ACC) conjugation/metabolism. The amino acid methionine is converted to S-adenosyl-L-methionine (SAM) by SAM-synthetase (SAMS) with the requirement of ATP. The general...
ACC AND ETHYLENE BIOSYNTHESIS
As mentioned above, ACC is produced from SAM, releasing MTA. This reaction is catalyzed by the enzyme ACC-synthase (ACS; Boller et al., 1979). ACS is a member of the PLP-dependent enzymes, which use vitamin B6 as a co-factor for its enzymatic function. ACS was localized in the cytosol by activity assays on extracts retrieved after differential centrifugation (Boller et al., 1979). ACS genes were first characterized in zucchini by Sato and Theologis (1989) and in tomato by Van Der Straeten et al. (1990). ACS is encoded by a multigene family of 12 members in Arabidopsis, eight of which encode functional ACC synthases [ACS2 (named ACS1 in Van Der Straeten et al., 1992), ACS4-9, ACS11]. In addition, there is one inactive isoform (AtACS1) and one pseudogene (AtACS3; Yamagami et al., 2003). ACS was found to form functional dimers of which the 3D structure was determined by Capitani et al. (1999). The formation of heterodimers increases the structural and functional complexity of the ACS protein family (Tsuchisaka et al., 2009). The large ACS gene family displays a tissue-specific and differential expression pattern in Arabidopsis (Tsuchisaka and Theologis, 2004). Using single and multiple acs knock-out mutants, it was demonstrated that there are specific developmental and physiological roles for individual members of the ACS gene-family, but also that there is a complex combinatorial interplay amongst them (Tsuchisaka et al., 2009). A diverse group of internal and external signals modulate the level of ethylene biosynthesis in numerous plant species, acting at the level of ACS gene expression. These inducers include auxin, cytokinin, brassinosteroids, ethylene, copper, mechano-stimuli, ozone, pathogens and wounding (Van Der Straeten et al., 1992; Rodrigues-Pousada et al., 1993; Botella et al., 1995; Cary et al., 1995; Liang et al., 1996; Vahala et al., 1998; Woeste et al., 1999).
Three types of ACS proteins are recognized, based on their C-terminal structure. Type I ACS proteins contain in their C-terminal domain one putative calcium-dependent protein kinase (CDPK) phosphorylation target site and three mitogen-activated protein kinase (MAPK) phosphorylation sites (Yoon and Kieber, 2013). Type II ACS proteins only contain the MAPK phosphorylation sites, while type III ACS do not contain any phosphorylation sites (Yoon and Kieber, 2013). These post-translational phosphorylation sites play an important role in the stability of the ACS protein (Chae and Kieber, 2005). Both in Arabidopsis (Chae et al., 2003; Kim et al., 2003; Liu and Zhang, 2004; Wang et al., 2004; Yoshida et al., 2005, 2006; Joo et al., 2008; Christians et al., 2009; Lyzenga et al., 2012) and in tomato (Tatsuki and Mori, 2001; Kamiyoshihara et al., 2010) it was shown that differential phosphorylation of certain ACS members directed the protein for proteasomal degradation. Protein stability of certain ACS members is further regulated by the protein phosphatase 2A (PP2A; Skottke et al., 2011) and PP2C (Ludwikow et al., 2014), demonstrating a complex balance between phosphorylation and dephosphorylation to secure protein activity and stability.
The second ethylene biosynthesis protein is ACC-oxidase (ACO), which converts ACC to ethylene in the presence of oxygen. It took a long time before ACO activity could be demonstrated in vitro. The key aspect in isolating ACO was the addition of ascorbic acid (vitamin C) to the extraction media, as was first reported by Ververidis and John (1991) who isolated ACO from melon tissue and quantified in vitro ACO activity. Although the exact role of ascorbic acid for protein stability/activity remained uncertain for a long time (Rocklin et al., 2004), it was recently clarified that ascorbic acid participates in the ring opening of ACC, by providing a single-electron to the active site (Murphy et al., 2014). This catalytic reaction releases ethylene and a cyanoformate ion [NCCO2]-, which is subsequently decomposed into CO2 and CN- (Murphy et al., 2014). The reactive cyanide (CN-) is subsequently detoxified by β-cyanoalanine synthase to produce β-cyanoalanine (Miller and Conn, 1980). ACO belongs to the superfamily of dioxygenases that require iron (Fe2+) as co-factor and bicarbonate as activator (Dong et al., 1992; Zhang et al., 2004). The subcellular localization of ACO remains vague, as some studies localize ACO in the cytosol (Reinhardt et al., 1994; Chung et al., 2002; Hudgins et al., 2006), while other localize ACO at the plasma membrane (Rombaldi et al., 1994; Ramassamy et al., 1998). Although the ACO protein sequence does not contain any predicted transmembrane domains, it is still possible that the protein associates with the plasma membrane via (in)direct interactions. ACO is also encoded by a multigene family of five members in Arabidopsis [ACO1, 2, 4, At1g12010 (AOC3) and At1g77330 (ACO5)]. Expression of different members of the tomato ACO family in Escherichia coli showed that each isoform had a specific in vitro enzyme activity (Bidonde et al., 1998).
It is well accepted that ACS is the rate limiting step of ethylene biosynthesis in plants (Yang and Hoffman, 1984) although there are examples where ACO is the rate limiting step, e.g., during post-climacteric ripening of tomato fruit (Van de Poel et al., 2012). Three ACO genes are also auto-regulated by ethylene in Arabidopsis (De Paepe et al., 2004). This might suggest that the regulation of ACO expression and/or activity is more complex than anticipated. There are some hints for a putative post-transcriptional and/or post-translational regulatory mechanisms of ACO as suggested by Dilley et al. (2013) and as investigated through mathematical modeling by Van de Poel et al. (2014a).
Both ACS and ACO are two well-studied enzymes that exclusively participate in the ethylene biosynthesis pathway. Both proteins are well characterized, but many more questions remain unanswered. While the post-translational regulation of ACS has been revealed, the biochemical and mechanistic details of this protein modification are still unclear. Transcriptional and functional characterization of the different ACS gene-family members has shone light on the combinatorial interplay, nevertheless, much more work is needed to elucidate the exact role of each isoform and how they interact with each other. Much less is known about the post-translational regulation and combinatorial interplay of ACO. A lack of genetic studies focusing on ACO, raises the question whether or not ACO shares a similar structural, biochemical and post-translational complexity as ACS.
ACC AS A PIVOTAL MOLECULE: ACC CONJUGATES AND THE CONTROL OF ETHYLENE BIOSYNTHESIS
1-Aminocyclopropane-1-carboxylic acid is best known as the direct precursor of ethylene in the ethylene biosynthesis pathway. However, there also exist three different conjugates of ACC, suggesting that the biochemical regulation of the available ACC pool is more complex than anticipated, which in turn can possibly affect the eventual levels of the plant hormone ethylene.
Shortly after the identification of ACC as the intermediate between SAM and ethylene in the ethylene biosynthesis pathway, a first conjugated form of ACC, called malonyl-ACC (MACC), was discovered by Amrhein et al. (1981) in buckwheat seedlings and by Hoffman et al. (1982) in wheat leaves. MACC is formed by ACC-N-malonyl transferase (AMT) which was purified from tomato extracts (Martin and Saftner, 1995), although not structurally characterized. It was shown that the conjugation of ACC into MACC was stimulated by ethylene in preclimacteric tomatoes (Liu et al., 1985a), grapefruit flavedo (outer peel; Liu et al., 1985b) and tobacco leaves (Philosoph-Hadas et al., 1985), indicative for a feedback control of ethylene biosynthesis. Martin and Saftner (1995) also showed that the activity of AMT was ethylene inducible and that its activity correlated with the increase in ethylene production during climacteric ripening of tomato (Martin and Saftner, 1995). The exact amino acid and gene sequences of AMT are not yet known, and no putative AMT gene is annotated in the Arabidopsis genome, limiting more in-depth genetic and molecular studies. Because MACC does not participate in any other known biological conversions, MACC formation might be a mechanism to control the available ACC pool. This hypothesis was further strengthened by the observation that MACC could be translocated from the cytosol into the vacuole (and back) by ATP-dependent tonoplast carriers (Bouzayen et al., 1988, 1989; Tophof et al., 1989). Nonetheless, the reconversion of MACC toward ACC by an unknown MACC-hydrolase was reported twice in literature (Jiao et al., 1986; Hanley et al., 1989). The ability to hydrolyze MACC back into ACC and the ability to ‘store’ MACC in the vacuole is an interesting mechanism to regulate the cellular availability of MACC. Moreover, because MACC has no other biochemical role besides being an ACC conjugate, the regulation of MACC levels can also affect the available pool of ACC and possibly ethylene production levels. This hypothesis was investigated by in silico mathematical modeling, showing that the reconversion of MACC to ACC could have a potential stimulating effect on ethylene production during climacteric fruit ripening of tomato (Van de Poel et al., 2014a).
A second important derivative of ACC is γ-glutamyl-ACC (GACC), which was discovered in crude tomato extracts of ACC-N-malonyltransferase (Martin et al., 1995). These crude protein extracts were able to form a new ACC derivative, which could be identified as GACC (Martin et al., 1995). GACC is formed by the reaction of ACC with the tripeptide glutathione (GSH) by a γ-glutamyl-transpeptidase (also called γ-glutamyl-transferase, GGT; Martin et al., 1995; Martin and Slovin, 2000). While a first report, based on in vitro studies, stated that GACC was the most abundant ACC derivative in tomato (Martin and Saftner, 1995; Martin et al., 1995), another study found that MACC was the most abundant ACC derivative in vivo in tomato fruit during climacteric ripening (Peiser and Yang, 1998). The Arabidopsis genome contains four genes (GGT1-4), of which only GGT1 and GGT2 are catalytically active (Martin et al., 2007). Both GGT1 and GGT2 are co-expressed predominantly in rapidly growing tissue, and are localized extracellularly, which raised the question about the role of extracellular GACC (Martin et al., 2007). A knock-out mutant of GGT1 shows rapid senescence, while GGT3 knock-outs have a reduced rosette size and silique number (Martin et al., 2007). The effect of GACC formation by GGT on ACC availability and possibly ethylene biosynthesis remains to be investigated.
A third derivative of ACC is jasmonyl-ACC (JA-ACC). This molecule was discovered by screening for amino acid conjugates of JA, using GC-MS (Staswick and Tiryaki, 2004). Four amino acid conjugates of JA (JA-Ile, JA-Val, JA-Leu, and JA-Phe) were quantified in Arabidopsis tissue (Staswick and Tiryaki, 2004). Interestingly, the same authors also demonstrated that JA forms a conjugate with ACC (JA-ACC) in Arabidopsis leaves (Staswick and Tiryaki, 2004). Recombinant JAR1 enzyme was found to be able to form JA-ACC in vitro. Strangely, levels of JA-ACC were higher in the leaves of jar1 mutants compared to wild-type plants. It was also shown that JA-ACC inhibits root growth in Arabidopsis. Elegant genetic experiments with JA signaling mutants (coi1-35) showed that the JA-ACC-induced root inhibition was independent of JA signaling. Furthermore, the ethylene signaling mutant etr1-1 and the double mutant etr1-1 jar1-1 were insensitive to the JA-ACC treatment and displayed no inhibition of root growth, indicating that JA-ACC acts via the ethylene signaling pathway (Staswick and Tiryaki, 2004). Most likely the ACC moiety of JA-ACC is responsible for an increase in ethylene production which results in the root growth inhibition response. These experiments suggest that JA-ACC might serve as a pivotal molecule which can function as a modulator of the hormonal cross-talk between the ethylene and jasmonic acid pathway, although the exact molecular and biochemical mechanism of JA-ACC function remains unclear.
The three above-mentioned derivatives of ACC (MACC, GACC, and JA-ACC) are perhaps not the only ones. Future metabolic studies might reveal additional conjugates of ACC. Nonetheless, these three derivatives can potentially play an important role in the regulation of the pool of available ACC, which in turn can affect eventual ethylene production levels, with physiological and developmental consequences. Genetic perturbations of the formation of ACC derivatives could be a useful tool to unravel their exact roles. In addition, a more detailed structural and biochemical characterization of the enzymes involved in the formation of ACC derivatives is essential.
ACC DEAMINASE AND PLANTS
As mentioned above, plants possess several mechanisms to control their pool of ACC, for example by converting it to ethylene or to conjugates like MACC, GACC, or JA-ACC. Another unique way to metabolize ACC, is the deamination of ACC. ACC deaminase was first discovered in bacteria. Some plant growth-promoting bacteria are capable of processing the plant-borne ACC by converting it into ammonia and α-ketobutyrate using the enzyme ACC deaminase (Honma and Shimomura, 1978). ACC deaminase was retrieved in for example, Pseudomonas sp. strain ACP (Honma and Shimomura, 1978), Pseudomonas chloroaphis 6G5 (Klee et al., 1991), Pseudomonas putida GR12-2 (Jacobson et al., 1994) and Pseudomonas putida UW4 (Hontzeas et al., 2004).
The bacterial ACC deaminase is a PLP-dependent enzyme with a rather low affinity for ACC (the reported Km value is 1.5–15 mM; Glick et al., 2007). Nonetheless, relatively low concentrations of ACC (100 nM) can already induce the expression of the ACC deaminase gene (acdS), but so do other amino acids like L-alanine, DL-alanine, and DL-valine (Jacobson et al., 1994). acdS expression is also under the regulation of the nitrogen fixation (nif) promotor of some Rhizobia, linking ACC deaminase with nodule formation (Nukui et al., 2006; Nascimento et al., 2012).
Plant growth-promoting bacteria that harbor ACC deaminase must interact with the root environment in order to access plant-produced ACC. It was shown that root exudates contain certain amounts of ACC, which might attract ACC deaminase containing bacteria and establish the rhizosphere interaction (Penrose et al., 2001). It has been proposed that bacterial ACC deaminase can reduce the endogenous ethylene levels of plant roots by limiting the amount of available ACC, which will in turn prevent ethylene-induced root growth inhibition, and thus promote plant growth (Glick et al., 1998, 2007; Glick, 2014). Another model proposes that plant growth-promoting bacteria produce IAA which can be taken up by the plant, and can induce the expression of ACS, resulting in an increase in ACC production, providing a nitrogen supply for the bacteria (Glick et al., 1998, 2007; Glick, 2014).
There are many beneficial effects of ACC deaminase containing bacteria on plant growth, particularly in relation to stress tolerance. For instance, ACC deaminase containing bacteria can reduce stress susceptibility of plants during flooding (Barnawal et al., 2012; Li et al., 2013), drought (Mayak et al., 2004a), salinity (Mayak et al., 2004b; Nadeem et al., 2007, 2010), flower senescence (Nayani et al., 1998; Ali et al., 2012), metal pollution (Glick, 2010), organic pollution (Gurska et al., 2009) and pathogens (Glick, 2014 and references therein). In addition, it has been reported that the presence of ACC deaminase can increase the symbiotic performance of Rhizobial strains (Ma et al., 2003).
Hence, bacterial ACC deaminase is also used as a biotechnological tool to control endogenous ACC levels and consequently lower ethylene production in plants. Transgenic plants overexpressing bacterial ACC deaminase were shown to be more resistant to growth inhibition when confronted with fungal pathogens (Lund et al., 1998; Robison et al., 2001), salt stress (Sergeeva et al., 2006), and metals (Grichko et al., 2000; Nie et al., 2002).
Plants themselves also contain a homolog of the bacterial ACC deaminase. In Arabidopsis, it was demonstrated that the previously known enzyme D-cysteine desulfydrase also possesses ACC deaminase activity (McDonnell et al., 2009). Antisense lines showed a decreased ACC deaminase activity, an increased sensitivity to ACC and produced more ethylene (McDonnell et al., 2009). These results indicate that the plant-specific ACC deaminase might be another metabolic shunt regulating ACC levels and ethylene production in plants.
ACC TRANSPORT AND ITS ROLE DURING ROOT STRESS
Besides the biosynthesis, conjugation, and catabolism of a hormone, short or long range transport is another important aspect to regulate proper dosage of a hormonal signal within an organism. Often, hormones are synthesized at one site and transported to another site for their action. Hormonal transport from one cell to another is an advanced process, that facilitates tissue specific or long-distance physiological processes or stress responses. Because ethylene is a gaseous molecule, it can freely diffuse from one cell to a neighboring cell, evoking mainly local responses. The presence of aerenchyma or large intercellular voids facilitates rapid long-distance transport of ethylene gas in plant organs. But long-distance ethylene responses can also be achieved by transport of its precursor ACC. Often, but not always, ACC is transported from the roots to the shoot, when the roots are exposed to stress (McManus, 2012). Yet, local ACC transport between cells of the same tissue type and intracellular transport is also possible, illustrating the molecular complexity of ACC transport.
One of the best characterized ACC transport systems is the translocation of ACC from the roots to the shoots of tomato plants suffering from flooding or root hypoxia. A lack of oxygen in the rhizosphere will induce the expression of ACS in the roots (Olson et al., 1995; Shiu et al., 1998) resulting in an increased ACS activity (Bradford et al., 1982; Wang and Arteca, 1992). The excess of ACC in the roots is not converted to ethylene due to a lack of oxygen and the absence of ACO in the roots. Rather, ACC is loaded into the xylem and transported to the shoots (Bradford and Yang, 1980). Once arrived at the shoots, ACC is converted into ethylene by ACO, which is already present in the leaves (English et al., 1995). In tomato, root hypoxia will result in an epinastic response of leaves due to an increased ethylene production (Figure 2A; Doubt, 1917; Jackson and Campbell, 1976). Differential expression of both ACS and ACO during hypoxia was also observed in Arabidopsis (Peng et al., 2005), sunflower seedlings (Finlayson et al., 1991), maize (Atwell et al., 2006; Geisler-Lee et al., 2010), and rice (Zarembinski and Theologis, 1993, 1997; Van Der Straeten et al., 1997, 2001; Zhou et al., 2001). Interestingly, long-distance transport of ACC has also been suggested to occur during other root stress conditions as for example drought (Davies et al., 2000; Sobeih et al., 2004; Skirycz et al., 2011), rehydration after drought (Tudela et al., 1992), nutrient stress (Lynch and Brown, 1997) and salinity (Feng and Barker, 1992; Ghanem et al., 2008).
1-Aminocyclopropane-1-carboxylic acid transport in plants. (A) The epinastic response of tomato plants treated with ethylene gas as first observed by Doubt (1917). Figure reproduced from Doubt (1917). (B) ACC translocation via the phloem was demonstrated...
Another long-distance ACC transport system is achieved via the phloem. Foliar applied radioactive ACC was found to be transported via the phloem to other aerial parts in tomato (Amrhein et al., 1982) and in cotton plants (Morris and Larcombe, 1995; Figure 2B). It should be noted that the foliar applied ACC was also rapidly converted into MACC, which was not found to be transported via the phloem (Morris and Larcombe, 1995). This immobility of MACC is in accordance with the earlier observations that MACC is actively transported from the cytosol into the vacuoles, where it could be subsequently stored (Bouzayen et al., 1988, 1989; Tophof et al., 1989). Interestingly, there could be a link between phloem transport of ACC and the Yang cycle. Pommerrenig et al. (2011) showed that Yang cycle genes were specifically expressed in phloem, indicating that recycling of MTA toward SAM is preferentially carried out in this tissue. Perhaps MTA recycling is stimulated by high rates of ACC synthesis in phloem cells, or the recycled SAM forms a pool for phloem-specific ACC production. In roots, spatiotemporal gene expression profiling demonstrated that different ACS isoforms are expressed (but not exclusively) in the vascular tissue (Brady et al., 2007; Dugardeyn et al., 2008). Moreover, the loading of ACC to this tissue could affect the homeostasis of SAM and consequently, polyamines (Pommerrenig et al., 2011).
The exact molecular mechanism by which ACC is loaded into the xylem and/or the phloem and subsequently transported throughout the plant is still unknown, but is an important element in our understanding of long-distance ethylene signaling via ACC and how plants deal with different root and leaf stress conditions.
Long- or medium–long-distance ACC transport was also observed (or speculated) during different developmental processes. Tissue specific gene expression profiling of maize root cells showed that there are differences between ACS and ACO expression patterns (Gallie et al., 2009). ACO was predominantly expressed in the protophloem sieve elements and the companion cells, while ACS was expressed only in the root cortex. This discrepancy led the authors to hypothesize that ACC could be transported from the site of synthesis to the site of consumption, in order to ensure the ethylene production levels observed (Gallie et al., 2009). Differences in ACS and ACO expression patterns predicted in silico in Arabidopsis roots support the same hypothesis (Dugardeyn et al., 2008). A similar reasoning was made by Jones and Woodson (1997, 1999), who observed differences in ACO and ACS transcripts in different cell-types of carnation flower. They also postulated that ACC transport from sites with a high ACS expression (in e.g., petals and styles) secured the ability of ACO to produce ethylene in cells with a high ACO expression (for example the ovaries; Jones and Woodson, 1997, 1999).
Of course one should take into account that gene expression levels not always reflect actual protein levels, and that post-translational modifications can play an important role in protein stability and activity. A targeted metabolomics and proteomics study by Van de Poel et al. (2014b) investigated the tissue specificity of the ethylene biosynthesis pathway in tomato fruit. They observed that the pericarp tissue produced the highest amount of ethylene (and high ACO activity), while the pericarp had the lowest ACS activity and ACC content compared to other tissues. Perhaps ACC is transported from neighboring tissues with a high ACS activity or ACC content such as the locular gel, toward the pericarp in order to secure high rates of ethylene production during climacteric ripening of tomato (Van de Poel et al., 2014b).
Besides long-distance, short-distance intracellular transport of ACC was also observed in barley and wheat mesophyll cells (Tophof et al., 1989) and maize mesophyll cells (Saftner and Martin, 1993). ACC is transported across the tonoplast by carriers that rely on an electrochemical potential gradient of protons, and which are stimulated by the supplementation of ATP (Saftner and Martin, 1993). This intracellular compartmentalization of ACC allows the plant to precisely regulate the cellular pool of ACC, possibly also affecting ethylene biosynthesis.
Clearly, more research is needed to further unravel the exact molecular and biochemical mechanisms which assure intracellular, inter- and intra-tissue and long-distance ACC transport in plants, and their corresponding physiological effects.
ACC AS A SIGNALING MOLECULE
1-Aminocyclopropane-1-carboxylic acid holds a key position in many physiological processes as it is the direct precursor in the biosynthesis of ethylene. A balanced supply and consumption of ACC is essential to achieve the necessary production level of ethylene within a given spatial and temporal context. As shown above, the pool of ACC is regulated by a complex interaction of production, consumption, modification, and transport. Interestingly, recent findings have suggested a perhaps even more important role for ACC, as a signaling molecule independent from ethylene (Yoon and Kieber, 2013).
A first report by Xu et al. (2008) investigated the role of ACC signaling in relation with FEI1 and FEI2, which are leucine-rich repeat receptor-like kinases. The fei1 fei2 mutant displays a severe defect in anisotropic root growth due to a reduced cellulose microfiber content in the cell wall at the root tip (Figure 3). The fei1 fei2 phenotype can be reversed by the application of ethylene biosynthesis inhibitors, but not by ethylene signaling inhibitors. The application of both aminooxy-acetic acid (AOA) or α-aminoisobutyric acid (AIB) specifically inhibits ethylene biosynthesis and can reverse the phenotype of the fei1 fei2 mutant (Figure 3). AOA is an inhibitor of PLP-dependent enzymes, and will affect the activity of ACS resulting in a reduced ethylene production. AIB on the other hand is a structural analog of ACC and acts as a competitive inhibitor of ACC preventing ethylene production at the level of ACO. Ethylene signaling inhibitors such as 1-methylcyclopropane (1-MCP) and silver ions, did not affect the fei1 fei2 phenotype (Figure 3). Similarly, genetics showed that the fei1 fei2 mutant crossed with etr1-3 (a mutation in the ethylene receptor causing severe ethylene insensitivity), nor ein2-50 (a central regulator of ethylene signaling causing ethylene insensitivity) could reverse the phenotype. All together this study showed that the typical fei1 fei2 phenotype was not affected by ethylene signaling, but could be reversed by ethylene biosynthesis inhibitors. This suggests that the signal reversing the fei1 fei2 phenotype originated independent from ethylene signaling, involved ACS and is possibly ACC itself (Xu et al., 2008).
Role of ACC/ethylene on the fei phenotype. Root phenotypes of seedlings grown on MS medium containing 0% sucrose for 4 days and then transferred to MS medium containing 4.5% sucrose, or additionally supplemented with AOA (0.375 mM) or AIB (1 mM) as indicated....
A second report by Tsang et al. (2011) linked ACC signaling with cell elongation and cell wall composition of roots. Specific ethylene biosynthesis inhibitors [AVG, AOA, and 2-anilino-7-(4-methoxyphenyl)-7,8-dihydro-5(6H)-quinazolinone (7303)] could reverse the inhibition of root cell expansion which was induced by an isoxaben treatment (a cellulose biosynthesis inhibitor causing cell wall stress). Similarly, as observed by Xu et al. (2008), an ethylene signaling inhibitor (silver ions) could not reverse the isoxaben-induced reduction in root cell elongation. Also, the ein3 eil1 ethylene insensitive mutant responded to ACC and isoxaben, providing genetic evidence of an ACC signaling mechanism independent of ethylene signaling. They also showed that the application of ACC without isoxaben, inhibited root cell elongation and was partially ethylene-dependent and partially ethylene-independent. Altogether, their results demonstrate that monitoring of cell wall integrity requires an ACC sensing/signaling mechanism, which can result in a reduction of root cell elongation, when disrupted. In addition, Tsang et al. (2011) showed that this inhibition of root cell elongation required auxin and reactive oxygen species (ROS) signaling, downstream of ACC signaling.
In a third report (Tsuchisaka et al., 2009) an octuple acs mutant was made to study the interplay between different ACS isoforms. The octuple line was created by introduction of two amiRNA lines (ACS8 and ACS11) into the hexuple mutant acs2,4,5,6,7,9 creating an octuple mutant line with complete or severe inhibition of ACS function. The lines that showed a very strong silencing of ACS8 and ACS11, displayed embryo lethality. This suggests that ethylene biosynthesis (or ACC biosynthesis) is essential for Arabidopsis viability, while this is not the case for the single (ctr1 and ein2) and double (ctr1 ein2) ethylene signaling mutants (Kieber et al., 1993; Roman et al., 1995; Alonso et al., 1999). This phenotypic discrepancy between ethylene biosynthesis and signaling once more suggests that ACC can acts as a signaling molecule itself, independent from ethylene, at least during embryo development and Arabidopsis viability. In addition, Tsuchisaka et al. (2009) characterized a wide variety of physiological and developmental processes of their single and multiple acs knock-out lines, and observed many phenotypes which were similar as for previously described ethylene signaling mutants. But interestingly they also observed several phenotypes like reduced branching, which were not observed in ethylene-insensitive mutants. These discrepancies could again be caused by ACC acting as a signaling molecule. However, it is also possible that individual ACS members have unique roles in developmental processes, and that knocking-out multiple ACS members might result in pleiotropic effects irrelevant to ACC or ethylene metabolism. Finally, it must be noted that such a severe genetic interference in ACC metabolism might also affect upstream SAM or MTA levels or the pool of downstream ACC conjugates, which in turn could signal themselves and affect cell physiology to contribute to the phenotypic differences observed.
All together these reports suggest a role for ACC as a signaling molecule to regulate plant development and growth, independent from ethylene. The exact molecular mechanism by which ACC signaling operates, and whether or not there is an ACC receptor and downstream signaling components, remains to be investigated. Future biochemical studies with specific ethylene biosynthesis and signaling inhibitors, in combination with genetics to create higher order ethylene biosynthesis/signaling mutants (like etr1ers1etr2ein4ers2ctr1ein2 or multiple aco knock-outs), could shed light on the role of ACC as a signaling molecule. It also still needs to be elucidated whether this unique title of “signaling molecule” is to be awarded to ACC, or rather to one if its (unknown) downstream derivatives.
Over the years, a lot of work has been done on ACC since its discovery in 1979, and it has become clear that ACC is more than just the precursor of ethylene. Its role in ethylene biosynthesis is well characterized, although there are still many questions concerning the two unique proteins that are associated with ACC in ethylene biosynthesis: ACS and ACO. Pioneering work on the characterization of post-translational modifications and the combinatorial interplay of ACS isoforms, has opened our eyes to the complex regulation of ethylene biosynthesis at the protein level. More mechanistic details are to be uncovered, probably including a complex post-translational control of ACO. Furthermore, ACC is conjugated into MACC, GACC, and JA-ACC. These derivatives are an elegant biochemical shunt to regulate the pool of ACC available for ethylene production, although it remains rather speculative what the exact biological roles are for these ACC conjugates. A better characterization of the participating enzymes is necessary to further elucidate the importunateness of ACC derivatives. Furthermore, ACC can also be used by bacterial (and plant) ACC-deaminase, adding another layer of metabolic complexity to the regulation of ACC levels. It is also well established that ACC can be easily transported over short (intracellular and intra-tissue) and long-distances (via the xylem and phloem), providing the plant with an elaborate system to control local and remote ethylene responses. Last but not least, ACC has been identified as a potential signaling molecule, independent of ethylene. This property of ACC is perhaps the most exciting, opening new avenues in ACC research, with potentially profound effects on plant physiology. The molecular mechanism by which ACC is signaling and the identity of other putative signaling components in such an ‘ACC pathway’ remain to be discovered.
Bram Van de Poel and Dominique Van Der Straeten conceived the topic and wrote the manuscript
Conflict of Interest Statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
This work was supported by a Belgian American Educational Foundation Fellowship, a CBMG/CMNS Merit Postdoctoral Fellowship (University of Maryland) and a BOF Postdoctoral Fellowship (Ghent University) to Bram Van de Poel. Dominique Van Der Straeten gratefully acknowledges support from the Research Foundation Flanders (FWO) and Ghent University.
- Adams D. O., Yang S. F. (1977). Methionine metabolism in apple tissue – implication of S-adenosylmethionine as an intermediate in conversion of methionine to ethylene.Plant Physiol.60 892–896 10.1104/pp.60.6.892 [PMC free article][PubMed][Cross Ref]
- Adams D. O., Yang S. F. (1979). Ethylene biosynthesis – identification of 1-aminocyclopropane-1-carboxylic acid as an intermediate in the conversion of methionine to ethylene.Proc. Natl. Acad. Sci. U.S.A.76 170–174 10.1073/pnas.76.1.170 [PMC free article][PubMed][Cross Ref]
- Ali S., Charles T. C., Glick B. R. (2012). Delay of carnation flower senescence by bacterial endo- phytes expressing ACC deaminase.J. Appl. Microbiol.113 1139–1144 10.1111/j.1365-2672.2012.05409.x [PubMed][Cross Ref]
- Alonso J. M., Hirayama T., Roman G., Nourizadeh S., Ecker J. R. (1999). EIN2, a bifunctional transducer of ethylene and stress responses in Arabidopsis.Science284 2148–2152 10.1126/science.284.5423.2148 [PubMed][Cross Ref]
- Amrhein A., Schneebeck D., Skorupka H., Stockigt J., Tophof S. (1981). Identification of a major metabolite of the ethylene precursor 1-aminocyclopropane-l-carboxylic acid in higher Plants.Naturwissenschaften68 619–620 10.1007/BF00398617 [Cross Ref]
- Amrhein N., Breuing F., Eberle J., Skorupka H., Tophof S. (1982). “The metabolism of 1-aminocyclopropane-1-carboxylic acid,” inPlant Growth Substances ed. Wareigng P. H., editor. (London: Academic Press; ) 249–258
- Atwell B. J., Drew M. C., Jackson M. B. (2006). The influence of oxygen deficiency on ethylene synthesis, 1-aminocyclopropane-1-carboxylic acid levels and aerenchyma formation in roots of Zea mays.Physiol. Plant.71 15–22.
- Barnawal D., Bharti N., Maji D., Chanotiya C. S., Kalra A. (2012). 1-Aminocyclopropane-1-carboxylic acid (ACC) deaminase-containing rhizobacteria protect Ocimum sanctum plants during waterlogging stress via reduced ethylene generation.Plant Physiol. Biochem.58 227–235 10.1016/j.plaphy.2012.07.008 [PubMed][Cross Ref]
- Bidonde S., Ferrer M. A., Zegzouti H., Ramassamy S., Latche A., Pech J. C., et al. (1998). Expression and characterization of three tomato 1-aminocyclopropane-1-carboxylate oxidase cDNAs in yeast.Eur. J. Biochem.253 20–26 10.1046/j.1432-1327.1998.2530020.x [PubMed][Cross Ref]
- Boller T., Herner R. C., Kende H. (1979). Assay for and enzymatic formation of an ethylene precursor, 1-aminocyclopropane-l-carboxylic acid.Planta145 293–303 10.1007/BF00454455 [PubMed][Cross Ref]
- Botella J. R., Arteca R. N., Frangos J. (1995). A mechanical strain-induced 1-aminocyclopropane-1-carboxylic acid synthase gene.Proc. Natl. Acad. Sci. U.S.A.92 1595–1598 10.1073/pnas.92.5.1595 [PMC free article][PubMed][Cross Ref]
- Bouzayen M., Latche A., Alibert G., Pech J. C. (1988). Intracellular sites of synthesis and storage of 1-(malonylamino)cyclopropane-1-carboxylic acid in acer-pseudoplatanus cells.Plant Physiol.88 613–617 10.1104/pp.88.3.613 [PMC free article][PubMed][Cross Ref]
- Bouzayen M., Latche A., Pech J. C., Marigo G. (1989). Carrier-mediated uptake of 1-(malonylamino)cyclopropane-1-carboxylic acid in vacuoles isolated from catharanthus-roseus cells.Plant Physiol.91 1317–1322 10.1104/pp.91.4.1317 [PMC free article][PubMed][Cross Ref]
- Bradford K. J., Hsiao T. C., Yang S. F. (1982). Inhibition of ethylene synthesis in tomato plants subjected to anaerobic root stress.Plant Physiol.70 1503–1507 10.1104/pp.70.5.1503 [PMC free article][PubMed][Cross Ref]
- Bradford K. J., Yang S. F. (1980). Xylem transport of 1-aminocyclopropane-1-carboxylic acid, an ethylene precursor, in waterlogged tomato plants.Plant Physiol.65 322–326 10.1104/pp.65.2.322 [PMC free article][PubMed][Cross Ref]
- Brady S. M., Orlando D. A., Lee J.-Y., Koch J., Dinney J. R., Mace D., et al. (2007). A high-resolution root spatiotemporal map reveals dominant expression patterns.Science318 801–806 10.1126/science.1146265 [PubMed][Cross Ref]
- Capitani G., Hohenester E., Feng L., Storici P., Kirsch J. F., Jansonius J. N. (1999). Structure of 1-aminocyclopropane-1-carboxylate synthase, a key enzyme in the biosynthesis of the plant hormone ethylene.J. Mol. Biol.294 745–756 10.1006/jmbi.1999.3255 [PubMed][Cross Ref]
- Cary A. J., Liu W., Howell S. H. (1995). Cytokinin action is coupled to ethylene in its effects on the inhibition of root and hypocotyl elonagion.Plant Physiol.107 1075–1082 10.1104/pp.107.4.1075 [PMC free article][PubMed][Cross Ref]
- Chae H. S., Faure F., Kieber J. J. (2003). The eto1, eto2, and eto3 mutations and cytokinin treatment increase ethylene biosynthesis in Arabidopsis by increasing the stability of ACS protein.Plant Cell15 545–559 10.1105/tpc.006882 [PMC free article][PubMed][Cross Ref]
- Chae H. S., Kieber J. J. (2005). Eto Brute? Role of ACS turnover in regulating ethylene biosynthesis.Trends Plant Sci.10 291–296 10.1016/j.tplants.2005.04.006 [PubMed][Cross Ref]
- Christians M. J., Gingerich D. J., Hansen M., Binder B. M., Kieber J. J., Vierstra R. D. (2009). The BTB ubiquitin ligases ETO1, EOL1, and EOL2 act collectively to regulate ethylene biosynthesis in Arabidopsis by controlling type-2 ACC synthase levels.Plant J.57 332–345 10.1111/j.1365-313X.2008.03693.x [PMC free article][PubMed][Cross Ref]
- Chung M. C., Chou S. J., Kuang L. Y., Charng Y. Y., Yang S. F. (2002). Subcellular localization of 1-aminocyclopropane-1-carboxylic acid oxidase in apple fruit.Plant Cell Physiol.43 549–554 10.1093/pcp/pcf067 [PubMed][Cross Ref]
- Davies W. J., Bacon M. A., Thompson D. S., Sobeih W., Rodriguez L. G. (2000). Regulation of leaf and fruit growth in plants growing in drying soil: exploitation of the plants’ chemical signalling system and hydraulic architecture to increase the efficiency of water use in agriculture.J. Exp. Bot.51 1617–1626 10.1093/jexbot/51.350.1617 [PubMed][Cross Ref]
- De Paepe A., Vuylsteke M., Van Hummelen P., Zabeau M., Van Der Staeten D. (2004). Transcriptional profiling by cDNA-AFLP and microarray analysis reveals novel insights into the early response to ethylene in Arabidopsis.Plant J.39 537–559 10.1111/j.1365-313X.2004.02156.x [PubMed][Cross Ref]
- Dilley D. R., Wang W., Kadirjan-Kalbach D., Ververidis F., Beaudry R., Padmanabhan K. (2013). 1-Aminocyclopropane-1-carboxylic acid oxidase reaction mechanism and putative post-translational activities of the ACCO protein.AoB Plants5:plt031 10.1093/aobpla/plt031 [PMC free article][PubMed][Cross Ref]
- Dong J. G., Fernandezmaculet J. C., Yang S. F. (1992). Purification and characterization of 1-aminocyclopropane-1-carboxylate oxidase from apple fruit.Proc. Natl. Acad. Sci. U.S.A.89 9789–9793 10.1073/pnas.89.20.9789 [PMC free article][PubMed][Cross Ref]
- Doubt S. L. (1917). The response of plants to illuminating gas.Bot. Gaz.63 209–224 10.1086/332006 [Cross Ref]
- Dugardeyn J., Vandenbussche F., Van Der Straeten D. (2008). To grow or not to grow: what can we learn on ethylene-gibberellin cross-talk by in silico gene expression analysis.J. Exp. Bot.59 1–16 10.1093/jxb/erm349 [PubMed][Cross Ref]
- English P. J., Lycett G. W., Roberts J. A., Jackson M. B. (1995). Increased 1-aminocyclopropane-1-carboxylic acid oxidase activity in shoots of flooded tomato plants raises ethylene production to physiologically active levels.Plant Physiol.109 1435–1440. [PMC free article][PubMed]
- Feng J. N., Barker A. V. (1992). Ethylene evolution and ammonium accumulation by tomato plants under water and salinity stresses.J. Plant Nutr.15
Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide and its incidence in the United States and other countries has been steadily increasing over the past 25 years [1,2]. The primary risk factors for HCC include infection with hepatitis B and hepatitis C viruses, and long-term exposure to aflatoxin . In the United States, chronic alcoholism leading to chronic liver disease is a significant risk factor . The prognosis for HCC is poor and lack of good diagnostic markers and treatment options have rendered this disease a major health problem .
At the molecular level, several proteins have been identified as de-regulated during HCC. These include p53, Retinoblastoma (Rb), Insulin-like growth factor receptor 1 (IGFR-1), β-catenin, and cyclin D1 . Other signaling pathways relevant in HCC development include Phosphatidyl inositol-3 Kinase (PI3-K) and c-jun N-terminal kinase (JNK) [7,8]. This paper reviews the role of methionine adenosyltransferase (MAT) genes in the development and possible treatment of HCC.
2. Hepatic Methionine Metabolism
The liver is the main source of biosynthesis and consumption of the principle biological methyl donor, S-adenosylmethionine (AdoMet, also often abbreviated as SAMe and SAM). The synthesis of AdoMet from methionine and ATP is catalyzed by MAT isoenzymes. AdoMet is also a precursor for polyamine biosynthesis and in hepatocytes, a precursor for cysteine, the rate-limiting amino acid for the synthesis of the antioxidant glutathione (GSH) (Figure 1) . Under normal physiological conditions, most of the AdoMet is utilized in transmethylation reactions and is converted to S-adenosylhomocysteine (AdoHcy, often abbreviated as SAH) (Figure 1) . AdoHcy is a potent competitive inhibitor of transmethylation reactions. Both an increase in AdoHcy level and a decrease in the AdoMet:AdoHcy ratio are known to inhibit transmethylation reactions . For this reason, the removal of AdoHcy is essential. The reaction that converts AdoHcy to homocysteine and adenosine is reversible and catalyzed by AdoHcy hydrolase . The thermodynamics of this reaction favor the synthesis of AdoHcy. In vivo, the reaction proceeds in the direction of hydrolysis only if the products, adenosine and homocysteine, are rapidly removed .
In the liver, there are three pathways that metabolize homocysteine. One is the transsulfuration pathway, which converts homocysteine to cysteine via a two-step enzymatic process catalyzed by cystathionine β-synthetase (CBS) and γ-cystathionase, both requiring vitamin B6 . The other two pathways in which homocysteine are metabolized lead to the re-synthesis of methionine. One reaction is catalyzed by methionine synthase (MS), which requires normal levels of folate and Vitamin B12. The other pathway is catalyzed by betaine homocysteine methyltransferase (BHMT), which requires betaine, a metabolite of choline [9,11]. Remethylation of homocysteine via MS requires 5-methyltetrahydrofolate (5-MTHF), which is derived from 5,10-methylenetetrahydrofolate (5,10-MTHF) in a reaction catalyzed by methylenetetrahydrofolate reductase (MTHFR). 5-MTHF is then converted to tetrahydrofolate (THF) as it donates its methyl group and THF is converted to 5,10-MTHF to complete the folate cycle. In the liver, AdoMet plays a regulatory role on methionine metabolism by inhibiting MTHFR and MS and activating CBS [9, 11]. Thus, when AdoMet is depleted, homocysteine is channeled to remethylation to regenerate AdoMet; whereas when AdoMet level is high, homocysteine is channeled to the transsulfuration pathway through AdoMet-mediated activation of CBS. Another contributing factor for this is that the Km of CBS for AdoMet is 1-2.5 mM, whereas the Km of MS for AdoMet is 60 μM . Thus, when AdoMet level rises, it favors channeling of AdoMet to the transsulfuration pathway. AdoMet is also utilized in the biosynthesis of polyamines that are required for cell growth . During the synthesis of polyamines, methylthioadenosine (MTA) is generated as a byproduct that is a known inhibitor of methylation .
3. Expression and Regulation of MAT Genes in Healthy and Diseased Liver
Mammalian systems express two MAT genes, namely MAT1A and MAT2A that encode the catalytic subunits of the enzyme (Table 1) . The gene MAT1A encodes the α1 catalytic subunit, which organizes into dimers (MATIII) or tetramers (MATI) . The gene MAT2A encodes for the α2 catalytic subunit found in the MATII isoform. A third gene, MAT2B, encodes for a β regulatory subunit that regulates the activity of MATII by lowering the inhibition constant (Ki) for AdoMet and the Michaelis constant (Km) for methionine . MAT1A is expressed mostly in adult liver [14,16] with low level of expression in extrahepatic tissues . MAT1A is also expressed by pancreas with high level of expression seen in pancreatic acinar cells [17,18]. MAT2A is widely expressed in extrahepatic tissues and is also expressed in fetal liver but is replaced by MAT1A during development [16,19]. Different isoforms of MAT differ in kinetic and regulatory properties and sensitivities to inhibitors of MAT . MAT II has the lowest Km (∼4–10 μM), MAT I has intermediate Km (23 μM–1 mM), and MAT III has the highest Km (215 μM–7 mM) for methionine [20-22]. Although MAT isoenzymes catalyze the same reaction, they are differentially regulated by their product, AdoMet. AdoMet strongly inhibits MATII (50% inhibitory concentration (IC50) = 60 μM), which is close to the normal intracellular AdoMet concentration , whereas it minimally inhibits MATI (IC50 = 400 μM) and stimulates MATIII (up to eightfold at 500 μM AdoMet) . Thus, the type of MAT isoform expressed in cells controls the steady state AdoMet level. Consistently we showed that hepatic cell lines over-expressing MAT1A have increased accumulation of AdoMet as compared to cells expressing MAT2A . Increased expression of MAT2B can further lower steady state AdoMet level due to its influence on the Ki of MATII for AdoMet .
Adult differentiated liver expresses the MAT1A-encoded isoforms while increased hepatic expression of MAT2A is associated with increased growth, de-differentiation, and malignant degeneration [9,23]. Increased expression of MAT2B also provides a growth advantage to hepatoma cells, and although it is not expressed in normal liver, its expression is increased in liver cirrhosis and HCC .
MAT1A is silenced in HCC and during de-differentiation by both transcriptional and post-transcriptional mechanisms. MAT1A gene expression in the human hepatoblastoma cell line (HepG2) is regulated by histone acetylation and methylation. Treatment of these cells with demethylating agents or histone deacetylase inhibitors induces MAT1A mRNA expression and decreased MAT1A expression is associated with hypermethylation of a HpaII site at position -977 of the promoter . Promoter hypermethylation also correlated with reduced MAT1A expression in cirrhotic patients and HCC . Recently we reported that MAT1A 3′-UTR binds to the AU-rich RNA binding factor 1 (AUF1) . AUF1 is one of the hnRNP proteins known to destabilize target mRNAs . Interestingly we found that HCC specimens express higher AUF1 protein levels and knockdown of AUF1 increased MAT1A mRNA level . AUF1 expression is also high in fetal liver and falls during liver development, which coincides with increased MAT1A expression .
MAT2A expression is also regulated at both transcriptional and post-transcriptional levels. We identified four cis-acting elements and trans-activating factors (Sp1, c-Myb, NFκB and AP-1) to participate in MAT2A transcriptional up-regulation in HCC [29,30]. We have also found promoter methylation and histone acetylation regulate the human MAT2A gene. The human MAT2A promoter is hypomethylated in HCC but hypermethylated in normal liver . Histone acetylation status correlates with MAT2A expression in both human and rat so that a hyperacetylated status correlates with high MAT2A expression and vice versa [31,32]. More recently, we reported that MAT2A mRNA level is regulated by HuR and methylated HuR . HuR is a ubiquitously expressed mRNA binding protein known to stabilize its target mRNAs, whereas methylated-HuR exerts the opposite effect . Interestingly, during hepatocyte de-differentiation and in HCC, there is a switch from methylated-HuR to HuR binding to the 3′-UTR of MAT2A, resulting in increased MAT2A mRNA level . AdoMet treatment results in higher methylated-HuR level, which contributes to its known inhibitory effect on MAT2A expression .
Relatively little is known about regulation of MAT2B expression. We recently reported that MAT2B has two dominant splicing variants, variant 1 (V1) and variant 2 (V2) . Both variants are highly induced in HCC. Tumor necrosis factor α (TNFα) induces MAT2B V1 expression (but not V2) at the transcriptional level by mechanisms that involve AP-1 and NFκB . Leptin increases while AdoMet inhibits MAT2B V1 promoter activity and expression by mechanisms that involve ERK and AKT signaling . Whether MAT2B is regulated post-transcriptionally is unknown.
4. MAT Genes and HCC
Accumulating evidence support the notion that hepatic AdoMet deficiency is a risk factor in the development of HCC. Decrease in AdoMet content in preneoplastic and neoplastic liver may depend on changes in MAT isoenzyme pattern . Fall in MAT1A expression and MATI/III activity with concomitant up-regulation of MAT2A occurs in hepatoma cell lines and rodent HCC as well as in human liver cirrhosis and HCC [37,38]. The MATII isoform is inhibited by its reaction product AdoMet so that its up-regulation does not lead to increase in AdoMet liver content . Furthermore, MAT2B expression is induced in cirrhosis  so that most patients with chronic liver disease have hepatic AdoMet deficiency.
The MAT1A knockout (KO) mouse model was developed nearly 10 years ago to address how chronic AdoMet deficiency and deregulation of methionine metabolism may predispose to HCC. This model has provided invaluable insight into the pathogenesis of HCC in the setting of chronic AdoMet deficiency. This is highly relevant to human liver disease as the expression of MAT1A is markedly reduced . MAT1A KO mice are viable since MAT1A is expressed shortly after birth and in the absence of MAT1A, MAT2A is induced . Hepatic AdoMet level fell by nearly 75% and GSH level fell by 40%. At 3 months of age, MAT1A KO mice had body weights similar to wild-type littermates. However, their liver weights were significantly increased. At this age, the liver is histologically normal in the MAT1A KO mice fed a normal diet. Feeding a choline-deficient diet to the KO animals for six days induced severe macrovesicular steatosis compared to wild type controls. The livers of the 8-month-old wild-type littermates remained normal histologically, but the livers of 8-month old KO animals fed a normal diet exhibited macrovesicular steatosis involving 25–50% of hepatocytes and mononuclear cell infiltration, mainly in the periportal areas. By 18 months of age, many of the MAT1A KO mice developed liver cancer.
We have identified several mechanisms that can contribute to HCC development in the MAT1A KO mouse model. First is the existence of liver cancer stem cell population in the aging MAT1A KO mice. Methyl-deficient diets have been used to induce oval cell proliferation and HCC formation in susceptible models such as p53 knockout mice . Oval cells are liver stem cells found in the non-parenchymal fraction of the liver and reside near the terminal bile ducts, at the hepatocyte-cholangiocyte interface. In normal adult liver, oval cells are quiescent and few in number and proliferate only during severe, prolonged liver injury and in various models of experimental carcinogenesis . Our recent findings have shown that MAT1A KO mice have expansion of a population of oval cells that behave like liver cancer stem cells as they age . These CD49f+ cells have markedly increased expression of several oncogenes such as K-ras and survivin. Moreover, a subpopulation of the CD49f+ cells that are also CD133+ possess tumorigenic potential when injected into immune deficient mice. This is the first demonstration of adult liver stem cells possessing tumorigenic potential without the use of a carcinogen or manipulation of tumor-suppressor or oncogene expression. Further work has shown that liver cancer stem cells from MAT1A KO mice possess highly enhanced mitogen-activated protein kinase (MAPK) signaling with increased level and activity of the extracellular signal regulated kinase (ERK) , known to be associated strongly with HCC development . This is consistent with previous findings showing alterations in the MAPK pathway in MAT1A KO mice . As compared to the CD133- cell populations, CD133+ CD49f+ cells use their constitutive ERK activation to evade the apoptotic effect of transforming growth factor-β (TGF-β), a well-known growth inhibitor in hepatocytes [43,45].
Another mechanism is the role of genomic instability (GI). Differential expression of MAT1A and MAT2A genes can potentially influence DNA methylation and growth of human HCC [9,23]. DNA hypomethylation may generate GI during carcinogenesis . The work of Calvisi et al. indicates that early changes in methionine/AdoMet metabolism and global DNA methylation may have a prognostic value for hepatocarcinogenesis in the majority of individuals, probably acting through a modulation of GI . They have also shown that molecular alterations linked to AdoMet metabolism and DNA methylation are necessary for the development of the majority, but not all, human HCCs . HCC demonstrates a high incidence of GI and the level of GI correlates with tumor stage . The cellular defense pathway against GI includes several components, one of them being the Apurinic/Apyrimidinic Endonuclease 1 (APEX1), which is a multifunctional protein possessing both DNA repair and redox regulatory activities . APEX1 is induced by oxidative stress and this is part of the defense mechanism against GI [50,51]. MAT1A KO mice exhibit increased oxidative stress and malignant transformation . Based on this fact it would be expected that DNA repair pathways like APEX1 should be induced. On the contrary, it was observed that the expression of APEX1 was down-regulated in MAT1A KO mice and there was increased GI . This decrease in APEX1 has been attributed to AdoMet deficiency in MAT1A KO mice. Primary hepatocytes placed in culture rapidly de-differentiate and show a decline in MAT1A expression and intracellular AdoMet level. Exogenous treatment of these cells with pharmacological doses of AdoMet prevents AdoMet depletion, blunts the fall in MAT1A expression, and importantly, stabilizes APEX1 protein . Therefore AdoMet depletion can lead to decreased APEX1 protein stability and increased GI, contributing to malignant degeneration.
A third mechanism has to do with uncontrolled ERK activation. ERK activation is one of the several growth signals associated with highly malignant HCC phenotypes and it is tightly regulated in normal liver cells. One way through which ERK activity is kept under control is by the action of the dual-specificity MAPK phosphatase (DUSP1). DUSP1 is the first member of a family of dual-specificity MAPK phosphatases which can dephosphorylate both serine/threonine and tyrosine residues . There is a reciprocal regulation between DUSP1 and ERK. Prolonged activation of ERK promotes phosphorylation at the Ser296 residue of its inhibitor, DUSP1 . Phosphorylation of this specific residue renders the DUSP1 protein susceptible to proteasomal degradation by two substrate recognition protein belonging to a large S-phase kinase-associated protein/cullin/F box ubiquitin ligase: the S-phase kinase associated protein 2 (SKP2) and CDC28 protein kinase b1 complex. In contrast, transient activation of ERK leads to catalytic activation of DUSP1 followed by inactivation of ERK . Thus DUSP1 feedback inhibits ERK and this activity of DUSP1 is crucial for the regulation of ERK activity in liver cells. In human HCC, DUSP1 expression is inversely correlated with proliferation index and microvessel density, and directly correlated with apoptotic index and survival . Our group has shown that hepatic DUSP1 expression is low in MAT1A KO mice both at the mRNA and protein level, with protein level falling to lower level than mRNA . Correcting AdoMet deficiency in MAT1A KO mice by exogenous AdoMet treatment normalized DUSP1 mRNA and protein levels . AdoMet exerts its effect both transcriptionally and post-transcriptionally. AdoMet's ability to normalize DUSP1 mRNA level may be a p53-dependent effect because in MAT1A KO livers, binding of p53 to a p53 element in DUSP1 promoter was reduced compared to wild type livers and feeding AdoMet to these animals partially corrected p53 binding to the DUSP1 promoter. The increase in p53 binding in AdoMet fed animals is attributed to the fact that AdoMet stabilizes the APEX1 protein, which is a known transactivator of p53 . The reason for the drastic drop in DUSP1 at the protein level in MAT1A KO mice is due to increased proteasomal activity, causing rapid degradation of the DUSP1 protein. Moreover MAT1A KO mice also have an increase in expression of SKP2 protein, an E3 ligase responsible for ubiquitination of DUSP1. This can further contribute to the decline in DUSP1 protein level. AdoMet appears to exert a direct effect on proteasomal activity as incubation of purified proteasomes with AdoMet led to increased degradation of some of its subunits and decreased proteasomal activity. Consistently, AdoMet treatment in MAT1A KO mice normalized proteasomal activity, increased DUSP1 protein level and reduced ERK activity back to baseline [56