Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

13.1 Introduction

Vitamin C or ascorbate is not synthesised by humans and is consequently needed in the diet with the best dietary sources of ascorbate being green vegetables and fruits. Of the commercially significant and traded dessert fruit, kiwifruit has the highest concentration of vitamin C. While other fruit have higher ascorbate (e.g. acerola), they are either not significantly traded or mainly processed (frozen or juiced, e.g. blackcurrants and dried, e.g. acerola). Actinidia chinensis var. deliciosa ‘Hayward’ kiwifruit has 85 mg ascorbate/100 g fresh weight (FW); Actinidia chinensis var. chinensis ‘Hort16A’ marketed as Zespri® Gold Kiwifruit has 105–110 mg ascorbate/100 g FW while Actinidia eriantha can have over 800 mg/100 g FW. On a fresh weight basis, green ‘Hayward’ and gold ‘Hort16A’ kiwifruit contain 50 % more vitamin C than an orange, five or six times as much as a banana or ten times as much as an apple (Huang and Ferguson 2007). High vitamin C concentration is one of the reasons that kiwifruit are rated as very healthy (Ferguson and Ferguson 2003).

There is considerable variation between Actinidia species in ascorbate concentration, from low values in fruit of Actinidia henryi (4.4 mg/100 g FW) and Actinidia rudis (5) to very high in Actinidia latifolia (671–2140) and A. eriantha (500–1379 mg/100 g FW) (Huang et al. 2004). Fruit of Actinidia kolomikta also contain high amounts of ascorbate (650–850) (Chesoniene et al. 2004). When the total amount of vitamin C concentration per fruit is calculated, the best sources are the fruit of A. eriantha, A. chinensis var. chinensis and A. chinensis var. deliciosa, all of which have larger fruit. There is also large variation in the vitamin C concentration within a species (Huang and Ferguson 2007). Accessions of A. chinensis var. chinensis range from 50 to 420 mg ascorbate/100 g FW (Huang et al. 2004) while 143 fruiting plants from a single cross of A. chinensis var. chinensis ranged from 49 to 209 mg/100 g FW [A.R. Ferguson, unpublished, (Bulley et al. 2009)]. Values for ascorbate concentrations in fruit of accessions of A. chinensis var. deliciosa range from 30 to ca. 400 mg/100 g FW (Ferguson 1991). While the levels of the acids citrate, quinate and malate are usually much higher than ascorbate (Marsh et al. 2009) in cultivars of A. chinensis var. chinensis, A. chinensis var. deliciosa and Actinidia arguta, this is not the case in ultra-high ascorbate Actinidia genotypes.

In addition, in some Actinidia species, ascorbate content has been shown to have a high heritability (Cheng et al. 2004) which is linked to increased soluble sugar content. In the A. chinensis var. chinensis cross mentioned above, a significant QTL for ascorbate was detected in the 143 progeny (M.A. McNeilage et al. personal communication) on the genetic map (Fraser et al. 2009). In segregating back-cross populations (A. eriantha × A. chinensis var. chinensis or A. chinensis var. deliciosa back-crossed to either A. chinensis var. deliciosa or A. chinensis var. chinensis), a simple genetic model with two main loci will fit the measured ascorbate data (W.A. Laing and A.G. Seal unpublished).

13.2 Biochemical Pathways of Ascorbate Biosynthesis

One main pathway of ascorbate biosynthesis has been fully documented in various species of higher plants, named the l-Galactose pathway (Wheeler et al. 1998; Linster and Clarke 2008; Wheeler et al. 2015), starting from glucose, and all the genes in the pathway have been identified (Fig. 13.1) and shown to have activity as proteins (including kiwifruit, W.A. Laing et al. unpublished). The genes for this pathway have been identified, cloned and validated from various Actinidia species (Laing et al. 2004a, 2007; Richardson et al. 2004; Linster et al. 2008; Bulley et al. 2009; Torabinejad et al. 2009; Li et al. 2011; Bulley et al. 2012; Li et al. 2013a, b, c, 2014). Alternative pathways through galacturonate and glucuronate (Li et al. 2010b) have been proposed but not all the genes and enzymes have been identified. The glucuronate pathway could either derive from glucose via UDP-glucose or from conversion of myo-inositol conversion to glucuronate by myo-inositol oxygenase. The most likely precursor to the galacturonate pathway would be either pectin or UDP-glucuronate epimerisation to UDP-galacturonate. However, polygalacturonase gene expression (Wang et al. 2000) and kiwifruit cell wall breakdown occurs later in fruit development and ripening when ascorbate concentrations are stable (Bulley et al. 2009).

Fig. 13.1
figure 1

Reactions, enzymes and context of ascorbic acid biosynthesis and regeneration in plants. (A) l-Galactose pathway, reactions 2–9. (B) myo-Inositol/glucuronate pathway, reactions 7, 18–26. (C) Galacturonate pathway, reactions 14–17. (D) l-Gulose pathway, possible reactions 5, 6, 7, 8 and 10. Reactions with question marks after the number are hypothetical and the exact enzyme is yet to be identified. Underlined chemical names are those that appear in more than one position in the diagram. Gene expression of transcripts of numbered enzymes in bold type was analysed. 1, glucose-6-phosphate isomerase; 2, mannose-6-phosphate isomerase; 3, phosphomannomutase; 4, GDP-mannose pyrophosphorylase; 5, GDP-mannose-3′,5′-epimerase; 6, GDP-l-galactose transferase; 7, l-galactose-1-phosphate phosphatase; 8, l-galactose dehydrogenase; 9, l-galactono-1,4-lactone dehydrogenase; 10, l-gulono-1,4-lactone oxidase; 11, GDP-d-mannose-4,6-dehydratase; 12, GDP-l-fucose synthase; 13, UDP-galacturonate epimerase; 14, polygalacturonate 4-α-galacturonosyltransferase; 15, galacturonate-1-phosphate uridylyltransferase and galacturonate-1-phosphate phosphatase (hypothetical); 16, d-galacturonic acid reductase; 17, aldonolactonase; 18, l-myo-inositol 1-phosphate synthase; 19, myo-inositol oxygenase; 20, d-glucurono-1-phosphate phosphatase; 21, glucuronate reductase; 22, gulonolactonase; 23, phosphoglucomutase; 24, UDP-glucose-pyrophosphorylase; 25, UDP-glucose dehydrogenase; 26, glucuronate-1-phosphate uridylyltransferase; 27, monodehydroascorbate reductase; 28, dehydroascorbate reductase; vtc, vitamin C content (Bulley et al. 2009. With the permission of Oxford University Press)

Full size image

13.2.1 l-Galactose Pathway

13.2.1.1 Conversion of Glucose-6-P to Gannose-1-P

The genes encoding the enzymes for these conversions have all been identified in kiwifruit and do not appear to be limiting ascorbate formation, although overexpression of phosphomannomutase (PMM) has been shown to increase ascorbate in leaves of other species (Qian et al. 2007; Badejo et al. 2009). It is possible that, under some circumstances, carbon supply to ascorbate biosynthesis might limit ascorbate synthesis. Two phosphomannose isomerase (PMI) genes exist in Arabidopsis with PMI1 being shown to be essential for ascorbate synthesis and the only one to be expressed constitutively, with PMI2 not being expressed in light (Maruta et al. 2008). Both PMI isoforms were feedback inhibited by ascorbate showing a higher level general cut-off mechanism to be in place before metabolism enters ascorbate biosynthesis proper (Maruta et al. 2008).

Phosphoglucoisomerase (PGI) , PMI and PMM are all represented in the kiwifruit genome (Huang et al. 2013) and found in the kiwifruit EST collection (Crowhurst et al. 2008) (Table 13.1).

Table 13.1 Ascorbate-related genes in kiwifruit. BLASTp searches based on identified or postulated Arabidopsis peptide sequences were run against Actinidia chinensis var. chinensis-predicted protein sequences (Huang et al. 2013)
Full size table

13.2.1.2 Conversion of Mannose-1-P to Galactose-1-P

GDP-mannose pyrophosphorylase (GMP) converts mannose-1-P to GDP-mannose . GDP-mannose provides carbon both for ascorbate synthesis and for cell wall components and protein modification (Zablackis et al. 1996; Keller et al. 1999; Handford et al. 2003). In some circumstances, it is probable that the supply of carbon skeletons may limit ascorbate biosynthesis which may be why transformation with the GMP gene has variable effects of ascorbate concentration (Badejo et al. 2007; Wang et al. 2011; Cronje et al. 2012; Imai et al. 2012; Zhou et al. 2012; Zhang et al. 2015a). It is possible that upstream PMI expression levels may also be limiting and this suggests that coexpression of PMI might overcome such limitations.

GDP-mannose epimerase (GME) is another key gene in that it results in the formation of GDP-l-galactose which is almost all used for ascorbate production, with little l-galactose being found elsewhere in the metabolome (Gilbert et al. 2009). However, by itself GME has little effect of ascorbate concentrations in plant tissues (see below).

GDP-galactose phosphorylase (GGP) which converts GDP-galactose to galactose-1-phosphate is the key enzyme in the ascorbate pathway and has been shown to control ascorbate in a wide range of species (Laing et al. 2007; Bulley et al. 2009, 2012; Zhang et al. 2011, 2015a; Zhou et al. 2012; Li et al. 2013a, b, c; Huang et al. 2014; Ma et al. 2014). The evidence is either from the studying patterns of gene expression of these genes during fruit development in kiwifruit or from the overexpression of kiwifruit genes in other species.

Transformation with GGP results in very significant increases in tissue ascorbate concentrations (Laing et al. 2007; Bulley et al. 2012; Zhou et al. 2012; Zhang et al. 2015a). While transformation with GME alone has small effects on ascorbate in a range of species (Zhang et al. 2011, 2015b; Huang et al. 2014; Ma et al. 2014), coexpression of GME and GGP results in a strong synergistic increase on ascorbate concentration (Bulley et al. 2009; Laing et al. 2015). GME gene transcription also tracks ascorbate production along with GGP transcription (Bulley et al. 2009; Li et al. 2013c).

13.2.1.3 Conversion of Galactose-1-P to Ascorbate

Galactose-1-P phosphatase (GPP) (Laing et al. 2004b), galactose dehydrogenase (GalDH) (Laing et al. 2004a; Li et al. 2010b) and galactono lactone dehydrogenase (GalLDH) (W.A. Laing unpublished) have all been identified in kiwifruit, but again do not appear to regulate ascorbate formation, except when activity is reduced (Tabata et al. 2001; Gatzek et al. 2002; Conklin et al. 2006; Alhagdow et al. 2007; Imai et al. 2009; Torabinejad et al. 2009; Zhou et al. 2012; Zhang et al. 2015b). However, it has been suggested GPP may be involved in light and abiotic stress responses (Li et al. 2013b). In tomato, GPP has been suggested to play an important role in regulating ascorbate accumulation during fruit development (Ioannidi et al. 2009), and while its expression was the highest of the l-galactose pathway genes, GME and more particularly GGP expression also correlated well with ascorbate accumulation. A later study, also in tomato, found that ‘translocation from source leaves and biosynthesis via the d-mannose/l-galactose pathway are dominant sources in immature fruits, while the alternative d-galacturonate pathway contributes to AsA [ascorbate] accumulation in ripened Micro-Tom fruits’ (Badejo et al. 2012). It therefore appears that a possible explanation for the very high expression of GPP in ripening tomato fruit may be it is there to convert the l-galactose-1P postulated to be derived from the d-galacturonate pathway via the conversion of UDP-d-galacturonate to l-galactose-1P (Fig. 13.1).

13.2.2 l-Glucuronate Pathway and Myo-Inositol

A. chinensis var. deliciosa and A. arguta contain high concentrations of myo-inositol in the leaves and fruit (Bieleski et al. 1997; Klages et al. 1998) which potentially could serve as a substrate of ascorbate production by conversion of myo-inositol to glucuronate by myo-inositol oxygenase. Myo-inositol-1-P is synthesised from glucose-6-P by myo-inositol synthase and then, myo-inositol-1-P is dephosphorylated by the enzyme myo-inositol phosphatase (Gillaspy et al. 1995). This latter enzyme is the same as that encoded by GPP, l-galactose-1-P phosphatase (Laing et al. 2004b) in both kiwifruit and Arabidopsis (Torabinejad et al. 2009). Good homologues of the oxygenase and the synthase have also been identified in a kiwifruit EST collection (Crowhurst et al. 2008). The glucuronate is then converted to l-gulonate by an unknown enzyme, then to the gulonate 1,4 lactone and thence to ascorbate, again by enzymes encoded by unknown genes. However, possible candidates for these genes are found in the kiwifruit EST database (Crowhurst et al. 2008) although they are not validated, even in other species. The other genes that encode enzymes that convert glucose to glucuronate (UDP-glucose dehydrogenase , glucuronate-1-P uridyltransferase and glucurono-1-P phosphatase ) are validated in other species and good homologues are present in the kiwifruit genome and EST databases.

13.2.3 l-Galacturonate Pathway

This pathway is proposed to result in ascorbate production from the breakdown of cell wall pectin into galacturonate (Di Matteo et al. 2010) or from the conversion of UDP-glucuronate into UDP-galacturonate by UDP-glucuronate epimerase (Gu and Bar-Peled 2004; Usadel et al. 2004). The UDP-galacturonate is then hydrolysed to galacturonate (this requires a uridylyltransferase and a phosphatase, no enzymes characterised) and then reduced to galactonate by galacturonate reductase (Agius et al. 2003). A homologue to the validated strawberry galacturonate reductase is found in kiwifruit ESTs (Crowhurst et al. 2008) and in the kiwifruit genome (Huang et al. 2013) with 65–67 % identity and 83 % similarity, respectively. The kiwifruit gene has been cloned and expressed in Escherichia coli, but it showed no reductase activity (Li et al. 2010b, 2011). While the strawberry gene has been shown to increase ascorbate in tomato (Amaya et al. 2014), potato (Hemavathi et al. 2009) and Arabidopsis (Agius et al. 2003), this gene is not thought to control ascorbate in kiwifruit (Li et al. 2010b, 2011).

13.3 Ascorbate Recycling Enzymes

A reactive oxygen species or an oxidising agent such as H2O2 oxidises ascorbate to monodehydroascorbate (MDHA), which is either directly reduced back to ascorbate by monodehydroascorbate reductase (MDHAR) or non-enzymatically disproportionated into ascorbate and dehydroascorbate (DHA) . DHA is then reduced back to ascorbate through reducing equivalents provided by glutathione either chemically or by dehydroascorbate reductase (DHAR) . Oxidised glutathione is finally reduced by glutathione reductase . This redox hub maintains the ascorbate in a mostly reduced state (Foyer and Noctor 2011). There are five MDHAR genes and three DHAR genes in Arabidopsis.

Different studies have overexpressed DHAR (Chen et al. 2003; Kwon et al. 2003; Eltayeb et al. 2006; Goo et al. 2008; Naqvi et al. 2009; Yin et al. 2010; Haroldsen et al. 2011; Qin et al. 2011; Huang et al. 2014) and MDHAR (Kavitha et al. 2010; Li et al. 2010a; Yin et al. 2010; Haroldsen et al. 2011; Gest et al. 2012) in a range of plants and, except for maize kernels (only very low ascorbate is present in seeds), the concentration of ascorbate was mostly little affected. None of these studies was performed in kiwifruit. However, many papers reported an increase in the ratio of reduced to oxidised ascorbate and an improvement in stress resistance.

Various versions of the two ascorbate recycling genes are found in the kiwifruit genome (Table 13.1). However, there appears to be no published work on MDHAR and DHAR genes in kiwifruit. We have constitutively overexpressed a kiwifruit DHAR in Arabidopsis which resulted in changes to the ratio of reduced/oxidised ascorbate and also appeared to increase salinity tolerance (S.M. Bulley, unpublished).

13.4 Oxalate in Kiwifruit

Kiwifruit fruit also contain high amounts of oxalate, much in crystalline (raphide) form (Rassam and Laing 2005; Rassam et al. 2007). In other species, it has been proposed that ascorbate breaks down into oxalate (Keates et al. 2000; Kostman et al. 2001) although little is known about genes involved in the pathway (Green and Fry 2005). The relationship between the amount of oxalate and ascorbate in a plant tissue is not straightforward, suggesting that the breakdown of ascorbate is a regulated process (Rassam and Laing 2005; Rassam et al. 2007).

13.5 Regulation of Ascorbate Biosynthesis

13.5.1 Regulation Through Gene Expression of Pathway Genes

There have been several studies of expression of vitamin C-related genes in kiwifruit (Bulley et al. 2009; Li et al. 2010b, 2013a, c, 2014) and these have shown changes during fruit development and between kiwifruit species and accessions. For example, Bulley et al. 2009 showed large differences during fruit development with a peak in ascorbate accumulation around four to seven weeks after anthesis and gene expression maxima around the same time. However, the biggest differences between taxa (A. chinensis var. chinensis, A. chinensis var. deliciosa and A. eriantha) were for GGP and GME, and to some extent for GMP, with the high ascorbate A. eriantha having much higher expression levels of these genes. Through transient and stable transformation of Arabidopsis thaliana and Nicotiana benthamiana with kiwifruit genes, it was concluded that GGP and GME synergistically controlled ascorbate biosynthesis. Other results (Li et al. 2014) also supported the difference in GGP expression between A. eriantha (high ascorbate) and Actinidia rufa (low ascorbate). On the other hand, Li et al. 2010b showed a similar pattern of ascorbate accumulation and gene expression during fruit development, for the one cultivar A. chinensis var. deliciosa ‘Qinmei’. Based on correlations between gene expression and ascorbate concentrations, they concluded GPP controlled ascorbate biosynthesis. Various light and stress conditions have also been shown to affect gene expression of GGP, GME (Li et al. 2013c) and GPP (Li et al. 2013b) in kiwifruit, but these studies do not determine where the control of ascorbate biosynthesis lies in fruit. Correlations between gene expression and a metabolite support but do not prove that a gene product regulates the metabolite, and it is likely some degree of coordination between genes in a pathway would occur. Some recent sequencing of an A. chinensis var. chinensis kiwifruit during fruit development is shown in Fig. 13.2. These data (L. Luo et al. unpublished) are consistent with other published studies mentioned above.

Fig. 13.2
figure 2

Next-generation sequencing study of gene expression during fruit development in an Actinidia chinensis var. chinensis genotype. Data unpublished from L. Luo et al.

Full size image

Various studies have established GGP as a key regulator of ascorbate biosynthesis in many plants, and most work has been done on this gene (Laing et al. 2007; Bulley et al. 2009; Li et al. 2013a, b, 2014). This is reasonable given that the GGP enzyme catalyses the first committed step in ascorbate biosynthesis, although others have suggested GMP as the key control gene (Wang et al. 2013a). Gene expression of GGP and GME in a range of species is strongly regulated by light (Dowdle et al. 2007; Yabuta et al. 2007; Gao et al. 2011; Massot et al. 2012) and abiotic stress (Li et al. 2013a), both factors regulate ascorbate concentrations. In addition, as discussed above, expression of these two genes relates to the ascorbate concentrations in different kiwifruit species with different ascorbate concentrations (Bulley et al. 2009; Li et al. 2014). More significantly, transient transformation of Nicotiana benthamiana with various kiwifruit genes from the ascorbate biosynthetic pathway shows that GGP strongly affects ascorbate, and GME has little effect while a combination of GME and GGP synergistically stimulates ascorbate concentration (Bulley et al. 2009, 2012; Laing et al. 2015). Other genes in the pathway except GMP have little effect on ascorbate concentration ((Zhou et al. 2012), W.A. Laing unpublished). In addition, genetic mapping in apple has established that the QTLs for ascorbate colocate with GGP orthlogs (Mellidou et al. 2012) and that transformation of strawberry, potato and tomato with kiwifruit or Arabidopsis GGP results in significantly increased ascorbate (Bulley et al. 2012). GGP gene expression also shows a strong diurnal trend, peaking during the morning and falling during the day in Arabidopsis (Dowdle et al. 2007), suggesting that ascorbate biosynthesis potential is primed for the period of the day when peak ascorbate is required, at midday maximum light intensity. We have also observed similar diurnal trends in gene expression in the leaves of Arabidopsis (W.A. Laing unpublished).

13.5.2 Regulation Through Translation of GGP

It appears well established that transcriptional regulation of the key genes GGP and GME controls ascorbate concentrations. However, translational regulation of ascorbate concentrations also plays a significant part. Recently, it was shown that the 5′ untranslated region (UTR) of GGP from a wide range of species including kiwifruit contains a highly conserved upstream open reading frame (uORF) (Laing et al. 2015). This uORF is unusual in that it starts with a non-canonical start codon, ACG. The model proposed was that under high ascorbate concentrations, the uORF is translated and acts as an inhibitor of GGP translation. Under low ascorbate, the uORF is skipped and GGP is translated. This model provides a direct link between the ascorbate concentration and the production of GGP, the key regulatory enzyme in ascorbate biosynthesis, and allows rapid and feedback responsive control of ascorbate biosynthesis under demand conditions (e.g. high light and low temperatures).

13.5.3 Other Ascorbate Affecting Genes

Four genes that regulate ascorbate in Arabidopsis have been identified and characterised. These are a protein kinase/phosphatase (VTC3) (Conklin et al. 2013), a transcription factor (AtERF98) (Zhang et al. 2012), photomorphogenic factor COP9 signalosome subunit 5B (CSN5B) (Wang et al. 2013b) and an F box protein (AMR1) (Zhang et al. 2009). Actinidia homologues of the first three genes have been found but not of the F box protein. Little is known how VTC3 regulates ascorbate except that it is constitutively expressed and does not vary much under a wide range of conditions, suggesting that regulation may be at a post-transcriptional level (Conklin et al. 2013). It is possible that VTC3 may be a factor in the uORF regulation described above (Laing et al. 2015). AMR1 appears to be a negative regulator of ascorbate and a negative regulator of the l-galactose pathway of ascorbate biosynthesis genes (Zhang et al. 2009). Inspection of this paper shows that GGP and GME are the genes most affected by AMR1. An obvious hypothesis is that AMR1 targets for degradation a transcription factor that increases transcription of the ascorbate genes. The AtERF98 transcription factor increases the transcription of many of the genes in the l-galactose pathway, GMP especially and GGP (Zhang et al. 2012) and binds with the promoter of GMP. However, the authors did not explore how AtERF98 interacted with other promoters or whether it was a target for AMR1. Lastly, CSN5B was shown to interact with the N terminus of GMP and target it for degradation (Wang et al. 2013b). CSN5B appears to promote degradation of GMP protein in the dark through ubiquitination and the proteasome. However, the authors did not test whether CSN5B interacted with other genes in the l-galactose pathway. Interestingly, knockout of another COP9 signalosome subunit, CSN8, increased ascorbate even further than a knockout of CSN5B (Wang et al. 2013b).

In addition, an ascorbate transporter located in the chloroplast has also been identified (Miyaji et al. 2015) and an Actinidia homologue is present. The transporter would serve to transport ascorbate into the chloroplast from its point of synthesis in the mitochondria.

13.6 Conclusions

The ascorbate biosynthetic pathway in kiwifruit is similar to that found in other species. The most likely pathway to provide the great bulk of ascorbate in kiwifruit is the l-Galactose pathway and regulation of ascorbate concentration is through the GGP gene and enzyme, with possibly some higher level feedback by ascorbate against PMI1. Ascorbate content plant tissues appear to be mainly the result of variation in transcription of GGP (and GME in synergy with GGP) as well as at the translational level through feedback regulation by ascorbate of the translation of GGP. Regulation through GGP makes sense as it the first committed step in the biosynthesis of ascorbate.