Folate biofortification in food crops

Available online at www.sciencedirect.com ScienceDirect Folate biofortification in food crops Simon Strobbe and Dominique Van Der Straeten Folates are essential vitamins in the human diet. Folate deficiency is still very common, provoking disorders such as birth defects and anemia. Biofortification via metabolic engineering is a proven powerful means to alleviate folate malnutrition. A variety of metabolic engineering approaches have been successfully implemented in different crops and tissues. Furthermore, ensuring folate stability is crucial for longterm storage of crop products. However, the current strategies, shown to be successful in rice and tomato, will need to be finetuned to enable adequate biofortification of other staples such as potato, wheat and cassava. Thus, there is a need to overcome remaining hurdles in folate biofortification. Overall, biofortification, via breeding or metabolic engineering, will be imperative to effectively combat folate deficiency. Address Laboratory of Functional Plant Biology, Department of Biology, Ghent University, K.L. Ledeganckstraat 35, B-9000 Ghent, Belgium Corresponding author: Van Der Straeten, Dominique (Dominique. VanDerStraeten@ugent.be) Current Opinion in Biotechnology 2017, 44:202–211 This review comes from a themed issue on Plant biotechnology Edited by Dominique Van Der Straeten, Hans De Steur and Teresa B Fitzpatrick http://dx.doi.org/10.1016/j.copbio.2016.12.003 ensuring its cellular retention [4]. Polyglutamylated folates can therefore be considered more stable than monoglutamates in vivo. The chemical diversity of folates reflects a perfect adaptation to their varied biological function as C1-donors and acceptors, rendering them a pivotal role in primary metabolism of nearly all organisms. Folate-dependent enzymes play a key role in thymidylate and purine synthesis, as well as pantothenate (vitB5) formation [3]. 5-methylTHF donates its methyl group to homocysteine to form methionine by the action of the cobalamin (vitB12)dependent methionine synthase [5]. In plants, folates have an additional essential role in photorespiration, as well as in chlorophyll, plastoquinone, tocopherol, pectin and lignin synthesis [6]. Due to their central role in primary metabolism, detrimental physiological effects arise upon folate deficiency [7]. Animals, unable to synthesize folates de novo, rely primarily on their diet for an adequate folate supply. Decreased folate levels result in impeded erythrocyte development, causing megaloblastic anemia. During embryogenesis, folate deficiency provokes aberrant neurulation, leading to the onset of neurodegenerative disorders such as anencephaly and spina bifida [8]. Together, folate deficiency induced Neural Tube Defects (NTDs) are estimated to account for over 150 000 birth defects each year, predominantly in the developing world [9]. 0958-1669/ã 2017 Elsevier Ltd. All rights reserved. Folates are a group of water soluble B-vitamins (vitB9), consisting of a pteridine ring, a para-aminobenzoate moiety ( p-ABA) and a g-linked tail with one or more L-glutamates (Figure 1) [1]. Folates are labile compounds, prone to (photo-)oxidative cleavage [2]. Specific folate entities are chemically distinguished by three different structural modifications. First, folates exist in varying oxidation states, with tetrahydrofolate (THF) being the most reduced form. THF is the bioactive vitamin, functioning as an essential co-enzyme in numerous metabolic reactions. Second, folates can harbor a range of one-carbon (C1) units on the pteridine (N5) and p-ABA (N10) moiety, influencing their stability. Third, the length of the glutamate tail is highly variable [3]. A longer glutamate tail facilitates binding of the vitamin to folate-dependent enzymes, as well as Current Opinion in Biotechnology 2017, 44:202–211 Fermented foods, leguminous and leafy vegetables can be considered rich sources of folates. However, some massively consumed staple crops, such as rice, corn, wheat, potato and cassava, contain inadequate folate levels (Table 1). The recommended daily intake (RDI) of folate is 400 mg for an adult, increasing to 600 mg during pregnancy [4]. Unfortunately, many diets, in developing as well as developed countries, fail to reach these standards. A combined strategy of technical, socio-economical and biotechnological solutions will be essential to relieve this global burden. In this review, current state-of-the-art on biotechnological approaches for folate biofortification will be discussed. Biosynthesis In plants, folate biosynthesis is characterized by subcellular compartmentation (Figure 2) [3]. The pterin branch of folate biosynthesis takes place in the cytosol, yielding 6-hydroxymethyldihydropterin (HMDHP). Secondly, www.sciencedirect.com Folate biofortification Strobbe and Van Der Straeten 203 Figure 1 R1 R2 H H CH3 H H CHO HC=NH H CHO H - CH2 = CH - THF 5 - methyITHF 5 - formyITHF 5 - formiminoTHF 10 - formyITHF 5,10 - methyleneTHF 5,10 methenyITHF n O COOH NH HC CH2 R2 O N 5 NH N10 R1 H2C CH2 C H O H H 2N N N H OH H Current Opinion in Biotechnology Chemical structure of folates. Folates consist of three moieties: a pteridine (green), a p-aminobenzoate molecule (blue) and a glutamate tail (red). The green/blue transition reflects a (photo)-oxidation-labile bond. The folate shown is a polyglutamylated tetrahydrofolate (THF). Plant folates carry up to eight glutamates [4,67]. Different folate forms are distinguished by different C1-substituents, at different levels of oxidation, on N5 or N10. the p-ABA branch resides in plastids, consuming chorismate as a substrate. The resulting p-ABA, together with HMDHP, are assumed to enter the mitochondria by passive diffusion and carrier mediated transport, respectively [2]. Condensation of the two moieties occurs in mitochondria, followed by polyglutamylation of the resulting folate [10]. However, polyglutamated folates are retained in mitochondria, as they are intracellularly transported as monoglutamates, with vacuolar import being the only known exception [11]. Table 1 Folates in foods. Different food products are ranked according to folate content. Most staples contain inadequate levels of folate (RDI: 400 mg for adults; 600 mg for pregnant women). Brie cheese is an example of a fermented food. Data on folate content were derived from the USDA National Nutrient Database for Standard Reference (Release 28, September 2015, revised in May 2016). Based on these data, the fold increase, needed to obtain a sufficient amount of folate to reach the RDI in 100 g of raw food material, was calculated. As adequate folate levels are most critical during pregnancy, 600 mg was set as the target level. Possible losses during processing and variation of folate bioavailability – both shown to notably decrease the amount of bioeffective folate, as for instance in rice endosperm [35,69] – are not accounted for) Food Folate content (mg/100 g) Fold increase to reach RDI in 100 g Global supplya (g/capita.day) Rice, white, long-grain, regular, raw Tomatoes, red, ripe, raw Potatoes, flesh and skin, raw Corn grain, yellow Plantains, raw Cassava, raw Lettuce, green leaf, raw Wheat, soft white Cheese, Brie Spinach, raw Beans, white, mature seeds, raw Lentils, raw Turkey, liver, raw 8 15 15 19 22 27 38 41 65 194 388 479 677 75 40 40 32 27 22 16 15 9 3 2 2 – 148.2 55.4 94.9 48.2 9.6 40.3 / 178.8 / / 6.8 / / a Data on average global supply (if available) of the corresponding (wet) crop product are derived from FAOSTAT, 2011 (http://faostat.fao.org)(Food Supply—Crops Primary Equivalent). For rice, milled equivalent is presented. www.sciencedirect.com Current Opinion in Biotechnology 2017, 44:202–211 204 Plant biotechnology Figure 2 GTP GTPCHI Cytosol 3 DHN-P3 4 pterin branch DHN-P Chorismate 5 p-ABA branch DHN 1 ADCS 6 (DHM) HMDHP ADC 2 p-ABA THF-GIun p-ABA 11 p-ABA THF HMDHP 7 Plastid HMDHP-P2 8 DHPA DHP 14 9 p-ABA-GIun THF-GIun HMDHP DHF THF 10 11 THF THF-Glun 11 12 THF p-ABA-GIun FPGS THF-GIun GIu 12 Mitochondrion THF-GIun p-ABA-GIu 13 p-ABA Vacuole Current Opinion in Biotechnology Overview of folate biosynthesis and salvage. Biosynthesis (black). In plastids, para-aminobenzoate ( p-ABA) is synthesized from chorismate, originating from the shikimate pathway. The first step in the plastidial p-ABA branch of folate biosynthesis is aminodeoxychorismate (ADC) formation by the action of ADC synthase (ADCS). In the cytosol, 6-hydroxymethyldihydropterin (HMDHP) is formed (pterin branch), the first step of which is the conversion of guanosine triphosphate (GTP) to dihydroneopterin triphosphate (DHN-P3) by GTP cyclohydrolase I (GTPCHI). Pterin and p-ABA moieties are condensed and reduced in the mitochondria. The final step of folate biosynthesis is polyglutamylation of tetrahydrofolate (THF) by folylpolyglutamate synthetase (FPGS) to folate polyglutamates (THF-Glun). Genes applied in successful biofortification approaches are indicated in bold and encircled. Salvage (green). Serrated arrows symbolize (photo-) oxidative cleavage of folates. Dihydropterin-6-aldehyde (DHPA) and p-aminobenzoyl-polyglutamate ( p-ABAGlun) originate from (photo-)oxidative cleavage of folates and require salvage reactions to ensure recycling of folate biosynthetic intermediates. HMDHP is recovered by pterin aldehyde reductase (PTAR), reducing DHPA. Two subsequent deglutamylation reactions in the vacuole convert pABA-Glun to p-ABA, the first of which is catalyzed by g-glutamyl hydrolase (GGH), yielding p-ABA-Glu. The remaining glutamate residue is removed by p-ABA-Glu hydrolase (PGH). Transport and storage (blue). THF is able to exit the mitochondrion, where it can be polyglutamylated by cytosolic FPGS (ctFPGS). Plastids can take up THF from the cytosol, followed by their polyglutamylation by plastidial FPGS. THF-Glun are able to enter the vacuole where they can be converted to THF by GGH, or retained in a storage form (dotted ellipse) [11]. THF-Glun serve as a cofactor in one-carbon metabolism in different subcellular compartments. Barrels represent transporter proteins, corresponding with known (solid) or unknown (transparent) genes. Regulation (red). Dihydropteroate synthase (DHPS) is known to be feedback inhibited by the three subsequent folate biosynthesis intermediates [68]. Abbreviations. DHN-P, dihydroneopterin monophosphate; DHN, dihydroneopterin; DHM, dihydromonapterin; HMDHP-P2, 6-hydroxymethyldihydropterin pyrophosphate; DHP, dihydropteroate; DHF, dihydrofolate; Glu, glutamate. Enzymes. 1, ADC synthase (ADCS); 2, ADC lyase; 3, GTP cyclohydrolase I (GTPCHI); 4, DHN-P3pyrophosphatase; 5, non-specific phosphatase; 6, DHN aldolase; 7,HMDHP pyrophoshokinase; 8, DHPsynthase; 9, DHFsynthetase; 10, DHFreductase; 11, folylpolyglutamate synthetase (FPGS); 12, g-glutamyl hydrolase (GGH); 13, p-ABA-Glu hydrolase (PGH); 14, pterin aldehyde reductase (PTAR). Salvage Plants require salvage reactions to enable recycling of dihydropterin-6-aldehyde (DHPA) and p-aminobenzoylCurrent Opinion in Biotechnology 2017, 44:202–211 (poly)glutamate ( p-ABA-Glun), originating from oxidative cleavage of folates. DHPA is reduced in the cytosol to retrieve HMDHP [12]. When fully oxidized, however, www.sciencedirect.com Folate biofortification Strobbe and Van Der Straeten 205 pterins can no longer be recycled into folate [13]. To allow salvage of p-ABA, the glutamate tail is removed in the vacuole. Fighting folate deficiency A varied diet containing folate-rich foods is the optimal approach in combating folate deficiency. However, this requires global educational efforts combined with dietary interventions. As an alternative, application of the synthetic folate analog, folic acid, via supplementation or fortification of processed foods such as flour, has been implemented in many countries [14]. Although health benefits of folic acid fortification on NTD incidence stand undisputed [15,16], adverse effects may emerge upon high intake. Excessive supplementation results in high levels of unmetabolized folic acid in blood plasma, which has been linked to an increased risk of colorectal cancer in men [17] and impaired immunity in women [18]. Therefore, supplementation of synthetic folic acid, although proven to be very successful, should be dealt with caution [18–20]. In addition, the use of probiotic gut bacteria, overproducing folates, has been suggested [21]. These interventions are, however, difficult to implement in poor rural regions, where they are most needed [4]. Fortunately, biofortification (the enhancement of the natural folate content in crops) promises to be a costeffective complementary strategy in the battle against micronutrient malnutrition [22]. Advances in folate biofortification A flow-chart of different approaches towards biofortification is presented in Figure 3. Breeding Conventional breeding for nutritional enhancement relies on inheritance of favorable quantitative trait loci Figure 3 Folate biofortification of a target crop Breeding: is there sufficient folate variation in crop germplasm? Yes Metabolic engineering No Success Yes Identify QTL(s) via GWAS Is GA-engineering strategy sufficient? No Yes Is stability during storage acceptable? Yes No Employ QTL(s) in MAB Is additional engineering of other biosynthesis genes successful? Success Implementation of novel targets No Yes Are the desired folate levels and stability reached? No No Enhancing folate stability Introducing FBPs Extending polyglutamate tail Enhancing antioxidant levels Augmenting folate salvage Current Opinion in Biotechnology Flow-chart of possible approaches towards folate biofortification. The flowchart indicates the different requirements that need to be assessed in the process towards biofortification of a target crop, as well as the order in which these should be addressed. Ellipses represent steps in biofortification via breeding (purple) or metabolic engineering (blue) towards a successfully biofortified target crop/tissue (green). Dotted arrow indicates the possible implementation of targets, identified via QTL annotation, in metabolic engineering approaches. www.sciencedirect.com Current Opinion in Biotechnology 2017, 44:202–211 206 Plant biotechnology (QTLs) from sexually compatible parental lines. Therefore, biofortification by breeding is constrained by the natural variation of the desired trait present in the available collection of crop germplasm, as well as by being time-consuming [23]. However, genome-wide association studies (GWAS) combined with marker assisted breeding (MAB) prove to be a powerful tool in biofortification, demonstrated by the identification of multiple maize QTLs responsible for 3.22 and 5.76-fold increase in b-carotene (provitamin A) and a-tocopherol (vitamin E) content, respectively [24,25]. Although previous studies suggested insufficient folate variation in rice and wheat accessions [3,26], screening vast collections of germplasm might reveal greater diversity and thereby favor the applicability of breeding strategies [27]. Indeed, by investigating 78 rice germplasms, up to 7.6-fold difference in milled rice folate content was observed [28]. Similarly, considerable variation in folate levels was found in different potato, spinach and dry bean accessions [29,30,31]. The availability of germplasm containing sufficient folate variation has enabled identification of three rice QTLs, influencing grain folate content [32]. The identified rice QTLs do not correspond to folate biosynthesis genes, which are known to play a decisive role in folate accumulation [33,34,35]. Interestingly, three rice genes were attributed to one of these QTLs [32]. These include a rice gene, homologous to human folate hydrolase (catalyzing the shortening of the glutamate tail), as well as a homolog of an Arabidopsis plastidial folate transporter and a serine hydroxymethyl transferase. This indicates that QTL-mapping, apart from its high potency to be used in marker-assisted breeding (MAB), enables assignment of certain genes to be implicated in folate accumulation and possible discovery of novel factors in folate metabolism (Figure 3). Metabolic engineering Boosting folate biosynthesis Boosting folate biosynthesis via metabolic engineering was the first proposed strategy to biofortify plants [36–38], as it had proven its potency in lactic acid bacteria [39]. This was first assessed in G-engineered Arabidopsis, where heterologous expression of a bacterial GTP cyclohydrolase I (GTPCHI), the enzyme executing the first committed step in cytosolic pterin synthesis (push-strategy), yielded up to a fourfold increase in total folate content [36]. The sole introduction of GTPCHI has resulted in a similar level of enhancement in tomato, maize, lettuce and Mexican common bean [37,40,41,42] (Table 2). The modest enrichment of folate levels by this strategy (up to ninefold), together with the strong accumulation of pterin precursors [36,37,42], suggested the existence of an additional bottleneck in folate biosynthesis [43]. This is probably the consequence of a depleted p-ABA pool, as feeding the transgenic plants with p-ABA resulted in additional folate enhancement [37,42]. The ability of a depleted p-ABA pool to constrain folate Current Opinion in Biotechnology 2017, 44:202–211 accumulation is further highlighted in rice by combining expression of different folate biosynthesis genes (except ADCS) with GTPCHI, rendering no further improvement compared to the G-engineered parental lines [44]. Surprisingly, the p-ABA levels in the engineered Mexican common bean lines were elevated, though still shown to limit folate accumulation. This phenomenon has not been detected in previous G-engineered crops and reveals the possible existence of a feedforward regulatory mechanism. Likewise, single gene approaches using aminodeoxychorismate synthase (ADCS), performing the first step towards p-ABA formation in plastids, resulted in a decrease or insignificant enhancement of folate levels in rice and potato [33,45]. Furthermore, an assessment of different one-gene approaches in rice endosperm revealed that ectopic expression of dihydrofolate synthase (DHFS) and folylpolyglutamate synthase (FPGS), which represent two of the three final mitochondrial folate biosynthesis steps (pull-strategy), ensure a very modest increase in folate content [44]. The same study was unable to confirm previously reported 1.4-fold folate enhancement in rice seeds by the sole introduction HMDHP pyrophosphokinase/dihydropteroate of synthase (HPPK/DHPS) [46], catalyzing the first steps in mitochondrial folate biosynthesis. These findings reveal that single gene approaches will likely remain insufficient for high folate accumulation and indicate the need for multi-gene strategies. Indeed, combined usage of GTPCHI and ADCS transgenes, thereby stimulating both pterin and p-ABA branches of folate biosynthesis, enables folate overproduction in tomato fruit and rice endosperm, reaching the desired target levels [33,47]. However, the enrichment of folates by this GTPCHI/ADCS (GA)-strategy results in a shift of the polyglutamylation ratio, as monoglutamates are more prevalent. Despite the success of the GA-strategy in tomato and rice, extrapolation towards biofortification of Arabidopsis or potato tubers appeared ineffective [45]. The increased levels of pterins and p-ABA in these GA-engineered plants indicated the presence of an additional restriction in folate biosynthesis, possibly linked to more strict regulation in meristematic tissues [4]. Moreover, accumulation of tetrahydrofolates in rice endosperm is proposed to be solely restricted by GTPCHI and ADCS activity, since basal expression of the endogenous folate biosynthesis genes remain unaltered in GA-engineered rice [34]. Conversely, endogenous folate biosynthesis is upregulated in GA-tomato [48]. Boosting folate stability Folate stability, though often neglected, is problematic, as folate levels drop more than 50% during 4 month storage of GA-engineered rice grains [35]. Moreover, www.sciencedirect.com Folate biofortification Strobbe and Van Der Straeten 207 Table 2 Overview of metabolic engineering strategies to enhance folate levels in plants Engineering approach Genes engineered Gene origin Target species Folate change Reference Biosynthesis Single-gene ADCS ADCS GTPCHI GTPCHI GTPCHI GTPCHI GTPCHI GTPCHI GTPCHI GTPCHI GTPCHI HPPK/DHPS HPPK/DHPS HPPK/DHPS DHFS GTPCHI + other biosynthesis genes a GTPCHI + ADCS GTPCHI + ADCS GTPCHI + ADCS Arabidopsis thaliana Arabidopsis thaliana Escherichia coli Arabidopsis thaliana Mus musculus Arabidopsis thaliana Escherichia coli Gallus gallus Arabidopsis thaliana Arabidopsis thaliana Arabidopsis thaliana Triticum aestivum Triticum aestivum Arabidopsis thaliana Arabidopsis thaliana Arabidopsis thaliana Rice seeds Potato Arabidopsis Arabidopsis Tomato Potato Maize Lettuce Mexican common bean Rice seeds Rice seeds Rice leaves Rice seeds Rice seeds Rice seeds Rice seeds (1/6)-fold insignificant fourfold insignificant twofold twofold twofold ninefold threefold insignificant 6.1-fold twofold 1.4-fold insignificant 1.27-fold 6.1-fold Storozhenko et al. [33] Blancquaert et al. [45] Hossain et al. [36] Blancquaert et al. [45] de la Garza et al. [37] Blancquaert et al. [45] Naqvi et al. [40] Nunes et al. [41] Ramı́rez Rivera et al. [42] Storozhenko et al. [33] Dong et al. [44] Gillies et al. [46] Gillies et al. [46] Dong et al. [44] Dong et al. [44] Dong et al. [44] Arabidopsis thaliana Arabidopsis thaliana Mus musculus (GTPCHI), Arabidopsis thaliana (ADCS) Arabidopsis thaliana Rice seeds Arabidopsis Tomato 100-fold insignificant 25-fold Storozhenko et al. [33] Blancquaert et al. [45] de la Garza et al. [47] Potato threefold Blancquaert et al. [45] Multi-gene GTPCHI + ADCS Polyglutamylation FPGS FPGS GTPCHI + ADCS + FPGS Arabidopsis thaliana Oryza sativa Arabidopsis thaliana Rice seeds Rice seeds Rice seeds 1.45-fold 4.7-fold 100-fold Dong et al. [44] Abilgos Ramos, 2010 b Blancquaert et al. [35] Folate binding proteins FBP Bos taurus Rice seeds 6.2-fold Abilgos Ramos, 2010 b GNMT GTPCHI + ADCS + FBP Rattus norvegicus Arabidopsis thaliana (G + A) Bos taurus (FBP) Rice seeds Rice seeds 8.8-fold 150-fold Abilgos Ramos, 2010 b Blancquaert et al. [35] GGH RNAi 5-FCL ablation / / Arabidopsis Arabidopsis 1.3-fold twofold Akhtar et al. [11] Goyer et al. [51] Homeostasis a b A two-gene approach was conducted, combining GTPCHI with respectively ADCL, DHNA, HPPK/DHPS, DHFS, DHFR and FPGS. Abilgos Ramos R, PhD thesis, University of Nottingham, 2010. as folates are labile compounds, enhancing their stability could further boost folate build-up (Figure 3). Polyglutamylation Polyglutamylation positively influencing folate stability, together with the observed elevated ratio of monoglutamates in folate accumulating crops [33,47], were incentives for engineering of the glutamate tail length. The polyglutamylation state could be manipulated towards accumulation of polyglutamates by knock-down of g-glutamyl hydrolase (GGH), which removes the glutamate tail in folate homeostasis in vacuoles. This concept has been addressed in Arabidopsis and tomato, confirming that vacuolar GGH expression has a negative influence on folate content and polyglutamate levels [11]. Consequently, the level of polyglutamated folates is enriched upon GGH suppression, due to an enlarged vacuolar sink [11]. Conversely, overexpression of FPGS, responsible for www.sciencedirect.com the addition of the polyglutamate tail, resulted in a 4.7fold increase in total folate of rice endosperm (Abilgos Ramos R, PhD thesis, University of Nottingham, 2010). Furthermore, ectopic expression of mitochondrial FPGS was proven successful in the endosperm of GA-engineered rice, resulting in an increase of polyglutamylated folates, exhibiting enhanced stability upon storage [35]. This engineering strategy is, however, metabolically different, as it implies trapping an enlarged pool of folate polyglutamates in mitochondria, where they are more likely to be exposed to reactive oxygen species (ROS). Folate binding proteins Folate binding proteins (FBPs) have been of major interest for metabolic engineering strategies [38], as they are known to greatly stabilize folates in mammals [49]. Because plant derived FBPs remain to be characterized, current biofortification approaches rely on their Current Opinion in Biotechnology 2017, 44:202–211 208 Plant biotechnology mammalian counterparts. In this perspective, synthetic codon-optimized bovine FBP was introduced in GAengineered rice endosperm, enabling sequestration of folate polyglutamates in the cytosol [35]. Rice lines obtained by this intervention possess improved folate stability, as well as folate levels exceeding those in the ‘first generation’ of folate biofortified rice [33]. Interestingly, engineering the folate binding glycine N-methyltransferase (GNMT) from rat liver was shown to be the most successful one-gene approach in rice endosperm, exhibiting 8.8-fold folate enhancement (Abilgos Ramos R, PhD thesis, University of Nottingham, 2010). Taken together, mammalian FBPs are suggested to augment folate levels by promoting their sequestration (creating a folate sink), additional to their ability to prolong crop storage by increasing folate stability. Engineering folate homeostasis Driving folate homeostasis towards accumulation of more stable folate forms has been proposed as an alternative engineering strategy [50]. 5-formylTHF is the most stable naturally occurring folate, with a presumed storage function [1,4]. Mutation of formylTHF cycloligase (5FCL), the sole enzyme known to consume 5-formylTHF, led to enrichment of Arabidopsis leaves with folate, unfortunately coinciding with reduced growth rate [51]. complete active folate biosynthesis pathway in seeds [54]. This strategy, however, is not a guarantee for success, as its implementation remained ineffective in Arabidopsis and potato [45]. Hence, future biofortification strategies should include a back-up strategy, applicable to these crops in which GA-engineering approaches are ineffective (Figure 3). The latter should tackle the remaining hurdles impeding folate biofortification downstream of pABA and pterin accumulation, the existence of which has been observed in Arabidopsis and potato [45]. Co-occurrence of pterins and p-ABA has also been detected in Gengineered Mexican common bean [42], suggesting intracellular transport of these intermediates or HPPK/ DHPS activity to be possible constraints. The extent to which intracellular transport determines folate accumulation could be examined by assessing subcellular localization of folates, together with its biosynthetic intermediates, in the engineered plants. In this regard, characterization of a mitochondrial pterin importer could mean a leap forward towards optimized metabolic engineering of folate content. To date, only plastidial and vacuolar folate monoglutamate transporters have been identified [55–57]. On the other hand, the role of HPPK/DHPS and other biosynthesis genes, could be examined in transgene-stacking approaches, in combination with GA-engineering. Future research challenges Folate stability The occurrence of folate deficiency is predominantly caused by low folate levels in popular staples such as rice, potato, maize, plantain, cassava and wheat (Table 1). Biofortification via metabolic engineering or breeding holds the potential to reach the required folate levels in these crops, the concept of which has been proven in rice [35]. Breeding Future breeding strategies should focus on the pursuit of sufficient folate variation in target crop germplasm, followed by identification of the underlying QTLs. Despite the limitation of breeding strategies, high-resolution QTL-mapping in model species will enable identification of novel engineering targets for folate biofortification [25,32,52] (Figure 3). Metabolic engineering Folate biosynthesis Simultaneous activation of the p-ABA and pterin branches of folate biosynthesis appears essential to reach substantial folate enhancement, considering that singlegene approaches have only resulted in modest folate increase, due to inadequate supply of pterins and/or pABA. Therefore, a GA-engineering push-strategy will remain a prerequisite in folate biofortification [43,53]. Wheat is a good candidate crop for assessing this engineering approach, as it has been shown to harbor the Current Opinion in Biotechnology 2017, 44:202–211 Concerning folate stability, different strategies need further assessment: (1) introduction of FBPs, (2) engineering glutamate tail length, (3) enhancing antioxidant levels, (4) augmenting folate salvage (Figure 3). Considering the power of folate sequestration and protection by transgenic FBPs, targeting to different subcellular compartments could be tested, thereby fine-tuning the engineering strategy. Furthermore, different nonplant FBPs could show to be more successful than thus far implemented proteins, as their efficacy appears variable (Abilgos Ramos R, PhD thesis, University of Nottingham, 2010). Extending the glutamate tail by heterologous expression or overexpression of cytosolic FPGS is of particular interest [35], as it ensures folate accumulation in the cytoplasm (even stronger in combination with FBP), guarded from detrimental reactive oxygen species in the mitochondria or GGH activity in the vacuoles. Similarly, lowering GGH activity could be beneficial, given its ability to counteract the accumulation of polyglutamates in the cytosol [11,35]. Novel techniques in genome editing, for example, the CRISPR/Cas9 technology [58], enable such directed manipulations of gene activity. However, GGH suppression should be approached with caution, given its role in folate salvage, forming p-ABAGlu from p-ABA-Glun released upon (photo)-oxidative cleavage of THF-Glun. Therefore, in future metabolic www.sciencedirect.com Folate biofortification Strobbe and Van Der Straeten 209 engineering, GGH could be altered to favor p-ABA-Glun as a substrate over THF-Glun, using directed evolution [59,60]. In biofortification, the most desired folate vitamer is the methylated derivative of the fully reduced THF (5methyl-THF), given its stability and bio-activity. In this respect, enhancing antioxidant levels could assist in accumulation of the targeted folate vitamer together with a possible protection from oxidative cleavage [3]. This strategy could be reinforced by the capacity of folates to influence the cellular redox state, thereby ensuring sufficient replenishment of reduced antioxidants [61]. Antioxidant candidates are ascorbate (vitamin C), pyridoxine (vitamin B6) and a-tocopherol (vitamin E), which could further enhance the nutritional value of the engineered crop [62,63]. Similarly, folate salvage, more particularly HMDHP recovery from DHPA by pterin aldehyde reductase (PTAR), has emerged as another possible target for folate biofortification. This is supported by the occurrence of pterin aldehydes in the G-engineered Mexican common beans [42]. Interestingly, folate salvage could be extended, by introduction of a protozoan pterin reductase gene, capable of reducing fully oxidized pterins [64,65]. Conclusion Folate biofortification exemplifies how metabolic engineering strategies enable the acquisition of fundamental knowledge on the complex matter of folate biosynthesis, salvage and homeostasis, as well as its regulation, part of which remains to be elucidated. The main goal is to design an effective biofortification strategy, considering both folate accumulation and stability, adaptable to the specific metabolism of different target tissues in crops cultivated in regions troubled with folate malnutrition. A strategy successful for biofortification of potato, combined with further fundamental research in Arabidopsis, could provide keys to effective folate enhancement in other staples. This could form a cornerstone for multi-biofortified crops [66], as these promise to be a powerful tool to reduce the global burden of micronutrient deficiency. Acknowledgement S.S. is indebted to the Agency for Innovation by Science and Technology in Flanders (IWT) for a predoctoral fellowship. D.V.D.S. acknowledges support from Ghent University (Bijzonder Onderzoeksfonds, BOF2004/GOA/012 and BOF2009/G0A/004), and the Research Foundation—Flanders (FWO, projects 3G012609 and 35963). References and recommended reading Papers of particular interest, published within the period of review, have been highlighted as:  of special interest  of outstanding interest www.sciencedirect.com 1. Rebeille F, Ravanel S, Jabrin S, Douce R, Storozhenko S, Van Der Straeten D: Folates in plants: biosynthesis, distribution, and enhancement. Physiol Plant 2006, 126:330-342. 2. Hanson AD, Gregory JF: Folate biosynthesis, turnover, and transport in plants. Annu Rev Plant Biol 2011, 62:105-125. 3. Blancquaert D, Storozhenko S, Loizeau K, De Steur H, De Brouwer V, Viaene J, Ravanel S, Rebeille F, Lambert W, Van Der Straeten D: Folates and folic acid: from fundamental research toward sustainable health. CRC Crit Rev Plant Sci 2010, 29:1435. 4. Blancquaert D, De Steur H, Gellynck X, Van Der Straeten D: Present and future of folate biofortification of crop plants. J Exp Bot 2014, 65:895-906. 5. Banerjee RV, Matthews RG: Cobalamin-dependent methionine synthase. FASEB J 1990, 4:1450-1459. 6. Ravanel S, Douce R, Rebeille F: Metabolism of folates in plants. Adv Bot Res 2011, 59:67-106. 7. Araujo JR, Martel F, Borges N, Araujo JM, Keating E: Folates and aging: role in mild cognitive impairment, dementia and depression. Ageing Res Rev 2015, 22:9-19. 8. Geisel J: Folic acid and neural tube defects in pregnancy—a review. J Perinat Neonatal Nurs 2003, 17:268-279. 9. Flores AL, Vellozzi C, Valencia D, Sniezek J: Global burden of neural tube defects, risk factors, and prevention. Indian J Community Health 2014, 26:3-5. 10. Rebeille F, Macherel D, Mouillon JM, Garin J, Douce R: Folate biosynthesis in higher plants: purification and molecular cloning of a bifunctional 6-hydroxymethyl-7,8-dihydropterin pyrophosphokinase/7,8-dihydropteroate synthase localized in mitochondria. EMBO J 1997, 16:947-957. 11. Akhtar TA, Orsomando G, Mehrshahi P, Lara-Nunez A, Bennett MJ, Gregory JF, Hanson AD: A central role for gammaglutamyl hydrolases in plant folate homeostasis. Plant J 2010, 64:256-266. 12. Orsomando G, Bozzo GG, de la Garza RD, Basset GJ, Quinlivan EP, Naponelli V, Rebeille F, Ravanel S, Gregory JF, Hanson AD: Evidence for folate-salvage reactions in plants. Plant J 2006, 46:426-435. 13. Noiriel A, Naponelli V, Bozzo GG, Gregory JF, Hanson AD: Folate salvage in plants: pterin aldehyde reduction is mediated by multiple non-specific aldehyde reductases. Plant J 2007, 51:378-389. 14. Atta CAM, Fiest KM, Frolkis AD, Jette N, Pringsheim T, St Germaine-Smith C, Rajapakse T, Kaplan GG, Metcalfe A: Global birth prevalence of spina bifida by folic acid fortification status: a systematic review and meta-analysis. Am J Public Health 2016, 106:E24-E34. 15. Williams J, Mai CT, Mulinare J, Isenburg J, Flood TJ, Ethen M, Frohnert B, Kirby RS: Updated estimates of neural tube defects prevented by mandatory folic acid Fortification—United States, 1995–2011. MMWR Morb Mortal Wkly Rep 2015, 64:1-5. 16. Wang HC, De Steur H, Chen G, Zhang XT, Pei LJ, Gellynck X, Zheng XY: Effectiveness of folic acid fortified flour for prevention of neural tube defects in a high risk region. Nutrients 2016, 8:152. 17. Cho E, Zhang XH, Townsend MK, Selhub J, Paul L, Rosner B, Fuchs CS, Willett WC, Giovannucci EL: Unmetabolized folic acid in prediagnostic plasma and the risk for colorectal cancer. J Natl Cancer Inst 2015:107. 18. Selhub J, Rosenberg IH: Excessive folic acid intake and relation to adverse health outcome. Biochimie 2016, 126:71-78. 19. Smith AD, Refsum H, Selhub J, Rosenberg IH: Decision on folic acid fortification in Europe must consider both risks and benefits. BMJ 2016, 352. 20. Mills JL, Dimopoulos A, Bailey RL: What is standing in the way of complete prevention of folate preventable neural tube defects? Birth Defects Res A Clin Mol Teratol 2016, 106:517-519. Current Opinion in Biotechnology 2017, 44:202–211 210 Plant biotechnology 21. LeBlanc JG, Milani C, de Giori GS, Sesma F, van Sinderen D, Ventura M: Bacteria as vitamin suppliers to their host: a gut microbiota perspective. Curr Opin Biotechnol 2013, 24:160-168. 22. De Steur H, Blancquaert D, Strobbe S, Lambert W, Gellynck X, Van  Der Straeten D: Status and market potential of transgenic biofortified crops. Nat Biotechnol 2015, 33:25-29. This review provides an overview of the socio-economic potential of transgenic nutritionally enhanced staple crops. Particular attention is given to pro-vitamin A accumulating ‘Golden Rice’, as well as folate enhanced rice. 23. Shimelis H, Laing M: Timelines in conventional crop improvement: pre-breeding and breeding procedures. Aust J Crop Sci 2012, 6:1542-1549. 24. Azmach G, Gedil M, Menkir A, Spillane C: Marker-trait association analysis of functional gene markers for provitamin A levels across diverse tropical yellow maize inbred lines. Bmc Plant Biol 2013, 13:227. 25. Lipka AE, Gore MA, Magallanes-Lundback M, Mesberg A, Lin HN, Tiede T, Chen C, Buell CR, Buckler ES, Rocheford T et al.: Genome-wide association study and pathway-level analysis of tocochromanol levels in maize grain. G3 2013, 3:1287-1299. 26. Piironen V, Edelmann M, Kariluoto S, Bedo Z: Folate in wheat genotypes in the HEALTHGRAIN diversity screen. J Agric Food Chem 2008, 56:9726-9731. 27. Jha AB, Ashokkumar K, Diapari M, Ambrose SJ, Zhang HX, Tar’an B, Bett KE, Vandenberg A, Warkentin TD, Purves RW: Genetic diversity of folate profiles in seeds of common bean, lentil, chickpea and pea. J Food Compost Anal 2015, 42:134140. 28. Dong W, Cheng ZJ, Wang XL, Wang B, Zhang HZ, Su N, Yamamaro C, Lei CL, Wang J, Wang JL et al.: Determination of folate content in rice germplasm (Oryza sativa L.) using trienzyme extraction and microbiological assays. Int J Food Sci Nutr 2011, 62:537-543. 29. Robinson BR, Sathuvalli V, Bamberg J, Goyer A: Exploring folate  diversity in wild and primitive potatoes for modern crop improvement. Genes 2015, 6:1300-1314. This studies reports on a 10-fold folate variation within different Solanum species, demonstrating the potential of different germplasm accessions for breeding of folate-enhanced potatoes. 30. Shohag MJI, Wei YY, Yu N, Zhang J, Wang K, Patring J, He ZL, Yang XE: Natural variation of folate content and composition in spinach (Spinacia oleracea) germplasm. J Agric Food Chem 2011, 59:12520-12526. 31. Khanal S, Shi C, Xue J, Shi J, Rajcan I, Pauls KP, Navabi A: Quantitative trait loci analysis of folate content in common beans (Phaseolus vulgaris L.). Can J Plant Sci 2011, 91:375-376. 32. Dong W, Cheng ZJ, Xu JL, Zheng TQ, Wang XL, Zhang HZ,  Wang J, Wan JM: Identification of QTLs underlying folate content in milled rice. J Integr Agric 2014, 13:1827-1834. Using 264 recombinant inbred lines and 182 backcross inbred lines, the authors have pinpointed 3 QTLs on the rice genome explaining up to 25.3% of the folate variation in milled rice. This resulted in the discovery of two new genes, considered to influence folate levels in rice grains. The study exemplifies the vast potential of QTL mapping studies, enabling identification of new targets for marker assisted breeding or metabolic engineering. 33. Storozhenko S, De Brouwer V, Volckaert M, Navarrete O, Blancquaert D, Zhang GF, Lambert W, Van Der Straeten D: Folate fortification of rice by metabolic engineering. Nat Biotechnol 2007, 25:1277-1279. 34. Blancquaert D, Van Daele J, Storozhenko S, Stove C, Lambert W, Van Der Straeten D: Rice folate enhancement through metabolic engineering has an impact on rice seed metabolism, but does not affect the expression of the endogenous folate biosynthesis genes. Plant Mol Biol 2013, 83:329-349. 35. Blancquaert D, Van Daele J, Strobbe S, Kiekens F, Storozhenko S,  De Steur H, Gellynck X, Lambert W, Stove C, Van Der Straeten D: Improving folate (vitamin B-9) stability in biofortified rice through metabolic engineering. Nat Biotechnol 2015, 33:10761078. Current Opinion in Biotechnology 2017, 44:202–211 This study reports on the second generation of folate-engineered rice, created with the purpose of stabilizing folatesin vivo, as well as ensuring sufficient folate levels upon long-time storage. Stimulating polyglutamylation of folates via expression of mitochondrial FPGS is shown to prolong shelf-life of folate-enriched rice. Furthermore, introduction of folate binding protein has enabled further enhancement of folates, exceeding previously reached folate levels, in addition to enhancing stability. 36. Hossain T, Rosenberg I, Selhub J, Kishore G, Beachy R, Schubert K: Enhancement of folates in plants through metabolic engineering. Proc Natl Acad Sci U S A 2004, 101:5158-5163. 37. de la Garza RD, Quinlivan EP, Klaus SMJ, Basset GJC, Gregory JF, Hanson AD: Folate biofortification in tomatoes by engineering the pteridine branch of folate synthesis. Proc Natl Acad Sci U S A 2004, 101:13720-13725. 38. Storozhenko S, Ravanel S, Zhang GF, Rebeille F, Lambert W, Van Der Straeten D: Folate enhancement in staple crops by metabolic engineering. Trends Food Sci Technol 2005, 16:271281. 39. Sybesma W, Starrenburg M, Kleerebezem M, Mierau I, de Vos WM, Hugenholtz JR: Increased production of folate by metabolic engineering of Lactococcus lactis. Appl Environ Microbiol 2003, 69:3069-3076. 40. Naqvi S, Zhu CF, Farre G, Ramessar K, Bassie L, Breitenbach J, Conesa DP, Ros G, Sandmann G, Capell T et al.: Transgenic multivitamin corn through biofortification of endosperm with three vitamins representing three distinct metabolic pathways. Proc Natl Acad Sci U S A 2009, 106:7762-7767. 41. Nunes ACS, Kalkmann DC, Aragao FJL: Folate biofortification of lettuce by expression of a codon optimized chicken GTP cyclohydrolase I gene. Transgenic Res 2009, 18:661-667. 42. Ramı́rez Rivera NG, Garcı́a-Salinas C, Aragão FJ, Dı́az de la  Garza RI: Metabolic engineering of folate and its precursors in Mexican common bean (Phaseolus vulgaris L.). Plant Biotechnol J 2016, 14:2021-2032. A single-gene engineering approach was conducted in Mexican common bean, boosting the pterin branch of folate biosynthesis. The profile of different pterin forms accumulating was examined, providing insight in folate regulation. Unique to this study is the occurrence of enhanced para-aminobenzoate ( p-ABA) levels in G-engineered lines, which points to the existence of a feed-forward regulatory system in folate biosynthesis. 43. Farre G, Blancquaert D, Capell T, Van Der Straeten D, Christou P,  Zhu CF: Engineering complex metabolic pathways in plants. Annu Rev Plant Biol 2014, 65:187-223. This review provides a general overview of the principles in metabolic engineering. It highlights the different approaches and bottlenecks upon engineering plant metabolic pathways, including the folate pathway. 44. Dong W, Cheng ZJ, Lei CL, Wang XL, Wang JL, Wang J, Wu FQ, Zhang X, Guo XP, Zhai HQ et al.: Overexpression of folate  biosynthesis genes in rice (Oryza sativa L.) and evaluation of their impact on seed folate content. Plant Foods Hum Nutr 2014, 69:379-385. This study shows an evaluation of eight differentArabidopsis folate biosynthesis genes on folate accumulation in rice endosperm upon their constiutive ectopic expression. The authors show that two genes (DHFS and FPGS) enable slight but significant folate enhancement, aside from to GTPCHI and ADCS (which is supported by previous work). A two-gene metabolic engineering strategy, combining GTPCHI with six other folate biosynthesis genes (excluding ADCS), confirmed the necessity of boosting p-ABA supply to accomplish folate accumulation. 45. Blancquaert D, Storozhenko S, Van Daele J, Stove C, Visser RGF, Lambert W, Van Der Straeten D: Enhancing pterin and paraaminobenzoate content is not sufficient to successfully biofortify potato tubers and Arabidopsis thaliana plants with folate. J Exp Bot 2013, 64:3899-3909. 46. Gillies SA, McIntosh SR, Henry RJ: A transgenic cereal crop with enhanced folate: rice expressing wheat HPPK/DHPS. 11th International Wheat Genetics Symposium 2008. 47. de La Garza RID, Gregory JF, Hanson AD: Folate biofortification of tomato fruit. Proc Natl Acad Sci U S A 2007, 104:4218-4222. www.sciencedirect.com Folate biofortification Strobbe and Van Der Straeten 211 48. Waller JC, Akhtar TA, Lara-Nunez A, Gregory JF, McQuinn RP, Giovannoni JJ, Hanson AD: Developmental and feedforward control of the expression of folate biosynthesis genes in tomato fruit. Mol Plant 2010, 3:66-77. 49. Jones ML, Nixon PF: Tetrahydrofolates are greatly stabilized by binding to bovine milk folate-binding protein. J Nutr 2002, 132:2690-2694. 50. Scott J, Rebeille F, Fletcher J: Folic acid and folates: the feasibility for nutritional enhancement in plant foods. J Sci Food Agric 2000, 80:795-824. 51. Goyer A, Collakova E, de la Garza RD, Quinlivan EP, Williamson J, Gregory JF, Shachar-Hill Y, Hanson AD: 5Formyltetrahydrofolate is an inhibitory but well tolerated metabolite in Arabidopsis leaves. J Biol Chem 2005, 280:2613726142. 52. Dell’Acqua M, Gatti DM, Pea G, Cattonaro F, Coppens F, Magris G, Hlaing AL, Aung HH, Nelissen H, Baute J et al.: Genetic properties of the MAGIC maize population: a new platform for high definition QTL mapping in Zea mays. Genome Biol 2015, 16:167. 53. Saini RK, Nile SH, Keum Y-S: Folates: chemistry, analysis,  occurrence, biofortification and bioavailability. Food Res Int 2016, 89:1-13. This review discusses the important issues for folate biofortification in plants. It provides a detailed table of folate content in different crops. The authors discuss enhancing folate content in crops by elicitation using methyl-jasmonate, salicylate and ethylene. Furthermore, methods of folate analysis are covered comprehensively. 54. McIntosh SR, Henry RJ: Genes of folate biosynthesis in wheat. J Cereal Sci 2008, 48:632-638. 55. Klaus SMJ, Kunji ERS, Bozzo GG, Noiriel A, de la Garza RD, Basset GJC, Ravanel S, Rebeille F, Gregory JF, Hanson AD: Higher plant plastids and cyanobacteria have folate carriers related to those of trypanosomatids. J Biol Chem 2005, 280:38457-38463. 56. Bedhomme M, Hoffmann M, McCarthy EA, Gambonnet B, Moran RG, Rebeille F, Ravanel S: Folate metabolism in plants— an Arabidopsis homolog of the mammalian mitochondrial folate transporter mediates folate import into chloroplasts. J Biol Chem 2005, 280:34823-34831. 57. Raichaudhuri A, Peng MS, Naponelli V, Chen S, SanchezFernandez R, Gu HL, Gregory JF, Hanson AD, Rea PA: Plant vacuolar ATP-binding cassette transporters that translocate folates and antifolates in vitro and contribute to antifolate tolerance in vivo. J Biol Chem 2009, 284:8449-8460. 58. Belhaj K, Chaparro-Garcia A, Kamoun S, Patron NJ, Nekrasov V: Editing plant genomes with CRISPR/Cas9. Curr Opin Biotechnol 2015, 32:76-84. www.sciencedirect.com 59. Arnold FH: The nature of chemical innovation: new enzymes by evolution. Q Rev Biophys 2015, 48:404-410. 60. Packer MS, Liu DR: Methods for the directed evolution of proteins. Nat Rev Genet 2015, 16:379-394. 61. Fan J, Ye JB, Kamphorst JJ, Shlomi T, Thompson CB,  Rabinowitz JD: Quantitative flux analysis reveals folatedependent NADPH production. Nature 2014, 510:298-302. This study is the first to assign a role for folates in the control of the redox state of mammalian cells. It was demonstrated that folates play an indispensable role in NADPH production and glutathione reduction, thus lowering cellular oxidative stress. 62. Vanderschuren H, Boycheva S, Li KT, Szydlowski N, Gruissem W, Fitzpatrick TB: Strategies for vitamin B6 biofortification of plants: a dual role as a micronutrient and a stress protectant. Front Plant Sci 2013, 4:143. 63. Fitzpatrick TB, Basset GJC, Borel P, Carrari F, DellaPenna D, Fraser PD, Hellmann H, Osorio S, Rothan C, Valpuesta V et al.: Vitamin deficiencies in humans: can plant science help? Plant Cell 2012, 24:395-414. 64. Bekaert S, Storozhenko S, Mehrshahi P, Bennett MJ, Lambert W, Gregory JF, Schubert K, Hugenholtz J, Van Der Straeten D, Hanson AD: Folate biofortification in food plants. Trends Plant Sci 2008, 13:28-35. 65. Moreira DD, Ferreira RF, Murta SMF: Molecular characterization and functional analysis of pteridine reductase in wild-type and antimony-resistant Leishmania lines. Exp Parasitol 2016, 160:60-66. 66. Blancquaert D, De Steur H, Gellynck X, Van Der Straeten D: Metabolic engineering of micronutrients in crop plants. Ann N Y Acad Sci 2016 http://dx.doi.org/10.1111/nyas.13274. 67. Garratt LC, Ortori CA, Tucker GA, Sablitzky F, Bennett MJ, Barrett DA: Comprehensive metabolic profiling of mono- and polyglutamated folates and their precursors in plant and animal tissue using liquid chromatography/negative ion electrospray ionisation tandem mass spectrometry. Rapid Commun Mass Spectrom 2005, 19:2390-2398. 68. Mouillon JM, Ravanel S, Douce R, Rebeille F: Folate synthesis in higher-plant mitochondria: coupling between the dihydropterin pyrophosphokinase and the dihydropteroate synthase activities. Biochem J 2002, 363:313-319. 69. Kiekens F, Blancquaert D, Devisscher L, Van Daele J, Stove VV, Delanghe JR, Van Der Straeten D, Lambert WE, Stove CP: Folates from metabolically engineered rice: A long-term study in rats. Mol Nutr Food Res 2015, 59:490-500. Current Opinion in Biotechnology 2017, 44:202–211