Bioresource Technology 258 (2018) 345–353

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Review

Current advance in biological production of malic acid using wild type and T
metabolic engineered strains
Zhongxue Daia, Huiyuan Zhoua, Shangjie Zhanga, Honglian Gua, Qiao Yanga, Wenming Zhanga,b,
⁎
Weiliang Donga,b, Jiangfeng Maa,b, Yan Fanga,b, Min Jianga,b, , Fengxue Xina,b
a
State Key Laboratory of Materials-Oriented Chemical Engineering, College of Biotechnology and Pharmaceutical Engineering, Nanjing Tech University, Nanjing 211800,
PR China
b
Jiangsu National Synergetic Innovation Center for Advanced Materials (SICAM), Nanjing Tech University, Nanjing 211800, PR China

A R T I C LE I N FO A B S T R A C T

Keywords: Malic acid (2-hydroxybutanedioic acid) is a four-carbon dicarboxylic acid, which has attracted great interest due
L-Malic acid to its wide usage as a precursor of many industrially important chemicals in the food, chemicals, and phar-
Biosynthesis maceutical industries. Several mature routes for malic acid production have been developed, such as chemical
Renewable resources synthesis, enzymatic conversion and biological fermentation. With depletion of fossil fuels and concerns re-
Metabolic engineering
garding environmental issues, biological production of malic acid has attracted more attention, which mainly
Carbon dioxide
consists of three pathways, namely non-oxidative pathway, oxidative pathway and glyoxylate cycle. In recent
decades, metabolic engineering of model strains, and process optimization for malic acid production have been
rapidly developed. Hence, this review comprehensively introduces an overview of malic acid producers and
highlight some of the successful metabolic engineering approaches.

1. Introduction fumaric acid using either immobilized or isolated enzymes of fumarase

could produce pure L-malic acid production, which can also be used as a
Malic acid (2-hydroxybutanedioic acid), one of C4-dicarboxylic precursor of amino acid infusions for the treatment of hyper-
acids has been wildly used in various industries, such as food, chemi- ammonemia and liver disfunction (Battat et al., 1991). However, the
cals, pharmaceutical and agriculture (Liu et al., 2015). Additionally, it sensitivity of fumarase to temperature and the substrate inhibition ef-
shows a bright future in metal cleaning, textile finishing, water treat- fect restrict the large scale production of L-malic acid (Giorno et al.,
ment, fabric dying as well as chemical synthesis of poly β-L-malic acid 2001; Kimura et al., 1986). Compared with chemical synthesis or en-
(PMA), which can be applied as biodegradable plastic (Cheng et al., zymatic conversion, considerable interests have shifted to the microbial
2017; Wei et al., 2017). The U.S. Department of Energy has classified fermentation, which could synthesize pure L-malic acid from sustain-
malic acid as one of top 12 building block chemicals, which has the able and eco-friendly feedstocks, such as lignocellulose instead of pet-
potential to replace maleic anhydride with predicted market of ex- roleum-based feedstocks. Especially, with depletion of fossil fuels and
ceeding 200,000 metric tons annually (Alonso et al., 2015; Sauer et al., concerns with regard to environmental issues, biological production of
2008; Stojkovic and Znidarsic-Plazl, 2012; Zou et al., 2015). L-malic acid attracted more attention together with other advantages,
Malic acid contains three types including D-, L- and mixture of DL- such as the metrics of mild reaction condition, high selectivity and
malic acid based on the optical isomer of a symmetric carbon atom, controllable product distribution.
which plays different roles in practical application. As the main method In this review, we specifically focus on metabolic pathways re-
to produce malic acid, chemical synthesis route mainly produces DL- sponsible for malic acid production and also highlights different malic
malic acid through the hydration reaction with petroleum-derived acid producers including wild type and metabolically engineered
maleic acid or fumaric acid as the substrate, yielding the racemate. strains. The cost of substrates has always been an important factor for
However, the harsh conditions of high temperature and pressure make large scale fermentation. So far, various feedstocks, such as starch,
it economically unrealistic due to the high operation cost and en- degrease corn powder, glycerol, stillage and syngas have been used for
vironmental issues (Goldberg et al., 2006). Enzymatic conversion of malic acid production instead of monosaccharides, like glucose (Battat

⁎
Corresponding author at: State Key Laboratory of Materials-Oriented Chemical Engineering, College of Biotechnology and Pharmaceutical Engineering, Nanjing Tech University,
Puzhu South Road 30#, Nanjing 211800, PR China.
E-mail address: [email protected] (M. Jiang).

https://doi.org/10.1016/j.biortech.2018.03.001
Received 13 January 2018; Received in revised form 27 February 2018; Accepted 1 March 2018
Available online 09 March 2018
0960-8524/ © 2018 Elsevier Ltd. All rights reserved.
Z. Dai et al. Bioresource Technology 258 (2018) 345–353

Fig. 1. Three possible pathways responsible for malic acid production, using oxaloacetic acid and/or acetyl-CoA as precursors. A: Reductive tricarboxylic acid (rTCA) pathway. B:
Tricarboxylic acid (TCA) cycle pathway. C: Glyoxylate cycle pathway.

et al., 1991; Khan et al., 2014). Hence, this review also introduces the covalently bound to the polypeptide chain of the enzyme. As a key
current status of malic acid production from different substrates. enzyme for malic acid production and accumulation, MDH could fur-
ther catalyze the reduction of oxaloacetic acid to malic acid coupled
2. Metabolic pathways and secretion systems involved in malic with the conversion of NADH to NAD+ in the cytoplasm (Brown et al.,
acid production 2013; Gharib et al., 2016). Hence, this reductive pathway requires both
biotin and NADH during biosynthesis of malic acid. As the first step to
Various microbes have been isolated, characterized and applied for convert pyruvic acid into malic acid, the expression levels of PYC
L-malic acid production, including Aspergillus spp. (Battat et al., 1991), should be strictly regulated, as pyruvic acid located at the branch point
Penicillium spp. (Khan et al., 2014), Zygosaccharomyces rouxii (Taing and could be metabolized into various end products. The cytosolic locali-
Taing, 2007), and Ustilago trichophora (Zambanini et al., 2016a). L-malic zation appears to provide assistance to these organisms in accumulating
acid can be either directly converted from pyruvic acid by one-step high concentrations of organic acids (Goldberg et al., 2006).
conversion, or several steps through reductive pathway or oxidative As a ubiquitous enzyme, ME can catalyze the reversible decarbox-
pathways, et al. ylation of malic acid to pyruvic acid and CO2 with simultaneous re-
duction of primarily NADP+ or NAD+ in one step, which can be found
2.1. Reductive tricarboxylic acid (rTCA) pathway in the cytosol or mitochondria (Ohno et al., 2008). However, the phy-
siological function of ME still remains uncertain and mysterious in most
In rTCA pathway (Fig. 1A), pyruvic acid was firstly carboxylzed to microorganisms despite decades of study. In the present study, three
oxaloacetic acid, followed by the reduction of oxaloacetic acid to malic kinds of MEs have been characterized based on their coenzyme speci-
acid in the cytosol. It should be noted that if pyruvic acid is originated ficities and abilities to decarboxylate OAA. In both prokaryotes and
from glycolysis process, this rTCA pathway is ATP neutral. Meanwhile, eukaryotes, the well-conserved MEs from cytoplasmic, mitochondrial,
1 mol CO2 is fixed with formation of 1 mol malic acid, leading to a and plastidic have been characterized together with the resolution of
maximal theoretical yield of 2 mol/mol glucose. rTCA pathway mainly the crystal structures of some typical MEs (Bologna et al., 2007).
takes place in the cytosol and requires two steps of enzymatic conver- Generally, most microorganisms possess a NADP-dependent ME, and
sion from pyruvic acid conducted by pyruvate carboxylase (PYC), some bacteria, such as E. coli, B. subtilis, Rhizobium sp., Pseudomonas sp.,
which catalyzes the ATP-dependent condensation of pyruvic acid and Alcaligenes faecalis, and Agrobacterium tumefaciens et al. possess both
CO2 to form oxaloacetic acid, and malate dehydrogenase (MDH), which (an) NAD- and an NADP-specific enzyme(s) (Hao et al., 2012; Iwakura
converts oxaloacetic acid to malic acid under aerobic conditions, re- et al., 1978). In bacteria, as portion of the phosphoenolpyruvate (PEP)-
spectively. Especially, the co-metabolism of CO2 is very crucial for the pyruvate-OAA node, MEs connect glycolysis/gluconeogenesis with the
achievement of a high malic acid yield in rTCA cycle. Therefore, the tricarboxylic acid (TCA) cycle. In E. coli, two genes possessing
cytosolic localization is generally considered to be important for ac- homology to ME: sfcA (or maeA) and maeB (or ypfF) as well as sfcA have
cumulation of high concentrations of organic acids by scavenging un- been used to metabolically construct malic acid producer (Sauer and
wanted acids in the cytosol. Owning to the high yield and relative Eikmanns, 2005; Stols and Donnelly, 1997). Although ME shows both
simplicity, rTCA cycle has been considered as the most economic one activities of pyruvate carboxylation (malate-forming) and pyruvate re-
for malic acid production (Yin et al., 2015). Hence, this pathway has duction (lactate-forming), the reaction specificity lies in obtained malic
been comprehensively introduced into the model organisms, such as acid, which was used as the primary product (Zheng et al., 2009).
Escherichia coli, Saccharomyces cerevisiae etc. to achieve malic acid
production. 2.2. Tricarboxylic acid (TCA) cycle pathway
Malic acid synthesis related enzymes are exclusively localized in the
cytosolic fraction due to the lack of a mitochondrial-targeting peptide in Tricarboxylic acid (TCA) cycle is the second metabolic pathway for
some malic acid producers, such as Penicillium spp., Aspergillus spp., malic acid production, in which oxaloacetic acid and acetyl-coenzyme
Rhizopus sp. and S. cerevisiae (Bercovitz et al., 1990; Chi et al., 2016; A (acetyl-CoA) are catalyzed to citric acid, followed by several oxida-
Pines et al., 1997). During these enzymes, PYC was firstly identified in tive reactions in the mitochondria (Fig. 1B). If acetyl-CoA is formed by
1959, whose structure and function have been clearly elaborated over pyruvate dehydrogenase and oxaloacetic acid is formed by pyruvate
the last decade (Attwood, 1995). PYC is considered as biotin-dependent carboxylase, the conversion of glucose to malic acid in TCA cycle will
and tetrameric enzyme, which catalyzes the MgATP-dependent con- result in the release of 2 CO2, which will reduce the maximum theo-
densation of pyruvic acid and CO2 to form oxaloacetic acid (Jitrapakdee retical malic acid yield to 1 mol/mol glucose. The TCA cycle also serves
et al., 2008; Kenealy et al., 1986). The biotin prosthetic group is a dual role in the catabolism and anabolism by catalyzing complete

346
Z. Dai et al. Bioresource Technology 258 (2018) 345–353

Table 1
Comparison of malic acid production by using different wild type strains.

Strains Description Substrates Concentration Productivity Yield References

(g/L) (g/L/h) (g/g)

Aspergillus flavus – 100 g/L glucose 58 0.27 0.58

Aspergillus flavus ATCC 13,697 – 120 g/L glucose 113 0.59 0.94 Battat et al. (1991)
Aspergillus oryzae NRRL 3488 – 45 g/L glucose 30.27 0.64 0.67 Knuf et al. (2013)
Aspergillus oryzae DSM 1863 – 120 g/L glucose 58.16 0.16 0.51 Ochsenreither et al. (2014)
Aspergillus oryzae DSM 1863 – 111 g/L xylose 39.40 0.11 0.44 Ochsenreither et al. (2014)
Aspergillus oryzae DSM 1863 – 109 g/L glycerol 45.43 0.13 0.54 Ochsenreither et al. (2014)
Aspergilus oryzae DSM 1863 – syngas 1.83 0.02 0.22 Oswald et al. (2016)
Aspergillus niger ATCC 9142 – thin stillage (3.4 g/L glucose; 17.1 g/L 17 0.09 0.80 West (2011)
glycerol; 15.8 g/L lactic acid)
Aspergillus niger ATCC 10,577 – thin stillage (3.4 g/L glucose; 17.1 g/L 19 0.10 – West (2011)
glycerol; 15.8 g/L lactic acid)
Penicillium oxalicum CBPS-3F-Tsa Using AlPO4 as P source potato dextrose 32.16 0.45 – Gadagi et al. (2007)
Penicillium viticola 152 Using corn steep liquor as 140 g/L glucose 131 1.36 1.00 Khan et al. (2014)
N source
Penicillium sclerotiorum K302 – 140 g/L glucose 71.67 1.00 0.69 Wang et al. (2013)
Schizophyllum commune IFO- – 50 g/L glucose 18 0.16 0.36 Kawagoe et al. (1997)
4928
Zygosaccharomyces rouxii 0.5% glutamic acid as 300 g/L glucose 74.90 0.21 0.39 Taing and Taing (2007)
precursors
Ustilago trichophora TZ1 – 250 g/L glycerol 196 0.39 0.82 Zambanini et al. (2016b)
Ustilago trichophora TZ1 – 360 g/L glycerol 195 0.74 0.43 Zambanini et al. (2016a)
Rhizopus delemar – 125 g/L corn straw hydrolyte 120 2.00 0.96 Li et al. (2014)
Monascus araneosus – 100 g/L glucose 27.90 0.23 0.37 Lumyong and Tomita
(1993)

oxidation of acetyl-CoA to CO2 for respiratory ATP formation and by due to their common substrate: isocitrate which will split into succinic
providing carbon precursor metabolites and NADPH for biosynthetic acid. Glyoxylate was further converted into malic acid through the
processes. Most aerobic microorganisms possess the pyruvate kinase overexpression of aconitase (ACN), iso-citrate lyase (ICL) and malate
and the pyruvate dehydrogenase complex that feed acetyl-CoA into the synthase (MS). MS requires Mg2+ for its activity. However, it was found
TCA cycle (Sauer and Eikmanns, 2005). that the glyoxylate cycle cannot be involved in malic acid biosynthesis,
There are two forms of MDH found in eukaryotic cells. One is cy- as the cycle will be repressed by high concentration of glucose in the
tosolic MDH catalyzing the first step in gluconeogenesis from pyruvic medium.
acid, and the other is a mitochondrial enzyme that functions in the Enzymes in the noncyclic pathway are also involved in the glyox-
oxidative TCA cycle catalyzing the NAD(H)-dependent reversible con- ylate cycle, such as PYC, citrate synthase (CS), ACN, ICL and MS.
version of malic acid to oxaloacetic acid. Similarly, fumarase, a tetra- However, there is difference that oxaloacetic acid is replenished by
meric protein belonging to the group of hydratases could catalyze the PYC, resulting in a theoretical maximum yield of 1.33 mol/mol glucose.
reversible hydration of fumaric acid to malic acid. It includes oxygen-
labile homodimeric fumarase hydratases and stable homotetrameric
enzymes, which have been characterized in E. coli, Euglena gracilis, and 2.4. Secretion system
Bacillus stearothermophilus (Shibata et al., 1985). In some prokaryote,
such as E. coli, three different fumarase genes, including FUMA, FUMB Accumulation of high concentration of malic acid in vivo would
and FUMC were described. FUMA and FUMB belong to oxygen-labile cause toxicity to microbes; hence, it is important to enhance diffusion of
homodimeric fumarase hydratases and FUMC belongs to stable homo- the synthesized malic acid across the plasma membrane into the
tetrameric enzymes. The role of fumarase in yeast for malic acid ac- medium. A series of dicarboxylate transporters, involved in the secre-
cumulation has been thoroughly studied, which may shed light on the tion of dicarboxylates from some yeasts, such as Candida sphaerica, C.
characterization of fumarase originating from other organisms. For utilis, Hansenula anomala and Kluyveromyces marxianus have been well
example, production of malic acid with negligible formation of fumaric described (Cassio and Leao, 1993; Cortereal et al., 1989; Queiros et al.,
acid were found in A. flavus, which was explained by the function of the 1998). Recently, one dicarboxylate transporter gene designated as mae1
fumarase, as fumarase catalyzes the conversion of fumaric acid to L- has been identified from Schizosaccharomyces pombe, which shows high
malic acid rather than the reverse reaction (Pines et al., 1996). activities towards malic acid transportation. It was observed that mae1
Generally, under aerobic condition, TCA cycle functions to convert of S. pombe is a proton-dicarboxylate symporter that allows uphill
oxaloacetic acid to malic acid via succinic acid. Meanwhile, the pro- transport and accumulation as a function of △pH (Grobler et al., 1995;
duced malic acid will not be further converted to succinic acid, because Sousa et al., 1992; Zelle et al., 2008). Furthermore, the Sp mae1 gene
the enzymes (anaerobic fumarase and fumarate reductase) which con- encodes a permease for malic acid and other C4 dicarboxylic acids and
vert malic acid to succinic acid only function under anaerobic condi- its protein behaves as a proton symporter, which is not subjected to
tions. glucose repression (Osothsilp and Subden, 1986; Sousa et al., 1992).
Although some researchers thought that malic, succinic and fumaric
acids in fungi were secreted to the broth from the hyphae, it is still
2.3. Glyoxylate cycle pathway
unknown if the transporter gene can function in other malic acid-pro-
ducing fungi. As mentioned above, A. oryzae shows high malic acid
Another pathway for malic acid production is glyoxylate pathway,
production. By analyzing its genomic DNA and proteins, a C4-di-
in which the maximum malic acid yield is constrained to 1 mol/mol
carboxylic acid transporter named C4T318 was found (Machida et al.,
glucose due to the carbon loss occurred in the oxidative decarboxyla-
2005). However, it is still being under investigation whether it takes
tion reaction (Fig. 1C). In the glyoxylate cycle, iso-citrate lyase (ICL)
part in L-malic acid and other C4-dicarboxylate transportation.
and TCA cycle enzyme isocitrate dehydrogenase (IDH) are competitive

347
Z. Dai et al. Bioresource Technology 258 (2018) 345–353

3. Malic acid production by using wild type microorganisms malic acid were produced by using A. niger ATCC 9142 and A. niger
ATCC 10577 respectively, suggesting the potential of thin stillage for
Originally, L-malic acid was mainly produced through the extraction the industrialized production of malic acid (Kim et al., 2010; West,
from the apple juice or egg shells, which contained less than 1% malic 2011; Kim et al., 2008). Nevertheless, biological processes using As-
acid (Lin et al., 2012; Peleg et al., 1988). With the rapid development pergillus species still have some drawbacks, which need to be addressed
and matureness of chemical synthetic and enzymic routes, the tradi- in the further studies, such as the limited oxygen supply due to fila-
tional methods were completely substituted. Recently bio-based malic mentous growth (Klement and Buchs, 2013).
acid production through microbial fermentation from renewable sub-
strates has attracted great attention (Table 1). 3.2. Penicillium spp.

3.1. Aspergillus spp. Penicillium spp. were the reported second largest malic acid produ-
cers and several species have been reported. P. oxalicum could produce
As a native L-malic acid producer, Aspergillus spp., including A. 32.16 g/L of L-malic acid from potato dextrose broth, which was iso-
flavus, A. niger and A. oryzae has been widely used for L-malic acid lated from rice rhizosphere soil (Gadagi et al., 2007). P. viticola 152
production through the rTCA pathway. One of advantages for malic screened from marine algae was able to use corn steep liquor (CSL) rich
acid production through the reductive cytosolic TCA branch is the in biotin as nitrogen source to produce malic acid (Nonaka et al., 2011;
complete balance in terms of ATP and NADH levels gained from gly- Sharma et al., 2013). Under optimized conditions, 103 g/L of malic acid
colysis to pyruvic acid and consumption in the subsequent steps to in flask and 131 g/L of malic acid in a 10 L fermentation were obtained,
malic acid. It was found that high malic acid production generally oc- demonstrating the highest malic acid titer by Penicillium spp. (Khan
curred under stress conditions, such as abundant carbon sources, lim- et al., 2014). Wang et al. further isolated P. sclerotiorum K302 from
ited nitrogen sources, together with the existence of neutralizer, such as marine environments, which could also produce high-level of malic
CaCO3 (Goldberg et al., 2006; Klement et al., 2012; Peleg et al., 1988). acid. By optimizing the production conditions, 69 g/L of malic acid
For instance, malic acid accumulation through rTCA pathway only occurred in flask fermentation. In the meantime, 71.67 g/L of malic
occurred only under nitrogen limited conditions. In addition, the sup- acid with yield of 0.69 g/g were reached within 72 h in 10 L fermentor
plementation of CaCO3 will improve the production of malic acid, (Wang et al., 2013).
which plays dual roles as both buffering agency and sources of CO2
(Goldberg et al., 2006). Actually, Ca2+ could change the plasma 3.3. Other eucaryon
membrane permeability, which will further facilitate the secretion of
malic acid (Cassio and Leao, 1993). In addition, the activity of PYC can Some other eucaryons were also found to produce malic acid in-
also be improved (Zelle et al., 2010). digenously. For example, Schizophyllum commune IFO-4928 produced
In 1960s, Abe et al. patented the malic acid production process 18 g/L of malic acid with productivity of 0.18 g/L/h from 50 g/L of
using one isolated A. flavus strain, in which the final malic acid titer of glucose with supplementation of 50 g/L of CaCO3 at 27 °C (Kawagoe
58 g/L was obtained with yield and productivity of 0.58 g/g and 0.27 g/ et al., 1997). As a sugar-tolerant yeast, Zygosaccharomyces rouxii was
L/h, respectively. Through further process optimization, 113 g/L of found to produce 74.90 g/L of malic acid with yield of 0.39 g/g from
malic acid was obtained with productivity of 0.59 g/L/h from 120 g/L 193 g/L glucose within 15 days. The optimized parameters were: 30%
of glucose by using A. flavus ATCC 13697 in 16 L fermenter for 6–8 days glucose concentration, initial pH 5.0, and 25 °C incubation temperature.
(Battat et al., 1991). However, it should be noticed that due to the co- Additionally, as the precursors of malic and succinic acid, glutamic acid
production of carcinogenic aflatoxins and special growth requirements showed the strongest effect on malic acid production. When adding
during the fermentation process, A. flavus is disqualified as the malic 0.5% glutamic acid in culture, malic acid production increased dra-
acid producer (Brandl and Andersen, 2015; Payne et al., 2006). Given matically (Taing and Taing, 2007).
the high homology between A. flavus and A. oryzae, Knuf et al. chose A. Recently, 68 Ustilaginaceae sp. has been screened by Geiser et al.
oryzae NRRL 3488 for malic acid production, which is generally re- During these, Ustilago trichophora TZ1 was found possessing the cap-
garded as safe (GRAS). Under high glucose concentration and nitrogen ability to synthesize malic acid from glucose or glycerol. Generally,
starvation conditions, 30.27 g/L of malic acid was gained with the yield glucose would be the preferred substrate for malic acid production and
of 0.67 g/g within 47.5 h. Further investigation found that peptone was higher yield from glucose (2 mol malic acid/mol glucose) than that
a more suitable nitrogen source than ammonium for malic acid pro- from glycerol (1 mol malic acid/mol glycerol)) was obtained. However,
duction (Knuf et al., 2013). Meanwhile, A. oryzae shows different malic U. trichophora TZ1 tend to utilize glycerol to produce malic acid.
acid production profiles when using different carbon source. For ex- Combined directed evolution with fermentation optimization, U. tri-
ample, when using xylose or glycerol as the substrate, A. oryzae DSM chophora was able to produce 196 g/L of malic acid with yield and
1863 only produced 39.40 or 45.43 g/L of malic acid, which is ob- productivity of 0.82 g/g and 0.39 g/L/h from glycerol, respectively in
viously lower than 58.16 g/L of malic acid produced from glucose shake flasks cultivations (Zambanini et al., 2016b). However, some
(Ochsenreither et al., 2014). Oswald et al. further found that A. oryzae problems still occurred, such as the oxygen transfer resulting from
DSM 1863 has the ability to utilize acetic acid as the sole carbon source viscous culture broth, which would reduce the overall yield and pro-
and convert acetic acid to malic acid directly. As the main metabolic ductivity. Accordingly, a fed-batch fermentation was carried out to
product of Clostridium ljungdahlii in acetogenic syngas fermentation, decrease the broth viscosity by reducing the concentration of glycerol
acetic acid can act as a bridge from syngas to malic acid. The microbial retained in the fermentation broth. Meanwhile, the boundary condi-
co-culturing system was proposed including A. oryzae DSM 1863 and C. tions of an eventual industrial process was determined by optimizing
ljungdahlii, which could produce 1.83 g/L of malic acid from syngas the initial concentrations of ammonium and glycerol with pH con-
within 96 h, demonstrating the feasibility of malic acid production from trolled at 6.5. Under the optimized conditions, the overall production
syngas through integration of anaerobic syngas fermentation and rate was increased to 0.74 g/L/h with a titer of 195 g/L and a yield of
aerobic malic acid production (Oswald et al., 2016). 0.43 g/g (Zambanini et al., 2016a).
In addition to A. flavus and A. oryzae, A. niger is another malic acid It was found that 27.90 g/L of malic acid was synthesized by an
producer, which can produce malic acid from organic wastes. For in- albino mutant isolated from Monascus araneosus from 100 g/L of glu-
stance, West investigated the ability of A. niger to produce malic acid cose after incubation under aerobic conditions for 5 days at 37 °C
when using the thin stillage as the carbon source, which contains 3.4 g/ (Lumyong and Tomita, 1993). Recently, R. delemar HF-121 was found
L of glucose and 17.1 g/L of glycerol. The optimal 17 g/L and 19 g/L of to produce more than 120 g/L of malic acid from biomass hydrolysate

348
Z. Dai et al. Bioresource Technology 258 (2018) 345–353

Table 2
Comparison of malic acid production by metabolically engineered strains.

Strains Genetic modifications Substrates Concentration Productivity Yield References

(g/L) (g/L/h) (g/g)

Aspergillus oryzae NRRL Overexpression of PYC, MDH and a C4-dicarboxylate transporter 160 g/L glucose 154 0.94 1.03 Brown et al. (2013)
3488
Aspergillus oryzae NRRL Overexpression of PYC, MDH, PFK and a C4-dicarboxylate 110 g/L glucose 165 1.38 0.68 Liu et al. (2017)
3488 transporter; Expression of EcoPPC, EcoPCK and SpMAE1
Aspergillus oryzae 2103a- Overexpression of PYC, MDH and a C4-dicarboxylate transporter 100 g/L glucose 66.30 0.86 1.11 Knuf et al. (2014)
68
Ustilago trichophora TZ1 Overexpression of PYC, MDH (cytoplasmic Mdh1 and 200 g/L glycerol 134 0.56 0.42 Zambanini et al.
mitochondrial Mdh2) and two malic acid transporters (2017)
Saccharomyces cerevisiae Overexpression of PYC2, MDH3; Expression of SpMAE1 200 g/L glucose 59 0.19 0.31 Zelle et al. (2008)
Saccharomyces cerevisiae Expression of Af PYC, RoPYC, Af MDH, RoMDH and SpMAE1 100 g/L glucose 30.25 0.32 0.30 Chen et al. (2017)
Pichia pastoris Overexpression of PYC, MDH1 100 g/L glucose 42.28 0.44 0.42 Zhang et al. (2015)
Torulopsis glabrata Expression of RoPYC, RoMDH and SpMAE1 60 g/L glucose 8.50 0.18 0.14 Chen et al. (2013)
Escherichia coli WGS-10 Expression of MsPCK; deletion of pta 20 g/L glucose 9.25 0.74 0.42 Moon et al. (2008)
Escherichia coli KJ071 deletion of ldhA, adhE, ackA, focA, pflB and mgsA 100 g/L glucose 83.88 – – Jantama et al.
(2008)
Escherichia coli XZ658 deletion of ldhA, ackA, adhE, pflB, mgsA, poxB, frdBC, sfcA, maeB, 50 g/L glucose 34 0.47 1.06 Zhang et al. (2011)
fumB, fumAC
Escherichia coli W3110 deletion of ldhA, poxB, pflB, pta, ackA, frdBC, fumABC; 60 g/L glucose 21.65 0.30 0.36 Dong et al. (2017)
Overexpression of ME and POS5
Escherichia coli Expression of T. kME glucose 0.35 0.12 1.09 Ye et al. (2013)
Escherichia coli Expression of A.fPYC, S.cMS; Overexpression of CS, CAN and ICL 65 g/L glucose 36.05 0.58 0.55 Gao et al. (2017)
Bacillus subtilis Expression of E.cPPC, S.cMDH; deletion of ldh 18 g/L glucose 2.10 0.03 0.12 Mu and Wen (2013)
Thermobifida fusca muC- Expression of C.gPYC 100 g/L cellulose 62.76 0.51 0.63 Deng et al. (2016)
16
Thermobifida fusca muC- Expression of C.gPYC 50 g/L milled corn 21.47 0.18 0.43 Deng et al. (2016)
16 stover

in pilot-scale jar within 60 h. Meanwhile, metabolic pathway analysis production of malic acid using the model organisms like Aspergillus spp.,
showed that R. delemar HF-121 can simultaneously utilize glucose and S. cerevisiae (Fig. 2A) or E. coli (Fig. 2B).
xylose in spite of the preferential use of glucose. The high malic acid
production from biomass hydrolysate highlights the application pro- 4.1. Metabolic construction of eukaryote for malic acid production
spect of industrialization when using this strain (Li et al., 2014).
4.1.1. Aspergillus spp.
4. Metabolic engineering strategies for the production of malic As described above, A. oryzae has competitive production ability of
acid malic acid, and it is also known as an industrial enzyme producer. An
additional advantage is that Aspergillus strains are able to simulta-
To improve the titer, yield and productivity of malic acid, many neously utilize sugars derived from lignocellulosic materials including
research groups are active in metabolic engineering, which exhibits a glucose and xylose (Begum and Alimon, 2011; Duarte and
great potential in improving microbial production processes (Table 2). Costaferreira, 1994; Prathumpai et al., 2003). A. oryzae NRRL 3488 has
It was also possible to establish processes with metabolically en- been investigated via overexpression of a native cytosolic alleles of
gineering, which can re-design the metabolic network for the enhanced PYC, MDH and a native C4-dicarboxylate transporter, leading to an

Fig. 2. Metabolic pathways of L-malic acid. Metabolites present in both compartments are differentiated by the indicators cytosol and mitochondrion. A: Metabolic pathways in
eukaryocyte; B: Metabolic pathways in prokaryote.

349
Z. Dai et al. Bioresource Technology 258 (2018) 345–353

improved production of malic acid. The strain overexpressing all three 4.1.3. Saccharomyces cerevisiae
genes produced 154 g/L of malic acid with productivity and yield of Different from the filamentous fungi, S. cerevisiae shows a series of
0.94 g/L/h and 1.03 g/g (69% of the theoretical maximum) in 164 h advantages for malic acid production, such as (1) robust tolerance to
(Brown et al., 2013). As noted, a small amount of citric acid is also high substrate concentration, (2) less sensitive to metal ions, (3) uni-
produced, whose metabolic route does not exist in these genetically cellular fungi allowing for better process control, (4) capability of uti-
modified strains. Although the exact metabolic route has not been lizing a wider range of carbon sources, (5) safety in food industry. S.
known clearly, it has been reported that malic acid accumulation may cerevisiae is one of the safest and accepted eukaryotic microorganisms,
trigger citric acid excretion in A. niger through a process mediated by which could be potential malic acid producer through genetic mod-
mitochondrial tricarboxylate transporters (Karaffa and Kubicek, 2003). ification. Indeed, rTCA cycle has been enhanced in S. cerevisiae through
Basis on this hypothesis, Liu et al. recently constructed the ox- overexpression of the native pyc, an allele of the peroxisomal mdh3,
aloacetic acid anaplerotic reaction in order to improve the synthesis of which had been retargeted to the cytosol by deletion of the C-terminal
malic acid by stimulating TCA cycle. In addition, phosphoenolpyruvate targeting sequence, and malic acid transporter gene from S. pombe. The
(PEP) can be converted to OAA by either phosphoenolpyruvate car- recombinant S. cerevisiae produced 59 g/L of malic acid with yield and
boxykinase (pck) or phosphoenolpyruvate carboxylase (ppc), which is a productivity of 0.31 g/g and 0.19 g/L/h, respectively. Further scaling
limiting step for malic acid synthesis with CO2 fixation. Then wild type up and optimization in bioreactor increased malic acid yield to 0.36 g/
malic acid producer A. oryzae was modified by heterologous expression g, 19% higher than that in the flask. This was explained by that mod-
of pck and ppc from E. coli together with a C4-dicarboxylate transporter erate oxygen limitation benefitted malic acid production (Zelle et al.,
gene from S. pombe (Tan et al., 2013; Zhu et al., 2015). Moreover, the 6- 2010; Zelle et al., 2008). Recently, universal strategies integrated with
phosphofructokinase encoded by the pfk gene, which can adjust gly- pathway optimization and transporter modification were applied for
colysis pathway was identified as a potential limiting step for malic acid reconstructing or intensifying malic acid biosynthesis pathway and
synthesis, which was further overexpressed with the strong sodM pro- transportation system in S. cerevisiae. It was found that that key en-
moter. The final modified A. oryzae strain gave an malic acid con- zymes originating from different microbes will greatly affect malic acid
centration of 165 g/L with yield and productivity of 0.68 g/g and production. For example, through overexpression of PYC from A. flavus
1.38 g/L/h in 3 L bioreactors using a fed-batch fermentation mode, and MDH from R. oryzae, the recombinant could produce the highest
which represents the highest malic acid production (Liu et al., 2017). malic acid production. Then the malic acid efflux system was enhanced
Except for A. oryzae NRRL 3488, another wild strain A. oryzae through modification of the malic acid transporter from S. pombe by
2103a-68, which derived from NRRL3488 was also modified to produce removing the ubiquitination. Thus, the malic acid pathway was well
malic acid. By overexpressing PYC, MDH, and a malic acid transporter, modularized and optimized by regulating the distribution of metabolic
the engineered strain finally produced 66.30 g/L of malic acid at a rate flow. The recombinant could finally produce 30.25 g/L of malic acid
of 0.86 g/L/h, with a yield of 1.11 g/g. Besides malic acid, a major by- with yield and productivity of 0.30 g/g and 0.32 g/L/h, respectively
product of succinic acid was produced to final titers of 9.03 g/L from (Chen et al., 2017).
glucose. This suggests a direct correlation between the increase of the
cytosolic malic acid pool and the secretion of succinic acid, although no 4.1.4. Pichia pastoris
reaction can be found that allows for a continuation of the rTCA to- Compared to S. cerevisiae, P. pastoris, a methylotrophic yeast pos-
wards succinic acid in the cytosol (Vongsangnak et al., 2008). The sessed better property for commercial bio-transformations due to its
production of succinic acid indicates that the pool of intracellular malic efficient heterologous gene expression, physical robustness and well-
acid increased and the flux continues further down to succinic acid established fermentation methods (Cregg et al., 1993). Additionally, the
(Knuf et al., 2014; Regev-Rudzki et al., 2009). When using xylose as the dissimilatory pathway of P. pastoris in methanol utilization can produce
sole carbon source or glucose/xylose mixtures as the carbon source, two molecules of NADH, which provides the redox equivalent of NADH
12 g/L or 45 g/L of malic acid was produced, respectively. Despite the for malic acid production. Zhang et al. expressed three expression
molar yield of malic acid on xylose fell below the yield of the wild type cassettes carrying the homologous pyc, the cytoplasmic mdh and the
on glucose, it paved the way for establishing a biorefinery-based pro- retargeted fum. The cassettes were integrated into the chromosome of P.
cess for the production of this important dicarboxylic acid. pastoris GS115 to obtain the high and stable expression. In the batch
However, it should be noticed that some disadvantages limit their fermentation, a final concentration of 42.28 g/L malic acid was pro-
use, for example, no native replicating extrachromosomal vectors have duced within 96 h, and the malic acid accumulation within 24 h was
been found in contrast to E. coli and S. cerevisiae. Even if there are ar- negligible until it grew into the stationary phase at 48 h (Zhang et al.,
tificial vectors which have been developed, its utility has been limited 2015).
by the instability (Aleksenko and Clutterbuck, 1997; Aleksenko et al.,
1996). 4.1.5. Torulopsis glabrata
Except for S. cerevisiae and P. pastoris, T. glabrata CCTCC M202019,
4.1.2. Ustilago trichophora a kind of multi-vitamin auxotroph yeast for pyruvic acid production,
U. trichophora TZ1 is a good natural malic acid producer, which has was also engineered to produce malic acid by manipulating carbon flux
showed the potential for the production of malic acid; however, the from pyruvic acid to malic acid. After identifying the bottleneck in
yield of 0.23 g/g was relatively too low, only one-third of the theore- malic acid production using model iNX804, genes Ro pyc, Ro mdh and
tical maximum. Recently, after overexpression of PYC, two MDHs Sp mae1 were simultaneously introduced, leading to the final malic acid
(mitochondrial and cytoplasmic) and two malic acid transporters si- concentration of 8.5 g/L within 60 h under the optimal vitamin con-
multaneously, high yield of 0.42 g/g and productivity of 0.56 g/L/h centrations. However, the yield of malic acid on glucose in this study is
were gained. Moreover, malic acid production still maintained at high only 0.14 g/g, which is much lower than theoretical one via reduction
levels of 134 g/L (Zambanini et al., 2017). of oxaloacetic acid (1.49 g/g) or the actual yield achieved with S. cer-
These metabolic engineering strains clearly showed that rTCA plays evisiae (0.30 g/g) (Chen et al., 2013). However, the pyruvic acid flux
the most important role among the four identified pathways for malic does not translate into malic acid efficiently owing to the presence of a
acid production. It is also believed that such engineered and modified kinetic bottleneck within the malic acid biosynthesis pathway (Xu et al.,
stains have the potential for malic acid production application. 2013).
However, the unclear genome sequence and the immatureness of ge- Up to now, many researchers have developed various methods for
netic modification tools present obstacles in engineering its metabolic breeding high-malic acid-producing yeasts, but genes responsible for
pathway for optimal malic acid production. the high-acidity phenotype are unknown. To address this problem,

350
Z. Dai et al. Bioresource Technology 258 (2018) 345–353

Negora et al. identified VID24 as the gene responsible for the high pathways for malic acid production have been identified, but only few
acidity phenotype in industrial sake yeast and provided a useful in- studies have focused on the glyoxylate pathway. Gao et al. reported the
dication to control the ability of malic acid production in yeasts multiplexed tuning of a five-enzyme cascade reaction in the glyoxylate
(Negoro et al., 2016). pathway, leading to the conversion of malic acid from pyruvic acid by
integrating in vitro modular engineering with in vivo CRISPRi tech-
4.2. Metabolic construction of prokaryote nology. The metabolic burden caused by heterologous expression of
multiple genes in the operon context could be minimized using CRISPRi
4.2.1. Escherichia coli technology (Kim et al., 2016). The introduction of CRISPRi could solve
The metabolic pathways of E. coli is remarkably adaptable and a the problems of unbalance between different modules, such as the ac-
variety of genetic approaches have been used to engineer E. coli to re- cumulation of citrate and α-ketoglutarate. A final malic acid titer of
design the metabolic network for enhancement of malic acid produc- 36.05 g/L with a yield of 0.55 g/g glucose, was achieved at 50 h post-
tion. Various strains have been genetically modified through deleting or induction performance in 3.6 L fed-batch fermentation culture (Gao
weakening competing pathways and expressing or overexpressing key et al., 2017).
enzymes. For instance, by predicting the optimal malic acid production
pathway using metabolic flux analysis (MFA), the gene of pck from 4.2.2. Bacillus subtilis
Mannheimia succiniciproducens MBEL55E, which is responsible for con- B. subtilis, a well characterized microorganism, which has also been
version of PEP to oxaloacetic acid was introduced into the pta gene applied in biological industries due to its mature biosafety, genetic tools
mutant E. coli. The final strain E. coli W3110 could produce up to and fermentative technology. The impact of fumarase on both bacterial
9.25 g/L of malic acid with the yield of 0.42 g/g within 12.5 h. Even growth and malic acid production has been investigated. An efficient
though no by-product was detected, another two ldhA and adhE genes heterologous biosynthesis pathway composed of ppc from E. coli and
were inactivated, attempting to increase the malic acid accumulation; mdh from S. cerevisiae was successfully introduced into B. subtilis. The
however results showed a slightly lower productivity of 0.69 g/L/h recombinant could produce 0.81 g/L of malic acid. Furthermore, the
inversely (Moon et al., 2008). engineered B. subtilis produced up to 2.01 g/L malic acid through knock
After deletion of six central fermentation genes, including ldhA, out of ldh gene under two-stage fermentation conditions. The pro-
adhE, ackA, focA, pflB, mgsA, strain KJ071 can synthesize 83.88 g/L of ductivity and yield were 0.03 g/L/h and 0.11 g/g, respectively. Though
malic acid in simple batch fermentations with mineral salts medium the final production of malic acid was relatively low, this represents the
containing 10% glucose (Jantama et al., 2008). To further increase the first successfully engineered B. subtilis for malic acid production, which
production of malic acid, multiple genes were modified. Deletion of paves the way for further improvement of malic acid production in B.
fumarate reductase in KJ060 did not cause accumulation of fumaric subtilis (Mu and Wen, 2013).
acid, but eliminated over 90% of the succinic acid production and
promoted the accumulation of malic acid along with an increase in 4.2.3. Thermobifida fusca
pyruvic acid and a decline in acetic acid. Then scfA and maeB were T. fusca is an aerobic, moderately thermophilic, filamentous soil
deleted in order to eliminate pyruvic acid production followed by a bacterium and a great platform for consolidate bioprocessing, because
series of deletion of genes mgsA, ldhA, pkyA and pykF. The optimal it can convert untreated biomass directly with high efficiency.
strain XZ658 produced malic acid as the major fermentation product Previously, a mutant strain of T. fusca: muC obtained by adaptive
with a yield of 1.06 g/g using a two-stage batch fermentation process evolution was found to accumulate a decent amount of malic acid from
(aerobic growth followed by anaerobic fermentation). The final con- sugars (Deng and Fong, 2011). In this study, T. fusca muC was en-
centration and productivity were 34 g/L and 0.47 g/L/h, respectively gineered by expressing an exogenous pyruvate carboxylase gene to
(Zhang et al., 2011). produce the high yield of malic acid on cellulose and lignocellulosic
To meet the increasing demand for a simpler and more efficient biomass. The final strain showed the highest malic acid production of
fermentation mode, E. coli W3110 was further genetically modified and 62.76 g/L in batch fermentation culture using 100 g/L of cellulose at
optimized for malic acid production through one-step synthesis 124 h. When using 50 g/L of minimal treated milled corn stover
pathway. First, ldhA, poxB, pflB, pta and ackA were deleted for the (18.45 g/L of glucan and 9.65 g/L of xylan) as the carbon source,
purpose of accumulating plenty of pyruvic acid. Then, overexpression 21.47 g/L of malic acid was produced after cultivation for 5 days. The
of NADP+-dependent malic enzyme and NADH kinase (pos5) together flux analysis discovered that malic acid in the muC strain was produced
with deletion of the genes responsible for succinic acid biosynthesis from oxaloacetic acid (non-oxidative pathway) and TCA cycle in the
resulted in the conversion to malic acid from pyruvic acid. The final low oxygen tension condition (oxidative pathway) (Deng et al., 2016).
fermentation was implemented in a 5 L bioreactor with the concentra-
tion, yield and productivity of 21.65 g/L, 0.36 g/g and 0.30 g/L/h, re- 5. Future perspectives
spectively. As by-products, 16.54 g/L of pyruvic acid, 0.98 g/L of suc-
cinic acid, and 0.19 g/L of fumaric acid were detected (Dong et al., Significant improvements for malic acid production have been
2017). Meanwhile, the importance of NADPH was mentioned, because achieved through microbial fermentation using wild type or genetically
the biosynthesis of 1 mol of L-malic acid requires 1 mol of NADPH. engineered microbes. However, the low yield caused by the formation
NADPH is known to play an important role in the biosynthesis of sev- of by-products, such as fumaric acid, citric acid and the loss of CO2
eral high-value chemicals, and NADPH availability is essential for in- lowers the economics of large scale production. Further breakthroughs
creasing the yield of target product (Chemler et al., 2010; Ng et al., of L-malic acid yield should be brought to the research in further stu-
2015). Although malic acid could be accumulated in aerated fermen- dies. The driving force was primarily scientific curiosity to learn more
tation, the highest malic acid production was still achieved under how they control and keep the imbalance of biosynthesis system, which
anaerobic situation (Battat et al., 1991; Zelle et al., 2008). Through might restrain the enzymes function and cause a suboptimal product.
anaerobic fermentation, NADH/NADPH could be used for malic acid Taken together, the challenge in metabolic engineering for malic acid
production instead of the respiration chain to produce ATP (Song and production involves three aspects: (1) identifying more efficient syn-
Lee, 2006). thetic pathways, (2) enhancing intermediate metabolites transmission
Introduction of malic enzyme derived from Thermococcus kodakar- efficiency in metabolic or synthetic pathways, (3) reducing carbon flux
ensis into E. coli was carried out. The modified strain could direct glu- loss by repressing by-products production. Hence, modular pathway
cose to malic acid, resulting in the accumulation of 0.35 g/L from engineering for the optimization of metabolic modules might be future
0.32 g/L glucose within 3 h (Ye et al., 2013). Four major metabolic solution to realize the balance among L-malic acid, metabolic

351
Z. Dai et al. Bioresource Technology 258 (2018) 345–353

intermediate, and biomass. Chen, X.L., Wang, Y.C., Dong, X.X., Hu, G.P., Liu, L.M., 2017. Engineering rTCA pathway
There is an expectation that malic acid fermentation will not only and C4-dicarboxylate transporter for L-malic acid production. Appl. Microbiol. Biot.
101 (10), 4041–4052.
stay in fundamental researches, but also have a high industrial im- Cheng, C., Zhou, Y.P., Lin, M., Wei, P.L., Yang, S.T., 2017. Polymalic acid fermentation by
plementation potential. To obtain a simple and efficient system in terms Aureobasidium pullulans for malic acid production from soybean hull and soy mo-
of yield, concentration and energy consumption, it is necessary to lasses: fermentation kinetics and economic analysis. Bioresour. Technol. 223,
166–174.
minimize the cost of production together with the separation in order to Chi, Z., Wang, Z.P., Wang, G.Y., Khan, I., Chi, Z.M., 2016. Microbial biosynthesis and
enhance the competitiveness with enzymic and petrochemical process. secretion of L-malic acid and its applications. Crit. Rev. Biotechnol. 36 (1), 99–107.
However, it is worth mentioning that a high portion of more than 50% Cortereal, M., Leao, C., Vanuden, N., 1989. Transport of L(-)malic acid and other di-
carboxylic-acids in the yeast Candida sphaerica. Appl. Microbiol. Biotechnol. 31 (5–6),
in the total costs will be used in the separation and purification of 551–555.
biologically produced malic acid from fermentation broth in the mi- Cregg, J.M., Vedvick, T.S., Raschke, W.C., 1993. Recent advances in the expression of
crobial production. Various methods for separating malic acid have foreign genes in Pichia pastoris. Bio-Technology 11 (8), 905–910.
Deng, Y., Fong, S.S., 2011. Laboratory evolution and multi-platform genome re-sequen-
been studied in the past decades. This part will be discussed in next
cing of the cellulolytic actinobacterium Thermobifida fusca. J. Biol. Chem. 286 (46),
review about the technology development and technological challenge 39958–39966.
in the recovery and purification of malic acid. Deng, Y., Mao, Y., Zhang, X.J., 2016. Metabolic engineering of a laboratory-evolved
Thermobifida fusca muC strain for malic acid production on cellulose and minimal
treated lignocellulosic biomass. Biotechnol. Progr. 32 (1), 14–20.
6. Conclusion Dong, X.X., Chen, X.L., Qian, Y.Y., Wang, Y.C., Wang, L., Qiao, W.H., Liu, L.M., 2017.
Metabolic engineering of Escherichia coli W3110 to produce L-malate. Biotechnol.
Generally, biological malic acid production only occurred under the Bioeng. 114 (3), 656–664.
Duarte, J.C., Costaferreira, M., 1994. Aspergilli and lignocellulosics – enzymology and
conditions of high carbon-to-nitrogen ratio, which is usually limited by biotechnological applications. FEMS Microbiol. Rev. 13 (2–3), 377–386.
its low yield, titer, and productivity due to the end-product inhibition. Gadagi, R.S., Shin, W.S., Sa, T.M., 2007. Malic acid mediated aluminum phosphate so-
In addition, the formation of by-products, like succinic and acetic acid lubilization by Penicillium oxalicum CBPS-3F-Tsa isolated from Korean paddy rhizo-
sphere soil. First Int. Meeting Microb. Phos. Sol. 102, 285–290.
renders the downstream process of separation and purification. Hence, Gao, C., Wang, S., Hu, G., Guo, L., Chen, X., Xu, P., Liu, L., 2017. Engineering Escherichia
further work is still needed to develop more robust strains and design coli for malate production by integrating modular pathway characterization with
rational processes to achieve high malic acid production. CRISPRi-guided multiplexed metabolic tuning. Biotechnol. Bioeng. http://dx.doi.
org/10.1002/bit.26486.
Gharib, G., Rashid, N., Bashir, Q., Gardner, Q.T.A.A., Akhtar, M., Imanaka, T., 2016.
Acknowledgements Pcal_1699, An extremely thermostable malate dehydrogenase from hyperthermo-
philic archaeon Pyrobaculum calidifontis. Extremophiles 20 (1), 57–67.
Giorno, L., Drioli, E., Carvoli, G., Cassano, A., Donato, L., 2001. Study of an enzyme
This work was supported by the Jiangsu Province Natural Science
membrane reactor with immobilized fumarase for production of L-malic acid.
Foundation for Youths (No. BK20170993), the Jiangsu Synergetic Biotechnol. Bioeng. 72 (1), 77–84.
Innovation Center for Advanced Bio-Manufacture, the Key Science and Goldberg, I., Rokem, J.S., Pines, O., 2006. Organic acids: old metabolites, new themes. J.
Technology Project of Jiangsu Province (BE2016389), the National Chem. Technol. Biotechnol. 81 (10), 1601–1611.
Grobler, J., Bauer, F., Subden, R.E., VanVuuren, H.J.J., 1995. The mae1 gene of
Natural Science Foundation of China (No. 21706125, No. 21727818, Schizosaccharomyces pombe encodes a permease for malate and other C-4 di-
No. 21706124, No. 31700092), the Project of State Key Laboratory of carboxylic acids. Yeast 11 (15), 1485–1491.
Materials-Oriented Chemical Engineering (KL16-08), and Top-notch Hao, Z., Liu, S.K., Yang, C.P., 2012. Expression, purification and characterization of a
NAD(+)-malic enzyme from Oryza sativa L. in Escherichia coli. Res. J. Biotechnol. 7
Academic Programs Project of Jiangsu Higher Education Institutions (1), 5–10.
PPZY2015B155, TAPP. Iwakura, M., Tokushige, M., Katsuki, H., Muramatsu, S., 1978. Studies on regulatory
functions of malic enzymes comparative studies of malic enzymes in Bacteria. J.
Biochem-Tokyo. 83 (5), 1387–1394.
References Jantama, K., Haupt, M.J., Svoronos, S.A., Zhang, X.L., Moore, J.C., Shanmugam, K.T.,
Ingram, L.O., 2008. Combining metabolic engineering and metabolic evolution to
Aleksenko, A., Clutterbuck, A.J., 1997. Autonomous plasmid replication in Aspergillus develop nonrecombinant strains of Escherichia coli C that produce succinate and
nidulans: AMA1 and MATE elements. Fungal Genet. Biol. 21 (3), 373–387. malate. Biotechnol. Bioeng. 99 (5), 1140–1153.
Aleksenko, A., Gems, D., Clutterbuck, J., 1996. Multiple copies of MATE elements support Jitrapakdee, S., St Maurice, M., Rayment, I., Cleland, W.W., Wallace, J.C., Attwood, P.V.,
autonomous plasmid replication in Aspergillus nidulans. Mol. Microbiol. 20 (2), 2008. Structure, mechanism and regulation of pyruvate carboxylase. Biochem. J.
427–434. 413, 369–387.
Alonso, S., Rendueles, M., Diaz, M., 2015. Microbial production of specialty organic acids Karaffa, L., Kubicek, C.P., 2003. Aspergillus niger citric acid accumulation: do we under-
from renewable and waste materials. Crit. Rev. Biotechnol. 35 (4), 497–513. stand this well working black box? Appl. Microbiol. Biotechnol. 61 (3), 189–196.
Attwood, P.V., 1995. The structure and the mechanism of action of pyruvate-carboxylase. Kawagoe, M., Hyakumura, K., Suye, S.I., Miki, K., Naoe, K., 1997. Application of bubble
J. Biochem. Cell B 27 (3), 231–249. column fermenters to submerged culture of Schizophyllum commune for production
Battat, E., Peleg, Y., Bercovitz, A., Rokem, J.S., Goldberg, I., 1991. Optimization of L- of L-malic acid. J. Ferment. Bioeng. 84 (4), 333–336.
malic acid production by Aspergillus flavus in a stirred fermenter. Biotechnol. Bioeng. Kenealy, W., Zaady, E., Dupreez, J.C., Stieglitz, B., Goldberg, I., 1986. Biochemical as-
37 (11), 1108–1116. pects of fumaric-acid accumulation by Rhizopus arrhizus. Appl. Environ. Microb. 52
Begum, M.F., Alimon, A.R., 2011. Bioconversion and saccharification of some lig- (1), 128–133.
nocellulosic wastes by Aspergillus oryzae ITCC-4857.01 for fermentable sugar pro- Khan, I., Nazir, K., Wang, Z.P., Liu, G.L., Chi, Z.M., 2014. Calcium malate overproduction
duction. Electron. J. Biotechnol. 14 (5) 3–3. by Penicillium viticola 152 using the medium containing corn steep liquor. Appl.
Bercovitz, A., Peleg, Y., Battat, E., Rokem, J.S., Goldberg, I., 1990. Localization of pyr- Microbiol. Biotechnol. 98 (4), 1539–1546.
uvate-carboxylase in organic acid producing Aspergillus strains. Appl. Environ. Kim, S.K., Han, G.H., Seong, W., Kim, H., Kim, S.W., Lee, D.H., Lee, S.G., 2016. CRISPR
Microb. 56 (6), 1594–1597. interference-guided balancing of a biosynthetic mevalonate pathway increases ter-
Bologna, F.P., Andreo, C.S., Drincovich, M.F., 2007. Escherichia coli malic enzymes: Two penoid production. Metab. Eng. 38, 228–240.
isoforms with substantial differences in kinetic properties, metabolic regulation, and Kim, Y., Mosier, N.S., Hendrickson, R., Ezeji, T., Blaschek, H., Dien, B., Cotta, M., Dale, B.,
structure. J. Bacteriol. 189 (16), 5937–5946. Ladisch, M.R., 2008. Composition of corn dry-grind ethanol by-products: DDGS, wet
Brandl, J., Andersen, M.R., 2015. Current state of genome-scale modeling in filamentous cake, and thin stillage. Bioresour. Technol. 99 (12), 5165–5176.
fungi. Biotechnol. Lett. 37 (6), 1131–1139. Kim, Y., Hendrickson, R., Mosier, N.S., Ladisch, M.R., Bals, B., Balan, V., Dale, B.E., Dien,
Brown, S.H., Bashkirova, L., Berka, R., Chandler, T., Doty, T., McCall, K., McCulloch, M., B.S., Cotta, M.A., 2010. Effect of compositional variability of distillers' grains on
McFarland, S., Thompson, S., Yaver, D., Berry, A., 2013. Metabolic engineering of cellulosic ethanol production. Bioresour. Technol. 101 (14), 5385–5393.
Aspergillus oryzae NRRL 3488 for increased production of L-malic acid. Appl. Kimura, T., Kawabata, Y., Sato, E., 1986. Enzymatic production of L-malate from maleate
Microbiol. Biot. 97 (20), 8903–8912. by Alcaligenes Sp. Agric. Biol. Chem. 50 (1), 89–94.
Cassio, F., Leao, C., 1993. A comparative study on the transport of L(-)malic acid and Klement, T., Buchs, J., 2013. Itaconic acid – a biotechnological process in change.
other short-chain carboxylic-acids in the yeast Candida Utilis evidence for a general Bioresour. Technol. 135, 422–431.
organic-acid permease. Yeast 9 (7), 743–752. Klement, T., Milker, S., Jager, G., Grande, P.M., de Maria, P.D., Buchs, J., 2012. Biomass
Chemler, J.A., Fowler, Z.L., McHugh, K.P., Koffas, M.A., 2010. Improving NADPH pretreatment affects Ustilago maydis in producing itaconic acid. Microb. Cell
availability for natural product biosynthesis in Escherichia coli by metabolic en- Fact. 11.
gineering. Metab. Eng. 12 (2), 96–104. Knuf, C., Nookaew, I., Brown, S.H., McCulloch, M., Berry, A., Nielsen, J., 2013.
Chen, X.L., Xu, G.Q., Xu, N., Zou, W., Zhu, P., Liu, L.M., Chen, J., 2013. Metabolic en- Investigation of malic acid production in Aspergillus oryzae under nitrogen starvation
gineering of Torulopsis glabrata for malate production. Metab. Eng. 19, 10–16. conditions. Appl. Environ. Microb. 79 (19), 6050–6058.

352
Z. Dai et al. Bioresource Technology 258 (2018) 345–353

Knuf, C., Nookaew, I., Remmers, I., Khoomrung, S., Brown, S., Berry, A., Nielsen, J., 2014. Sauer, U., Eikmanns, B.J., 2005. The PEP-pyruvate-oxaloacetate node as the switch point
Physiological characterization of the high malic acid-producing Aspergillus oryzae for carbon flux distribution in bacteria. FEMS Microbiol. Rev. 29 (4), 765–794.
strain 2103a–68. Appl. Microbiol. Biotechnol. 98 (8), 3517–3527. Sauer, M., Porro, D., Mattanovich, D., Branduardi, P., 2008. Microbial production of
Li, X.J., Liu, Y., Yang, Y., Zhang, H., Wang, H.L., Wu, Y., Zhang, M., Sun, T., Cheng, J.S., organic acids: expanding the markets. Trends Biotechnol. 26 (2), 100–108.
Wu, X.F., Pan, L.J., Jiang, S.T., Wu, H.W., 2014. High levels of malic acid production Sharma, N., Prasad, G.S., Choudhury, A.R., 2013. Utilization of corn steep liquor for
by the bioconversion of corn straw hydrolyte using an isolated Rhizopus delemar biosynthesis of pullulan, an important exopolysaccharide. Carbohyd. Polym. 93 (1),
strain. Biotechnol. Bioprocess. E 19 (3), 478–492. 95–101.
Lin, S.Y., Wang, L.Y., Jones, G., Trang, H., Yin, Y.G., Liu, J.B., 2012. Optimized extraction Shibata, H., Gardiner, W.E., Schwartzbach, S.D., 1985. Purification, characterization, and
of calcium malate from eggshell treated by PEF and an absorption assessment in vitro. immunological properties of fumarase from euglena-gracilis var Bacillaris. J.
Int. J. Biol. Macromol. 50 (5), 1327–1333. Bacteriol. 164 (2), 762–768.
Liu, J.J., Xie, Z.P., Shin, H.D., Li, J.H., Du, G.C., Chen, J., Liu, L., 2017. Rewiring the Song, H., Lee, S.Y., 2006. Production of succinic acid by bacterial fermentation. Enzyme
reductive tricarboxylic acid pathway and L-malate transport pathway of Aspergillus Microb. Technol. 39 (3), 352–361.
oryzae for overproduction of L-malate. J. Biotechnol. 253, 1–9. Sousa, M.J., Mota, M., Leao, C., 1992. Transport of malic-acid in the yeast
Liu, G.L., Zhou, Y., Luo, H.P., Cheng, X., Zhang, R.D., Teng, W.K., 2015. A comparative Schizosaccharomyces pombe evidence for a proton-dicarboxylate symport. Yeast 8
evaluation of different types of microbial electrolysis desalination cells for malic acid (12), 1025–1031.
production. Bioresour. Technol. 198, 87–93. Stojkovic, G., Znidarsic-Plazl, P., 2012. Continuous synthesis of L-malic acid using whole-
Lumyong, S., Tomita, F., 1993. L-malic acid production by an albino strain Of Monascus cell microreactor. Process Biochem. 47 (7), 1102–1107.
araneosus. World J. Microb. Biot. 9 (3), 383–384. Stols, L., Donnelly, M.I., 1997. Production of succinic acid through overexpression of NAD
Machida, M., Asai, K., Sano, M., Tanaka, T., Kumagai, T., Terai, G., Kusumoto, K.I., (+)-dependent malic enzyme in an Escherichia coli mutant. Appl. Environ.
Arima, T., Akita, O., Kashiwagi, Y., Abe, K., Gomi, K., Horiuchi, H., Kitamoto, K., Microbiol. 63 (7), 2695–2701.
Kobayashi, T., Takeuchi, M., Denning, D.W., Galagan, J.E., Nierman, W.C., Yu, J.J., Taing, O., Taing, K., 2007. Production of malic and succinic acids by sugar-tolerant yeast
Archer, D.B., Bennett, J.W., Bhatnagar, D., Cleveland, T.E., Fedorova, N.D., Gotoh, Zygosaccharomyces rouxii. Eur. Food Res. Technol. 224 (3), 343–347.
O., Horikawa, H., Hosoyama, A., Ichinomiya, M., Igarashi, R., Iwashita, K., Juvvadi, Tan, Z.G., Zhu, X.N., Chen, J., Li, Q.Y., Zhang, X.L., 2013. Activating phosphoenolpyr-
P.R., Kato, M., Kato, Y., Kin, T., Kokubun, A., Maeda, H., Maeyama, N., Maruyama, J., uvate carboxylase and phosphoenolpyruvate carboxykinase in combination for im-
Nagasaki, H., Nakajima, T., Oda, K., Okada, K., Paulsen, I., Sakamoto, K., Sawano, T., provement of succinate production. Appl. Environ. Microb. 79 (16), 4838–4844.
Takahashi, M., Takase, K., Terabayashi, Y., Wortman, J.R., Yamada, O., Yamagata, Vongsangnak, W., Olsen, P., Hansen, K., Krogsgaard, S., Nielsen, J., 2008. Improved
Y., Anazawa, H., Hata, Y., Koide, Y., Komori, T., Koyama, Y., Minetoki, T., Suharnan, annotation through genome-scale metabolic modeling of Aspergillus oryzae. BMC
S., Tanaka, A., Isono, K., Kuhara, S., Ogasawara, N., Kikuchi, H., 2005. Genome se- Genom. 9.
quencing and analysis of Aspergillus oryzae. Nature 438 (7071), 1157–1161. Wang, Z.P., Wang, G.Y., Khan, I., Chi, Z.M., 2013. High-level production of calcium
Moon, S.Y., Hong, S.H., Kim, T.Y., Lee, S.Y., 2008. Metabolic engineering of Escherichia malate from glucose by Penicillium sclerotiorum K302. Bioresour. Technol. 143,
coli for the production of malic acid. Biochem. Eng. J. 40 (2), 312–320. 674–677.
Mu, L., Wen, J.P., 2013. Engineered Bacillus subtilis 168 produces L-malate by hetero- Wei, P.L., Cheng, C., Lin, M., Zhou, Y.P., Yang, S.T., 2017. Production of poly(malic acid)
logous biosynthesis pathway construction and lactate dehydrogenase deletion. World from sugarcane juice in fermentation by Aureobasidium pullulans: kinetics and process
J. Microb. Biotechnol. 29 (1), 33–41. economics. Bioresour. Technol. 224, 581–589.
Negoro, H., Kotaka, A., Matsumura, K., Tsutsumi, H., Hata, Y., 2016. Enhancement of West, T.P., 2011. Malic acid production from thin stillage by Aspergillus species.
malate-production and increase in sensitivity to dimethyl succinate by mutation of Biotechnol. Lett. 33 (12), 2463–2467.
the VID24 gene in Saccharomyces cerevisiae. J. Biosci. Bioeng. 121 (6), 665–671. Xu, N., Liu, L.M., Zou, W., Liu, J., Hua, Q., Chen, J., 2013. Reconstruction and analysis of
Ng, C.Y., Farasat, I., Maranas, C.D., Salis, H.M., 2015. Rational design of a synthetic the genome-scale metabolic network of Candida glabrata. Mol. Biosyst. 9 (2),
Entner-Doudoroff pathway for improved and controllable NADPH regeneration. 205–216.
Metab. Eng. 29, 86–96. Ye, X.T., Honda, K., Morimoto, Y., Okano, K., Ohtake, H., 2013. Direct conversion of
Nonaka, K., Masuma, R., Iwatsuki, M., Shiomi, K., Otoguro, K., Omura, S., 2011. glucose to malate by synthetic metabolic engineering. J. Biotechnol. 164 (1), 34–40.
Penicillium viticola, a new species isolated from a grape in Japan. Mycoscience 52 Yin, X., Li, J.H., Shin, H.D., Du, G.C., Liu, L., Chen, J., 2015. Metabolic engineering in the
(5), 338–343. biotechnological production of organic acids in the tricarboxylic acid cycle of mi-
Ochsenreither, K., Fischer, C., Neumann, A., Syldatk, C., 2014. Process characterization croorganisms: advances and prospects. Biotechnol. Adv. 33 (6), 830–841.
and influence of alternative carbon sources and carbon-to-nitrogen ratio on organic Zambanini, T., Kleineberg, W., Sarikaya, E., Buescher, J.M., Meurer, G., Wierckx, N.,
acid production by Aspergillus oryzae DSM1863. Appl. Microbiol. Biotechnol. 98 Blank, L.M., 2016a. Enhanced malic acid production from glycerol with high-cell
(12), 5449–5460. density Ustilago trichophora TZ1 cultivations. Biotechnol. Biofuels 9.
Ohno, Y., Nakamori, T., Zheng, H.T., Suye, S.I., 2008. Reverse reaction of malic enzyme Zambanini, T., Sarikaya, E., Kleineberg, W., Buescher, J.M., Meurer, G., Wierckx, N.,
for HCO3- fixation into pyruvic acid to synthesize L-malic acid with enzymatic Blank, L.M., 2016b. Efficient malic acid production from glycerol with Ustilago tri-
coenzyme regeneration. Biosci. Biotechnol. Biochem. 72 (5), 1278–1282. chophora TZ1. Biotechnol. Biofuels 9.
Osothsilp, C., Subden, R.E., 1986. Malate transport in Schizosaccharomyces pombe. J. Zambanini, T., Hosseinpour Tehrani, H., Geiser, E., Sonntag, C.K., Buescher, J.M., Meurer,
Bacteriol. 168 (3), 1439–1443. G., Wierckx, N., Blank, L.M., 2017. Metabolic engineering of Ustilago trichophora
Oswald, F., Dorsam, S., Veith, N., Zwick, M., Neumann, A., Ochsenreither, K., Syldatk, C., TZ1 for improved malic acid production. Metab. Eng. Commun. 4, 12–21.
2016. Sequential mixed cultures: from syngas to malic acid. Front. Microbiol. 7. Zelle, R.M., de Hulster, E., van Winden, W.A., de Waard, P., Dijkema, C., Winkler, A.A.,
Payne, G.A., Nierman, W.C., Wortman, J.R., Pritchard, B.L., Brown, D., Dean, R.A., Geertman, J.M.A., van Dijken, J.P., Pronk, J.T., van Maris, A.J.A., 2008. Malic acid
Bhatnagar, D., Cleveland, T.E., Machida, M., Yu, J., 2006. Whole genome comparison production by Saccharomyces cerevisiae: engineering of pyruvate carboxylation, ox-
of Aspergillus flavus and A. oryzae. Med. Mycol. 44, S9–S11. aloacetate reduction, and malate export. Appl. Environ. Microb. 74 (9), 2766–2777.
Peleg, Y., Stieglitz, B., Goldberg, I., 1988. Malic-acid accumulation by Aspergillus flavus Zelle, R.M., De Hulster, E., Kloezen, W., Pronk, J.T., van Maris, A.J.A., 2010. Key process
.1. Biochemical aspects of acid biosynthesis. Appl. Microbiol. Biotechnol. 28 (1), conditions for production of C4-dicarboxylic acids in bioreactor batch cultures of an
69–75. engineered Saccharomyces cerevisiae strain. Appl. Environ. Microb. 76 (3), 744–750.
Pines, O., EvenRam, S., Elnathan, N., Battat, E., Aharonov, O., Gibson, D., Goldberg, I., Zhang, T., Ge, C.Y., Deng, L., Tan, T.W., Wang, F., 2015. C4-dicarboxylic acid production
1996. The cytosolic pathway of L-malic acid synthesis in Saccharomyces cerevisiae: by overexpressing the reductive TCA pathway. FEMS Microbiol. Lett. 362 (9).
the role of fumarase. Appl. Microbiol. Biotechnol. 46 (4), 393–399. Zhang, X., Wang, X., Shanmugam, K.T., Ingram, L.O., 2011. L-malate production by
Pines, O., Shemesh, S., Battat, E., Goldberg, I., 1997. Overexpression of cytosolic malate metabolically engineered Escherichia coli. Appl. Environ. Microb. 77 (2), 427–434.
dehydrogenase (MDH2) causes overproduction of specific organic acids in Zheng, H.T., Ohno, Y., Nakamori, T., Suye, S., 2009. Production of L-malic acid with
Saccharomyces cerevisiae. Appl. Microbiol. Biotechnol. 48 (2), 248–255. fixation of HCO3- by malic enzyme-catalyzed reaction based on regeneration of
Prathumpai, W., Gabelgaard, J.B., Wanchanthuek, P., van de Vondervoort, P.J.I., de coenzyme on electrode modified by layer-by-layer self-assembly method. J. Biosci.
Groot, M.J.L., McIntyre, M., Nielsen, J., 2003. Metabolic control analysis of xylose Bioeng. 107 (1), 16–20.
catabolism in Aspergillus. Biotechnol. Progr. 19 (4), 1136–1141. Zhu, L.W., Zhang, L., Wei, L.N., Li, H.M., Yuan, Z.P., Chen, T., Tang, Y.L., Liang, X.H.,
Queiros, O., Casal, M., Althoff, S., Moradas-Ferreira, P., Leao, C., 1998. Isolation and Tang, Y.J., 2015. Collaborative regulation of CO2 transport and fixation during suc-
characterization of Kluyveromyces marxianus mutants deficient in malate transport. cinate production in Escherichia coli. Sci. Rep. 5.
Yeast 14 (5), 401–407. Zou, X., Wang, Y.K., Tu, G.W., Zan, Z.Q., Wu, X.Y., 2015. Adaptation and transcriptome
Regev-Rudzki, N., Battat, E., Goldberg, I., Pines, O., 2009. Dual localization of fumarase is analysis of Aureobasidium pullulans in corncob hydrolysate for increased inhibitor
dependent on the integrity of the glyoxylate shunt. Mol. Microbiol. 72 (2), 297–306. tolerance to malic acid production. PLos One 10 (3).

353