See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/332114629

Bioreactors for succinic acid production processes

Article  in  Critical Reviews in Biotechnology · March 2019

DOI: 10.1080/07388551.2019.1592105

CITATIONS READS

9 753

4 authors:

Mariateresa Ferone Francesca Raganati

University College Dublin University of Naples Federico II
10 PUBLICATIONS   46 CITATIONS    40 PUBLICATIONS   453 CITATIONS   

SEE PROFILE SEE PROFILE

Giuseppe Olivieri Antonio Marzocchella

Wageningen University & Research University of Naples Federico II
133 PUBLICATIONS   2,151 CITATIONS    167 PUBLICATIONS   2,741 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Galdieria sulphuraria for industrial utilization View project

Microalgae as factory of high value bioproducts View project

All content following this page was uploaded by Mariateresa Ferone on 08 April 2019.

The user has requested enhancement of the downloaded file.

 Critical Reviews in Biotechnology

ISSN: 0738-8551 (Print) 1549-7801 (Online) Journal homepage: https://www.tandfonline.com/loi/ibty20

Bioreactors for succinic acid production processes

Mariateresa Ferone, Francesca Raganati, Giuseppe Olivieri & Antonio

Marzocchella

To cite this article: Mariateresa Ferone, Francesca Raganati, Giuseppe Olivieri & Antonio
Marzocchella (2019): Bioreactors for succinic acid production processes, Critical Reviews in
Biotechnology, DOI: 10.1080/07388551.2019.1592105

To link to this article: https://doi.org/10.1080/07388551.2019.1592105

Published online: 31 Mar 2019.

Submit your article to this journal

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ibty20
CRITICAL REVIEWS IN BIOTECHNOLOGY
https://doi.org/10.1080/07388551.2019.1592105

REVIEW ARTICLE

Bioreactors for succinic acid production processes

Mariateresa Feronea,b, Francesca Raganatia, Giuseppe Olivieria,c and Antonio Marzocchellaa
a
Dipartimento di Ingegneria Chimica, dei Materiali e della Produzione Industriale, Universit
a degli Studi di Napoli Federico II, Napoli,
Italy; bUCD School of Agriculture & Food Science, University College Dublin, Dublin, Ireland; cDepartment of Agrotechnology and
Food Sciences, Wageningen University and Research Centre, Wageningen, The Netherlands

ABSTRACT ARTICLE HISTORY

Succinic acid (SA) has been recognized as one of the most important bio-based building block Received 5 September 2018
chemicals due to its numerous potential applications. Fermentation SA production from renew- Revised 4 February 2019
able carbohydrate feedstocks can have the economic and sustainability potential to replace pet- Accepted 10 February 2019
roleum-based production in the future, not only for existing markets, but also for new larger
KEYWORDS
volume markets. Design and operation of bio-reactors play a key role. During the last 20 years, Succinic acid; bioreactor;
many different fermentation strategies for SA production have been described in literature, fermentation; batch and
including utilization of immobilized biocatalysts, integrated fermentation and separation systems continuous process;
and batch, fed-batch, and continuous operation modes. This review is an overview of different biorefinery
fermentation process design developed over the past decade and provides a perspective on
remaining challenges for an economically feasible succinate production processes. The analysis
stresses the idea of improving the efficiency of the fermentation stage by improving bioreactor
design and by increasing bioreactor performance.

Introduction succiniciproducens, Corynebacterium glutamicum, and

recombinant Escherichia coli [5–10]. Bio-based produc-
Organic acids are amongst the first metabolites pro-
tion of SA was also achieved by employing several fun-
duced through biotechnological processes. Succinic
gal or yeast strains, such as Penicillium simplicissimum,
acid (SA) has been listed among one of the 12 chemical
Aspergillus niger, Saccharomyces cerevisiae, and Yarrowia
building blocks that can potentially be produced from
lipolytica [11,12].
biomass, according to the US Department of Energy in
A. succinogenes, B. succiniciproducens, and M. succini-
2004 [1]. Currently, SA finds application as a surfactant,
ciproducens are the most promising wild-type bacterial
an ion chelator, an additive in agricultural and food,
strains for application in an industrial process. All these
and in the pharmaceutical industries [2]. The market
strains have been isolated from the rumen, an ecosys-
size for this acid was USD 157.2 Million in 2015 [3], but
tem where the bacteria obtain their energy from the
it is expected to grow rapidly during the next years due
to SA’s potential as a building block chemical. In fact, decarboxylation of succinate to propionate, which rep-
SA can also be used for the synthesis of the biodegrad- resents a nutrient for the ruminant [13]. These microor-
able plastic polybutylene succinate (PBS), polyester pol- ganisms produce SA by the reductive branch of TCAs
yols, plasticizers, polyurethanes and 1,4-butanediol (1,4- cycle, from the intermediate oxaloacetate (OOA) [14].
BDO) [4]. Through this pathway, theoretically, 1 mol of SA can be
SA is an intermediate of the tricarboxylic acid (TCA) produced from 0.5 mol of glucose, 0.6 mol of xylose, or
cycle and one of the end products of anaerobic micro- 1 mol of glycerol, and the fixation of 1 mol of CO2.
organisms performing mixed acid fermentation. However, from 0.5 mol of glucose only 1 mol of NADH
The most suitable microbial hosts for industrial- can be synthesized, and the production of 1 mol of SA
scale SA production belong to bacteria, fungi or requires 2 mol of NADH. This additional reducing power
yeast. The most investigated bacterial strains are: (i.e. NADH) requirement in the C4 pathway needs to be
Anaerobiospirillum succiniciproducens, Actinobacillus suc- provided by other parts of overall metabolism (e.g. gly-
cinogenes, Mannheimia succiniciproducens, Basfia colysis and C3 pathway) [4].

CONTACT Mariateresa Ferone [email protected] Dipartimento di Ingegneria Chimica, dei Materiali e della Produzione Industriale,
Universit
a degli Studi di Napoli Federico II, P.le V. Tecchio 80, 80125 Napoli, Italy
ß 2019 Informa UK Limited, trading as Taylor & Francis Group
2 M. FERONE ET AL.

Bradfield and Nicol [15] suggested that the max- biochemistry, alternative substrates, acid recovery, fer-
imum glucose to SA conversion yield can be calculated mentation modes, and economics.
from the following stoichiometric Equation (1): In this review, a wide range of reactor configuration
6 12 6 and operation in different fermentation processes for
C6 H12 O6 þ CO2 þ NADH ! C4 H6 O4 þ NADþ þ ATP þ H2 O (1)
7 7 7 SA production, with special focus on the biofilm reac-
The above equation results in a maximum theoret- tors, are discussed.
ical SA yield of 1.12 g SA per gram glucose (in mass
units), assuming that no biomass or by-products are Bioreactor system
formed and considering the effect of the reduction
Batch fermentation
potential. The carbon diverted to biomass and by-prod-
uct formation leads to the lower experimental val- Batch fermentation is the simplest mode of operation.
ues reported. The fermentation is carried out in a mechanically stirred
Bio-based succinate production may replace the vessel with other accessories, such as gas sparger and
petrochemical-based route with marked environmental heating/cooling jacket. Substrate concentration for SA
benefits: (1) the use of renewable feedstock as substrate production in batch process typically ranges from 20 to
for fermentation [16] and (2) CO2 is fixed into SA during 100 g/L. The fermentation temperature is kept around
fermentation [17]. 37–39  C [29] and the optimal pH is near-neutral
However, the main drawback of bio-based processes (6.8–7.2) [5] for most bacterial species, while it can be
is that they are usually characterized by low specific lowered to 4.5–6 for yeasts [30–32]. In most cases,
conversion rates. Therefore, large volume reactors are anaerobic conditions are needed for SA fermentation;
necessary for processes operated at high throughput. this is ensured by sparging oxygen-free nitrogen or car-
The development of industrial-scale bio-based proc- bon dioxide over the surface of broth. Some substrates
esses needs innovative technologies aimed at intensi- need pretreatment or upstream processing such as mill-
fied operations. ing, acid/enzymatic hydrolysis, dilution, and filtration
In this regards, design and operation of bio-reactors [33,34]. The typical batch time for fermentation
(depending on the microbial host used and operating
play an important role in the biotech industry. An
conditions) is 48–100 h. A survey of the most relevant
improvement of reactor performances and productiv-
batch studies is summarized in Table 1.
ities, and, consequently, of the whole process econom-
The earliest work regarding SA production was
ics, can be obtained knowing: the rate of reactions,
focused on studying the metabolic characteristics of
transport phenomena, hydrodynamics, and the operat-
the succinogenes bacteria and the effect of different
ing conditions [18,19].
parameters on the process, such as pH, CO2 concentra-
This review focuses on the design and operation of
tion, redox potential, and carbon source. The most
bioreactors in order to produce SA.
important studies regarding physiology and metabol-
Several reviews, previously published, have
ism of succinogenes bacteria were conducted by
addressed various aspects of SA fermentation. The first
McKinlay et al. [52], Nghiem et al. [35], Samuelov et al.
reviews were published in 2000s by Zeikus et al. [2],
[5], Van der Werf et al. [17], and Lee et al. [9].
McKinlay et al. [4] and Song and Lee [20] were mainly The highest SA concentration from glucose in a
focused on the market and the application of SA and batch fermentation was obtained with an engineered E.
on the potential of the biotechnological process for its coli strain [8]. This strain, named KJ122, was obtained
production. Several reviews concern the genetic engin- by deleting the genes encoding the fermentation path-
eering and manipulation and strain improvement for ways for by-products and two genes involved in the
SA and other organic acid production [21–23]. Kurzrock oxaloacetate metabolism. Strain KJ122 produced SA at
and Weuster-Botz [24] and Lo pez-Garzon and Straathof high titer and yield during simple, anaerobic, batch fer-
[25] have reviewed the developments with respect to mentations employing a mineral salts medium.
various techniques (precipitation, electrodialysis, and The highest SA productivity in a batch process was
liquid–liquid extraction) used for SA recovery from fer- obtained by Marinho et al. [49]. They used the macro-
mentation broth. More recent reviews by Akhtar et al. alga Saccharina latissima as a feedstock for SA fermen-
[26] and Tan et al. [27] have addressed the issue of the tation by A. succinogenes; a maximum productivity of
use of renewable feedstock for fermentation. Pateraki 3.9 g/Lh was achieved in a 1.5 L batch fermentation
et al. [28] addressed several important factors of SA fer- with 15% solid loading (%w/w) and a hydrolysate – to –
mentation by the bacterium Actinobacillus succinogenes: medium ratio of 75:25.
Table 1. Summary of literature on bioreactors and fermentation processes (batch and fed-batch) with free cells.
Initial C
Operation concentration CO2 CSA YSA/sub Productivity
Microorganism mode Carbon source g/L supply g/L g/g g/Lh Notes Ref.
A. succiniciproducens Batch Glucose 20 Gas 32.2 0.99 1.2 [35]
M. succiniciproducens Batch Glucose 20 Gas 13.5 0.68 1.8 [9]
A. succinogenes 130Z Batch Glucose 55 Gas and Na2CO3 33.8 0.62 1.35 [36]
A. succinogenes 130Z Batch Glucose 85 Gas and MgCO3 47.6 1.5 Kinetic study, see [37]
details in Table 2
E. coli Batch Glucose – K2CO3 71.5 1.00 0.9 Genetically modi- [8]
fied strain
A. succinogenes 130Z Batch Wheat 140 MgCO3 64.2 0.81 1.19 [38]
A. succinogenes 130Z Batch Cheese whey 50 Gas 27.9 0.44 0.6 [34]
A. succinogenes 130Z Batch Glycerol 30 Gas and MgCO3 28 0.75 0.4 Kinetic study, see [39]
details in Table 2
A. succinogenes 130Z Fed-Batch Glucose – MgCO3 60.2 0.75 1.3 [40]
M. succiniciproducens Batch Glucose 66 Gas 33.8 0.62 1.35 Kinetic study, see [41]
details in Table 2
M. succiniciproducens Batch Crude glycerol – – 8.4 1.2 0.9 [42]
A. succinogenesNJ113 Batch Corn fiber 50 Gas 35.4 0.72 [43]
hydrolysate
A. succinogenes Fed-Batch Glucose – Gas 145.0 0.52 1.3 In situ product [44]
removal by
expanded-
bed adsorption
A. succiniciproducens Fed-Batch Glucose – Gas 31.8 0.87 – [45]
A. succinogenes130Z Fed-Batch Glucose – Gas 95.6 0.79 1.99 Genome shuffling to [46]
enhance
SA production
E. coli Batch Microalgae 20 Gas 17.4 0.87 – Dual-phase [47]
hydrolysate fermentation
A. succinogenes Batch Xylose-rich media 100 Gas 27.4 0.69 0.6 Kinetic study, see [48]
130Zand B. 26.0 0.76 0.66 details in Table 2
succiniciproducens
B. succiniciproducens Batch Corn stover 60 Gas 30.0 0.43 0.69 [16]
hydrolysate
A. succinogenes 130Z Batch Microalgae 40 MgCO3 36.8 0.92 3.9 [49]
hydrolysate
A. succinogenes 130Z Batch Microalgae 50 MgCO3 33.8 0.63 0.7 [50]
hydrolysate
A. succinogenes 130Z Batch Glucose-Mannose- 60 MgCO3 27.0 0.55 0.22 [51]
Arabinose-Xylose
(GMAX) sugars
Missing data are either not mentioned or not investigated in the literature-cited references.
The yield is molSA/molGLY.
The yield is referred to the total sugars consumed.
The first results are referred to A. succinogenes and the second to B. succiniproducens.
CRITICAL REVIEWS IN BIOTECHNOLOGY
3
4 M. FERONE ET AL.

Kinetic studies

[53]
[51]
[51]
[51]
[51]
[37]
[48]
[54]
[36]
[48]
[55]
[56]
[41]
Ref.
SA production by fermentation is affected by both sub-

0.53
nFA

2.3

2.3
strate and product inhibition. Several batch studies

–
–
–
–

–
–

–
–
–
1
regard the development of kinetic models to simulate,

FAmax (g/L)
control, and optimize this bioprocess. Most of these

45.6

16.0
18.0

22.0
works used modified Monod equations that considered

–
–
–
–

–
–

–
–
–
both substrate and product inhibition to estimate the
kinetic parameters and predict the behavior of the

nAA
0.1

2.3

2.3
most important fermentation variables for various sub-

–
–
–
–

–
–

–
–
–
1
strate (glucose, glycerol, and xylose) and different bac-

The table reports a summary of the literature data available about the kinetic parameters assessed for different succinic acid producers grown on different carbon sources.
terial strains (A. succinogenes, M. succiniproducens, and

AAmax (g/L)
80.0

44.2
38.0

38.0
B. succiniproducens) [37, 39, 41, 48].

–
–
–
–

–
–

–
–
–
The main investigations on the kinetics of SA produc-
tion are reported hereinafter and summarized in Table 2.
In a paper by Corona-Gonzalez et al. [36], a kinetic

1.07
nSA

2.3

2.3
–
–
–
–
–

–
–
–
1
study of SA production from glucose by A. succinogenes
was presented. They proposed a Jerusalimsky expres-

SAmax (g/L)

17.3
sion to describe the inhibition effect caused by the ini-

104.2
55.0
45.6

55.0
–
–
–
–
–

–
–
tial glucose and acid concentration. However, their
model did not elucidate the individual effects of each
acid produced, but the inhibition was explained in
KiSA (g/L)

22.0
4.10
terms of the total acid concentration produced.

–
–
–
–
–
–
–

–
–
–
–
The same kinetic model was used to explain A. succi-

The inhibition constant is referred to inhibition caused by the acid mixture (SA, AA, and FA) on specific growth rate.
nogenes [37] and M. succiniproducens [41] batch fermen-

0.6
nS
–
–
–
–
–

–
–
–
–
–
–
–
tation profiles on glucose. An extended Monod model
was used to describe cell growth when inhibition by
CSmax (g/L)

both substrate and products occurs. Moreover, a

Table 2. Kinetic parameters assessed according to different models from the literature.

155
–
–
–
–
–

–
–
–
–
–
–
–

Kinetic parameters referred to the mixed sugars medium, where xylose is the most abundant one.
Luedeking–Piret expression was able to simulate the
end-product production rates, considering both
growth-associated and non-growth-associated contri-
Ki (g/L)

15.5

55.5
15.4

15.2

88.3
–

–
–
–
–

–
–
30

butions. These models adequately described cell

growth, substrate consumption, and product formation
during SA fermentations using both microorganisms
KS (g/L)
6.96

4.82
9.49
7.14
2.03
0.70
2.90

1.55
0.1

1.1
–
–

and allowed the identification of the critical concentra-

tion of glucose and of the acids produced.
Recently Pateraki et al. [48] developed an unstructured
(h1)
lmax

1.43
0.39
0.10
0.17
0.11
0.50
0.39
0.12
0.78
0.93

1.3
–
–

model to predict the fermentation behavior of both A.

succinogenes and B. succiniproducens in a medium con-
Operation mode

taining a mixture of hexose and pentose sugars. The

Continuous

Continuous
Continuous

model was also validated by bench-scale fermentation

experiments with a medium miming the sugar compos-
Batch
Batch
Batch
Batch
Batch
Batch
Batch
Batch
Batch

Batch

ition contained in spent sulfite liquor, the liquid waste

stream from the sulfite pulping process, showing a good
Arabinose

agreement between experimental and predicted data.

Mannose

Glycerol

Glycerol
Sugar
Glucose
Glucose

Glucose

Glucose

Glucose

Glucose
Xylose

Xylose

AA: acetic acid; FA: formic acid

Xylose

SA from renewable feedstocks

Some other batch studies addressed the possibility to
A. succiniproducens
B. succiniproducens

M.succiniproducens
A. succinogenes

use different renewable raw materials such as industrial

Microorganism

waste and by-product streams (e.g. cheese whey, crude

glycerol from biodiesel production, sugar cane molas-
ses, and wheat milling by-products), lignocellulosic
 CRITICAL REVIEWS IN BIOTECHNOLOGY 5

materials (e.g. sugarcane bagasse, corn fiber and corn- Liu et al. [40]. The SA concentration, productivity,
cob, and bio-waste cotton) and microalgae hydrolysate and yield were 60.2 g/L, 1.3 g/Lh, and 0.75 gSA/g,
to reduce process costs. respectively.
Glycerol, a by-product of biodiesel production, is an Yan et al. [61] investigated SA production by A. succi-
inexpensive and abundant carbon source. Moreover, nogenes in a fibrous bed reactor. The process included
given the more reduced state of carbon in glycerol, it 4 fed-batch fermentations cycles that led to the produc-
results in higher product yields compared with com- tion of 98.7 g/L of SA with a yield of 0.89 gSA/g and a
mon sugars. Vlysidis et al. [54] reported the production productivity 2.77 g/Lh. Bretz and Kabasci [45] estab-
of 26.7 g/L succinate from crude glycerol, at a product- lished a fed-batch process for SA production by
ivity yield of 0.96 g/g and 0.23 g/Lh, respectively. Anaerobiospirillum succiniproducens, obtaining a yield
Yuzbashev et al. [30] obtained an engineered Yarrowia increase of (0.88 gSA/g) compared to the batch process
lipolytica strain capable of accumulating succinate at a (0.60 gSA/g) with similar glucose concentrations.
concentration of 45 g/L in shaking flasks with buffering Li et al. [71] employed a fibrous bed reactor for SA
and around 17 g/L without buffering. The ability to production from glycerol by an engineered Yarrowia lip-
grow at low pH reduces the need for pH neutralization olytica strain. With a fed-batch fermentation strategy,
during SA fermentation, and thus, the generation of the highest SA concentration ever reported (up to
salts can be avoided, facilitating the following down- 198.2 g/L) was obtained.
stream process [57]. Batch and fed-batch processes aim to maximize the
Chen et al. [58] obtained 35.4 g/L SA with 0.72 g/g substrate conversion rate and the product yield and
yield and a productivity of 0.98 g/Lh using corn hydro- final SA concentration. However, batch and fed-batch
lysates obtained through sulfuric acid pretreatment, fol- cultivation modes do not aim to maximize the product-
lowed by neutralization with CaCO3 salt and activated ivity – i.e. how much a product can be produced in a
carbon absorption (to remove furfural). Du et al. [38] reactor having a specified size within a unit time. That
developed a solid state fermentation strategy for SA is, because batch or fed-batch is inefficient, and thus
production from a wheat-derived medium. The bran, a both systems have low productivity, due to the long
byproduct of the wheat milling industry, was used to dead time. Moreover, in a batch process, the maximum
produce amylolytic and proteolytic enzymes via SSF biomass concentration is achieved only at the end of
using two fungal strains. The resulting solutions were the process. This implies that when the system is at its
separately utilized to hydrolyze the starch and proteins greatest production capacity, the conditions for product
contained in wheat milling by-product to produce a production are often at their worst and thus, the pro-
glucose-rich stream, containing 14.0 g/L glucose and a ductivities are low.
nitrogen-rich stream of more than 3.5 g/L free amino
nitrogen. With the addition of MgCO3 into the wheat- Continuous fermentation
derived medium they obtained 64 g/L of SA.
Continuous fermentation processes have several advan-
In two studies with microalgae hydrolysate (main
tages over batch and fed-batch fermentation processes
sugars: glucose and mannitol) was used as a biomass
(with free cells): 1 – since fermentation can be operated
feedstock for SA production. A final SA concentration of
in continuous mode, only one inoculum is sufficient for
17.4 g/L was obtained after 72 h dual-phase fermenta-
an extended period; 2 – reduction of sterilization time;
tion with an engineered strain of E. coli [47], while
3 – enhanced overall productivity. The most common
24.4 g/L of SA were obtained using wild type A. succino-
strategies of continuous fermentation including free
genes [50].
immobilized cells, and cell recycling have been
attempted [64, 72].
Fed-batch fermentation The technique of continuous fermentation is aimed
at increasing the productivity of the process. In the fol-
This fermentation mode is employed when high con-
lowing sections, the most common strategies of con-
centrations of substrate are toxic to microbial biocata-
tinuous fermentation will be discussed.
lyst. Results of several studies with different microbial
strains and substrates in fed-batch mode are summar-
Free cell continuous reactors
ized in Tables 1 and 3.
Fed-batch operation keeping the glucose concentra- In a continuous fermentation process, a continuous
tion at 10–15 g/L during feeding could enhance SA flow of feed media is provided to the bioreactor in
concentration and productivity, as demonstrated by order to achieve a growth rate close to the maximum
 6

Table 3. Summary of literature fermentation processes (batch, repeated-batch, fed-batch, and continuous) with immobilized cells.
Reactor and operating conditions Results
Microorganism Notes Ref.
D SA Yield Productivity
Mode of operation Support Bioreactor Substrate h1 g/L gSA/gS g/Lh
A. succinogenes Continuous/ Plastic composite sup- BioFlo 3000 BioReactor Glucose 0.2–1.2 10.4 at D ¼ 0.2 h1 0.72 2.1 For D <0.8 h1, sus- [59]
repeated- batch port (PCS), polypro- pended cells were pre-
M. FERONE ET AL.

pylene based (50% sent in the medium

PP), or PP tubes (160 g/L dry wt.). For D
> 0.8 h1, discrete
clumps of cells formed
in the medium.
A. succinogenes Continuous Genulite Groperl, Aluminum top and Glucose 0.56 12.0 0.8 6.7 – [60]
expanded perlite bottom section and
particles consisting a glass tube with a
of amorphous vol- length of 115 mm
canic glass. and an inner diam-
Diameters eter of 37.5 mm.
2–4 mm,40% of the Volume: 156 mL
reactor’s (including recycle)
open volume
A. succinogenes Batch/repeated- Cotton fibrous matrix, Stirred tank bioreactor Glucose – 98.7 (RFB) 0.89 (RFB) 2.27 (RFB) – [61]
batch (RB)/Fed- 100% cotton terry Volume: 3L
batch (RFB) cloth thickness of
1 mm, poros-
ity >95%
A. succinogenes Continuous Cotton cloth matrix Stirred tank bioreactor Glucose 0.05 55.3 0.8 2.77 – [62]
Volume: 3L
R
A. succinogenes Continuous PoraverVexpanded Glass cylinder (44 mm Glucose 0.054–0.72 32.5 at 0.90 1.8 Concentration of acids in [63]
glass particles, OD, 3.2 mm wall D ¼ 0.054 h1 the medium has min-
diameters 3–4 mm, thickness), alumi- imal influence on the
density 190 ± 20 kg num base, and biofilm amount at
m3, vol- head. An external steady-state.
ume: ±65 mL recycle line pro-
vided agitation.
Volume 165mL
(including recycle)
A. succinogenes Repeated-batch Zeolite, diatomite Glass bioreactor, Glucose – 21.4 0.6 2.6 Results of 5 repeated [64]
2
(43 m /g) and acti- Volume: 800 mL batches with immobi-
vated carbon lized cells in agar
(865 m2/g) were
used as supports for
immobilization by
adhesion.
Alginic acid, j-carra-
geenan, chitosan,
agar, and poly-acryl-
amide hydrogel
were used as sup-
ports for
(continued)
Table 3. Continued.
Reactor and operating conditions Results
Microorganism Notes Ref.
D SA Yield Productivity
Mode of operation Support Bioreactor Substrate h1 g/L gSA/gS g/Lh
immobilization
by entrapment.
A. succinogenes Continuous – 3 mm silicone tube of Glucose 0.5–2.2 18.0 at D ¼ 0.5 h1 0.70 9.2 – [65]
approximately 5 m
length
Volume 50 mL
A. succinogenes Continuous Stainless-steel wool Same as [65] Glucose 0.05–0.5 0.91 – Metabolic flux ana- [15]
lysis (MFA)
1
A. succinogenes Continuous Central porous poly- BioFlo 3000 fermenter Xylose-rich 0.02–0.05 39.6 at D ¼ 0.02 h 0.77 0.78 – [66]
propylene (PP) tube outfitted with a hydrolysate
perforated with a novel agitator fit-
multitude of holes ting to support bio-
into which porous film attachment;
PP or silicone arms surface area-to-vol-
were affix ume ratio: 1.36 cm2
cm3 exclud-
ing porosity
A. succinogenes Continuous Bound wooden sticks, Same as [65] Xylose 0.05–0.1–0.3 10.9 at D ¼ 0.3 h1 0.55 3.4 – [67]
silicone-tubing seg-
ments, and loosely
spaced
wooden sticks
A. succinogenes Continuous Four wooden sticks Same as [65] Glucose or Xylose 0.02–0.3 35.0 0.8 – – [68]
wrapped in mutton Volume: 360 mL
cloth and fixed to a
top distributor plate
were inserted into
the reactor. The
stick structure
served as a support
surface for biofilm
A. succinogenes Continuous A single wooden stick Same as [68] Glucose 0.044–0.131 12.7 0.55 at 1.64 – [69]
(diameter 6 mm, D ¼ 0.129 h1
length 120 mm)
covered
with strands of
fiberglass was
added to the center
of the reactor body
A. succinogenes Continuous Tygon rings, 3 mm 5 cm ID, 8 cm high -Glucose – 43.0 at D ¼ 0.5 h1 0.98 22.0 – [70]
glass tube with a -Xylose 7.1 at 0.93 7.4
4.5 cm bed -GAX (Glucose- D ¼ 1.04 h1 0.56 15.0
Total volume:40 mL Arabinose- 20.5 at
Xylose) D ¼ 0.7 h1
Y. lipolytica Batch/Fed- Cotton towel fixed 2.5-L bench- Glycerol – 198 (FB) – 0.84 – [71]
batch (FB) onto the baffle top fermenter
CRITICAL REVIEWS IN BIOTECHNOLOGY
7
8 M. FERONE ET AL.

growth rate for a prolonged time (from days to months glyoxylate shunt) pathway, but only occurs in non-
and depending on the process), thus offering improved growth conditions.
product production rates and reduced downtime. A However, there are only a few reports in the litera-
summary of continuous, free-cells SA fermentation ture regarding continuous, free-cell fermentation when
studies is given in Table 4. using Actinobacillus succinogenes, because of its natural
A continuous, free cells cultivation approach for the ability to form a biofilm when operated for long time
fermentative production of SA from crude glycerol was periods [59]. All continuous studies with A. succinogenes
developed with the bacterium Basfia succiniciproducens resulted in unavoidable biofilm formation.
DD1 [79]. In glycerol-limiting condition, the highest SA In the study by Kim et al. [72], a membrane separ-
concentrations, productivities and yields obtained were: ation recycle system was implemented to produce SA
5.21 g/L, 0.094 g/Lh, and 1.02 gSA/g, respectively. in continuous operation mode. Using this system, they
Another example of continuous process for SA produc- obtained a SA volumetric productivity of 6.63 g/Lh at
tion was reported by Kim et al. [75]. They used the dilution rate 0.5 h1. However, severe membrane
Mannheimia succiniciproducens MBEL55E as the biocata- fouling problems were experienced with this system.
lyst and wood hydrolysate as economic carbon source, In the work reported by Ferone et al. [53], the kinetic
obtaining SA yield and productivities of 0.55 gSA/g and of growth and SA production by A. succinogenes in a
3.19 g/Lh, respectively. continuous stirred tank reactor (CSTR) was investigated.
In the work conducted by van Heerden and Nicol Biofilm formation was unavoidable when the reactor
[80], batch and continuous fermentation strategies for was operated for more than 10 days. Therefore, fermen-
SA production by a genetically modified E. coli strain tations were stopped when biofilm formation occurred
have been compared. With this strain, the highest SA to prevent erroneous data interpretation.
selectivity (SA/AA) and yield were obtained in batch
cultures (SA:AA ¼ 5, SA yield ¼ 0.94 gSA/gG). From the
Immobilized cell reactors
chemostat runs it, was possible to determine that
the optimal pathway for SA production, where reduc- Biomass concentration is the rate-limiting step during
tion power is generated via the oxidative TCA (or anaerobic fermentations. Due to the slow growth rate

Table 4. Summary of literature on continuous, free-cells succinic acid fermentation studies.

D CSA YSA/sub Productivity
Microorganism Carbon source h1 g/L g/g g/Lh Notes Ref.
A. succiniproducens Whey 0.085–0.15 24.0 at D ¼ 0.085 h1 0.72 2.1 – [14]
A. succiniproducens Whey 0.03–0.14 12.3 at D ¼ 0.11 h1 0.93 1.4 – [73]
M. succiniciproducens Whey 0.1–0.7 5.6 at D ¼ 0.7 h1 0.60 3.9 Whey and corn steep [74]
liquor (CSL)
as substrates
M. succiniciproducens Wood hydrolysate 0.1–0.7 8.0 at D ¼ 0.4 h1 0.55 3.2 [75]
A. succinogenes Glucose 0.2–1.2 7.0 at D ¼ 1.0 h1 0.76 7.0 Results here reported [59]
are referred to the
suspended cells
experiments
A. succiniproducens Glucose 0.93 83.0 0.88 10.4 Integrated mem- [76]
brane-bioreactor
with an electrodi-
alysis system
A. succiniproducens Glucose 0.23 14.3 – 3.3 Internal membrane [77]
cell recycle reactor
M. succiniciproducens Glucose 0.1–0.3 8.9 at D ¼ 0.2 h1 0.25 1.8 Genetically modified [69, 78]
strain to reduce
by-products
A. succinogenes and Glucose 0.2–0.5 15.8 at 0.59 6.3 External membrane [72]
M. 0.1–0.3 D ¼ 0.4 h1 0.64 1.3 cell recycle reactor
succiniciproducens 12.8 at
D ¼ 0.1 h1
B. succiniciproducens Glycerol 0.004–0.018 5.2 1.02 0.094 Problem of wall [79]
cells growth
1
A. succiniproducens Glycerol 0.022–0.25 15.5 at D ¼ 0.14 h 1.37 2.2 Kineticstudy results [56]
referred to pH 6.2
E. coli KJ134 Glucose 0.021–0.187 12.9 at D ¼ 0.056 h1 0.77 0.72 Genetically modi- [80]
fied strain
The results in the first line are referred to A. succinogenes and the ones in the second line to M. succiniproducens.
 CRITICAL REVIEWS IN BIOTECHNOLOGY 9

of anaerobic microorganisms, production rates, and, repeated-batch fermentation strategy (SA concentration
consequently, productivity, are typically extremely low. 35 g/L and yield 0.68 gSA/gG).
Thus, to increase productivity, the most popular strat- Most of literature reporting studies on SA production
egy is to increase biomass concentration in the bioreac- by a A. succinogenes biofilm comes from a research
tor. Immobilized cell bioreactors are designed to retain group at University of Pretoria (South Africa). van
high biomass concentration. Heerden and Nicol [60] reported the continuous SA pro-
Multi-species biofilms are used in the industrial con- duction in an external recycle bioreactor, using
text to achieve several aims, including wastewater treat- expanded perlite particles as a support for A. succino-
ment for removal of organics, and heavy metals [81,82]. genes biofilm formation. High productivity (6.35 g/Lh
Single species biofilm is used industrially to produce at a dilution rate of 0.56 h1) and SA yield of 0.69 gSA/
chemicals [83]. gG were achieved. Bradfield and Nicol [15] used a simi-
The immobilized culture systems have various merits lar reactor configuration for metabolic flux distribution
over free cell systems as follows [83]: (i) physical reten- analysis of A. succinogenes biofilms in continuous oper-
tion of cells on the matrix; (ii) easier separation of cells ation mode. The production of 48.5 g/L of SA and a
from products; (iii) high cell concentrations; (iv) smaller yield of 0.91 gSA/gG were reported.
reactor volumes; (v) higher productivity; (vi) flexibility of In a subsequent study from the same research group
reactor design (such as fixed bed, trickle bed, and fluid- [65], a shear controlled fermenter was developed that
ized bed [FBR]) for continuous operation; (vii) control of enabled them to perform both suspended (free) cells
reaction kinetics; (viii) minimized nutrient depletion and and biofilm fermentation experiments. The volumetric
product inhibition; and (ix) better mass transfer charac- productivity of A. succinogenes biofilm fermentations
teristics because of reduced viscosity of the medium. (17.1 g/Lh) was about 10-fold higher than the chemo-
However, the demerits of the immobilized system are stat fermentations (1.83 g/Lh). Moreover, SA yields
mass transfer limitation of the substrate and activity obtained during the biofilm cycles were significantly
loss because of immobilization. Another operational higher (0.74 gSA/gG) compared to the suspended cell
problem that might occur is improper substrate utiliza- fermentations (0.48 gSA/gG). This difference in the yields
tion due to the accumulation of gas bubbles produced was interpreted as a metabolic difference between the
during fermentation within immobilization matrix, and maintenance and growth modes. In fact, in the biofilm
the consequent hindered contact between cell reactor, the maintenance SA production rate is predom-
and substrate. inant and is related to higher reducing power availabil-
The selection of the optimized reactor configuration ity for the SA producing pathway. It has been
should take into account the specific production rate hypothesized that the additional redox power is gener-
and the operating simplicity [84]. In particular, for a bio- ated by the pentose phosphate pathway in the form of
film reactor the specific production rate may be NADPH, subsequently converted to NADH by a transhy-
increased with the biofilm concentration. Therefore, drogenase [85] (an enzyme present in A. succinogenes).
particular attention must be paid to biofilm formation, The hypothesis that the oxidative pentose phos-
growth, detachment, and stability. phate pathway (OPPP) could serve as a source of add-
Biofilms are used in various types of reactor such as: itional reduction power in continuous glucose and
CSTR, packed bed (PBR), trickling bed (TBR), FBR, airlift xylose fermentations of A. succinogenes biofilms was
reactors (ALR), upflow anaerobic sludge blanket (UASB), subsequently demonstrated by Bradfield and Nicol [68]
and expanded bed reactors. Operating conditions in the following investigation.
depend on reactor typology. Maharaj et al. [63] operated a continuous biofilm
reactor using PoraverV beads as the packing and A. suc-
R
A summary of literature on the use of different
immobilized culture systems for the fermentation of cinogenes as the biocatalyst. The maximum volumetric
various substrates to produce SA is given in Table 3. productivity obtained was 10.8 g/Lh, corresponding to
The first contribution to this area was from a research the dilution rate 0.7 h1. Their process was character-
group at Iowa State University, United States. A plastic ized by high stability and repeatability when compared
composite material was used as a support for A. succi- to previous similar studies. Moreover, it was observed
nogenes biofilm growth in both batch and continuous that specific SA productivity (per mass of biomass)
fermentation. Urbance et al. [59] reported the produc- decreased with increasing concentrations of SA, which
tion of 10.4 g/L of SA concentrations in continuous bio- was attributed to possible changes in the composition
film fermentations at the dilution rate of 0.2 h1 (see of the biofilm or fraction of the metabolically active
Table 3), although optimal results were obtained with a cells within the biofilm.
10 M. FERONE ET AL.

Apart from pure glucose, xylose [67, 70] and xylose- fouling problems in such a system, in addition to lactic
enriched hydrolysate [66] were also used as a carbon acid bacteria contamination.
source to produce SA by A. succinogenes in biofilm reac- In work by Meynial-Salles et al. [76], the use
tors. Xylose was successfully converted to SA by A. suc- of a membrane bioreactor for cell recycle was
cinogenes biofilms, even though the yields and reported. Under the best cultural conditions, SA prod-
productivities were lower than those obtained in a pre- uctivity of 14.8 g/Lh was obtained using wild type
vious study on glucose. When xylose-rich corn stover Anaerobiospirillum succiniproducens.
hydrolysate was used a feedstock to produce SA, a
maximum concentration, yield and productivity of
Integrated fermentation and recovery systems
39.6 g/L, 0.78 gSA/gS, and 1.77 g/Lh were achieved,
respectively. As stated in the “Batch fermentation” section, high SA
In studies by Corona-Gonzalez et al. [64] both concentrations in the fermentation broth caused inhib-
entrapment and adsorption techniques for immobiliza- ition phenomena. One way to overcome this issue
tion of A. succinogenes were evaluated using different would be to remove the inhibitory product(s) directly
support materials. It was found that entrapment in agar from the ongoing fermentation by liquid–liquid extrac-
beads was the best immobilization technique, because tion, membranes, or adsorption.
the fermentation time was significantly reduced, while In the previously mentioned study by Meynial-Salles
yield, productivity, and the SA final concentration were et al. [76], these authors also investigated the perform-
increased in repeated batch fermentations. A fibrous ance of the cell-recycle reactor combined with an elec-
bed reactor was used for the production of SA during trodialysis system to remove organic acids in situ, the
continuous operation in a study by Yan et al. [62]. A SA aim being to increase the final product concentration.
concentration of 55.3 g/L, productivity of 2.77 g/Lh, With this system, very high SA concentration (83 g/L)
and yield of 0.8 gSA/gG were obtained, by feeding 80 g/ and a productivity of 10.4 g/Lh was obtained.
L glucose at the dilution rate set at 0.05 h1. Although the productivity was slightly lower compared
In a recent investigation conducted by Li et al. [71] to the membrane reactor alone, the in situ electrodialy-
the production of bio-succinate from glycerol by immo- sis system allowed to reduction of the products inhib-
bilized Yarrowia lipolytica cells in a fibrous bed bioreac- ition phenomena and obtains steady SA production at
tor was carried out. SA concentration up to 52 g/L and high concentrations.
productivities up to 1.5 g/Lh were achieved. Li et al. [44] developed an integrated system for SA
production and in situ product removal (ISPR). SA was
recovered using an expanded-bed adsorption (EBA)
Cell recycling and bleeding
separation process with an anion exchange resin. In
Compared to bioreactors where cells are immobilized, fed-batch fermentation, the maximum SA titer, yield
cell-recycling fermenters are advantageous due to and productivity obtained were: 145.2 g/L, 0.76 g/g, and
enhanced broth homogeneity, which facilitates diffu- 2.58 g/Lh, respectively. Besides improved fermentation
sion and total microorganism recycling [86]. performance, this integrated system also allowed over-
Lee et al. [77] used a membrane reactor system for coming of growth inhibition by products and to reduce
cell-recycling cultures of A. succiniciproducens and the total number of downstream processing steps.
studied the fermentation performance and the physio-
logical responses to a high-cell-density environment.
Bio-based production of SA: commercial status
The maximum SA productivity was about 3.3 g/Lh, a
and perspectives
value that is consistently higher than those obtained in
batch cultures with the same bacterium. There are currently four producers of bio-based SA –
An external membrane cell recycle reactor was used Myriant, BioAmber, Succinity, and Reverdia (Figure 1) –
by Kim et al. [72], to improve cell concentrations and which have formed joined ventures with several other
SA productivity with two different succinate-producing commercial entities.
bacteria, Actinobacillus succinogenes and Mannheimia Myriant is employing an engineered E. coli strain for
succiniproducens. The productivity increased up to the production of SA and it has been operating a 15
6.6 gSA/Lh for A. succinogenes (5 times higher than ktonnes capacity plant since 2013.
batch fermentation) and around 3.0 gSA/Lh for M. suc- BioAmber has initially selected E. coli as producing
ciniproducens (2.3 times higher). However, the higher organism, but more recently it appears to have
cell concentration achieved also lead to membrane switched to Candida krusei as current SA producer. It
 CRITICAL REVIEWS IN BIOTECHNOLOGY 11

Figure 1. Bio-based succinic acid production plants: (A) Myriant in United States; (B) BioAmber in Canada; (C) Succinity in Spain;
and (D) Reverdia in Italy.

has been operating a 30 ktonnes capacity plant since a proprietary S. cerevisiae strain with low pH fermenta-
2014. Their evolved strain (Pichia kudriavzevii 13723) tion technology. The engineered yeast strain produced
produced 48.2 g/L of SA from glucose with the yield 43 g/L of SA employing aerobic conditions, using dual
and productivity of 0.69 mol/mol glucose and 0.97 g/ phase fed- batch fermentation at pH 3.
Lh, respectively, by aerobic batch fermentation at pH These bio-based succinate-production facilities are
3 [87]. targeting specialized markets, but large-scale produc-
Succinity (a joint venture between BASF & Corbion- tion for bulk chemical synthesis is still a distant pro-
Purac, Gorinchem, Netherlands) is employing an engi- spect. To produce bio-based SA as a competitive
neered B. succiniciproducens strain for the production of platform chemical, its production cost should be
SA. It has been operating a 10 ktonne capacity plant around 1 $/kg of acid [91]. This implies that: 1 – the
since 2013. This wild-type strain – characterized by a investments in equipment as well as the operating
high yield of 0.75 mol of SA per mol of glucose – was costs of the industrial production process need to
further optimized through metabolic flux analysis (MFA) be extremely low; 2 – the bioreactors must be charac-
and subsequent metabolic engineering [88]. The engi- terized by high throughput. Therefore, the use of
neered strain obtained used glycerol and maltose to facultative anaerobic microorganisms – e.g. A. succino-
produce 70 g/L of SA with the yield and productivity of genes – can reduce bioreactor costs due to the absence
1.7 mol/mol and 2.91 g/Lh, respectively, by anaerobic of aeration that increases significantly capital and
fed-batch fermentation [89,90] operating costs.
Reverdia (a joint venture between Roquette and In Figure 2 is shown a map of succinic productivity
DSM) started production in a commercial plant with vs. concentration, similar to that proposed by Setlhaku
about 10 ktonnes per year in December 2012. They use et al. [92] for the biobutanol production process. This
12 M. FERONE ET AL.

Figure 2. Process performances diagram: SA productivity vs. SA concentration of the main studies on succinic acid production by
fermentation using different process configurations.

map includes batch, fed-batch and continuous bioreac- performance of the process toward productivity and
tor systems, with free-cell or immobilized cells and a concentration targets (Tables 1–4) .
couple of examples of fermentation and product recov-
ery integrated systems, as reported in the open litera-
ture. The minimum SA concentration and productivity Conclusions
requirement for its fermentation production on a large
scale are based on those obtained for glutamic acid: SA has received increased attention due to its potential
150 g/L and 5 g/Lh [4]. as a building block for the production of a number of
Cells immobilization or confinement has been dem- industrially valuable chemicals. Several companies have
onstrated to be a good strategy to keep the productiv- started producing bio-succinate by using proprietary
ity at values close or even higher than this target, organisms and strains, and sugars mainly derived from
Moreover, it has the additional advantage of facilitating corn or glycerol as a feedstock. However, these bio-
cell reuse. The concentration target can be reached based succinate-production facilities are targeting spe-
only if a recovery process is integrated into the fermen- cialized markets, while its large-scale production it is
tation system, as confirmed in Figure 2. This is most not yet being used to support bulk chemical markets.
likely because the process is characterized by product Parameters that influence the viability of the biopro-
inhibition and high concentrations of the acid cause cess are yield, concentration, and productivity. To keep
complete inhibition of cell growth, substrate consump- the productivity at reasonable values, a potential solu-
tion, and acid production [37]. Despite it is well-known tion is the confinement of cells in the reactor in order
that biofilms are characterized by higher tolerance to increase the cell concentration.
toward toxic compounds, using this strategy alone can- This review highlighted that continuous fermenta-
not overcome the inhibition problem [93]. tion process, in particular immobilized-cells reactors,
All these considerations suggest that a continuous coupled with ISPR, are more efficient than traditional
fermentation process with immobilized cells, coupled batch production and it deserves more attention in
with an in situ recovery system, might push the the future.
 CRITICAL REVIEWS IN BIOTECHNOLOGY 13

Considering the advances in efficient downstream [10] Kuhnert P, Scholten E, Haefner S, et al. Basfia succinici-
processes, strain development, reduced feedstock costs, producens gen. nov., sp. nov., a new member of the
family Pasteurellaceae isolated from bovine rumen. Int
reduced greenhouse gas emissions, it is expected that a
J Syst Evol Microbiol. 2010;60:44–50.
sustainable and economically viable technology for bio- [11] Kamzolova SV, Vinokurova NG, Shemshura ON, et al.
based SA production can be established in the The production of succinic acid by yeast Yarrowia lipo-
near future. lytica through a two-step process. Appl Microbiol
Biotechnol. 2014;98:7959–7969.
[12] Arikawa Y, Kuroyanagi T, Shimosaka M, et al. Effect of
Disclosure statement gene disruptions of the TCA cycle on production of
succinic acid in Saccharomyces cerevisiae. J Biosci
The authors report no declarations of interest.
Bioeng. 1999;87:28–36.
[13] Nagaraja TG. Microbiology of the Rumen.
Rumenology. Switzerland: Springer, Cham; 2016. p.
Funding
39–61.
The study was supported by the Ministero dell’Istruzione, [14] Samuelov NS, Datta R, Jain MK, et al. Whey fermenta-
delll’Universit
a e della Ricerca project “Development of green tion by Anaerobiospirillum succiniciproducens for pro-
technologies for production of BIOchemicals and their use in duction of a succinate-based animal feed additive.
preparation and industrial application of POLImeric materials Appl Environ Microbiol. 1999;65:2260–2263.
from agricultural biomasses cultivated in a sustainable way [15] Bradfield MFA, Nicol W. Continuous succinic acid pro-
in Campania Region – BIOPOLIS” PON03PE_00107_1/1. duction by Actinobacillus succinogenes in a biofilm
reactor: steady-state metabolic flux variation. Biochem
Eng J. 2014;85:1–7.
References [16] Salvachua D, Smith H, St. John PC, et al. Succinic acid
production from lignocellulosic hydrolysate by Basfia
[1] Bozell JJ, Petersen GR. Technology development for
succiniciproducens. Bioresour Technol. 2016;214:
the production of biobased products from biorefinery
558–566.
carbohydrates—the US department of energy’s “top
[17] Van der Werf MJ, Guettler MV, Jain MK, et al.
10” revisited. Green Chem. 2010;12:525–728. Environmental and physiological factors affecting the
Available from: http://xlink.rsc.org/?DOI¼b922014c succinate product ratio during carbohydrate fermenta-
[2] Zeikus JG, Jain MK, Elankovan P. Biotechnology of suc- tion by Actinobacillus sp. 130Z. Arch Microbiol. 1997;
cinic acid production and markets for derived indus- 167:332–342.
trial products. Appl Microbiol Biotechnol. 1999;51: [18] Qureshi N, Blaschek HP. ABE production from corn: a
545–552. recent economic evaluation. J Ind Microbiol
[3] MarketResearch.com [Internet]. New York (NY) Biotechnol. 2001;27:292–297.
MarketsandMarkets. [cited 2018 Aug 31] Available [19] Wang R, Koppram R, Olsson L, et al. Kinetic modeling
from: https://www.marketresearch.com of multi-feed simultaneous saccharification and co-fer-
[4] McKinlay JB, Vieille C, Zeikus JG. Prospects for a bio- mentation of pretreated birch to ethanol. Bioresour
based succinate industry. Appl Microbiol Biotechnol. Technol. 2014;172:303–311.
2007;76:727–740. [20] Song H, Lee SY. Production of succinic acid by bacter-
[5] Samuelov NS, Lamed R, Lowe S, et al. Influence of ial fermentation. Enzyme Microb Technol. 2006;39:
CO2-HCO3 levels and pH on growth, succinate pro- 352–361.
duction, and enzyme activities of Anaerobiospirillum [21] Beauprez JJ, De Mey M, Soetaert WK. Microbial suc-
succiniciproducens. Appl Environ Microbiol. 1991;57: cinic acid production: natural versus metabolic engi-
3013–3019. neered producers. Process Biochem. 2010;45:
[6] Guettler MV, Rumler D, Jain MK. Actinobacillus succino- 1103–1114.
genes sp. nov., a novel succinic-acid-producing strain [22] Yin X, Li J, Shin H, et al. Metabolic engineering in the
from the bovine rumen. Int J Syst Bacteriol. 1999;49: biotechnological production of organic acids in the
207–216. tricarboxylic acid cycle of microorganisms: advances
[7] Litsanov B, Brocker M, Bott M. Toward homosuccinate and prospects. Biotechnol Adv. 2015;33:830–841.
fermentation: metabolic engineering of Corynebacterium [23] Ahn JH, Jang YS, Lee SY. Production of succinic acid
glutamicum for anaerobic production of succinate from by metabolically engineered microorganisms. Curr
glucose and formate. Appl Environ Microbiol. 2012;12: Opin Biotechnol. 2016;42:54–66.
3325–3337. [24] Kurzrock T, Weuster-Botz D. Recovery of succinic acid
[8] Jantama K, Zhang X, Moore JC, et al. Eliminating side from fermentation broth. Biotechnol Lett. 2010;32:
products and increasing succinate yields in engi- 331–339.
neered strains of Escherichia coli C. Biotechnol Bioeng. [25] Lopez-Garzo n CS, Straathof A. Recovery of carboxylic
2008;101:881–893. acids produced by fermentation. Biotechnol Adv.
[9] Lee P, Lee S, Hong S, et al. Isolation and characteriza- 2014;32:873–904.
tion of a new succinic acid-producing bacterium, [26] Akhtar J, Idris A, Abd. Aziz R. Recent advances in pro-
Mannheimia succiniciproducens MBEL55E, from bovine duction of succinic acid from lignocellulosic biomass.
rumen. Appl Microbiol Biotechnol. 2002;58:663–668. Appl Microbiol Biotechnol. 2014;98:987–1000.
14 M. FERONE ET AL.

[27] Tan JP, Md. Jahim J, Wu TY, et al. Insight into biomass [42] Scholten E, D€agele D. Succinic acid production by a
as a renewable carbon source for the production of newly isolated bacterium. Biotechnol Lett. 2008;30:
succinic acid and the factors affecting the metabolic 2143–2146.
flux toward higher succinate yield. Ind Eng Chem Res. [43] Chen K, Jiang M, Wei P, et al. Succinic acid production
2014;53:16123–16134. from acid hydrolysate of corn fiber by Actinobacillus
[28] Pateraki C, Patsalou M, Vlysidis A, et al. Actinobacillus succinogenes. Appl Biochem Biotechnol. 2010;160:
succinogenes: advances on succinic acid production 477–485.
and prospects for development of integrated biorefi- [44] Li Q, Wang D, Hu G, et al. Integrated bioprocess for
neries. Biochem Eng J. 2016;112:285–303. high-efficiency production of succinic acid in an
[29] Litsanov B, Brocker M, Oldiges M, et al. Succinic acid. expanded-bed adsorption system. Biochem Eng J.
In: Kondo VSB, editor. Bioprocessing of renewable 2011;56:150–157.
resources to commodity bioproducts. Hoboken (NJ): [45] Bretz K, Kabasci S. Feed-control development for suc-
JohnWiley & Sons, Inc; 2014. p. 435–472. cinic acid production with Anaerobiospirillum succinici-
[30] Yuzbashev TV, Yuzbasheva EY, Sobolevskaya TI, et al. producens. Biotechnol Bioeng. 2012;109:1187–1192.
Production of succinic acid at low pH by a recombin- [46] Zheng P, Zhang K, Yan Q, et al. Enhanced succinic
ant strain of the aerobic yeast Yarrowia lipolytica. acid production by Actinobacillus succinogenes after
Biotechnol Bioeng. 2010;107:673–682. genome shuffling. J Ind Microbiol Biotechnol. 2013;40:
[31] Xiao H, Shao Z, Jiang Y, et al. Exploiting Issatchenkia 831–840.
orientalis SD108 for succinic acid production. Microb [47] Bai B, Zhou J. min, Yang M. hua, et al. Efficient pro-
Cell Fact. 2014;13:121–132. duction of succinic acid from macroalgae hydrolysate
[32] Castro RCdA, Mussatto SI, Roberto IC. A vertical ball by metabolically engineered Escherichia coli. Bioresour
mill as a new reactor design for biomass hydrolysis Technol. 2015;185:56–61.
and fermentation process. Renew Energ. 2017;114: [48] Pateraki C, Almqvist H, Ladakis D, et al. Modelling suc-
775–780. cinic acid fermentation using a xylose based substrate.
[33] Gunnarsson IB, Karakashev D, Angelidaki I. Succinic Biochem Eng J. 2016;114:26–41.
acid production by fermentation of Jerusalem arti- [49] Marinho GS, Alvarado-Morales M, Angelidaki I.
choke tuber hydrolysate with Actinobacillus succino- Valorization of macroalga Saccharina latissima as
genes 130Z. Ind Crops Prod. 2014;62:125–129.
novel feedstock for fermentation-based succinic acid
[34] Wan C, Li Y, Shahbazi A, et al. Succinic acid produc-
production in a biorefinery approach and economic
tion from cheese whey using Actinobacillus succino-
aspects. Algal Res. 2016;16:102–109.
genes 130 Z. Appl Biochem Biotechnol. 2008;145:
[50] Alvarado-Morales M, Gunnarsson IB, Fotidis IA, et al.
111–119.
Laminaria digitata as a potential carbon source for
[35] Nghiem NP, Davison BH, Suttle BE, et al. Production
succinic acid and bioenergy production in a biorefi-
of succinic acid by Anaerobiospirillum succiniciprodu-
nery perspective. Algal Res. 2015;9:126–132.
cens. Appl Biochem Biotechnol. 1997;63:565–576.
 [51] Ferone M, Raganati F, Olivieri G, et al. Biosuccinic Acid
[36] Corona-Gonzalez RI, Bories A, Gonzalez-Alvarez V,
from lignocellulosic-based hexoses and pentoses by
et al. Kinetic study of succinic acid production by
Actinobacillus succinogenes: characterization of the
Actinobacillus succinogenes ZT-130. Process Biochem.
2008;43:1047–1053. conversion process. Appl Biochem Biotechnol. 2017;
[37] Lin SKC, Du C, Koutinas A, et al. Substrate and prod- 183:1465–1477.
uct inhibition kinetics in succinic acid production by [52] McKinlay JB, Shachar-Hill Y, Zeikus JG, et al.
Actinobacillus succinogenes. Biochem Eng J. 2008;41: Determining Actinobacillus succinogenes metabolic
128–135. pathways and fluxes by NMR and GC-MS analyses of
[38] Du C, Lin SKC, Koutinas A, et al. A wheat biorefining 13C-labeled metabolic product isotopomers. Metab
strategy based on solid-state fermentation for fermen- Eng. 2007;9:177–192.
tative production of succinic acid. Bioresour Technol. [53] Ferone M, Raganati F, Olivieri G, et al. Continuous suc-
2008;99:8310–8315. cinic acid fermentation by actinobacillus succino-
[39] Vlysidis A, Du C, Webb C, et al. Experimental and genes: assessment of growth and succinic acid
modelling studies of the bioconversion of glycerol to production kinetics. Appl Biochem Biotechnol. 2018;
succinic acid by actinobacillus succinogenes. 178. DOI:10.1007/s12010-018-2846-8
AicheConfexCom [Internet]. 2008. Available from: [54] Vlysidis A, Binns M, Webb C, et al. Glycerol utilisation
http://aiche.confex.com/aiche/2008/webprogram/ for the production of chemicals: conversion to suc-
Paper140136.html cinic acid, a combined experimental and computa-
[40] Liu YP, Zheng P, Sun ZH, et al. Strategies of pH con- tional study. Biochem Eng J. 2011;58–59:1–11.
trol and glucose-fed batch fermentation for produc- [55] Lee PC, Lee SY, Chang HN. Kinetic study of organic
tion of succinic acid by Actinobacillus succinogenes acid formations and growth of Anaerobiospirillum suc-
CGMCC1593. J Chem Technol Biotechnol. 2008;83: ciniciproducens during continuous cultures. J Microbiol
722–729. Biotechnol. 2009;19:1379–1384.
[41] Song H, Jang SH, Park JM, et al. Modeling of batch [56] Lee PC, Lee SY, Chang HN. Kinetic study on succinic
fermentation kinetics for succinic acid production by acid and acetic acid formation during continuous cul-
Mannheimia succiniciproducens. Biochem Eng J. 2008; tures of Anaerobiospirillum succiniciproducens grown
40:107–115. on glycerol. Bioprocess Biosyst Eng. 2010;33:465–471.
 CRITICAL REVIEWS IN BIOTECHNOLOGY 15

[57] Cheng KK, Zhao XB, Zeng J, et al. Downstream proc- recycle system. J Microbiol Biotechnol. 2009;19:
essing of biotechnological produced succinic acid. 1369–1373.
Appl Microbiol Biotechnol. 2012;95:841–850. [73] Lee PC, Lee WG, Kwon S, et al. Batch and continuous
[58] Chen KQ, Li J, Ma JF, et al. Succinic acid production cultivation of Anaerobiospirillum succiniciproducens for
by Actinobacillus succinogenes using hydrolysates of the production of succinic acid from whey. Appl
spent yeast cells and corn fiber. Bioresour Technol. Microbiol Biotechnol. 2000;54:23–27.
2011;102:1704–1708. [74] Lee PC, Lee SY, Hong SH, et al. Batch and continuous
[59] Urbance SE, Pometto AL, DiSpirito AA, et al. cultures of mannheimia succiniciproducens MBEL55E
Evaluation of succinic acid continuous and repeat- for the production of succinic acid from whey and
batch biofilm fermentation by Actinobacillus succino- corn steep liquor. Bioprocess Biosyst Eng. 2003;26:
genes using plastic composite support bioreactors. 63–67.
Appl Microbiol Biotechnol. 2004;65:664–670. [75] Kim DY, Yim SC, Lee PC, et al. Batch and continuous
[60] van Heerden CD, Nicol W. Continuous succinic acid fermentation of succinic acid from wood hydrolysate
fermentation by Actinobacillus succinogenes. Biochem by Mannheimia succiniciproducens MBEL55E. Enzyme
Eng J. 2013;73:5–11. Microb Technol. 2004;35:648–653.
[61] Yan Q, Zheng P, Dong JJ, et al. A fibrous bed bioreac- [76] Meynial-salles I, Dorotyn S, Soucaille P. A new process
tor to improve the productivity of succinic acid by for the continuous production of succinic acid from
Actinobacillus succinogenes. J Chem Technol glucose at high yield, titer, and productivity.
Biotechnol. 2014;89:1760–1766. Biotechnol Bioeng. 2008;99:129–135.
[62] Yan Q, Zheng P, Tao ST, et al. Fermentation process [77] Lee PC, Lee SY, Chang HN. Cell recycled culture of
for continuous production of succinic acid in a fibrous succinic acid-producing Anaerobiospirillum succinicipro-
bed bioreactor. Biochem Eng J. 2014;91:92–98. ducens using an internal membrane filtration system. J
[63] Maharaj K, Bradfield MFA, Nicol W. Succinic acid-pro- Microbiol Biotechnol. 2008;18:1252–1256.
ducing biofilms of Actinobacillus succinogenes: repro- [78] Oh IJ, Lee HW, Park CH, et al. Succinic acid production
ducibility, stability and productivity. Appl Microbiol from continuous fermentation process using
Biotechnol. 2014;98:7379–7386. Mannheimia succiniciproducens LPK7. J Microbiol
[64] Corona-Gonzalez RI, Miramontes-Murillo R, Arriola- Biotechnol. 2008;18:908–912. Available from: http://
Guevara E, et al. Immobilization of Actinobacillus succi- www.ncbi.nlm.nih.gov/pubmed/18633290
nogenes by adhesion or entrapment for the produc- [79] Scholten E, Renz T, Thomas J. Continuous cultivation
tion of succinic acid. Bioresour Technol. 2014;164: approach for fermentative succinic acid production
113–118. from crude glycerol by Basfia succiniciproducens DD1.
[65] Brink HG, Nicol W. Succinic acid production with Biotechnol Lett. 2009;31:1947–1951.
Actinobacillus succinogenes: rate and yield analysis of [80] van Heerden CD, Nicol W. Continuous and batch cul-
chemostat and biofilm cultures. Microb Cell Fact. tures of Escherichia coli KJ134 for succinic acid fer-
2014;13:111. mentation: metabolic flux distributions and
[66] Bradfield MFA, Mohagheghi A, Salvach ua D, et al. production characteristics. Microb Cell Fact. 2013;12:
Continuous succinic acid production by Actinobacillus 80–90.
succinogenes on xylose-enriched hydrolysate. [81] Russo ME, Natale FD, Prigione V, et al. Adsorption of
Biotechnol Biofuels. 2015;8:181. acid dyes on fungal biomass: equilibrium and kinetics
[67] Bradfield MFA, Nicol W. Continuous succinic acid pro- characterization. Chem Eng J. 2010;162:537–545.
duction from xylose by Actinobacillus succinogenes. [82] Kumar A, Dewulf J, Langenhove H. Van Membrane-
Bioprocess Biosyst Eng. 2015;8:181. based biological waste gas treatment. Chem Eng J.
[68] Bradfield MFA, Nicol W. The pentose phosphate path- 2008;136:82–91.
way leads to enhanced succinic acid flux in biofilms [83] Qureshi N, Annous BA, Ezeji TC, et al. Biofilm reactors
of wild-type Actinobacillus succinogenes. Appl for industrial bioconversion processes: employing
Microbiol Biotechnol. 2016;100:9641–9652. potential of enhanced reaction rates. Microb Cell Fact.
[69] Herselman J, Bradfield MFA, Vijayan U, et al. The effect 2005;24:1–21.
of carbon dioxide availability on succinic acid produc- [84] Raganati F, Procentese A, Olivieri G, et al. Butanol pro-
tion with biofilms of Actinobacillus succinogenes. duction by clostridium acetobutylicum in a series of
Biochem Eng J. 2017;117:218–228. packed bed biofilm reactors. Chem Eng Sci. 2016;152:
[70] Ferone M, Ercole A, Olivieri G, et al. Continuous suc- 678–688.
cinic acid fermentation by Actinobacillus succinogenes [85] McKinlay JB, Laivenieks M, Schindler BD, et al. A gen-
in a packed bed reactor. Biotechnol Biofuels. 2018;11: omic perspective on the potential of Actinobacillus
138. Available from: https://doi.org/10.1186/s13068- succinogenes for industrial succinate production. BMC
018-1143-7 Genomics. 2010;11:680.
[71] Li C, Yang X, Gao S, et al. High efficiency succinic acid [86] Qureshi N, Schripsema J, Lienhardt J, et al. Continuous
production from glycerol via in situ fibrous bed bio- solvent production by Clostridium beijerinckii BA101
reactor with an engineered Yarrowia lipolytica. immobilized by adsorption onto brick. World J
Bioresour Technol. 2017;225:9–16. Microbiol Biotechnol. 2000;16:337–382.
[72] Kim MI, Kim NJ, Shang L, et al. Continuous production [87] Finley KR, Huryta M, Mastel BM, et al. Patent cargill-
of succinic acid using an external membrane cell bioamber pichia. US 20130302866. 2013. p. 237.
 16 M. FERONE ET AL.

[88] Becker J, Reinefeld J, Stellmacher R, et al. Systems- European Commission, contract No.ENER/C2/423-
wide analysis and engineering of metabolic pathway 2012/SI2.673791. 2015.
fluxes in bio-succinate producing Basfia succiniciprodu- [92] Setlhaku M, Heitmann S, Go rak A, et al. Investigation
cens. Biotechnol Bioeng. 2013;110:3013–3023. of gas stripping and pervaporation for improved feasi-
[89] Schroder H, Haefner S, Von Abendroth G, et al. BASF bility of two-stage butanol production process.
SE Microbial succinic acid producers and purification Bioresour Technol. 2013;136:102–108. Available from:
of succinic acid. United States Patent US8673598B2 http://dx.doi.org/10.1016/j.biortech.2013.02.046
2014 Mar 18. [93] van der Pol E, Springer J, Vriesendorp B, et al.
[90] Scholder H, Haefner S, Abendroth G, et al. Bacterial Precultivation of Bacillus coagulans DSM2314 in the
cells having a glyoxylate shunt for the manufacture of presence of furfural decreases inhibitory effects of
succinic acid. BASF SE United States Patent lignocellulosic by-products during l(þ)-lactic acid fer-
US8877466B2 2014 Nov 4. mentation. Appl Microbiol Biotechnol. 2016;100:
[91] E4tech, RE-CORD and WUR From the Sugar Platform 10307–10319. Available from: http://dx.doi.org/10.
to biofuels and biochemicals”.Final report for the 1007/s00253-016-7725-z

View publication stats