Biosynthesis of isocitric acid in repeated-batch culture and testing of its stress-protective activity

Article abstract of DOI:10.1007/s00253-019-09729-8

Biosynthesis of Ds(+)-threo-isocitric acid from ethanol in the Yarrowia lipolytica batch and repeated-batch cultures was studied. Repeated-batch cultivation was found to provide for a good biosynthetic efficiency of the producer for as long as 748?h, probably due to maintenance of high activities of enzymes involved in the biosynthesis of isocitric acid. Under optimal repeated-batch cultivation conditions, the producer accumulated 109.6?g/L Ds(+)-threo-isocitric acid with a production rate of 1.346?g/L?h. The monopotassium salt of isocitric acid isolated from the culture liquid and purified to 99.9% was found to remove neurointoxication, to restore memory, and to improve the learning of laboratory rats intoxicated with lead and molybdenum salts. Taking into account the fact that the neurotoxic effect of heavy metals is mainly determined by oxidative stress, the aforementioned favorable action of isocitric acid on the intoxicated rats can be explained by its antioxidant activity among other pharmacological effects.

Full text of DOI:10.1007/s00253-019-09729-8

Applied Microbiology and Biotechnology
https://doi.org/10.1007/s00253-019-09729-8
APPLIED MICROBIAL AND CELL PHYSIOLOGY
Biosynthesis of isocitric acid in repeated-batch culture and testing
of its stress-protective activity
1
1
2,3
4
2
Igor G. Morgunov
& Svetlana V. Kamzolova & Olga V. Karpukhina & Svetlana B. Bokieva & Anatoly N. Inozemtsev
Received: 4 January 2019 /Revised: 18 February 2019 /Accepted: 25 February 2019
Springer-Verlag GmbH Germany, part of Springer Nature 2019
#
Abstract
Biosynthesis of Ds(+)-threo-isocitric acid from ethanol in the Yarrowia lipolytica batch and repeated-batch cultures was studied.
Repeated-batch cultivation was found to provide for a good biosynthetic efficiency of the producer for as long as 748 h, probably
due to maintenance of high activities of enzymes involved in the biosynthesis of isocitric acid. Under optimal repeated-batch
cultivation conditions, the producer accumulated 109.6 g/L Ds(+)-threo-isocitric acid with a production rate of 1.346 g/L h. The
monopotassium salt of isocitric acid isolated from the culture liquid and purified to 99.9% was found to remove
neurointoxication, to restore memory, and to improve the learning of laboratory rats intoxicated with lead and molybdenum
salts. Taking into account the fact that the neurotoxic effect of heavy metals is mainly determined by oxidative stress, the
aforementioned favorable action of isocitric acid on the intoxicated rats can be explained by its antioxidant activity among other
pharmacological effects.
.
.
.
.
Keywords Microbial synthesis Ds(+)-threo-isocitric acid Repeated-batch cultivation Antioxidants Stress-protective activity
Introduction
There is an ever-increasing search for nontoxic original inno-
*
Igor G. Morgunov
vative compounds which can be produced by microbial synthe-
sis at a relatively large scale and which can find wide applica-
tion in medicine, food industry, and agriculture, that is, where
there are high requirements to the purity of such compounds, to
their stereoisomeric and isomeric composition, and to their abil-
ity to be metabolized in the cells to carbon dioxide and water
without accumulation (Stottmeister et al. 2005; Sauer et al.
2008; Groenewald et al. 2014; Zinjarde 2014; Papanikolaou
et al. 2017; Kawasaki and Ueda 2017; Morgunov et al. 2017;
Koller 2018). Many countries even prohibit the relevant usage
of chemically synthesized compounds.
morgunovs@rambler.ru
Svetlana V. Kamzolova
kamzolova@rambler.ru
Olga V. Karpukhina
karpukhina.msu@yandex.ru
Svetlana B. Bokieva
bokievasb@rambler.ru
Anatoly N. Inozemtsev
a_inozemtsev@mail.ru
1
2
G.K. Skryabin Institute of Biochemistry and Physiology of
Microorganisms, Federal Research Center BPushchino Center for
Biological Research of the Russian Academy of Sciences^, Prospect
Nauki 5, Pushchino, Moscow Region 142290, Russia
According to data available in the literature (Demain and
Martens 2017), up to 22,500 of biologically active compounds
are produced by microbiological synthesis. Among them,
there is Ds(+)-threo-isocitric acid or (1R,2S)-1-hydroxy-
Department of Higher Nervous Activity, Faculty of Biology,
Lomonosov Moscow State University, 1-12 Leninskie Gory,
Moscow 119234, Russia
1
,2,3-propanetricarboxylic acid, abbreviated as ICA, pro-
duced by the yeast Yarrowia lipolytica. ICA can be used (1)
as a raw material for the production of ascorbic acid, (2) as a
biochemical marker of some diseases, (3) as a therapeutic
agent in the treatment of Fe-deficiency anemia, (4) for the
resorption of blood clots, and (5) for treatment of drug and
3
4
Semenov Institute of Chemical Physics, Russian Academy of
Sciences, 4 Kosygina str, Moscow 119991, Russia
Khetagurov North Ossetian State University, 44-46 Vatutina str,
Vladikavkaz, North Ossetia 362025, Russia

Appl Microbiol Biotechnol
alcohol abuse (Finogenova et al. 2005; Aurich et al. 2012,
017; Heretsch et al. 2008; Bullin et al. 2018). Recently, we
also found that Ds(+)-threo-ICA is a good protector of damage
caused by hydrogen peroxide and salts of some heavy metals
substrate for growth of producers of practically valuable com-
pounds has several advantages over other substrates. It does
not contain harmful impurities, it is well assimilated by yeast
and dissolves in water in any proportions. Since ethanol is
used in the human diet, the products derived from it do not
require additional purification from the residual substrate. As
reported by Weusthuis et al. (2011), the several companies in
the USA and Switzerland created the food products based on
microbial biomass produced from ethanol. In our experiments
with the natural strain Y. lipolytica VKM Y-2373 with the
employment of a metabolic inhibitor—itaconic acid, we
reached the ICA concentration up to 90.5 g/L (Kamzolova
et al. 2018; Morgunov et al. 2018).
It should be noted that the known microbiological process-
es of ICA production are based on batch cultivation, when
microbial growth is limited by a deficiency of nutrients in
the medium and acid formation is inhibited by metabolites
gradually accumulated in the culture. As a result, after 5–
6 days of the fermentation process, its continuation becomes
unreasonable.
2
(
Cu, Pb, Zn, and Cd) in H O in the infusorian Paramecium
2 2
caudatum, being even more efficient than the classical antiox-
idant ascorbic acid (Morgunov et al. 2018). The antioxidant
properties of ICA need further studies because of a wide dis-
tribution of heavy metals in the environment and their adverse
biological effects, in particular, the disorder of mental func-
tions, such as learning and memory (Karpukhina et al. 2014;
Sripetchwandee et al. 2014; Inozemtsev et al. 2016; Karri
et al. 2016; Kulikova et al. 2016; Inozemtsev et al. 2017).
As found in the aforementioned works, the neurotoxicosis
caused by heavy metals inhibits learning and memory proba-
bly because of the excessive formation of active oxygen spe-
cies (in other words, heavy metals induce oxidative stress),
which damage cellular membranes and cause many dysfunc-
tions. It is the antioxidant activity of ICA that is probably
responsible for its neuroprotective effect on the learning and
memory inhibition in the laboratory rats intoxicated with
heavy metals.
The application of ICA as an antioxidant is limited by the
fact that chemically synthesized ICA represents an inseparable
mixture of four isomers, of which only Ds(+)-threo-ICA is a
naturally occurring central metabolite of all living cells,
whereas the other three isomers are nonnatural inhibitory
compounds (Finogenova et al. 2005). Thus, chemically syn-
thesized ICA cannot be used in the food and pharmaceutical
industries. It should be noted that the natural Ds(+)-threo-ICA
produced by Sigma-Aldrich (USA) for the assay of isocitrate
dehydrogenase is isolated from the juice of the specially cul-
tivated Sedum spectabile plant and has a sale price of 538
EUR per 1 g of monopotassium salt. Such a high price makes
this product unaffordable for medical applications.
The aim of this work was to optimize the process of ICA
production by Y. lipolytica in the repeated-batch mode and to
study the protective effect of ICA on the learning and memory
inhibition in the laboratory rats subjected to the neurotoxic
action of heavy metals.
Materials and methods
Microorganisms and reagents
Experiments were carried out with the yeast strain Y. lipolytica
VKM Y-2373, which was selected earlier and showed its abil-
ity to synthesize ICA from ethanol in sufficient amounts
(Kamzolova et al. 2018; Morgunov et al. 2018). All chemicals
were of analytical grade (Mosreactiv, Russia). Ethanol was
purchased from the BKupavnareaktiv^ (Russia). Ammonium
molybdate (para) tetrahydrate (Alfa Aesar, USA) and lead (II)
acetate trihydrate (Acros Organics, USA) were used in the
testing with ICA.
It is known that Ds(+)-threo-ICA can also be produced by
microbial synthesis. Some researchers selected efficient ICA-
producing Y. lipolytica strains from natural sources, while
others used for this purpose classical mutagenesis and gene
engineering (Förster et al. 2007; Heretsch et al. 2008; Holz
et al. 2009; Aurich et al. 2012, 2017; Kamzolova et al. 2013,
2
016, 2018; Liu et al. 2015; Morgunov et al. 2018; Timoumi
Batch cultivation
et al. 2018; Hu et al. 2019; Rzechonek et al. 2019). In the
above studies, it has been established that Y. lipolytica pro-
duced ICA and citric acid (CA) simultaneously, and the ratio
of excreted ICA to CA greatly depends on the carbon source
used for cultivation. For example, when the yeast is cultivated
on glucose and glycerol, it mainly produces CA, whereas
when grown on n-alkanes, ethanol, and fatty acids, it produces
ICA and CA in approximately equal amounts.
Batch cultivation was carried out in a 10-L ANKUM-2M
fermentor manufactured in a town of Pushchino (Russia).
The fermentor had 5 L of cultivation medium containing the
following (g/L): (NH ) SO , 3.0; МgSO ·7H O, 1.4;
4
2
4
4
2
Ca(NO ) , 0.8; NaCl, 0.5; KH PO , 2.0; K HPO , 0.2; double
3
2
2
4
2
4
amount of microelements (Burkholder et al. 1944); and yeast
autolysate BDifco,^ 0.5.
At present, ethanol is considered a promising source of
carbon, as it can be obtained from sugar cane, beet, maize,
lignocellulose, and other renewable sources. Ethanol as a
The cultivation temperature and the concentration of dis-
solved oxygen were 29.0 ± 0 .1 °C and 60% of saturation,
respectively. The pH of the medium was monitored and

Appl Microbiol Biotechnol
maintained at a level of 6.0 by adding the necessary amount of
and memory (Bures et al. 1983), was developed in rats by
application of stimuli for 5 days (25 trials per day).
Experiments were carried out using a 60 × 30 × 30-cm shuttle
box divided into two equal compartments by a wall with a 10 ×
10-cm hole. The conditioned stimulus was an acoustic signal
lasting 10 s. Whenever the rat crossed from one compartment to
the other, that is, made avoidance within this 10-s interval, the
sound was switched off and the electrical shock was not ap-
plied. Otherwise, the electrical shock was applied in the com-
partment occupied by the animal. The electrical shock caused
motor reaction and the rat crossed to the opposite compartment,
thus exhibiting shock escape response. After that, both stimuli
were terminated. The interstimulus interval was 30 s. CAAR
was evaluated as the percentage of successful avoidances rela-
tive to the number of applied stimuli.
2
0% KOH solution. The growth substrate ethanol was added
as it was consumed from the medium.
Continuous repeated-batch cultivation
Repeated-batch cultivation was performed according to the
following scheme: after 72 h of batch cultivation, the medium
was refreshed by 20% at intervals of 24 h (5 cycles); then by
5
0% at 48-h intervals (4 cycles) and 72-h intervals (2 cycles);
and then by 80% at 72-h intervals (2 cycles). Finally, the
medium was refreshed by 50% and cultivation was continued
for 72 h. Such cultivation regime provided for the maximum
concentration of ICA in the medium.
Analytical methods
On the fifth day of training, when avoidance behavior be-
came well established, CAAR was deliberately smashed. For
this purpose, the stimuli were not switched off even if the rat
crossed to the other compartment in response to the stimulus
(this was repeated successively five times) so that the rat was
exposed to current shocks regardless of the correct reaction.
After the fifth crossing, the electrical current was switched
off immediately and the sound 2 s later. Then, CAAR was
estimated as usually in 20 trials. As the effect of the smash
was most pronounced immediately after it, we compared the
results of the five last trials before the smash and the five first
trials after it.
The assay of biomass, nitrogen, ICA, and CA and the calcu-
lation of the specific growth rate (μ), process productivity
(Q ), specific rate of ICA synthesis (q ), and product yield
p p
(Yp) were described in detail earlier (Kamzolova et al. 2013).
Methods for determining activities of citrate synthase,
aconitate hydratase, NAD- and NADP-dependent isocitrate
dehydrogenases, isocitrate lyase were described in detail
(
Morgunov et al. 2013).
Testing of ICA
The results were expressed as the arithmetic means of
avoidance responses with the standard deviation of the mean.
The dynamics of training in the groups was estimated using
the Kruskal-Wallis test, and the difference between the groups
was estimated using the Wilcoxon test. The differences were
considered to be statistically significant at p < 0.05.
Experiments on laboratory animals were carried out according
to the rules of European Convention for the Protection of
Vertebrate Animals used for Experimental and Other
Scientific Purposes (Council of Europe 1986) and of
Directive 2010/63/EU of the European Parliament and of the
Council of 22 September 2010 on the protection of animals
used for scientific purposes.
Experiments were performed with outbred white male rats
divided into six groups of ten animals each. The body weight
of the rats was between 250 and 370 g. The rats were pur-
chased from the Central Nursery of Laboratory Animals
BStolbovaya^ (Russia). The rats were kept in a vivarium at
Results
Fed-batch cultivation
Figure 1 shows the dynamics of nitrogen consumption and the
accumulation of biomass, ICA, and CA in the Y. lipolytica
VKM Y-2373 culture grown under the condition of growth
limitation by nitrogen (panel a), as well as the logarithmic
growth curve, specific growth rate (μ), and the specific rate
of ICA biosynthesis (qICA) (panel b). The logarithmic growth
curve exhibited a clear-cut exponential phase from 0 to 12 h of
cultivation, growth retardation phase from 12 to 72 h, and
stationary phase (from 72 to 144 h). After 6 h of cultivation,
the specific growth rate (μmax) reached the maximum value
2
2–24 °С with free access to water and food.
Tested substances (ICA, heavy metals, and water) were
injected intraperitoneally. Rats of control group 1 were given
sterile distilled water (2 mL) 1 h before the start of the training
session. Members of group 2 were given ICA in a dose of
2
0 mg/kg. Groups 3 and 4 were given 2 mL of aqueous solu-
−7
tions of lead diacetate (10 М) and ammonium molybdate
−5
(
10 М), respectively, 5 h before the start of the training
session. Groups 5 and 6 were given lead diacetate and ammo-
nium molybdate as described above for groups 3 and 4 and
then (after 4 h) ICA in the same dose as for group 2.
−1
0.231 h . ICA appeared in the medium only in the growth
retardation phase (after 12 h of cultivation) caused by the
exhaustion of a nitrogen source. The accumulation of ICA
continued in the stationary phase until the medium was
The conditioned reaction of active avoidance response
(CAAR), which served as the experimental model of learning

Appl Microbiol Biotechnol
Fig. 1 Fig. 1 Time courses of
growth, nitrogen consumption,
and ICA and CA production by Y.
lipolytica VKM Y-2373 (a) and
the calculated parameters of the
process (b): I, exponential phase;
II, growth retardation phase; and
III, stationary phase
biomass (X)
N
ICA
CA
a
2
0
105
9
0
5
16
7
1
2
8
4
0
60
45
3
1
0
0
5
0
24
48
72
Time (h)
96
120
144
b
log X
µ
qp
I
II
III
2
1
0
.5
0.3
0.2
0.1
0
2
.5
1
.5
0
0
24
48
72
96
120
144
-0.5
Time (h)
supplied with the growth substrate ethanol. After 144 h of
cultivation, the concentration of ICA in the medium was
Y. lipolytica cells in the phase of active growth (12 h of culti-
vation) showed high activities of enzymes of the initial oxida-
tion of ethanol, that is, alcohol dehydrogenase and aldehyde
dehydrogenase (0.070 and 0.060 U/mg protein, respectively).
The activity of citrate synthase, which is involved in the for-
mation of CA, was considerably higher than that of the sub-
sequent enzymes of the tricarboxylic acid (TCA) cycle,
aconitate hydratase and NAD-dependent isocitrate dehydro-
genase (by 3.4 and 20 times, respectively). The high activity
of isocitrate lyase (0.200 U/mg protein) indicated the opera-
tion of the glyoxylate cycle upon the assimilation of ethanol.
In the phase of retarded growth (caused by a deficiency of
nitrogen in the medium) and active biosynthesis of ICA (24,
48, and 72 h of cultivation), alcohol dehydrogenase, aldehyde
dehydrogenase, and citrate synthase remained active. The ac-
tivity of aconitate hydratase (this enzyme is responsible for
ICA formation in the TCA cycle) increased by 1.3 times,
whereas the activity of isocitrate lyase (enzyme responsible
for ICA oxidation in the glyoxylate cycle) decreased by ap-
proximately 2.2 times. In this case, the activity of NAD-
dependent isocitrate dehydrogenase, which oxidizes ICA in
the TCA cycle, remained at a relatively high level of 0.118–
0.121 U/mg protein. In the stationary phase (120 and 144 h of
9
1.5 g/L. The biosynthesis of ICA was accompanied by for-
mation of the only byproduct CA at a concentration of 23.8 g/
L, the ICA to CA ratio being 3.8:1. The average productivity
of the fermentation process calculated with allowance for the
medium dilution with the KOH solution used for the control
of pH turned out to be 0.896 g/L h. The yield of the target
product (YICA) was 0.75 g/g ethanol.
As seen from Fig. 1b, the specific rate of ICA biosynthesis
(qICA) was maximum from 21 to 72 h of cultivation and com-
prised 0.095–0.104 g/g h. From 72 to 144 h, qICA gradually
decreased by approximately 4.5 times.
To find the reason of lower culture productivity during
batch cultivation, the major enzymes of ethanol and ICA me-
tabolism were assayed in the cell-free homogenate in the
course of cultivation. Cells for analysis were withdrawn in
the phase of active growth (12 h of cultivation), in the phase
of growth retardation and active ICA formation (24, 48, and
7
2 h), and in the phase of stationary growth and slow ICA
formation (120 and 144 h). The results of the assay of enzyme
activities in the Y. lipolytica VKM Y-2373 cells grown in the
batch mode are shown in Table 1. As seen from Table 1,

Appl Microbiol Biotechnol
Table 1 Dynamics of enzyme activities (U/mg protein) in Y. lipolytica cells cultivated in the batch mode
Enzymes
Time (h)
1
2
24
48
72
120
144
Alcohol dehydrogenase
Aldehyde dehydrogenase
Citrate synthase
0.070 ± 0.015
0.060 ± 0.010
2.210 ± 0.170
0.650 ± 0.130
0.110 ± 0.010
0.080 ± 0.02
0.060 ± 0.010
2.000 ± 0.160
0.73 ± 0.150
0.121 ± 0.011
0.075 ± 0.020
0.060 ± 0.010
1.800 ± 0.230
0.880 ± 0.160
0.118 ± 0.020
0.052 ± 0.010
0.060 ± 0.010
1.800 ± 0.230
0.880 ± 0.160
0.120 ± 0.031
0.014 ± 0.010
0.070 ± 0.010
0.900 ± 0.080
0.530 ± 0.119
0.060 ± 0.020
0.010 ± 0.010
0.070 ± 0.010
0.840 ± 0.080
0.450 ± 0.120
0.020 ± 0.010
Aconitate hydratase
NAD-dependent isocitrate
dehydrogenase
Isocitrate lyase
0.200 ± 0.060
0.100 ± 0.020
0.087 ± 0.010
0.080 ± 0.015
0.067 ± 0.015
0.040 ± 0.010
growth), all key enzymes of the ethanol and ICA metabolism
became less active (alcohol dehydrogenase by 7.2 times, cit-
rate synthase by 2.2 times, aconitate hydratase by 1.8 times,
NAD-dependent isocitrate dehydrogenase by 6 times, and
isocitrate lyase by 2.2 times) as compared with their activities
in the phase of active acid formation (24, 48, and 72 h of
growth).
To analyze the activity of the major enzymes involved in
the metabolism of the growth substrate ethanol and of the
target product ICA, microbial cells were sampled during the
batch period (48 h of cultivation) and in the course of different
stationary phases of the repeated-batch cultivation period
(192, 384, 528, 672, and 748 h of cultivation). The results of
the enzyme assay in the cell-free homogenate of Y. lipolytica
VKM Y-2373 are summarized in Table 2. As seen from
Table 2, the activities of alcohol dehydrogenase, aldehyde
dehydrogenase, citrate synthase, aconitate hydratase, NAD-
dependent isocitrate dehydrogenase, and isocitrate lyase
remained at a high level during the whole cultivation period.
By the end of the repeated-batch cultivation (Table 2, 748 h),
the activities of almost all assayed enzymes, except aldehyde
dehydrogenase, were higher than those by the end of the batch
cultivation (Table 1, 144 h). Namely, alcohol dehydrogenase
was more active by 7 times, citrate synthase by 2.1 times,
aconitate hydratase by 1.7 times, NAD-dependent isocitrate
dehydrogenase by 5 times, and isocitrate lyase by 1.5 times.
Table 3 shows data on the mean concentration of biomass
and ICA in the culture broth, specific rate of ICA biosynthesis
(q ), process productivity (Q ), and ICA yield relative to the
Repeated-batch cultivation
With the repeated-batch cultivation technique, a part of the
culture liquid is withdrawn from the fermentor and the same
volume of fresh cultivation medium which contains all neces-
sary for growth nutrients, including nitrogen, is added to the
fermentor. This procedure is repeated at regular intervals.
In experiments with the ICA-producing Y. lipolytica VKM
Y-2373 strain, we varied the volume of replaced liquid and the
cycle duration. Figure 2 shows the dynamics of biomass and
ICA during the continuous repeated-batch cultivation of the
ICA producer. As seen from Fig. 2, even after 748 h of culti-
vation, the concentration of ICA in the broth was as high as
p
p
1
03 g/L.
ethanol consumption (Yp) in different regimes of repeated-
biomass
ICA
Fig. 2 Biosynthesis of ICA in the
repeated-batch culture of
Y. lipolytica
1
1
20
00
80
60
40
20
0
0
48
96
144
192
240
288
336
384
Time (h)
432
480
528
576
624
672
720

Appl Microbiol Biotechnol
Table 2 Dynamics of enzyme activities (U/mg protein) in Y. lipolytica cells cultivated in the repeated-batch mode
Enzymes
Time (h)
4
8
192
384
528
672
748
Alcohol dehydrogenase
Aldehyde dehydrogenase
Citrate synthase
0.075 ± 0.020
0.060 ± 0.010
1.800 ± 0.230
0.880 ± 0.160
0.118 ± 0.020
0.052 ± 0.010
0.060 ± 0.010
1.630 ± 0.170
0.800 ± 0.220
0.120 ± 0.031
0.060 ± 0.015
0.060 ± 0.010
1.850 ± 0.100
0.790 ± 0.200
0.110 ± 0.020
0.080 ± 0.020
0.075 ± 0.010
1.700 ± 0.200
0.810 ± 0.200
0.110 ± 0.020
0.070 ± 0.020
0.060 ± 0.010
1.630 ± 0.200
0.800 ± 0.200
0.110 ± 0.020
0.070 ± 0.020
0.060 ± 0.010
1.780 ± 0.200
0.750 ± 0.200
0.110 ± 0.020
Aconitate hydratase
NAD-dependent isocitrate
dehydrogenase
Isocitrate lyase
0.087 ± 0.010
0.080 ± 0.015
0.077 ± 0.015
0.057 ± 0.010
0.057 ± 0.010
0.060 ± 0.010
batch cultivation. When the volume of replaced liquid was
0% with the cycle duration of 72 h, the concentration of
Testing of ICA in model experiments on learning
and memory in rats
5
ICA, the process productivity (Qp), and the ICA yield (Yp)
reached the maximum values of 109.6 g/L, 1.346 g/L h, and
The effect of ICA and heavy metals was studied in experi-
ments on the development of reaction of active avoidance
response (CAAR) in laboratory rats. The results are shown
in Fig. 3.
0
.8 g/g ethanol, respectively. The increase in the volume of
replaced liquid from 50 to 80% diminished the ICA concen-
tration by 1.5 times, process productivity (Q ) by 1.2 times,
p
and product yield by 20%, which probably was related to
excessive cell renewal. When the replacement cycle duration
was decreased from 72 to 48 h with the optimal volume of
replaced liquid equal to 50%, the ICA concentration slightly
decreased by 1.3 times, whereas the process productivity (Qp)
As seen from Fig. 3, ICA favorably influenced the training
of rats within the first 2 days, whereas then its influence be-
came negligible. The peculiarity of this period is that the an-
imals are exposed to more electric shocks than in other days.
For example, the control rats showed less than 10% of avoid-
ance reactions during the first training day (that is, they re-
ceived at least nine electrical shocks of the ten possible). On
the third day of training and later, the number of CAAR in-
creased and number of electrical shocks respectively
diminished.
As seen from Fig. 3, the heavy metals under study inhibited
the learning of the rats, lead being more inhibitory than mo-
lybdenum. In the first experimental day, half of the animals
injected with lead diacetate did not made a single correct re-
sponse; in the next 2 days, there was a slight acceleration of
learning, and then it began to slow down. As a result, on the
last fifth day of training, the rats received more than 20 elec-
trical shocks of the 25 possible. These results testify in favor
of the strong inhibition of learning and memory by lead. In
contrast to lead, molybdenum exerted an insignificant inhibi-
tory action on the learning response. Indeed, ammonium
and the product yield (Y ) practically did not change. The
p
decrease in the volume of replaced liquid from 50 to 20%
diminished the ICA concentration by 1.7 times, process pro-
ductivity (Q ) by 1.2 times, and product yield (Y ) by 15%,
p
p
which probably was related to an insufficient level of cell
renewal. It should be noted that the specific rate of ICA syn-
thesis (qICA) remained relatively high (within 0.065–0.102 g/
g h) during the whole period of repeated-batch cultivation. As
for the major byproduct CA, its content did not exceed 25% of
the sum of ICA and CA irrespective of the cultivation mode
used.
For further experiments on laboratory animals, ICA was
isolated from the culture liquid in the form of monopotassium
salt (Morgunov et al. 2018). The isolated and purified product
represented white crystalline preparation containing 99.9%
ICA.
Table 3 Biosynthesis of ICA by Y. lipolytica VKM Y-2373 cultivated in the repeated-batch mode
Volume of
Cycle duration (h)
Number of cycles
Biomass (g/L)
ICA (g/L)
Q
p
(g/L h)
qICA (g/g h)
p
Y (g/g)
replaced liquid (%)
20
50
50
80
24
48
72
72
5
4
3
2
11.20 ± 0.67
10.33 ± 0.59
9.17 ± 0.35
10.25 ± 0.35
63.60 ± 3.13
87.20 ± 0.82
109.60 ± 5.82
71.40 ± 0.85
1.09
0.065
0.102
0.094
0.086
0.68
0.74
0.80
0.64
1.300
1.346
1.119

Appl Microbiol Biotechnol
control
ICA
Pb
ICA+Pb
Mo
ICA+Mo
Fig. 3 Effect of ICA and heavy
metal salts on the development of
the reaction of active avoidance
response (CAAR)
1
00
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
1
2
3
Trial days
4
5
molybdate decreased the number of CAAR relative to the
control rats only during the three last days of training, its effect
being even positive in the first training day.
running, vocalization, and urination. The decrease in the num-
ber of avoidances after SAR and the increase in the number of
ISR in the presence of ICA were small (10.9% and 50%,
respectively) and statistically insignificant.
As seen from Fig. 3, ICA prevented the neurotoxic action
of the heavy metals on learning and memory. Indeed, the
number of CAAR in the rats given a combined injection of
ICA with lead and molybdenum salts was higher than in the
case of injection of the heavy metal salts without ICA.
Especially significant was the effect of ICA against the nega-
tive influence of lead. Thus, on the last day, rats coexposed to
ICA and lead performed an average of 4 times more CAAR
than rats without ICA. In the first training day, lead acetate
reduced the favorable effect of ICA on the learning and mem-
ory of rats. At the same time, the number of CAAR in the rats
injected with both ICA and ammonium molybdate was higher
than that in the control rats injected with water, in the rats
injected with ICA alone, and in the rats injected with molyb-
denum alone. These data are a strong indication that molyb-
denum, in contrast to lead, enhances the neuroprotective effect
of ICA.
The effect of ICAwas further investigated by analyzing the
reversible functional failure of CAAR caused by the unexpect-
ed change of causal relationships between the stimuli, re-
sponse, and its effect, named as Bsmash of avoidance
responses^ (Inozemtsev 2009). As seen from Table 4, the
smash of avoidance responses (SAR) caused a considerable
failure of the developed skill in the control rats so that the
number of CAAR after the smash was by 31.3% lower than
that before the smash. SAR also drastically increased the num-
ber of aimless motor reactions in animals so that the rats run to
the opposite chamber compartment not only in response to the
stimuli (for example, electrical shock) but also spontaneously.
This type of motor activity is named as intersignal responses
Discussion
This study shows that the batch experiments indicated that
ICA is actively produced within 72 h, and then the formation
of isocitrate decreases (Fig. 1). These data are in good agree-
ment with our earlier data (Kamzolova et al. 2013, 2016,
2018; Morgunov et al. 2018), as well as with the results of
other authors (Förster et al. 2007; Holz et al. 2009; Aurich
et al. 2012, 2017; Rzechonek et al. 2019), showing the exis-
tence of two clear-cut phases of ICA biosynthesis in the yeast
Y. lipolytica with different acid formation rates.
The data, presented in Table 1, clearly show that the de-
crease in the culture productivity (q ) during batch cultivation
p
well coincides with the decrease in the activity of the enzymes
involved in the ethanol oxidation and ICA biosynthesis. Makri
et al. (2010) also observed a decrease in enzyme activities, in
particular, NAD-dependent isocitrate dehydrogenase activity
to minimal levels during the citric acid production phase.
Table 4 Effect of ICA on the number of CAAR and ISR before and
after smash
Substance
Before smash
After smash
CAAR
Control (H
2
O)
O)
98.0 ± 6.3
92.0 ± 14.0
ISR
66.0 ± 28.4*
82.0 ± 14.8
ICA
(
ISR).
Besides the described intersignal responses in the form of
Control (H
ICA
2
6.0 ± 13.5
24.0 ± 24.6
54.0 ± 15.2*
36.0 ± 37.5
running from one chamber compartment to the other, the rats
exhibited a variety of other reactions typical to stressed ani-
mals, such as multiple jumps to the chamber ceiling, chaotic
Data are the mean and standard deviation
*Stands for p < 0.05 relative to the number before smash

Appl Microbiol Biotechnol
The low activities of enzymes, involved in metabolism of
ICA, can be related to the aging of microbial cells because of
inhibition of their growth by nitrogen limitation. Some mod-
ification of the cultivation technique presumably can provide
for microbial cell renewal. Recent research into this problem
showed promise of repeated-batch cultivation in this respect.
The advantage of this type of cultivation has been demonstrat-
ed on the CA-producing Y. lipolytica strains (Anastassiadis
and Rehm 2006; Arzumanov et al. 2000; Makri et al. 2010;
Rywińska and Rymowicz 2010; Rymowicz et al. 2010;
Moeller et al. 2011; Kamzolova et al. 2015).
Testing of ICA in the model of learning and memory showed
that it increased the number of CAAR relative to the control
value but only at the beginning of the experiment (Fig. 3),
when the control animals received the maximum number of
electric shocks. As shown earlier (Berezhnoy et al. 2016),
CAAR development induced both pain stress and oxidative
one; meanwhile, the antioxidant carnosine diminished the ox-
idative stress and enhanced learning and memory. Therefore,
the positive effect of ICA in our experiment can be explained
by its stress-protective properties. For this reason, when on the
third day of training and later, the number of CAAR increased
and the number of electrical shocks respectively diminished,
the favorable effect of ICA disappeared (Fig. 3).
As seen from Table 4, SAR diminished the number of
CAAR; meanwhile, it increased the value of ISR by nine times.
As shown elsewhere (Inozemtsev 2009), this behavior indi-
cates emotional stress development. ICA prevents the sharp
decrease in the number of CAAR and the increase in the num-
ber of ISR. Consequently, our data indicate that ICA counter-
acts emotional stress. Since anxiolytics prevent the develop-
ment of such responses (Inozemtsev et al. 1996), it should be
concluded that ICA prevents the development of emotional
stress and the failure of learning and memory caused by SAR.
As seen from Fig. 3, ammonium molybdate and lead ace-
tate inhibited the learning of the rats. This result is in agree-
ment with our earlier data (Inozemtsev et al. 2017). Many
researchers consider that the main cause of neurotoxic activity
of heavy metals is the oxidative stress which they induce (Jan
et al. 2015). According to earlier research, cadmium induces
oxidative stress in rats and in the cell culture, whose develop-
ment is prevented by the antioxidant carnosine (Kulikova
et al. 2016). ICA also promotes the survival of infusoria im-
paired by heavy metals, such as Cu, Pb, Zn, and Cd
(Morgunov et al. 2018). Therefore, the prevention of the neu-
rotoxic effect of the heavy metals by ICA shown in this paper
can reasonably be explained by its antioxidant activity.
Thus, the testing of ICA produced by microbial synthesis
in the model of learning and memory showed its efficiency in
the prevention of pain stress induced by electrical shock on the
first stage of CAAR development, of emotional stress induced
by smash of avoidance responses, and of oxidative stress
caused by lead acetate and ammonium molybdate.
Our data, presented in Fig. 2, show that the production of
ICA by the repeated-batch fermentation of ethanol is more
effective than that by batch fermentation. During the continu-
ous fermentation process, its productivity varied from 1.09 to
1
1
.346 g/L h and the ICA accumulation values from 63.6 to
09.60 g/L (Table 3). Moreover, the activities of enzymes
involved in the synthesis of ICA remained at a high level
during the whole cultivation period (Table 2).
It should be noted that the repeated-batch microbiological
process of ICA production developed by us does not have
analogs in the literature and that its parameters can be com-
pared only with those of the batch processes of ICA produc-
tion. There is evidence that the natural strain Y. lipolytica
EH59 cultivated in the batch mode can accumulate in the
medium up to 93 g/L ICA; however, this advantage is deteri-
orated by the high content of the byproduct CA (82.3 g/L)
(
Heretsch et al. 2008). Aurich et al. (2017) described the effi-
cient process of ICA production from rapeseed oil by the
recombinant strain Y. lipolytica with the superexpressed
aconitate hydratase gene АСО1. That process is characterized
by the following parameters: the ICA concentration equal to
6
8.4 g/L, the process productivity (Q ) equal to 0.47 g/L h,
p
and the ICA:CA ratio equal to 3.1:1. Citric acids were also
manufactured from biodiesel waste using the recombinant
Y. lipolytica AWG7 strain with the superexpressed Gut1 and
Gut2 genes. The achieved ICA concentration was 42.5 g/L
and the product cost was relatively low (Rzechonek et al.
2
019). In our experiments with the natural, mutant, and genet-
ically modified microbial strains with the employment of such
metabolic inhibitors as itaconic and oxalic acids, we reached
the ICA concentration from 70.6 to 90.5 g/L depending on the
carbon source used for fermentation (either rapeseed oil or
ethanol) (Kamzolova et al. 2013, 2016, 2018; Morgunov
et al. 2018). The process productivity in the aforementioned
works was between 0.97 and 1.15 g/L h, while in the present
process (by 50% at 72-h intervals), the process productivity
Funding information The reported research was funded by the Russian
Foundation for Basic Research and the government of the Moscow
Region of the Russian Federation, grant № 17-48-500446.
Compliance with ethical standards
(Qp) consisted of 1.346 g/L h.
In the present study, we continued to study the pharmaceu-
Conflict of interest The authors declare that they have no conflict of
interest.
tical properties of biosynthetic ICA. Earlier, we found that
ICA is a promising natural compound for prevention of oxi-
dative stress induced by hydrogen peroxide or heavy metals in
the infusorian Paramecium caudatum (Morgunov et al. 2018).
Ethical statement All applicable international, national, and/or institu-
tional guidelines for the care and use of animals were followed.

Appl Microbiol Biotechnol
Publisher’s note Springer Nature remains neutral with regard to jurisdic-
tional claims in published maps and institutional affiliations.
Yarrowia lipolytica. Appl Microbiol Biotechnol 81:1087–1096.
https://doi.org/10.1007/s00253-008-1725-6
Hu W, Li WJ, Yang HQ, Chen JH (2019) Current strategies and future
prospects for enhancing microbial production of citric acid. Appl
Microbiol Biotechnol 103(1):201–209. https://doi.org/10.1007/
s00253-018-9491-6
References
Inozemtsev AN (2009) Biological origins of protective mechanisms ac-
tivated by the disruption of higher nervous activity. Moscow Univ
Biol Sci 64:57–62. https://doi.org/10.3103/S0096392509020011
Inozemtsev AN, Bokieva SB, Voronina TA, Tushmalova NA (1996)
Comparative effects of tranquilizers and nootropes on the elabora-
tion and functional impairment of the avoidance response. Bull Exp
Biol Med 122:786–789. https://doi.org/10.1007/BF02445147
Inozemtsev AN, Bokieva SB, Karpukhina OV, Gumargalieva KZ,
Kamensky AA, Myasoedov NF (2016) Semax prevents learning
and memory inhibition by heavy metals. Dokl Biol Sci 468:112–
114. https://doi.org/10.1134/S0012496616030066
Inozemtsev AN, Bokieva SB, Karpukhina OV, Gumargalieva KZ,
Kamensky AA, Myasoedov NF (2017) Paradoxical influence of
combined effect of semax and ammonium molybdate on learning
and memory in rats. Mosc Univ Biol Sci Bull 72:151–154. https://
doi.org/10.3103/S0096392517030051
Anastassiadis S, Rehm HJ (2006) Citric acid production from glucose by
yeast Candida oleophila ATCC 20177 under batch, continuous and
repeated batch cultivation. Electron J Biotechnol 9:0–0:26–39.
https://doi.org/10.2225/vol9-issue1-fulltext-5
Arzumanov TE, Shishkanova NV, Finogenova TV (2000) Biosynthesis
of citric acid by Yarrowia lipolytica repeat-batch culture on ethanol.
Appl Microbiol Biotechnol 53:525–529. https://doi.org/10.1007/
s002530051651
Aurich A, Specht R, Müller RA, Stottmeister U, Yovkova V, Otto C, Holz
M, Barth G, Heretsch P, Thomas FA, Sicker D, Giannis A (2012)
Microbiologically produced carboxylic acids used as building
blocks in organic synthesis. In: Wang X, Chen J, Quinn P (eds)
Reprogramming microbial metabolic pathways, Subcellular bio-
chemistry, vol 64. Springer, Dordrecht, pp 391–423
Aurich A, Hofmann J, Oltrogge R, Wecks M, Glaser R, Blömer L,
Mauersberger S, Roland A, Müller RA, Sicker D, Giannis A
Jan AT, Azam M, Siddiqui K, Ali A, Choi I, Haq QM (2015) Heavy
metals and human health: mechanistic insight into toxicity and coun-
ter defense system of antioxidants. Int J Mol Sci 16:29592–29630.
https://doi.org/10.3390/ijms161226183
Kamzolova SV, Dedyukhina EG, Samoilenko VA, Lunina JN, Puntus IF,
Allayarov RK, Chiglintseva MN, Mironov AA, Morgunov IG
(2017) Improved isolation of microbiologically produced (2R,3S)-
isocitric acid by adsorption on activated carbon and recovery with
methanol. Org Process Res Dev 21:866–870. https://doi.org/10.
1
021/acs.oprd.7b00090
Berezhnoy DS, Fedorova TN, Stvolinskii SL, Inozemtsev AN (2016)
Carnosine modulates oxidative homeostasis and levels of neuro-
transmitters in the brain in models of learning with positive and
negative reinforcement. Neurochem J 10:273–279. https://doi.org/
(2013) Isocitric acid production from rapeseed oil by Yarrowia
lipolytica yeast. Appl Microbiol Biotechnol 97:9133–9144. https://
doi.org/10.1007/s00253-013-5182-5
Kamzolova SV, Vinokurova NG, Lunina JN, Zelenkova NF, Morgunov
IG (2015) Production of technical-grade sodium citrate from
glycerol-containing biodiesel waste by Yarrowia lipolytica.
Bioresour Technol 193:250–255. https://doi.org/10.1016/j.biortech.
1
0.1134/S1819712416040048
Bullin K, Hennig L, Herold R, Krautscheid H, Richter K, Sicker D (2018)
An optimized method for an (2R,3S)-isocitric acid building block.
Monatsh Chem 150:247–253. https://doi.org/10.1007/s00706-018-
2
015.06.092
2
319-3
Kamzolova SV, Allayarov RK, Lunina JN, Morgunov IG (2016) The
effect of oxalic and itaconic acids on threo-Ds-isocitric acid produc-
tion from rapeseed oil by Yarrowia lipolytica. Bioresour Technol
Bures JN, Buresova O, Huston JP (1983) Techniques and basic experi-
ments for the study of brain and behavior, 2nd edn. Elsevier,
Amsterdam, p 398
Burkholder P, McVeigh J, Moyer D (1944) Studies on some growth
factors of yeasts. J Bacteriol 48:385–391
Council of Europe (1986) European Convention for the Protection of
Vertebrate Animals Used for Experimental and Other Scientific
Purpose (ETS No. 123). Council of Europe: Strasbourg, France
Demain A, Martens E (2017) Production of valuable compounds by
molds and yeasts. J Antibiotics 70:347–360
2
06:128–133. https://doi.org/10.1016/j.biortech.2016.01.092
Kamzolova SV, Shamin RV, Stepanova NN, Morgunov GI, Lunina JN,
Allayarov RK, Samoilenko VA, Morgunov IG (2018) Fermentation
conditions and media optimization for isocitric acid production from
ethanol by Yarrowia lipolytica. Biomed Res Int 2018:2543210–
2
543219. https://doi.org/10.1155/2018/2543210
Karpukhina OV, Gumargalieva KZ, Bokieva SB, Inozemtsev AN (2014)
Heavy metals change effects of piracetam on learning and memory. J
Nat Sci Sustain Technol 8(3):439–444
Karri V, Schuhmacher M, Kumar V (2016) Heavy metals (Pb, Cd, As and
MeHg) as risk factors for cognitive dysfunction: a general review of
metal mixture mechanism in brain. Environ Toxicol Pharmacol 48:
Finogenova TV, Morgunov IG, Kamzolova SV, Chernyavskaya OG
(2005) Organic acid production by the yeast Yarrowia lipolytica: a
review of prospects. Appl Biochem Microbiol 41:418–425. https://
doi.org/10.1007/s10438-005-0076-7
Förster A, Jacobs K, Juretzek T, Mauersberger S, Barth B (2007)
Overexpression of the ICL1 gene changes the product ratio of citric
acid production by Yarrowia lipolytica. Appl Microbiol Biotechnol
2
03–213. https://doi.org/10.1016/j.etap.2016.09.016
Kawasaki H, Ueda K (2017) Microbial innovations in the world of food.
Biosci Biotechnol Biochem 81(1):48–53. https://doi.org/10.1080/
0
7
7:861–869. https://doi.org/10.1007/s00253-007-1205-4
9168451.2016.1248363
Groenewald M, Boekhout T, Neuvéglise C, Gaillardin C, van Dijck PV,
Wyss M (2014) Yarrowia lipolytica: safety assessment of an oleag-
inous yeast with a great industrial potential. Crit Rev Microbiol 40:
Koller M (2018) Biodegradable and biocompatible polyhydroxy-
alkanoates (PHA): auspicious microbial macromolecules for phar-
maceutical and therapeutic applications. Molecules 23(2):E362.
https://doi.org/10.3390/molecules23020362
Kulikova OI, Fedorova TN, Stvolinsky SL, Orlova VS, Inozemtsev AN
(2016) Carnosine prevents the development of oxidative stress un-
der the conditions of toxic action of cadmium. Mosc Univ Biol Sci
Bull 71:240–244
1
87–206. https://doi.org/10.3109/1040841X.2013.770386
Heretsch P, Thomas F, Aurich A, Krautscheid H, Sicker D, Giannis A
2008) Syntheses with a chiral building block from the citric acid
(
cycle: (2R,3S)-isocitric acid by fermentation of sunflower oil.
Angew Chem Int Ed Eng 47:1958–1960. https://doi.org/10.1002/
anie.200705000
Liu HH, Ji XJ, Huang H (2015) Biotechnological applications of
Yarrowia lipolytica: past, present and future. Biotechnol Adv
33(8):1522–1546. https://doi.org/10.1016/j.biotechadv.2015.07.010
Holz M, Förster A, Mauersberger S, Barth G (2009) Aconitase overex-
pression changes the product ratio of citric acid production by

Appl Microbiol Biotechnol
Makri A, Fakas S, Aggelis G (2010) Metabolic activities of biotechno-
logical interest in Yarrowia lipolytica grown on glycerol in repeated
batch cultures. Bioresour Technol 101(7):2351–2358. https://doi.
org/10.1016/j.biortech.2009.11.024
Rywińska A, Rymowicz W (2010) High-yield production of citric acid by
Yarrowia lipolytica on glycerol in repeated-batch bioreactors. J Ind
Microbiol Biotechnol 37(5):431–435. https://doi.org/10.1007/
s10295-009-0687-8
Moeller L, Grünberg M, Zehnsdorf A, Aurich A, Bley T, Strehlitz B
Rzechonek DA, Dobrowolski A, Rymowicz W, Mirończuk AM (2019)
Aseptic production of citric and isocitric acid from crude glycerol by
genetically modified Yarrowia lipolytica. Bioresour Technol 271:
340–344. https://doi.org/10.1016/j.biortech.2018.09.118
(
2011) Repeated fed-batch fermentation using biosensor online con-
trol for citric acid production by Yarrowia lipolytica. J Biotechnol
53(3–4):133–137. https://doi.org/10.1016/j.jbiotec.2011.03.013
1
Morgunov IG, Kamzolova SV, Lunina JN (2013) The citric acid produc-
tion from raw glycerol by Yarrowia lipolytica yeast and its regula-
tion. Appl Microbiol Biotechnol 97(16):7387–7397. https://doi.org/
Sauer M, Porro D, Mattanovich D, Branduardi P (2008) Microbial pro-
duction of organic acids: expanding the markets. Trends Biotechnol
2
6(2):100–108. https://doi.org/10.1016/j.tibtech.2007.11.006
1
0.1007/s00253-013-5054-z
Sripetchwandee J, Pipatpiboon N, Chattipakorn N, Chattipakorn S (2014)
Combined therapy of iron chelator and antioxidant completely re-
stores brain dysfunctioninduced by iron toxicity. PLoS One 9(1):
e85115. https://doi.org/10.1371/journal.pone.0085115
Morgunov IG, Kamzolova SV, Dedyukhina EG, Chistyakova TI, Lunina
JN, Mironov AA, Stepanova NN, Shemshura ON, Vainshtein MB
(2017) Application of organic acids for plant protection against phy-
topathogens. Appl Microbiol Biotechnol 101(3):921–932. https://
doi.org/10.1007/s00253-016-8067-6
Stottmeister U, Aurich A, Wilde H, Andersch J, Schmidt S, Sicker D
(2005) White biotechnology for green chemistry: fermentative 2-
Morgunov IG, Karpukhina OV, Kamzolova SV, Samoilenko VA,
Inozemtsev AN (2018) Investigation of the effect of biologically
active threo-Ds-isocitric acid on oxidative stress in Paramecium
caudatum. Prep Biochem Biotechnol 48(1):1–5. https://doi.org/10.
oxocarboxylic acids as novel building blocks for subsequent chem-
ical syntheses. J Ind Microbiol Biotechnol 32:651–664. https://doi.
org/10.1007/s10295-005-0254-x
Timoumi A, Guillouet SE, Molina-Jouve C, Fillaudeau L, Gorret N
1
080/10826068.2017.1381622
(
2018) Impacts of environmental conditions on product formation
and morphology of Yarrowia lipolytica. Appl Microbiol Biotechnol
02(9):3831–3848. https://doi.org/10.1007/s00253-018-8870-3
Papanikolaou S, Rontou M, Belka A, Athenaki M, Gardeli C,
Mallouchos A, Kalantzi O, Koutinas AA, Kookos IK, Zeng AP,
Aggelis G (2017) Conversion of biodiesel-derived glycerol into bio-
technological products of industrial significance by yeast and fungal
strains. Eng Life Sci 17:262–281
Rymowicz W, Fatykhova AR, Kamzolova SV, Rywinska A, Morgunov
IG (2010) Citric acid production from glycerol-containing waste of
biodiesel industry by Yarrowia lipolytica in batch, repeated batch,
and cell recycle regimes. Appl Microbiol Biotechnol 87:971–979.
https://doi.org/10.1007/s00253-010-2561-z
1
Weusthuis R, Aarts J, Sanders J (2011) From biofuel to bioproduct: is
bioethanol a suitable fermentation feedstock for synthesis of bulk
chemicals? Biofuels Bioprod Biorefin 5:486–494
Zinjarde SS (2014) Food-related applications of Yarrowia lipolytica.
Food Chem 152:1–10. https://doi.org/10.1016/j.foodchem.2013.
11.117