Content uploaded by Florbela Carvalheiro
Author content
All content in this area was uploaded by Florbela Carvalheiro
Content may be subject to copyright.
ORIGINAL PAPER
Mannitol production by lactic acid bacteria grown
in supplemented carob syrup
Florbela Carvalheiro •Patrı
´cia Moniz •
Luı
´s C. Duarte •M. Paula Esteves •
Francisco M. Gı
´rio
Received: 5 April 2010 / Accepted: 26 July 2010
ÓSociety for Industrial Microbiology 2010
Abstract Detailed kinetic and physiological character-
isation of eight mannitol-producing lactic acid bacteria,
Leuconostoc citreum ATCC 49370, L. mesenteroides
subsp. cremoris ATCC19254, L. mesenteroides subsp.
dextranicum ATCC 19255, L. ficulneum NRRL B-23447,
L. fructosum NRRL B-2041, L. lactis ATCC 19256, Lac-
tobacillus intermedius NRRL 3692 and Lb. reuteri DSM
20016, was performed using a carob-based culture med-
ium, to evaluate their different metabolic capabilities.
Cultures were thoroughly followed for 30 h to evaluate
consumption of sugars, as well as production of biomass
and metabolites. All strains produced mannitol at high
yields ([0.70 g mannitol/g fructose) and volumetric pro-
ductivities ([1.31 g/l h), and consumed fructose and glu-
cose simultaneously, but fructose assimilation rate was
always higher. The results obtained enable the studied
strains to be divided mainly into two groups: one for which
glucose assimilation rates were below 0.78 g/l h (strains
ATCC 49370, ATCC 19256 and ATCC 19254) and the
other for which they ranged between 1.41 and 1.89 g/l h
(strains NRRL B-3692, NRRL B-2041, NRRL B-23447
and DSM 20016). These groups also exhibited different
mannitol production rates and yields, being higher for the
strains with faster glucose assimilation. Besides mannitol,
all strains also produced lactic acid and acetic acid. The
best performance was obtained for L. fructosum NRRL
B-2041, with maximum volumetric productivity of 2.36
g/l h and the highest yield, stoichiometric conversion of
fructose to mannitol.
Keywords Carob Lactic acid bacteria Lactobacillus
Leuconostoc Mannitol production
Introduction
Mannitol, a naturally occurring polyol, is widely used in
medicine, and in pharmaceutical, food and chemical
industries. It can be produced by chemical, microbiological
or enzymatic processes [1,2]. Industrial chemical pro-
duction is based on hydrogenation of fructose/glucose
mixtures at high temperature and pressure using Raney-
nickel catalyst. This process yields a mixture of mannitol
and sorbitol, an isomer, which has less interesting proper-
ties and hence significantly lower market price. Moreover,
separation of mannitol and sorbitol is rather difficult. The
enzymatic process requires the use of redox co-factors
[1,3], which, in general, renders it unattractive. Because of
these problems, mannitol production by microbial route
has become attractive, even at the industrial scale.
Heterofermentative lactic acid bacteria (LAB) belonging
to the genera Lactobacillus,Leuconostoc and Oenococcus
produce mannitol from fructose [4,5] and are among the
most efficient microorganisms for mannitol production. In
general, they can specifically produce mannitol from
glucose/fructose mixtures, without making sorbitol as
by-product [2], making it unnecessary to use purified
substrates or complex processes for product purification.
Metabolically, under adequate oxygen availability condi-
tions, glucose can be used as an energy and carbon source,
and fructose as an electron acceptor, as it can be reduced to
This article is part of the BioMicroWorld 2009 Special Issue.
F. Carvalheiro (&)P. Moniz L. C. Duarte
M. P. Esteves F. M. Gı
´rio
Unidade de Bioenergia, LNEG-Laborato
´rio
Nacional de Energia e Geologia,
Estrada do Pac¸o do Lumiar, 22,
1649-038 Lisbon, Portugal
e-mail: florbela.carvalheiro@ineti.pt
123
J Ind Microbiol Biotechnol
DOI 10.1007/s10295-010-0823-5
mannitol by means of a specific mannitol dehydrogenase
(EC 1.1.1.67). Depending on the microorganism, up to
2 mol mannitol can be produced from 1 mol glucose (and
2 mol fructose), if sugar assimilation is simultaneous. This
will also lead to formation of lactic acid and acetic acid
and/or ethanol and carbon dioxide as other metabolic
products.
Carob (Ceratonia siliqua L.) is a Mediterranean
perennial tree producing pods that contain seeds (10%)
and sugar-rich pulp (90%), which exhibits higher sugar
content than sugar cane [6]. The pulp is currently an
under-utilised by-product of carob locust bean gum pro-
duction, and its main application is as animal feed, and to
a lesser extent in traditional bakery products and bever-
ages. Therefore higher-added-value applications are nee-
ded. Carob pulp mainly contains sucrose (the major
sugar), glucose and fructose that can be water extracted
and easily used to obtain sugar-rich syrups [7,8]. These
syrups are a cheap carbon source that has already been
used for formulation of culture media, e.g. for xanthan,
ethanol and citric acid production [9–11]. Given its sugar
content, a carob-based medium may be advantageous for
mannitol production.
In a previous work [12], 30 bacterial strains from genera
Lactobacillus,Leuconostoc and Weissella were screened
for mannitol production in carob-based syrups. From these,
a group of eight strains were identified as potential man-
nitol producers that must be studied deeply to unveil their
potential advantages and limitations concerning use as
industrial mannitol producers using carob syrup medium.
In this work, a detailed kinetic and physiological char-
acterisation of these eight strains using a carob-based cul-
ture medium was carried out, to evaluate their performance
and metabolic capabilities, and to select the most promis-
ing strain.
Materials and methods
Raw material
Carob pulp was obtained from a local de-seeding factory
(Chorondo e Filhos L.da, Loule
´, Portugal). Carob pods
were purposely collected to contain the two main varieties
grown in Portugal, Mulata and Galhosa (80% and 20%,
respectively) to produce a defined and representative lot.
The pods were industrially de-seeded into kibbles and then
stored at room temperature in burlap sacks. On average, the
kibbles present the following composition (g/100 g dry
weight): sucrose, 42.68; glucose, 8.54; fructose, 5.97;
pinitol, 5.22; lignin, 18.86; cellulose, 7.64; hemicellulose,
0.26; protein, 4.70; fat, 0.61; ash, 3.11; and others (by
difference), 2.41 [12].
Medium preparation
Carob syrup was obtained by aqueous extraction using an
orbital incubator (Infors Unitron HT, Switzerland), at 50°C
for 5 h as described elsewhere [12] and stored at -20°C.
Before use, precipitates were removed by centrifugation at
7,500 gfor 25 min in a Beckman Coulter centrifuge
(Fullerton, USA) followed by filtration using Whatman no.
41 filter paper.
Sucrose hydrolysis and sterilisation of carob syrup were
performed by correcting syrup pH to 3 using 5 M HCl,
followed by 15 min at 121°C in an autoclave. Carob syrups
were chemically characterised as described below.
The pH of hydrolysed carob syrups was corrected to 6.5
by addition of sterile NaOH. These hydrolysates were sup-
plemented with MRS nutrients, except ammonium citrate,
which was replaced by 50 mM sodium citrate buffer (pH
6.2) to keep pH values in a range acceptable for bacterial
growth. Besides sodium citrate, the final medium contained
(per litre): 10 g peptone, 8 g beef extract, 4 g yeast extract,
1 ml Tween 80, 5 g NaCH
3
COO, 0.2 g MgSO
4
, 0.05 g
MnSO
4
and 2 g K
2
HPO
4
. The supplements and citrate
buffer were sterilised separately (121°C, 15 min) and added
aseptically to the pH-corrected carob hydrolysates, yielding
a dilution of the original carob syrup hydrolysate of 1:2.
Microorganisms and culture conditions
Eight heterofermentative lactic acid bacteria (LAB) were
used in this study. The strains and the corresponding
growth temperatures are summarised in Table 1. Stock
cultures were maintained in liquid nitrogen, and inocula
were prepared using MRS growth medium. Pre-inoculum
was obtained by re-suspending the cells in MRS growth
medium and incubation at the required temperature for
48 h. For inocula preparation, 100-ml Erlenmeyer flasks
containing 50 ml MRS medium were inoculated with 1 ml
pre-inoculum. Both pre-inoculum and inoculum were
incubated without agitation.
One millilitre of 16 h inoculum culture was used to seed
100-ml Erlenmeyer flasks containing 50 ml culture medium.
All cultures were incubated in an orbital shaker (100 rpm)
for 30 h at the same temperature as the inoculum. Samples
were withdrawn at a specific time to determine cell growth,
consumption of sugars and formation of products.
All experiments were performed, at least, in duplicate.
Replicates always differed by less than 10%, and typically
by less than 5%.
Analytical methods
Glucose, fructose, mannitol and pinitol were analysed by
high-performance liquid chromatography (HPLC; Waters,
J Ind Microbiol Biotechnol
123
Milford, USA) using a Sugar-SP0810 column (Shodex,
Japan). The column was maintained at 80°C, and the sugars
were eluted with ultra-pure water at flow rate of 1 ml/min.
The HPLC system was a Waters LC1 module l plus
(Millfort, MA, USA) equipped with both a refractive index
(RI) and an ultraviolet (UV) detector set at 280 nm. For
lactic acid, acetic acid, iso-butyric acid, citric acid, ethanol
and hydroxymethylfurfural (HMF) analyses, an Aminex
HPX-87H column (Bio-Rad, Hercules, CA, USA) with a
de-ashing Micro-guard pre-column (Bio-Rad) was used.
The column was maintained at 50°C, and the mobile phase
was 5 mM H
2
SO
4
at flow rate of 0.6 ml/min. All com-
pounds were analysed with a RI detector, and HMF was
also analysed by the UV detector. All samples were filtered
through 0.22-lm Gelman membrane filters prior to
analysis.
Total phenolic compounds were determined by using the
Folin–Ciocalteu colorimetric method according to Single-
ton and Rossi [13]. Briefly, 100 ll sample was mixed with
5 ml 1/10 (v/v) diluted Folin–Ciocalteu reagent and 4 ml
7.5% Na
2
CO
3
. Absorbance was measured at 765 nm after
15 min incubation at 45°C. Total phenolic compounds are
expressed as g gallic acid equivalents (GAE)/l. Assays
were carried out in triplicate. Cell growth was evaluated by
measuring absorbance at 600 nm. At the beginning and at
the end of fermentations, biomass dry weight was deter-
mined gravimetrically. Two millilitres of each sample was
centrifuged in dried Eppendorf tubes, washed twice with
2 ml filtered water and dried overnight at 100°C.
Calculations
Volumetric consumption rate of sugars (Q
S
, g/l h) was
calculated as total monosaccharides (glucose and fructose)
consumed in a defined time interval. The specific growth
rate (l,h
-1
) was calculated by linear regression of the
curve of ln(OD/ODi) versus time for the exponential
growth phase. The biomass (Q
X
) and mannitol (Q
ManOH
)
volumetric production rates (g/l h), hereinafter referred to
as the biomass and mannitol productivities, were calculated
as the increase in cell mass and mannitol concentrations,
respectively, for a designated time interval. The mannitol
yield (Y
ManOH,
g/g) was calculated as the ratio between
mannitol increase and fructose consumption, for a desig-
nated time interval.
Results and discussion
Composition of hydrolysed carob pulp syrups
and carob pulp base medium
Carob syrups used in this work were obtained by water
extraction of carob pods under operational conditions
previously optimised, and typically contain sucrose
(accounting for more than 50% of total sugars) together
with free fructose, glucose and pinitol in a total close to
200 g/l [12]. As a high content of free fructose and glucose
are required for efficient mannitol production, carob syrups
have to be hydrolysed. Table 2summarises the composi-
tion of the hydrolysed carob pulp syrups used in this work
for culture medium preparation after the sterilising hydro-
lysis. More than 90% of the original sucrose was converted
to fructose and glucose, leading to an increase of concen-
tration of both sugars, at least twofold. Under these mild
conditions, some degradation of hexoses also occurred, and
HMF was found as the main degradation product, although
at low concentration. Phenolic compounds were also found,
but in relative low amounts.
After supplementation, the concentration of components
in the hydrolysates was reduced to a half, and carob-pulp-
based medium had average composition (per litre) as fol-
lows: 44.4 g fructose, 47.6 g glucose, 7.7 g pinitol, 2.4 g
acetic acid, 0.85 g iso-butyric acid and 0.34 g HMF.
Organic acids, such as iso-butyric acid, and furan deriva-
tives such as HMF, may act as fermentation-inhibiting
Table 1 Lactic acid bacteria
strains used in this study Microorganism Strain Growth
temperature (°C)
Lactobacillus
Lactobacillus intermedius NRRL B-3692 37
Lactobacillus reuteri DSM 20016 37
Leuconostoc
Leuconostoc fructosum NRRL B-2041 30
Leuconostoc ficulneum NRRL B-23447 30
Leuconostoc citreum ATCC 49370 30
Leuconostoc lactis ATCC 19256 30
Leuconostoc mesenteroides subsp. cremoris ATCC 19254 30
Leuconostoc mesenteroides subsp. dextranicum ATCC 19255 30
J Ind Microbiol Biotechnol
123
compounds. However, at these low levels and initial pH
they may not exert a strong inhibition effect, suggesting the
suitability of hydrolysates for fermentation purposes;
hence, no detoxification procedure was applied.
Kinetic profiles
To evaluate mannitol production by selected strains using a
carob-based medium, the kinetic profiles for growth, sug-
ars, mannitol and other co-products were studied. Figures 1
and 2show the results obtained for Lactobacillus and
Leuconostoc strains, respectively. Amongst Lactobacillus
strains, fructose and glucose were simultaneously con-
sumed, but fructose assimilation rate was always higher.
Lb. intermedius displayed the fastest growth and maximum
specific growth rate, two times higher than those for
Lb. reuteri, which was also confirmed by the higher final
cell density achieved. The maximum mannitol concentra-
tions obtained were similar for both microorganisms,
although mannitol accumulation by Lb. intermedius
occurred at a higher rate, which is consistent with fructose
depletion. As expected, lactic acid and acetic acid were
also produced, mainly when glucose consumption was
higher. Ethanol was also found, but only in minor amounts,
typically a maximum concentration around 1 g/l.
Lb. intermedius was an exception, as it presented more
significant production (5 g/l) for the final fermentation
stages, mainly after fructose depletion and onset of single
glucose metabolism. Ethanol production from glucose has
been already described for L. intermedius strain [4]
All Leuconostoc strains preferably consumed fructose
over glucose, in similar trend to lactobacilli. The remaining
glucose concentration varied depending on the strain.
Higher fructose consumption rates were found for L. ficul-
neum and L. fructosum, which together with Lactobacillus
strains were the best mannitol producers. Maximum specific
growth rate were similar for all Leuconostoc strains
(*0.2 h
-1
), but the final biomass concentration achieved
also depended on the strain.
As for Lactobacillus, these strains also produced lactic
acid and acetic acid. For all LAB tested, at the maximum
mannitol concentrations, aliphatic acids ranged between
9.9 and 18.5 g/l and 14.9 and 27.0 g/l for acetic acid and
lactic acid, respectively.
Kinetic and stoichiometric parameters
Table 3shows the results for kinetic and stoichiometric
parameters obtained for Lactobacillus and Leuconostoc
strains grown in carob-based medium. All strains produced
mannitol at high yields ([0.70 g mannitol/g fructose) and
relatively high volumetric productivities ([1.31 g/l h), and
displayed high fructose consumption ([91%). Independent
of the genus, strains could be divided into two groups: one
for which glucose assimilation rates were always below
1.05 g/l h (glucose consumption lower than 55%), and a
Table 2 Composition of the carob syrup hydrolysates used for cul-
ture medium preparation
Compound Concentration (g/l)
Sucrose 11.4 ±2.1
Fructose 89.2 ±1.9
Glucose 96.0 ±1.8
Pinitol 15.6 ±0.3
Acetic acid 0.70 ±0.10
Iso-butyric acid 2.05 ±0.08
Hydroxymethylfurfural 0.67 ±0.02
Phenolic compounds 1.62 ±0.04
The concentrations of these compounds in the culture medium are
half of those reported in this table, due to the addition of the required
supplements
A
Time (h)
Glc, Fru, ManOH, HAc, HLac (g/l)
0
10
20
30
40
50
EtOH (g/l), Ln (Abs/Abs0)
0
1
2
3
4
5
6B
Time (h)
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Glc, Fru, ManOH, HAc, HLac (g/l)
0
10
20
30
40
50
EtOH (g/l), Ln (Abs/Abs0)
0
1
2
3
4
5
6
Fig. 1 Fermentation profiles for mannitol production by Lactobacil-
lus strains in carob-syrup-based medium. Lb. intermedius (a);
Lb. reuteri (b). Glucose (filled squares), fructose (filled triangles),
mannitol (open triangles), lactic acid (open squares), acetic acid
(open diamonds), ethanol (open circles), cell density (plus symbols).
Data points are the average of at least two independent replicates.
Lines are guides to the eye only and do not have any statistical
significance
J Ind Microbiol Biotechnol
123
second for which they ranged between 1.41 and 1.89 g/l h
(glucose consumption ranging from 63% to 96%). The
group exhibiting the faster glucose assimilation
(Lb. intermedius,Lb. reuteri,L. fructosum and L. ficulne-
um) also exhibited higher mannitol production rates and
yields as well as ratio of mannitol to aliphatic acids (lactic
acid and acetic acid) above 1. In fact, the rates for glucose
consumption and mannitol production were positively
correlated (r=0.86, data not shown).
These results clearly indicate that these two groups may
have different metabolic capabilities, with strains that
exhibited slower glucose consumption producing, in
general, higher acetic acid concentrations. The highest ratio
of mannitol to other products was obtained for Lb. reuteri
DSM 20016, but this is not due to higher mannitol pro-
duction but rather to higher biomass production yield and
lower lactic acid production.
The ratios of mannitol to other co-products (lactic and
acetic acids) obtained in this work were, however, lower
than the values (higher than 2), reported by Saha [14]. It is
worth noting that those reported values were obtained for
higher fructose concentrations and different ratio of glu-
cose to fructose, which is also a key parameter for efficient
mannitol production. Von Weymarn et al. [15] described
A
Time (h)
0 5 10 15 20 25 30
Glc, Fru, ManOH, HAc, HLac (g/l)
0
10
20
30
40
50
EtOH (g/l), Ln (Abs/Abs
0
)
0.0
0.5
1.0
1.5
2.0 B
Time (h)
Glc, Fru, ManOH, HAc, HLac (g/l)
0
10
20
30
40
50
0.0
0.5
1.0
1.5
2.0
C
Time (h)
Glc, Fru, ManOH, HAc, HLac (g/l)
0
10
20
30
40
50
0.0
0.5
1.0
1.5
2.0 D
Time (h)
Glc, Fru, ManOH, HAc, HLac (g/l)
0
10
20
30
40
50
0.0
0.5
1.0
1.5
2.0
E
Time (h)
Glc, Fru, ManOH, HAc, HLac (g/l)
0
10
20
30
40
50
0.0
0.5
1.0
1.5
2.0 F
Time (h)
Glc, Fru, ManOH, HAc, HLac (g/l)
0
10
20
30
40
50
0.0
0.5
1.0
1.5
2.0
0 5 10 15 20 25 30
0 5 10 15 20 25 30 0 5 10 15 20 25 30
0 5 10 15 20 25 30
0 5 10 15 20 25 30
EtOH (g/l), Ln (Abs/Abs
0
)EtOH (g/l), Ln (Abs/Abs
0
)
EtOH (g/l), Ln (Abs/Abs
0
) EtOH (g/l), Ln (Abs/Abs
0
) EtOH (g/l), Ln (Abs/Abs
0
)
Fig. 2 Fermentation profiles for mannitol production by Leuconostoc
strains in carob-syrup-based medium. L. fructosum (a), L. ficulneum
(b), L. citreum (c), L. lactis (d), L. mesenteroides subsp. cremoris (e),
L. mesenteroides subsp. dextranicum (f). Glucose (filled squares),
fructose (filled triangles), mannitol (open triangles), lactic acid (open
squares), acetic acid (open diamonds), ethanol (open circles), cell
density (plus symbols). Data points are the average of at least two
independent replicates. Lines are guides to the eye only and do not
have any statistical significance
J Ind Microbiol Biotechnol
123
for L. mesenteroides an optimum ratio of glucose to fruc-
tose of 0.55:1. Also, Saha [16] reported that lactic acid
production by Lb. intermedius surpassed mannitol con-
centration when sugar cane molasses (ratio of glucose to
fructose of 1:1) were used for culture medium preparation.
The maximum mannitol concentration (43.7 g/l) was
obtained with L. fructosum (Fig. 2a). This strain also
exhibited the highest volumetric productivity (2.36 g/l h, at
15 h) and conversion efficiency of fructose to mannitol of
100%. This productivity is higher than the obtained for
L. pseudomesenteroides ATCC 12291 grown in media
containing glucose and fructose (100:50) under similar
culture conditions [17], and the yield was higher than
values previously reported for Lb. intermedius B-3693
grown in fructose-based medium [4] and the same
L. pseudomesenteroides [17]. Although stoichiometric
conversion of fructose to mannitol has been also previously
reported for Lb. sanfranciscensis [5,18], that strain
exhibited lower productivity. Given that the initial perfor-
mances with those strains were further optimised, namely
using pH-controlled bioreactors (preventing use of citrate
buffer that was typically partially consumed, data not
shown), or using different operation modes, such as fed-
batch, cell recycling and optimised culture media, it is
expected that L. fructosum may also present a significant
performance improvement [14,15,19,20].
The mannitol volumetric productivities obtained for the
group exhibiting lower mannitol production (L. citreum,
L. lactis, L. mesenteroides subsp. cremoris and L. mesen-
teroides subsp. dextranicum) were similar to values reported
for L. mesenteroides ATCC 9135, L. pseudomesenteroides
ATCC 12291, Lb. brevis ATCC 8287 and Lb. fermentum
NRRL 1932 [15]. According to the present results, these
syrups, although containing acetic and iso-butyric acids,
HMF and some phenolics that were also solubilised during
extraction process of sugars, did not have an apparent
inhibitory effect on growth of these microorganisms. For
this reason, increasing the syrup concentration seems to be a
possibility to achieve higher performance. Furthermore, the
levels attained by the produced lactic and acetic acids did not
hinder fructose consumption. As these acids can be recov-
ered from the fermentation broth (e.g. by electrodialysis [4]),
their co-production may be an advantageous trait, as it will
be possible to design a multi-product process.
Conclusions
The eight mannitol-producing LAB strains tested were able
to grow and produce mannitol with relatively high effi-
ciency in carob-syrup-based medium. Kinetic and physio-
logical characterisation of these strains was performed, and
the results obtained suggest that the studied strains can be
divided into two groups based on their glucose consump-
tion and mannitol production rates, as well as ratio of
mannitol to aliphatic acids. The highest volumetric pro-
ductivity was obtained for L. fructosum, which was also
one of the most robust strains tested. This strain is currently
being used for mannitol process optimisation in carob-
based media.
Given that the levels attained by the produced acids did
not hinder fructose consumption, their co-production may
be an advantageous trait. The design of a multi-product
process is now possible, as these acids are also marketable
compounds, in particular lactic acid, which has a large and
expanding market for industrial production of bioplastics.
Table 3 Kinetic and stoichiometric parameters for growth and formation of products for Lactobacillus and Leuconostoc strains in carob-syrup-
based medium
Microorganism Consumption (%) Q
S
(g/l h)
Q
X
(g/l h)
Q
ManOH
(g/l h)
Y
ManOH
(g/g)
ManOH/
HLac
ManOH/
(HLac ?HAc)
Glc Fru
Lb. intermedius 96 100 3.74 0.27 1.68 0.91 1.6 1.1
Lb. reuteri 69 98 3.27 0.16 1.79 0.96 2.3 1.4
L. fructosum 76 91 3.25 0.14 1.82 1.06 1.6 1.0
L. ficulneum 63 98 3.12 0.09 1.65 0.89 1.7 1.2
L. citreum 38 97 2.65 0.07 1.31 0.70 1.4 0.8
L. lactis 33 93 2.34 0.08 1.39 0.82 1.4 0.8
L. mesenteroides subsp. cremoris 32 95 2.27 0.003 1.37 0.83 1.4 0.8
L. mesenteroides subsp. dextranicum 55 94 2.73 0.003 1.50 0.89 1.4 0.8
Glc glucose, Fru fructose, Q
S
consumption rate of sugars, Q
X
biomass productivity, Q
ManOH
mannitol productivity, Y
ManOH
mannitol yield on
fructose, ManOH/HLac ratio of mannitol to lactic acid, ManOH/(HLAc ?HAc) ratio of mannitol to lactic acid ?acetic acid. All values
calculated after 24 h
J Ind Microbiol Biotechnol
123
Acknowledgments This work was supported by AdI, Project Val-
orAlfa (70/00326). ARS Culture Collection (National Centre for
Agricultural Utilization Research, Peoria, IL, USA) is gratefully
acknowledged for supplying bacterial strains. The authors thank
Ame
´lia Marques, Carlos Barata and Ce
´u Penedo for technical support.
References
1. Soetaert W, Vanhooren PT, Vandamme EJ (1999) The produc-
tion of mannitol by fermentation. In: Bucke C (ed) Carbohydrate
biotechnology protocols, vol 10. Humana, Totowa, pp 261–275
2. Wisselink HW, Weusthuis RA, Eggink G, Hugenholtz J, Grobben
GJ (2002) Mannitol production by lactic acid bacteria: a review.
Int Dairy J 12:151–161
3. Parmentier S, Arnaut F, Soetaert W, Vandamme EJ (2005)
Enzymatic production of D-mannitol with the Leuconostoc
pseudomesenteroides mannitol dehydrogenase coupled to a
coenzyme regeneration system. Biocatal Biotransformation
23:1–7
4. Saha BC, Nakamura LK (2003) Production of mannitol and lactic
acid by fermentation with Lactobacillus intermedius NRRL
B-3693. Biotechnol Bioeng 82:864–871
5. von Weymarn N, Hujanen M, Leisola M (2002) Production of
D-mannitol by heterofermentative lactic acid bacteria. Process
Biochem 37:1207–1213
6. Santos M, Rodrigues A, Teixeira JA (2005) Production of dextran
and fructose from carob pod extract and cheese whey by Leu-
conostoc mesenteroides NRRL B512(f). Biochem Eng J 25:1–6
7. Roseiro JC, Gı
´rio FM, Amaral-Collac¸o MT (1991) The influence
of storage stability on the use of carob pulp aqueous extract as
raw material for fermentation processes. Lebensm-Wiss Technol
24:508–512
8. Roseiro JC, Gı
´rio FM, Amaral-Collac¸o MT (1991) Yield
improvements in carob sugar extraction. Process Biochem
26:179–182
9. Roseiro JC, Costa DC, Amaral-Collac¸o MT (1992) Batch and
fed-batch cultivation of Xanthomnas campestris in carob extracts.
Lebensm-Wiss Technol 25:100–195
10. Roukas T (1993) Ethanol production from carob pods by
Saccharomyces cerevisiae. Food Biotechnol 7:159–176
11. Roukas T (1998) Citric acid production from carob pod extract by
cell recycle of Aspergillus niger ATCC 9142. Food Biotechnol
12:91–104
12. Moniz P, Carvalheiro F, Moura P, Pereira J, Duarte LC, Esteves
MP, Gı
´rio FM (2009) Screening and characterization of lactic
acid bacteria for the production of mannitol in carob based syr-
ups. BioMicroWorld2009, Lisbon
13. Singleton V, Rossi J (1965) Colorimetry of total phenolics with
phosphomolybdic-phosphotungstic acid reagents. Am J Enol
Vitic 16:144–158
14. Saha BC (2006) Production of mannitol from inulin by simulta-
neous enzymatic saccharification and fermentation with Lacto-
bacillus intermedius NRRL B-3693. Enzyme Microb Technol
39:991–995
15. von Weymarn N, Kiviharju K, Leisola M (2002) High-level
production of D-mannitol with membrane cell-recycle bioreactor.
J Ind Microbiol Biotechnol 29:44–49
16. Saha BC (2006) A low-cost medium for mannitol production by
Lactobacillus intermedius NRRL B-3693. Appl Microbiol Bio-
technol 72:676–680
17. Helanto M, Aarnikunnas J, von Weymarn N, Airaksinen U, Palva
A, Leisola M (2005) Improved mannitol production by a random
mutant of Leuconostoc pseudomesenteroides. J Biotechnol
116:283–294
18. Korakli M, Schwartz E, Wolf G, Hammes WP (2000) Production
of mannitol by Lactobacillus sanfranciscensis. Adv Food Sci
22:1–4
19. von Weymarn FNW, Kiviharju KJ, Jaaskelainen ST, Leisola
MSA (2003) Scale up of a new bacterial mannitol production
process. Biotechnol Prog 19:815–821
20. von Weymarn N (2006) Process development for mannitol pro-
duction by lactic acid bacteria. Ph.D Thesis, Helsinki University
of Technology
J Ind Microbiol Biotechnol
123
