Ann Microbiol (2014) 64:1211–1221

DOI 10.1007/s13213-013-0761-3

ORIGINAL ARTICLE

Optimization of nutritional supplements for enhanced lactic

acid production utilizing sugar refinery by-products
Abhinay Srivastava & Amrita Poonia &
Abhishek Dutt Tripathi & Ravi Pratap Singh &
Suresh Kumar Srivastava

Received: 25 April 2013 / Accepted: 30 October 2013 / Published online: 21 November 2013
# Springer-Verlag Berlin Heidelberg and the University of Milan 2013

Abstract The present study investigated the synergistic effect producing bacteria and characterization of lactic acid was also
of nutritional supplements (amino acid and Tween 80) on done.
lactic acid production by Lactobacillus delbruckii utilizing a
sugar refinery by product (cane molasses) in a submerged
Keywords Lactic acid . Lactobacillus delbrueckii .
fermentation process. Initially, the effect of individual factors
Nutritional supplement . Process optimization .
on lactic acid yield was studied by supplementing amino acids
Characterization of lactic acid
and their combinations, Tween 80 and cane molasses at vary-
ing concentrations in production medium. A combination of
L -phenylalanine and L -lysine gave a maximum lactic acid
yield of 47.89±0.1 g/L on a dry cell weight basis at individual Introduction
factor level. Similarly, maximum lactic acid yield was obtain-
ed by supplementing the production medium with 40.0 g/L Lactic acid (2-hydroxypropanoic acid, CH3–CH (OH)–
and 2.0 g/L Tween 80 and cane molasses, respectively, at COOH) is a natural organic acid with vast applications in
individual factor level. In order to further improve the lactic the food and non-food industries, including the cosmetic and
acid yield, nutritional supplements were optimized by central pharmaceutical industries, and for the production of oxygen-
composite rotatable design (CCRD) using Minitab 15 soft- ated chemicals, plant growth regulators, and special chemical
ware. Shake flask cultivation under optimized conditions, i.e., intermediates (Dumbrepatil et al. 2008; Tashiro et al. 2011).
cane molasses (32.40 g/L), Tween 80 (2.0 g/L) and L -phenyl- Lactic acid can be produced either by chemical synthesis or by
alanine and L -lysine (34.0 mg/L) gave a lactic acid yield of microbial fermentation. Chemical synthesis from petrochem-
64.86±0.2 g/L, corresponding to 95.0 % of the predicted yield ical resources always results in racemic mixture of lactic acid,
of 67.78±0.3 g/L. Batch cultivation performed in 7.5 L bio- which is disadvantageous (Hofvendahl and Hahn-Hagerdal
reactor (working volume: 3.0 L) under optimized conditions 2000). Biological methods have the advantage that optically
gave maximum lactic acid yield and productivity of 79.12± pure lactic acid can be obtained by choosing a suitable mi-
0.2 g/L and 3.40 g/L·h, which is higher than previous studies crobe (Ryu et al. 2003).
with reduced fermentation time. Screening of lactic acid Industrial production of lactic acid is expensive but can be
minimized by regulating various parameters such as substrate
A. Srivastava : A. Poonia : A. D. Tripathi (*) cost, suitable strain selection and recovery process
Centre of Food Science and Technology, Institute of Agricultural (Hofvendahl and Hahn-Hagerdal 2000). Production costs
Sciences, Banaras Hindu University, Varanasi 221005, U.P., India can be minimized by optimizing media components and
e-mail: [email protected]
physical variables using fast, efficient and appropriate design
R. P. Singh model based method like response surface methodology
Department of Genetics and Plant Breeding, Institute of Agricultural (RSM), D-optimal method and Taguchi (DOE) methodology.
Sciences, Banaras Hindu University, Varanasi 221005, U.P., India Design model based optimization approaches are superior to
conventional optimization methods since they allow us to
S. K. Srivastava
School of Biochemical Engineering, Indian Institute of Technology, understand the interactive effect of different process variables
Banaras Hindu University, Varanasi 221005, U.P., India on product yield (Fenice et al. 2012).
1212 Ann Microbiol (2014) 64:1211–1221

High substrate cost is one of the biggest challenges in Materials and methods
fermentative lactic acid production. Raw materials for use in
lactic acid production should be cheap, relatively uncontam- Bacterial strain and growth medium
inated, easily available, with high yield and productivity and
should be easily fermentable with little or no pre-treatment Lactobacillus delbrueckii NCIM 2025 used in the present
(Vickroy 1985). Cheap raw materials, such as molasses, rye, study was procured from National Collection of Industrial
corn, whey and potato starch, can be beneficial in reducing Microorganism (NCIM), National Chemical Laboratory
lactic acid production costs (Wee et al. 2004; Oh et al. 2005; (NCL) Pune, India. The stock cultures were maintained at
Buyukkileci and Harsa 2004; Yun et al. 2004). Different 4 °C in 250 mL MRS broth (DeMan’s Sharpe and Rogosa
industrial by-products or wastes such as sugarcane mo- Broth) with sub-culturing every 10 days. The medium for cell
lasses (Calabia and Tokiwa 2007) and beet molasses growth contained (g/L): peptone 10.0, meat extract 10.0, yeast
(Dumbrepatil et al. 2008) have been evaluated as sub- extract 5.0, D -glucose 20.0, Tween 80 1.0, K2HPO4 2.0,
strates for lactic acid production with the aim of de- sodium acetate 5.0, tri-ammonium citrate 2.0, MgSO4⋅7H2O
creasing the cost of the process. 0.2, MnSO 4 ⋅4H 2 O 0.05 in 1,000 mL distilled water.
Lactic acid yield can be enhanced by addition of fatty acid Lactobacillus delbrueckii cells from stock cultures were trans-
and surfactants like Tween 80 to the production medium ferred to sterile growth medium (MRS agar plates) and incu-
(Duggan et al. 1959; Qi et al. 2009). The nonionic surfactant bated at 37 °C for 24 h. The culture was inoculated into MRS
Tween 80 (polyoxyethylene sorbitn mono-oleate) is used broth in screw capped test tubes, incubated for 24 h at 37 °C
widely as an additive in foods, pharmaceutical preparations, and used for fermentation.
and as an emulsifier, dispersant, or stabilizer (Budavavi et al.
1996). Tween 80 is an anionic surfactant, which, when added Production medium and cultivation
to bacterial growth media, can dissolve membrane phospho-
lipids and facilitate nutrient uptake. Tween 80 addition en- Production medium was similar to growth medium and was
hanced mycelial growth and EPS production in Pleurotus prepared in 250 mL conical flasks. Inoculum (1.0 mL) was
tuberregium (Elisashvili 2012) and has found widespread added to conical flasks containing 100 mL production medium
use in extracellular enzyme production (Liu et al. 2006; and incubated at 150 rpm at 37 °C for 24 h in a shaker plus
Zeng et al. 2006; Silva et al. 2007; Gonzalez et al. 2008). incubator (Sigma, St. Louis, MO). The pH of the final culture
Oh et al. (2005) reported that the growth of Lactobacillus medium was adjusted to 7.0±0.5 using 0.1 N HCl/0.1 N NaOH.
casei was strongly affected by Tween 80. However, to date
only a few in depth studies on the effect of Tween 80 on the Analytical methods
fermentation performance of Lactobacillus have been
published. Dry cell mass
Lactic acid bacteria (LAB) also require complex nutrients
such as amino acid, peptides, vitamins and fatty acids for Culture broth (20.0 mL) was centrifuged at 3,900 g for 10 min
greater growth and high lactic acid yield. High cell densities at 4 °C and a cell pellet was obtained. The cell pellet was
of LAB can be achieved in complex media containing more washed with saline water (NaCl, 0.8 g/m3) and then dried in
easily convertible nitrogen such as amino acids or pro- an aluminum weighing dish at 90 °C for 24 h in a hot air oven
teins with small molecular weight. LAB are unable to (Perfett India, Ambala, India).
synthesize certain essential amino acids (Fitzpatrick and
Keeffe 2001). LAB lack certain amino acids such as Sugar analysis
phenylalanine, lysine and arginine at the active site of
lactate dehrdrogenase (ldh )—a potent enzyme in lactic Cane molasses was procured from National Sugar Institute
acid biosynthesis. Lactic acid yield can be further im- (NSI), Kanpur, India and was pre-treated with activated char-
proved by supplementing production media with certain coal (1:1) for 2 h in order to reduce its opacity and other
amino acids, e.g., phenylalanine, lysine and aspartic interfering compounds. The total sugar in cane molasses
acid (Sriphochanart et al. 2011). was determined by the phenol-sulfuric method (Dubois
In the present work, an attempt was made to enhance lactic et al. 1956) and reducing sugar was determined by the
acid yield of Lactobacillus delbrueckii under submerged fer- 3,5-dinitrosalicylic acid (DNS) method.
mentation process by optimizing the medium composition
using an agro-industrial by-product (cane molasses), Tween Lactic acid analysis
80 and amino acids. Shake flask cultivation was further scaled
up in a 7.5 L bioreactor in order to study the growth kinetics in High performance liquid chromatography (HPLC) was used
batch mode in the optimized production medium. for the quantitative and qualitative determination of lactic acid
Ann Microbiol (2014) 64:1211–1221 1213

(Oh et al. 2005). Aliquots of 2.0 μL of different samples were Statistical analysis
analyzed by HPLC (Shimadzu, Tokyo, Japan) with a UV
detector at 210 nm using an RP C-18 column. The operating The data obtained from the various experiments were recorded
conditions comprised: mobile phase 5 mM H2SO4, flow rate and subjected to statistical analysis by CCRD. The signifi-
0.4 mL/min, column temperature 60 °C, and injection volume cance difference between means was tested against the critical
10 μL. Quantification was based on an internal standard. difference at 1.0 % level of significance by using statistical
software tool Minitab 15 for data analysis. Three factors were
Amino acid supplementation chosen at five levels for the optimization process, i.e., amino
acids, Tween 80 and cane molasses concentration, and two
Seven different trials with different amino acids at three dif- responses, i.e., lactic acid and cell biomass, were observed.
ferent combinations, viz., L -aspartic acid (T 1), L -phenyl ala- The experimental data obtained from the design were ana-
nine (T 2), L -lysine (T 3), L -aspartic acid + L -phenylalanine lyzed by the response surface regression procedure using the
(T 4), L -phenylalanine + L-lysine (T 5), L -aspartic acid + L - following second-order quadratic equation:
lysine (T 6) and control (T 7, without supplementation) were Y i ¼ β 0 þ ∑βi xi þ ∑βii x2 i þ ∑βij xi ji
used for the production of lactic acid. In the basal medium,
50.0 mg/L amino acid was added for the production of lactic
acid. In treatments T 4, T 5 and T 6, two amino acids were added Where Y i is the predicted response, x ix j were independent
in equal proportion with a final concentration of 50.0 mg/L in variables, β 0 is the offset term, β i is the linear coefficient, β ii
basal medium. is the quadratic coefficient, and β ij the interaction coefficient.

Bioreactor study
Effect of varying cane molasses and Tween 80 concentration
on cell growth and lactic acid yield Lactic acid production with the optimal medium was scaled up
in a 7.5 L bioreactor (BioFlo/Celligen 115, New Brunswick
Cane molasses concentration was varied between 10 and Scientific, Edison, NJ) containing 3.0 L medium. Agitation
50 g/L in order to check its effect on lactic acid and speed and culture temperature were controlled at 250 rpm and
cell biomass yield. Tween 80 concentration was varied 37 °C. The bioreactor containing 3.0 L production medium was
in the range of 1–5 g/L. Samples (20.0 mL) were withdrawn sterilized in an autoclave (capacity 50 L) at 121 °C for 20 min,
after 24.0 h of cultivation from different trials and cell mass cooled and then inoculated with 5.0 % inoculum (v/v). The pH
and lactic acid yield were analyzed. of the culture broth was maintained at optimum pH by auto-
matic addition of acid or base by pH–mV controller (Mettller
Sample preparation for Fourier transform infrared Toledo, Columbus, OH). Dissolved oxygen (DO) was mea-
spectroscopy analysis sured with a DO probe (Mettller Toledo). DO concentration
was maintained at 30 % saturation value by cascading the speed
Sample preparation was carried out as described (Helm et al. of the agitator and air flow rate. Agitation speed and DO
1991). After cultivation on agar plates, a platinum loop was cascade was controlled by setting the minimum and maximum
used to remove bacterial biomass from the third quadrant of agitation speeds at 100 and 400 rpm, respectively, in order to
each agar plate, and the biomass was dissolved in 100 μL maintain the desired DO concentration. Aeration rate during
distilled water. The concentration of the suspension was ad- lactic acid production was kept at 1.0 L/min.
justed so that the intensity of the amide I band (1,655 cm−1) in
the IR spectrum was between 0.35 and 1.25, which is within
the linear range of the DTGS detector. From each suspension, Results and discussion
35 μL was transferred to an IR transparent optical crystal
(ZnSe) in a multisample cuvette (Bruker Optics, Ettlingen, Screening of lactic acid producing microbe (Lactobacillus
Germany). The samples were dried under moderate vacuum delbrueckii) using FTIR
(0.1 bar) using anhydrous Silica Gel (Prolabo, Paris, France)
in a desiccator to form films suitable for Fourier transform Lactic acid producing microbes were screened by FTIR.
infrared spectroscopy (FTIR) analysis. The FTIR measure- Figure 1 represents the FTIR spectrum of Lactobacillus
ment was performed with a biomodule (Bruker Optics) delbrueckii . Sharp peaks at wave numbers 3,000, 1,800,
coupled to an equinox 55 spectrometer (Bruker Optics). The 1,500, 1,200 and 900 cm−1 represented fatty acids, amide
spectra were recorded in the region between 4,000 and and polysaccharide functional groups located in the bacterial
500 cm−1 with a spectral resolution of 6 cm−1 and an aperture cell membrane and cell wall, which is in accordance with
of 5.0 mm. previous findings (Naumann et al. 1991).
1214 Ann Microbiol (2014) 64:1211–1221

7.0

6.5 1288.1
814.3
723.5 624.1
592.6
578.6 459.2
430.0
1087.0
672.1 537.0 414.4
6.0 1244.8
1774.9
5.5 1053.6 494.3

1150.2 600.1
5.0 1738.5
755.1
466.9
1021.6

4.5 1120.9 611.1

3431.4 1707.4
3246.8 1200.3 919.6
2995.3 992.1
A 4.0
3371.7
3306.2
3.5 3504.0
3188.7 871.6
962.3
2941.0 1324.8
3621.0 1454.0
3736.5 2619.0
3.0 3969.0
3856.2

2.5

2.0

1.5

1.0
3500
4000 3000 2500 2000 1500 1000 500

Wavenumbers (cm-1)
Fig. 1 Spectrum of lactic acid producing bacteria (Lactobacillus delbrueckii) by Fourier transform infrared spectroscopy (FTIR)

Effect of nutritional supplements (cane molasses, Tween 80 cane molasses, Tween 80 and amino acid on lactic acid
and amino acid) on cell mass and lactic acid production and biomass yield. Maximum lactic acid and biomass
yield of 48.15±0.2 and 63.24±0.3 g/L was obtained at
Initially, experiments were carried out to understand the 40.0 g/L cane molasses concentration (Fig. 2). Lactic
effect of varying concentration of chosen variables, i.e., acid yield was found to be maximum at 4.0 g/L Tween 80

Fig. 2 Effect of varying cane

molasses concentration on lactic
acid and biomass yield obtained
after 24.0 h of submerged
fermentation (SD±0.5)
Ann Microbiol (2014) 64:1211–1221 1215

Fig. 3 Effect of varying Tween

80 concentrations on lactic acid
and biomass yield obtained after
24.0 h of submerged fermentation
(SD±0.5)

concentration in batch cultivation (Fig. 3). Any further increase Statistical optimization of nutritional supplement
in Tween 80 concentration decreased the lactic acid yield, due to concentration for enhanced lactic acid yield
reduced ldh activity (Naveena et al. 2005; Nagarjun et al. 2005).
In order to study the effect of amino acid supplementation Design of experiments
on lactic acid yield, seven trials were performed taking into
account three amino acids, viz: control (T 1), L -aspartic acid Statistical optimization of nutritional supplements was
(T 2), L -phenylalanine (T 3), L -lysine (T 4) and their combina- achieved by employing RSM for enhanced lactic acid
tions such as L -aspartic acid + L -phenylalanine (T 5), L -phenyl production. Three process variables, i.e., amino acids
alanine + L -lysine (T 6) and L -aspartic acid + L -Lysine (T 7) (L -phenylalanine + L -lysine), Tween 80 and cane mo-
(Fig. 4). Amino acid supplementation study revealed that lasses were chosen at five levels and then 20 set of
maximum lactic acid and cell biomass of 47.89±0.1 and experiments were designed by minitab 15 software
63.98±0.2 g/L was obtained with treatment T 6 containing using a central composite rotatable design (CCRD)
L -phenylalanine + L -lysine (1:1). (Table 1).

Fig. 4 Effect of different amino

acid and their combination on
lactic acid and biomass yield after
24.0 h of cultivation under
submerged fermentation
(SD±0.5)
1216 Ann Microbiol (2014) 64:1211–1221

Table 1 Levels of independent variables in experimental plan Table 3 Coefficients and t-value for lactic acid production using CCRD

Independent variable Ranges and levels Term Coefficient Standard coefficient t-value P-value

−α −1 0 1 +α Constant 44.295 2.399 18.463 0.000a

A: Amino acid −4.6807 1.592 −2.941 0.015a
Amino acids (mg/L) 6.36 20 40 60 73.63
B : Tween 80 2.5173 1.592 1.581 0.145
Tween 80 (g/L) 1.31 2 3 4 4.68
C : Cane molasses 9.3265 1.592 5.859 0.000a
Cane molasses (g/L) 3.18 10 20 30 36.18
A2 −7.3957 1.550 −4.773 0.001a
B2 −5.3062 1.550 −3.424 0.006a
Model fitting C2 −4.0034 1.550 −2.603 0.026a
AB −1.5625 2.080 −0.751 0.470
The design matrix in actual terms with eight factorial AC −5.0875 2.080 −2.446 0.034a
points, six axial points and six central points is present- BC 0.3700 2.080 0.178 0.862
ed in the Table 2. Maximum and minimum lactic acid a
Significant terms in the model
yields of 55.86±0.07 and 12.15±0.04 g/L were obtained
in trial nos. 13 and 5, respectively, at different combi- yield fitted in the terms of coded variables was obtained as
nations of cane molasses (3.18–36.80 g/L), amino acids follows:
(6.36–73.63 mg/L) and Tween 80 (1.31–4.68 g/L)
(Table 2). Regression analysis of the experimental de- Y ¼ 44:29−4:6807A þ 2:5173B þ 9:3265C–7:3957A2 −
sign demonstrated that the linear model (A, C), interac-
tive model (AC) and quadratic model (A2, B2 and C2) 5:3062−4:0334C2 –1:5625AB–5:0875AC þ 0:3700BC
were significant (P <0.05) (Table 3). ð1Þ
Applying multiple regression analysis, the results were
fitted to a second order polynomial equation. Lactic acid Where Y is the response in terms of lactic acid yield. Coded
terms A, B and C represent amino acids, Tween 80 and cane
Table 2 Central composite rotatable design (CCRD) design for optimi- molasses, respectively. ANOVA of the results of the quadratic
zation of three variables for production of lactic acid by Lactobacillus model is represented in Table 4. The determination coefficient
delbrueckii NCIM 2025
(R 2) of the model was determined to be 0.897 (a value of >0.75
Experiment Amino Tween Cane Cell mass Lactic acid indicates fitness of the model) indicating that 89.7 % of the
no. acids 80 (g/L) molasses (g/L) (g/L) variation in the response (i.e., lactic acid) can be explained by
(mg/L) (g/L)
the model. The adjusted R 2 was 0.804, which accounted for the
1 40 3 20 68.33±0.13 45.70±0.05 number of predictors in the model. Both R 2 values obtained
2 40 3 20 64.22±0.39 43.25±0.10 suggested that the model fit the data well. The F statistic was
3 20 2 10 29.40±0.06 21.56±0.10 9.68 and corresponded to a value of P =0.001 (the confidence
4 40 3 20 62.22±0.04 43.43±0.14 interval was 0.05), which indicated that the model was both
5 40 3 3.18 56.23±0.06 12.15±0.04 adequate and significant (Table 4). For first order effects, the
6 73.63 3 20 65.45±0.10 19.19±0.08
regression coefficient and t-values concluded that the cane
7 20 2 30 39.87±0.11 40.90±0.15
Table 4 Analysis of variance for the production of lactic acid by Lacto-
8 40 3 20 59.45±0.10 45.67±0.06 bacillus delbrueckii. DF Degree of freedom; Seq SS sequential sum of
9 40 1.318 20 49.89±0.12 19.56±0.17 squares; Adj MS adjusted mean square
10 20 4 30 38.76±0.18 47.80±0.12
Source DF Seq SS Adj MS Adj MS F-value P-value
11 6.364 3 20 27.65±0.16 29.80±0.11
12 40 3 20 60.23±0.29 43.67±0.10 Regression 9 3,016.07 3,016.07 335.119 9.68 0.001a
13 40 3 36.81 58.90±0.12 55.86±0.07 Linear 3 1,573.68 1,573.68 524.560 15.16 0.000a
14 20 4 10 32.28±0.03 19.86±0.13 Square 3 1,214.70 1,214.70 404.900 11.70 0.001a
15 40 3 20 60.29±0.08 43.67±0.16 Interaction 3 227.69 227.69 75.896 2.19 0.152
16 60 2 30 63.50±0.14 25.89±0.17 Residual error 10 3,460.2 346.02 34.602
17 60 4 10 62.20±0.18 18.95±0.15 Lack of fit 5 339.56 339.56 67.913 52.55 0.000a
18 40 4.68 20 33.22±0.06 41.25±0.16 Pure error 5 6.46 6.46 1.292
19 60 2 10 51.23±0.11 19.78±0.16 Total 19 3,362.09
20 60 4 30 56.26±0.30 19.42±0.23
a
Significant terms in the model
Ann Microbiol (2014) 64:1211–1221 1217

Fig. 5 Response surface plot of lactic acid production by Lactobacillus delbrueckii NCIM 2025 showing interaction between a Tween 80 and cane
molasses, b amino acids and cane molasses, and c amino acids and Tween 80

molasses (A) concentration had the most significant effect on software. The cane molasses and Tween 80 interactive effect
the lactic acid production, followed by amino acid (C). The was significant (P <0.05) (Table 4). Lactic acid production
quadratic main effect of amino acid (P <0.001), Tween 80 increased with increase in cane molasses concentration until
(P <0.006) and cane molasses (P <0.026) were the significant 30 g/L (Fig. 5a). A further rise in cane molasses concentration
factors. Therefore, amino acid, Tween 80 and cane molasses act decreased the overall lactic acid yield, which is in agreement
as limiting factors and even a small variance in their concen- with previous findings (Yu et al. 2008). Similarly, lactic acid
tration can alter lactic acid production to a considerable extent. yield showed a marked increase up to 2.0 g/L Tween 80
concentration. Tween 80 facilitates the conversion of sucrose
Interactive effect of nutritional supplements on lactic acid to lactic acid, which agrees with previous reports (Naveena
yield et al. 2005).
The amino acids and cane molasses interactive effect was
The interactive effect of nutritional supplements on lactic acid significant (P <0.05) (Table 4). Thus, increasing the concen-
yield was revealed by 3D surface obtained from minilab 15 tration of cane molasses will increase lactic acid production,

Table 5 Lactic acid yield before

and after optimization Variables Before After Lactic acid yield

Before After

Predicted Experimental

Amino acids (mg/L) 40.0 34.0 55.86±0.07 67.78±0.20 64.86±0.20

Tween 80 (g/L) 4.0 2.0
Cane molasses (g/L) 36.80 32.40
1218 Ann Microbiol (2014) 64:1211–1221

Fig. 6 Time profile of lactic acid,

cane molasses utilization and
biomass production in batch
cultivation (SD±0.5)

which is similar to previous reports (Farooq et al. 2012) Fig. 5c that lactic acid yield increased up to 2.0 g/L Tween 80
(Fig. 5b). Lactic acid yield increased with increasing cane concentration. A further rise in Tween 80 concentration de-
molasses and amino acid concentration (Fig. 5b). However, creased lactic acid yield due to substrate inhibition effects.
higher amino acid concentrations (>40 mg/L) decreased lactic Unsaturated fatty acids such as Tween 80 act as essential
acid yield due to reduced cell growth (De Lima et al. 2009). growth factors in lactic acid production (Partanen et al. 2001).
Amino acid supplementation enhanced lactic acid yield since
ldh lacks the amino acids L -phenylalanine and L -lysine, and Model verification
biosynthesis of these amino acid requires more energy in the
form of ATP and NADPH. The model was verified by performing experiments in tripli-
Figure 5c showed the interactive effect of amino acids and cate under optimized conditions. A lactic acid yield of 64.86±
Tween 80 on lactic acid yield. It can be clearly deduced from 0.2 g/L was recorded against the predicted yield of 67.78±

7.0

6.5
738
650.8
644.6
631.9
600.3
587.6
611.7573.4
531.6
563.8
472.0
461.3
554.1425.9

6.0
455.5
5.5

5.0 702.7 440.0

3496.8
4.5 522.0

Absorbance 4.0 3383.4

3462.7 510.3

3.5
3333.7
3.0 3913.5 3185.9
803.8
C=o
3965.8
2.5
3859.9
2.0

1.5

1.0

4000 3500 3000 2500 2000 1500 1000 500

Wavenumbers (cm-1)
Fig. 7 FTIR spectrum of lactic acid recovered from Lactobacillus delbrueckii NCIM 2025 under a submerged fermentation process
Ann Microbiol (2014) 64:1211–1221 1219

Fig. 8 Gas chromatogram of a sample obtained after 24.0 h of cultivation showing different peaks. Lactic acid is represented by peak 5. Sharp peaks 1,
3, 6, 7 and 8 represent citric acid, sucrose, acrylic acid, cis-aconitic acid and para amino benzoic acid, respectively

0.3 g/L. Table 5 shows that the predicted and experimental cassava wastewater and starchy waste as substrate after 48 h
lactic acid yield after optimization was in good agreement. fermentation (Coelho et al. 2011; John et al. 2008).

Characterization of lactic acid

Growth kinetics under optimum conditions in a bioreactor
The lactic acid was characterized by FTIR and HPLC.
A shake flask study was then carried out in a laboratory scale Figure 7 shows the FTIR spectra of lactic acid derived from
bioreactor to enhance the lactic acid yield and to understand Lactobacillus delbruckii NCIM 2025. FTIR for lactic acid
the kinetics of lactic acid production under controlled condi- showed two different types of O–H bond—the one in the acid
tions of temperature, pH, agitation and aeration. Figure 6 and the simple “alcohol” type in the chain attached to the –
shows the lactic acid production under optimized conditions COOH group. The O–H bond in the acid group is absorbed
by Lactobacillus delbruckii utilizing cane molasses at initial between the range of 2,500 and 3,300 cm−1, and the one in the
concentrations of 32.0 g/L (total sugar concentration). Lactic chain between 3,230 and 3,550 cm−1. These two taken togeth-
acid and biomass yields of 79.12±0.2 g/L and 95.4±0.3 g/L er create an immense trough covering the whole range from 2,
were obtained after a lag phase of 12.0 h. Total sugar concen- 500 to 3,550 cm−1; lost in this trough will be absorption due to
tration decreased to 6.0 g/L at the end of the production phase C–H bonds. The presence of strong a C=O showed an absorp-
in comparison to an initial concentration of 32.0 g/L. Under tion peak at about 1,730 cm−1. Figure 8 shows the HPLC
controlled physical conditions, lactic acid yield increased from profile of lactic acid and other byproducts such as citric acid,
64.86±0.2 g/L in shake flasks to 79.12±0.2 g/L in the biore- sucrose, lactic acid, acrylic acid, cis-aconitic acid, and para
actor. Lactic acid yield (Y p/x) in terms of cell biomass pro- aminobenzoic acid produced on optimized medium. Peak 5, a
duced and substrate consumed were found to be 84.5 and with retention time of 9.8 min, represents lactic acid.
239.0 %, respectively, which is several fold higher than pre-
vious findings (Albino et al. 2012). A lactic acid yield of
35.0 g/L was obtained in Lactococcus lactis spp. lactis iso- Conclusions
lates under similar cultivation conditions (Serna-Cock and
Rodriguez-de Stouvenel 2006). Lactic acid yields of Lactobacillus delbrueckii gave enhanced lactic acid yield
41.65 g/L and 40.0 g/L were reported in Lactobacillus with L -phenylalanine and L -lysine supplementation, utilizing
rhamnosus B 103 and Lactobacillus delbrueckii utilizing an agro-industrial by-product (cane molasses) and Tween 80

Table 6 Comparison of lactic acid production under similar conditions

Microorganism Substrate Product yield (g/L) Productivity (g/L·h) Fermentation time (h) Reference

Lactobacillus helveticus R211 Whey 66.0 1.4 18.0 Schepers et al. 2002
Enterococcus faecalis RKY1 Corn 63.5 0.5 24.0 Oh et al. 2005
Lactobacillus delbrueckii NCIM 2025 Cane molasses 79.12 3.4 12.0 Present study
Lactobacillus plantarum Corn starch 73.20 0.80 48.0 Kondo et al. 2009
Lactococcus lactis Cassava starch 76.0 1.6 48.0 Ogbonna et al. 2003
1220 Ann Microbiol (2014) 64:1211–1221

as substrate, under a submerged fermentation process. Farooq U, Anjum FM, Zahoor T, Rahman S-U, Randhawa MA, Ahmed
A, Akram K (2012) Optimization of lactic acid production from
Optimized medium comprising (g/L): cane molasses 32.40,
cheap raw material: sugarcane molasses. Pak J Bot 44(1):333–338
Tween 80 2.0 and amino acids (L -phenylalanine and L -lysine Fenice M, Barghini P, Selbmann L, Federici F (2012) Combined effects
in equal proportion) 34 mg/L, gave a maximum lactic acid of agitation and aeration on the chitinolytic enzymes production by
yield of 64.86±0.2 g/L after 24.0 h of cultivation. Batch the Antarctic fungus Lecanicillium muscarium CCFEE 5003.
Microb Cell Factories 11:12
cultivation performed in a 7.5-L bioreactor under optimized
Fitzpatrick JJ, Keeffe UO (2001) Influence of whey protein hydrolyzate
conditions gave lactic acid yield and productivity of addition to whey permeate batch fermentations for producing lactic
79.12 ± 0.2 g/L and 3.40 g/L·h, which was 21.98 % acid. Process Biochem 37:183–186
higher than the lactic acid yield obtained in the shake flask Gonzalez CF, Farina JI, de Figueroa LIC (2008) Optimized amylolytic
enzymes production in Saccharomycopsis fibuligera DSM-70554:
study with decreased fermentation time from 24 to 12 h. The
an approach to efficient cassava starch utilization. Enzym Microb
lactic acid yield obtained in the present study was higher in Technol 42:272–277
terms of yield and productivity in comparison to previous Helm D, Labischinski H, Schallehn G, Naumann D (1991) Classification
reports (Table 6). The data obtained by batch cultivation and identification of bacteria by Fourier-transform infrared spectros-
copy. J Gen Microbiol 137:69–79
kinetics will be useful for metabolic flux analysis of lactic
Hofvendahl K, Hahn-Hagerdal B (2000) Factors affecting the fermenta-
acid production and development of mathematic models in tive lactic acid production from renewable resources. Enzym Microb
different cultivation strategies (fed batch/continuous) for over- Technol 26:87–107
production of lactic acid utilizing cheaper substrates. John RP, Gangadharan D, Nampoothir KM (2008) Genome shuffling of
Lactobacillus delbrueckii mutant and Bacillus amyloliquefaciens
through protoplasmic fusion for l-lactic acid production from
starchy wastes. Bioresour Technol 99:8008–8015
Kondo A, Okano K, Zhang Q, Shinkawa S, Yoshida S, Tanaka T, Fukuda
References H (2009) Efficient production of optically pure D -lactic acid from
raw corn starch by using a genetically modified L -lactate dehydro-
genase gene-deficient and amylase-secreting Lactobacillus
Albino BF, Arias L, Gomez J, Castillo A, Gimeno M, Shirai K (2012) plantarum strain. Appl Environ Microbiol 462–467
Chitin and L(+)-lactic acid production from crab (Callinectes Liu J, Yuan X, Zeng G, Shi J, Chen S (2006) Effect of biosurfactant on
bellicosus) wastes by fermentation of Lactobacillus sp. B2 using cellulase and xylanase production by Trichoderma viride in solid
sugar cane molasses as carbon source. Bioprocess Biosyst Eng 35: substrate fermentation. Process Biochem 41:2347–2351
1193–1200. doi:10.1007/s00449-012-0706 Nagarjun PA, Rao RS, Rajesham S, Rao LV (2005) Optimization of lactic
Budavavi S, O’Neal MJ, Heckelman PE, Kinneary JF (1996) An ency- acid production in SSF by Lactobacillus amylovorus NRRL B-4542
clopedia of chemicals, drugs, and biologicals, 12 edn using Taguchi methodology. J Microbiol 43:38–43
Buyukkileci AO, Harsa S (2004) Batch production of L (+) lactic acid Naumann D, Helm D, Labischinski H (1991) Microbiological characte-
from whey by Lactobacillus casei (NRRL B-441). J Chem Technol rizations by FT-IR spectroscopy. Nature 351:81–82
Biotechnol 79:1036–1040 Naveena BJ, Altaf M, Bhadryya K, Madhavendra SS, Reddy G (2005)
Calabia BP, Tokiwa Y (2007) Production of D -lactic acid from sugarcane Direct fermentation of starch to L (+) lactic acid in SSF by L.
molasses, sugarcane juice and sugar beet juice by Lactobacillus amylophilus GV6 using wheat bran as support and substrate: medi-
delbrueckii. Biotechnol Lett 29:1329–1332 um optimization using RSM. Process Biochem 40:681–690
Serna-Cock L, Rodriguez-de Stouvenel A (2006) Lactic acid production Ogbonna JC, Roble ND, Tanaka H (2003) L-lactic acid production from raw
by a strain of Lactococcus lactis subsp. lactis isolated from sugar cassava starch in a circulating loop bioreactor with cells immobilized in
cane plants. Electronic J Biotechnol 2006:9 loofa (Luffa cylindrica). Biotechnol Lett 25:1093–1098
Coelho LF, de Lima CJB, Rodovalho CM, Bernardo MP, Contiero J Oh H, Wee YJ, Yun JS, Han SH, Jung S, Ryu HW (2005) Lactic acid
(2011) Lactic acid production by new Lactobacillus plantarum production from agricultural resources as cheap raw materials.
LMISM6 grown in molasses: optimization of medium composition. Bioresour Technol 96:1492–1498
Braz J Chem Eng 28(1):27–36 Partanen L, Marttinen N, Alatossava T (2001) Fats and fatty acids as
De Lima CJB, Coelho LF, Blanco KC, Contiero J (2009) Response growth factors for Lactobacillus delbrueckii. Syst Appl Microbiol
surface optimization of D (−) lactic acid production by 24:500–506
Lactobacillus SMI8 using corn steep liquor and yeast autoly- Qi B-K, Yao RS, Lai M, Deng S-S (2009) Effect of Tween 80 on
sate as an alternative nitrogen source. Afr J Biotechnol 8:5842– production of lactic acid by Lactobacillus casei. Songklanakarin J
5846 Sci Technol 31:85–89
Dubois MKA, Gilles KA, Hamilton JK, Rebers PA, Smith F (1956) Ryu HW, Yun JS, Wee YJ (2003) Lactic acid. In: Pandey A (ed) Concise
Colorimetric method for determination of sugars and related sub- encyclopedia of bioresource technology. Haworth, New York
stances. Anal Chem 28:350–356 Schepers AW, Thibault J, Lacroix C (2002) Lactobacillus heveticus
Duggan DE, Anderon AW, Ellker PR (1959) A frozen concentration of growth and lactic acid production during pH-controlled batch cul-
Lactobacillus acidophilus for preparation of a palatable acidophilus tures in whey permeate/yeast extract medium. Part II: kinetic model-
milk. Food Technol 13:465–469 ing and model validation. Enzym Microb Technol 30:187–194
Dumbrepatil A, Adsul M, Chaudhari S, Khire J, Gokhale D (2008) Silva CC, Dekker RFH, Silva RSSF, Silva MDLCD, Barbosa AM (2007)
Utilization of molasses sugar for lactic acid production by Effect of soybean oil and Tween 80 on the production of
Lactobacillus delbrueckii subsp. delbrueckii mutant Uc-3 in batch botryosphaeran by Botryosphaeria rhodina MAMB-05. Process
fermentation. Appl Environ Microbiol 74:333–335 Biochem 42:1254–1258
Elisashvili V (2012) Submerged cultivation of medicinal mushrooms: Sriphochanart W, Skolpap W, Scharer JM, Moo-Young M, Douglas PL
bioprocesses and products (review). Int J Med Mushrooms 14(3): (2011) Afr J Microbiol Res 5(22):3815–3822
211–239 Tashiro Y, Kaneko W, Sun Y, Shibata K, Inokuma K, Zendo T, Sonomoto
Ann Microbiol (2014) 64:1211–1221 1221

K (2011) Continuous D -lactic acid production by a novel of L -(+)-lactic acid production using corn steep liquor as an alterna-
thermotolerant Lactobacillus delbrueckii subsp. lactis qu 41. Appl tive nitrogen source by Lactobacillus rhamnosus CGMCC 1466.
Microbiol Biotechnol 89(6):1741–1750 Biochem Eng J 39:496–502
Vickroy TB (1985) Lactic acid. In: Blanch HW, Drew S, Wang DIC (eds) Yun JA, Wee YW, Kim JN, Ryu HW (2004) Fermentative production of
The practice of biotechnology: commodity products. Pergamon, DL -lactic acid from amylase-treated rice and wheat brans hydroly-
Elmsford, pp 761–776 zate by a novel lactic acid bacterium, Lactobacillus sp. Biotechnol
Wee UJ, Kim JN, Yun JS, Ryu HW (2004) Utilization of sugar Lett 26:1613–1616
molasses for economical L (+) lactic acid production by bath Zeng GM, Shi JG, Yuan XZ, Liu J, Zhang ZB, Huang GH, Li JB, Xi BD,
fermentation of Eneterococcus faecalis . Enzym Microbiol Liu HL (2006) Effects of Tween 80 and rhamnolipid on the extra-
Technol 35:568–573 cellular enzymes of Penicillium simplicissimum isolated from com-
Yu L, Lei T, Ren X, Pei X, Feng X (2008) Response surface optimization post. Enzym Microbiol Technol 39(7):1451–1456