Determination of expression and activity of genes

International Journal of Food Microbiology 185 (2014) 103–111

Contents lists available at ScienceDirect

International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro

Determination of expression and activity of genes involved in starch metabolism in Lactobacillus plantarum A6 during fermentation of a cereal-based gruel Christèle Humblot ⁎, Williams Turpin 1, François Chevalier, Christian Picq, Isabelle Rochette, Jean-Pierre Guyot IRD, UMR Nutripass IRD/Montpellier2/Montpellier1, F-34394 Montpellier, France

a r t i c l e

i n f o

Article history: Received 29 August 2013 Received in revised form 6 May 2014 Accepted 10 May 2014 Available online 27 May 2014 Keywords: mRNA Real time PCR Pearl millet Kinetics

a b s t r a c t Traditional fermented gruels prepared from cereals are widely used for complementary feeding of young children in Africa and usually have a low energy density. The amylase activity of some lactic acid bacteria (LAB) helps increase the energy content of gruels through partial hydrolysis of starch, thus enabling the incorporation of more starchy material while conserving the desired semi-liquid consistency for young children. Even if this ability is mainly related to the production of alpha-amylase (E.C. 3.2.1.1), in a recent molecular screening, genes coding for enzymes involved in starch metabolism were detected in the efficient amylolytic LAB Lactobacillus plantarum A6: alpha-glucosidase (E.C. 3.2.1.20), neopullulanase (E.C. 3.2.1.135), amylopectin phosphorylase (E.C. 2.4.1.1) and maltose phosphorylase (E.C. 2.4.1.8). The objective of this study was to investigate the expression of these genes in a model of starchy fermented food made from pearl millet (Pennisetum glaucum). Transcriptional and enzymatic analyses were performed during the 18-h fermentation period. Liquefaction was mainly caused by an extracellular alpha amylase encoded by the amyA gene specific to the A6 strain among L. plantarum species and found in both Lactobacillus amylovorus and Lactobacillus manihotivorans. The second most active enzyme was neopullulanase. Other starch metabolizing enzymes were less often detected. The dynamic detection of transcripts of gene during starch fermentation in pearl millet porridge suggests that the set of genes we investigated was not expressed continuously but transiently. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Traditional fermented gruels prepared from cereals are widely used for complementary feeding of young children in Africa. In Burkina Faso and Ghana, ben-saalga and koko gruels are prepared from fermented pearl millet (Pennisetum glaucum) slurries, and usually have a low energy density (Lei and Jakobsen, 2004; Tou et al., 2006). The main actors of the fermentation are lactic acid bacteria (LAB), mainly represented by the Lactobacillus, Pediococcus and Weissella genera (Humblot and Guyot, 2009; Lei and Jakobsen, 2004). The amylase activity of some LAB helps increase the energy content of gruels through partial hydrolysis of starch in the food matrix enabling more starchy material to be incorporated while conserving the desired semi-liquid consistency for young children (6–24 months old) (Nguyen et al., 2007; Songré-Ouattara et al., 2008). These bacteria could also be used to produce fermented functional foods as a non-dairy substitute for commercial fermented dairy beverages for populations with particular needs, for example, consumers with lactose intolerance. ⁎ Corresponding author at: IRD, UMR Nutripass IRD/Montpellier2/Montpellier1, B.P. 64501, 911 Avenue Agropolis, 34394 Montpellier Cedex 5, France. Tel.: +33 467416466; fax: +33 467416157. E-mail address: [email protected] (C. Humblot). 1 Present address: University of Toronto, Toronto, ON M5S1A8, Canada.

http://dx.doi.org/10.1016/j.ijfoodmicro.2014.05.016 0168-1605/© 2014 Elsevier B.V. All rights reserved.

Lactobacillus plantarum is a facultative heterofermentative LAB found in numerous fermented foods, from dairy and meat products to a variety of vegetable and plant fermentations; it is also present in the human intestinal tract (de Vries et al., 2006). L. plantarum has been used not only as a starter culture for food, but also as a probiotic and as a delivery vehicle for therapeutic compounds thanks to its ability to survive and persist in the gastrointestinal tract (Amdekar et al., 2009). Among a range of potentially useful characteristics, some strains isolated from different fermented foods were shown to have the ability to hydrolyse starch. In this work, the amylolytic LAB (ALAB) L. plantarum A6, isolated from fermented cassava, was chosen as a model to investigate the expression of genes related to starch metabolism in a starchy food matrix. Indeed, L. plantarum A6 has already been used to produce high energy density fermented gruels made from pearl millet, rice, plantain and sweet potato, among others (Giraud et al., 1991, 1994; Haydersah et al., 2012; Nguyen et al., 2007; Songre-Ouattara et al., 2010), and its effectiveness may be linked to its ability to express its genetic potential in an amylaceous matrix. Recently genes and enzymes involved in the metabolism of oligosaccharides and starch in lactobacilli were the object of a review (Ganzle and Follador, 2012). In the case of ALAB, many studies focused on the enzyme alpha-amylase (E.C. 3.2.1.1), which plays a key role in starch hydrolysis since it catalyses the endohydrolysis of (1-4)-alpha-D-glucosidic linkages in polysaccharides containing three or more (1-4)-alpha-linked D-

104

C. Humblot et al. / International Journal of Food Microbiology 185 (2014) 103–111

glucose units (Rodriguez Sanoja et al., 2000). However, other genes encoding several enzymes involved in starch hydrolysis and metabolism were found in the genomes of different L. plantarum strains, including the A6 strain. Neopullulanase (E.C. 3.2.1.135) is able to hydrolyse pullulan to produce panose (6-alpha-D-glucosylmaltose); amylopectin phosphorylase (E.C. 2.4.1.1) acts on starch or amylose to release D-glucose-1phosphate; alpha-glucosidase (E.C. 3.2.1.20) hydrolyses the terminal non-reducing (1-4)linked alpha-D-glucose residue with the release of D-glucose and maltose phosphorylase (E.C. 2.4.1.8) liberates D-glucose and beta-D-glucose-1-phosphate (G1P) (Kleerebezem et al., 2003; Turpin et al., 2011; Zhang et al., 2009). We thus hypothesized that together with alpha-amylase, they may also contribute to starch hydrolysis and metabolism and therefore to the changes in rheological properties observed during the fermentation of starchy raw materials once they are gelatinized (Nguyen et al., 2007; Songre-Ouattara et al., 2010). Only a few authors have investigated the expression of genes of interest in food matrixes at both transcriptional and proteomic levels (Oguntoyinbo and Narbad, 2012; Plumed-Ferrer et al., 2008) and to date there has been no dynamic view of gene expression in food matrixes linked with fermentation kinetics. In this work, the dynamic expression of genes coding for enzymes involved in starch hydrolysis and metabolism in amylaceous foods was investigated at transcriptional and enzymatic levels during the fermentation of an amylaceous food matrix by L. plantarum A6. As sequences of genes and proteins of most of these enzymes are highly variable among Lactobacillus species, we focused our investigations on the genes described for the species L. plantarum (NCBI). The data obtained in the food matrix were compared to data obtained in a reference culture medium containing starch as the sole carbon source. 2. Material and methods 2.1. Chemicals and raw material MRS medium and agar were purchased from Difco (Le Pont de Claix, France) and soluble potato starch from VWR (Fontenay-sous-Bois, France). All other chemicals were purchased from Sigma-Aldrich (St-Quentin-Fallavier, France). Pearl millet was purchased on a local market in Ouagadougou (Burkina Faso) for the preparation of gelatinized starchy slurries, since the efficiency of starch hydrolysis by LAB in such food matrixes has already been characterized in previous studies (Songre-Ouattara et al., 2009, 2010). The same batch of pearl millet grains was used for all experiments.

was negative for this test and was then used as a negative control for the experiments with the gelatinized slurry. All the fermentations were incubated at 30 °C under static conditions and sampled at hourly intervals for 18 h. All samples were then kept at −80 °C until analysis. Viable L. plantarum A6 bacteria were counted at the beginning and at the end of the fermentation by plating on MRS agar. The fermentations were performed as three independent experiments.

2.3. Measurement of gruel consistency The consistency of the gruels fermented by each L. plantarum strain was measured after 18 h of fermentation using a Bostwick consistometer (CSC Scientific Company Inc., Fairfax, Virginia, USA) (Bookwalter et al., 1968) according to the procedure described by Mouquet et al. (2006). The Bostwick flow value was expressed in mm/30 s. This parameter ranges from 0 (almost solid) to 240 mm/30 s (very fluid, corresponding to a gruel that can be drunk). Results are expressed as mean ± standard error of the mean (SEM).

2.4. Analysis of starch and starch degradation products The residual starch content in the samples of MRS-starch and in the gruels fermented with L. plantarum A6 was determined at the beginning and at the end of the fermentation using the K-TSTA kit (Megazyme, Wicklow, Ireland) following the manufacturer's instructions. Mono- and disaccharides, maltodextrins with 3 to 7 glucose units (G3 to G7) and fermentation products were analysed in the supernatants of all the samples of MRS-starch fermented with L. plantarum A6 after centrifugation at 8000 g for 10 min at 4 °C. For the analysis of the slurries, 2 g of slurry fermented by L. plantarum A6 or WCFS1, was homogenized with 8 mL of distilled water, 0.2 mL of H2SO4 2 N was added to each 1.3 mL sub-sample in microtubes, which were then centrifuged at 8000 g for 10 min at 4 °C. All the supernatants were filtered through 0.20 μm pore size filters and after appropriate dilution, 25 μL was injected and analysed by high-performance anion exchange chromatography (HPAEC) using an anion exchange chromatograph (Dionex S.A., Voisins-Le-Bretonneux, France) equipped with a CarboPac PA1 column and amperometric detection. The following conditions were used: mobile phase: H2O, NaOH (150 mM), sodium acetate gradient of 0 to 300 mM; flow rate: 0.1 mL/min; temperature: 35 °C. Results are expressed as mean ± SEM.

2.2. Experimental design 2.5. Measurement of enzymatic activities An overnight culture of the amylolytic strain L. plantarum A6 (LMG 18053) (Giraud et al., 1991) was inoculated (1%) on a modified MRS medium (MRS-starch) with soluble potato starch as the sole carbon source (20 g/L) or on gelatinized pearl millet slurries (10% dry matter) following the modified method described in Tou et al. (2007), which included soaking of the kernels, grinding and settling (fermentation) steps (Tou et al., 2007). The gelatinization was done by cooking at 80 °C for 10 min, this also ensures a drastic lowering of the endogenous microbiota (b 10 cfu/g of slurry) allowing the specific growth of the inoculated bacteria. To inoculate the gelatinized pearl millet slurry, an overnight culture on MRS was centrifuged at 8000 g for 10 min at 4 °C, and the cells obtained were washed in sterile NaCl 0.9 g/L, centrifuged in the same conditions and suspended in sterile NaCl 0.9 g/L to form a cell suspension containing approximately 109 cfu/mL, 1% of this suspension was inoculated to the gelatinized slurry (final concentration approximately 107 cfu/g of slurry). The strain L. plantarum WCFS1 (Kleerebezem et al., 2003) was tested for its ability to hydrolyse starch using Lugol's solution (0.4%, v/v) by measuring the clear zones around the colonies on agar MRS-starch (Agati et al., 1998). The strain

All the enzymatic measurements were performed on each sample, which were taken at hourly intervals (from 0 h to 18 h) from MRSstarch or gruels fermented with L. plantarum A6. Alpha-amylase activity (E.C. 3.2.1.1, substrate para-nitrophenyl maltoheptaoside, product measured para-nitrophenol) was measured using the CERALPHA method (Megazyme) following the manufacturer's instructions. Neopullulanase activity (E.C. 2.2.1.135, substrate pullulan, product measured reducing sugars) was measured as described previously (Kim et al., 2009). Maltose phosphorylase (E.C. 2.4.1.8, substrate maltose, product measured NADPH) and amylopectin phosphorylase (E.C. 2.4.1.1, substrate starch, product measured NADPH) were analysed according a procedure previously described for sourdough LAB (Stolz et al., 1996). Alpha-glucosidase activity (E.C. 3.2.1.20, substrate maltose, product measured glucose) was measured as described previously (Bergmeyer et al., 1988). Results are expressed as mean ± SEM.


2.6. Total RNA extraction, quality evaluation and reverse transcription To limit RNA degradation, RNA extractions were performed on samples kept at −80 °C for less than three days, and the extraction was performed at 4 °C. For the extraction from food matrixes, as the slurries were very sticky, they were diluted 10 times in NaCl 0.9 g/L and centrifuged for 10 min at 1000 g twice to eliminate the gelatinized starch. The supernatant was then centrifuged for 10 min at 10,000 g to pellet the bacteria. For the extraction from MRS-starch experiments, samples were centrifuged (10 min, 10,000 g) to pellet the bacteria. The pellet was resuspended in TE buffer (1 mM EDTA, 10 mM Tris, pH 7; Promega, Charbonnières, France), and the resulting suspension was lysed using an amalgamator with zirconium beads (VWR, Fontenay-sous-bois, France) in acid phenol at pH 4 (Eurobio, Ulysse, France) to allow disruption of bacteria. After centrifugation, the aqueous phase was transferred in TRIzol® Reagent (Invitrogen, Carlsbad, USA) and incubated for 5 min at room temperature. After addition of chloroform (Carlo Erba, Val de Reuil, France), the solution was centrifuged (10,000 g, 15 min) and the nucleic acid was precipitated by the addition of isopropanol (Sigma, Saint Louis, USA). The pellet was washed in 70% ethanol (Carlo Erba), suspended in nuclease free water (Promega), and kept overnight at −80 °C. The quality of RNA was checked using Nano Vue (GE Healthcare, Vélizy Villacoublay, France) and electrophoresis on 2% (w/v) agarose gel. RNA concentration was 33 ± 9 ng/μL from food samples and 298 ± 20 ng/μL from MRS-starch experiments. The ratio A260/A280 was above 1.8 and the ratio A260/A230 was above 1.9. The DNA was removed with RQ1 RNase-Free DNase (Promega) and on the same day, RNA was converted into cDNA using the AMV Reverse Transcription System (Promega) following the manufacturer's instructions (Turpin et al., 2012). The absence of genomic DNA in treated RNA samples was checked by real time PCR using the following primers: 338f converted into its reverse complement, 5′-CTGCTGCCTCCCGTAGGAGT-3′ (Muyzer et al., 1993) and Lpla72f 5′-ATCATGATTTACATTTGAGTG-3′ (Chagnaud et al., 2001) specific to the 16S rRNA gene sequence of L. plantarum. The real time PCR conditions are described in the following paragraph. 2.7. Measurement of the expression of genes related to starch metabolism at mRNA level All the expression measurements were performed on each sample, which were taken at hourly intervals (from 0 h to 18 h) from MRSstarch or gruels fermented with L. plantarum A6. The expression of the following genes was investigated: glgP coding for amylopectin phosphorylase, agl coding for alpha-glucosidase, a-amy coding for alphaamylase, amyA coding for extra-cellular alpha-amylase, dexC coding for neopullulanase and malP coding for maltose phosphorylase. All measurements were performed in duplicate on samples taken from L. plantarum A6 experiments using the QPCR system (Stratagene, Mx3005p™) and Sybr Green technology (Eurogentec, Angers, France).

105

For each reaction, 1 μL of the cDNA template was added to 15 μL of PCR mix containing 1X MESA GREEN qPCR MasterMix Plus (Eurogentec, Angers, France) and 0.3 mM of each primer. The PCR conditions used were 10 min at 95 °C and 40 cycles of 30 s at 95 °C, then 30 s at 50 °C, then 30 s at 72 °C, followed by a dissociation gradient from 55 °C to 95 °C to check PCR specificity by examining the temperaturedependent melting curves of the PCR products. The absence of background contamination was checked in real-time PCR realised without template (no-template control; NTC). The absence of residual chromosomal DNA was checked using real-time PCR with primers targeting 16S rRNA coding genes on RNA extracts before reverse transcription. Data were analysed using MxPro QPCR software 2007 Stratagene version 4.10. Table 1 shows the efficiency of the real time PCR assays for each primer pair calculated from the standard curves made with four different log10 dilutions of DNA extracted from pure cultures of L. plantarum A6 grown in MRS broth. Correlation coefficients were above 0.99 for each calibration curve. As no constitutively expressed reference gene was identified and as the rRNA level fluctuates during fermentation due to bacterial growth, we chose to express our results as qualitative analysis following the recommendations of Bustin et al. (2009). This allows assessing whether or not a nucleic acid template is present, rather than to quantify it accurately (Bustin et al., 2009). As a cut-off, we chose a quantification cycle (Cq) of 32, taking into account that the maximum Cq detected throughout the analysis was 31, and that the NTC gave “no Cq detected”. The results are presented individually for each (A, B or C) of the three independent experiments. 3. Results 3.1. Effect of the fermentation of starch by the A6 strain in MRS-starch and pearl millet gruels on mono-, disaccharide and maltodextrin compositions and on the consistency of the gruel With L. plantarum A6 in the liquid cultivation medium (MRS-starch), after the consumption of simple carbohydrates within the first 3 h transitory production of glucose, maltose, G3 and G4 was observed (Fig. 1A). The production of G5, G6 and G7 followed the same pattern but was less important since they reached a maximum of 0.4, 0.1 and 0.2 mmol/L, respectively. The starch concentration decreased from 18.9 ± 0.8 g/L (105 ± 5 mmol glucose equivalent/L) to 4.1 ± 0.3 g/L (23 ± 2 mmol glucose equivalent/L) of MRS-starch from the beginning to the end of fermentation. The bacterial populations increased from 6.6 × 107 ± 7.8 × 106 to 1.5 × 108 ± 3.1 × 107 cfu/mL of MRS-starch from the beginning to the end of fermentation. The pearl millet slurry was thick and sticky at the beginning of the fermentation (Bostwick flow of 0 mm/30 s). The initial pH was 7 ± 0.1. At the end of fermentation by L. plantarum A6, the consistency of the gruel was liquid with a Bostwick flow of 168 ± 8 mm/30 s. The

Table 1 List of primers and associated real time PCR efficiencies used to measure the expression of genes related to starch metabolism in samples collected from pearl millet gruels and MRS-starch inoculated with L. plantarum A6. Gene

Predicted function

Primer sequences 5′ to 3′

Primer reference

Melting temperature (°C)

PCR efficiency

glgP

Amylopectin phosphorylase

Turpin et al. (2011)

60.0

128%

agl

Alpha-glucosidase


59.5

107%

a-amy

Alpha amylase


60.0

153%

amyA

Extracellular alpha-amylase

This study

60.0

108%

dexC

Neopullulanase


60.0

100%

malP

Maltose phosphorylase

F_GCGGGTGTTCAAAGTATCGT R_TCTCGAGGGCCTCTTGTAAA F_GCsAAAATGCTAGCGACymT R_CCACTGCATyGGyGTACGy F_AGATCAGGCGCAAGTTCAGT R_TTTTATGGGCACACCACTCA F_ TATTTTGCATGCATGGTGCT R_ GCAGCGCACATTGACTTAAA F_CCAGACGAGCAAGAACAACA R_ATTGGCGATACGCCACTTAC F_TGCCAyAAyGArTGGGArAT R_ACsCkATCwGCCCArAAAC


60.0

105%

[saccharides] (mmol/kg of sllurry)

106


6

final pH was 3.9 ± 0.1. Liquefaction was accompanied by a decrease in starch content from the onset to the end of the fermentation (110 ± 10 g/kg and 88 ± 5.3 g/kg, respectively) and by the production of starch degradation products with a very different pattern from that of the MRS-starch (Fig. 1B). After the total consumption of glucose, fructose, and sucrose within 8 h, maltooligosaccharides, maltose, and glucose were produced. The main maltooligosaccharides produced had polymerization degrees of 3 and 4 glucose units. The production of maltooligosaccharides with higher polymerization degrees followed the same pattern with a maximum of 0.7, 0.5 and 0.6 mmol/kg of slurry for G5, G6 and G7, respectively, at the end of the fermentation. In contrast, in the negative control, fermentation of the slurry using the nonamylolytic strain L. plantarum WCFS1 did not modify the very sticky paste (0 mm/30 s) and the strain only consumed the glucose and fructose in the paste (Fig. 1C) without producing any starch degradation molecules. The L. plantarum A6 bacterial populations in the pearl millet slurry varied between 9.6 × 106 ± 1.2 × 105 and 6.6 × 108 ± 3.3 × 107 cfu/g of slurry from the beginning to the end of fermentation.

A

5 4 3 2 1 0 0

3

6

9

12

15

18

time (h) glucose maltose G5

[saccharides] (mmol/kg of slurry)

50

fructose G3 G6

saccharose G4 G7

B

45

3.2. Expression of genes of L. plantarum A6 coding for enzymes involved in starch metabolism

40 35 30 25 20 15 10 5 0 0

3

6

9

12

15

18

time (h) glucose maltose G5

50

fructose G3 G6

saccharose G4 G7

C

[saccharides ] (mmol/kg of slurry)

45 40 35 30 25 20 15

All melting curves were specific and most of the PCR efficiencies of the real time PCR assays for each primer pair calculated from the standard curves were in the normal range, which met the experimental requirements of the real-time PCR. The PCR efficiencies corresponding to the primer pair targeting glgP and amyA were much higher than 100%. As we chose to express qualitatively our results this was not critical, but those primers should be used carefully in the case of quantitative analysis (Table 1, Figs. S1, S2). The detection of transcripts of genes coding for the six enzymes involved in starch metabolism, glgP, agl, a-amy, amyA, dexC and malP, in L. plantarum A6 inoculated onto the MRS-starch medium are presented in Figs. 2 and S1. The transcripts of genes coding for the six enzymes investigated were variable between the repeated independent experiments, with the exception of maltose phosphorylase (malP) which was detected throughout the fermentation. The transcripts of most of the genes were detected at the beginning of the fermentation. The transcripts in the three repeated experiments were all detected between 7 and 9 h after inoculation, and then were less frequently detected between 11 and 16 h of fermentation. At the end of the fermentation, the transcripts of glgP, a-amy, amyA, dexC and malP were still detected, while the transcript of on the whole the less often detected agl was not detected anymore. During pearl millet fermentation there is even more variability in gene expression between the three independent experiments (Figs. 3, S2). The transcripts most often detected are from amyA and malP even if the transcripts of all the genes coding for the six enzymes were also less often detected than in MRS-starch.

10

3.3. Enzymatic activities of L. plantarum A6 in the food matrix

5 0 0

3

6

9

12

15

18

time (h) glucose

fructose

saccharose

maltose

G3

G4

G5

G6

G7

Fig. 1. Mono-, disaccharide, and maltodextrin (G3 and G4) concentrations during the incubation of MRS-starch inoculated with L. plantarum A6 (A), during the incubation of slurry inoculated with L. plantarum A6 (B) and during the incubation of slurry inoculated with L. plantarum WCFS1 (C). Results are expressed as the mean ± SEM of three independent experiments.

Enzymatic activities were measured in the MRS-starch and in the pearl millet gruels. In MRS-starch, the activities of maltose phosphorylase and alpha-glucosidase were near the detection limit. Alphaamylase activity increased continuously from 6 h to 15 h of incubation and then levelled off (Fig. 4). Transient activity of neopullulanase was observed; activity peaked after 8 h of incubation and then after 13 h, decreased to its original level (Fig. 4). Measurements of extra-cellular alpha-amylase activity in the supernatants of fermented pearl millet slurries showed weak activity (data not shown). In contrast, tests of activity in the slurry without separation of bacteria from the matrix enabled detection of much higher alphaamylase activity, which increased after 6 h and continued up to 12 h of fermentation (Fig. 5). Neopullulanase activity was also measured from 7 h to the end of the fermentation. Alpha-glucosidase activity


107

Fig. 2. Transcriptional expression of genes related to starch metabolism in L. plantarum A6 inoculated on MRS-starch. Measurements were performed on each sample, taken at hourly intervals (from 0 h to 18 h). Results are expressed individually from three independent experiments named A, B and C. Grey boxes represent an absence of detection of the transcript coding for the corresponding gene. White boxes represent a positive detection of the transcript coding for the corresponding genes. On the right of each figure is represented the range of Cq from each experiment.

was also detected but at a very low level, although a slight increase occurred during the course of fermentation (Fig. 5). 4. Discussion To date, the ability of amylolytic LAB to liquefy starchy matrixes during fermentation and to grow in such conditions has mainly been linked to the production of alpha-amylase (Diaz-Ruiz et al., 2003; Nguyen et al., 2007; Songré-Ouattara et al., 2008). However, based on the analysis of different enzyme activities, a tentative pattern of starch metabolism was proposed for the amylolytic LAB Lactobacillus fermentum Ogi E1 (Calderon Santoyo et al., 2003). This pattern involved not only the role of alpha-amylase but also a role for other enzymes including alphaglucosidase. In the same way, recent molecular screening of genes involved in functions of interest in human nutrition revealed a more complete picture of the metabolic potential for starch metabolism by LAB, in particular L. plantarum A6 (Turpin et al., 2011). Genes coding for the following enzymes: alpha-amylase (E.C. 3.2.1.1), alpha-glucosidase (E.C. 3.2.1.20), neopullulanase (E.C. 3.2.1.135), amylopectin phosphorylase (E.C. 2.4.1.1) and maltose phosphorylase (E.C. 2.4.1.8) were detected on genomic DNA by PCR (Turpin et al., 2011) leading us to investigate whether or not they were expressed in a starchy food matrix. Two genes coding for alpha-amylases were identified in the strain L. plantarum A6: the a-amy gene, which codes for an intracellular alpha-amylase and is found in several different species of Lactobacilli,

and the amyA gene, which codes for an extracellular alpha-amylase and is found only in amylolytic strains of species L. plantarum, Lactobacillus manihotivorans and Lactobacillus amylovorus (Table 2). It is interesting to note that the genes glgP, agl, a-amy, dexC, and malP coding for amylopectin phosphorylase, alpha-glucosidase, alpha-amylase, neopullulanase and maltose phosphorylase, respectively, were also detected in the non-amylolytic strain L. plantarum WCFS1, which was unable to liquefy the pearl millet slurry (Turpin et al., 2011). In addition, the malS gene, which was not found in L. plantarum A6 (data not shown), was annotated as coding for another alpha-amylase in the genome of L. plantarum WCFS1 and was named amy2 (similarity N 99%) in L. plantarum JDM1 and ST-III. Furthermore, genome analysis of these strains suggests that the genes listed above are complete and could be functional (Kleerebezem et al., 2003). So, one major difference between the amylolytic A6 strain and the WCFS1 strain is the presence of the amyA gene in L. plantarum A6, which codes for an extracellular alphaamylase (Giraud and Cuny, 1997), suggesting that this enzyme is responsible for the efficiency of L. plantarum A6 in liquefying the slurry. The apuA gene, which codes for an amylopullulanase with both alpha-amylase and neopullulanase activities, has been shown to be responsible for starch hydrolysis by the amylolytic strain of L. plantarum L137 (Kim et al., 2008). This gene was not detected in the genome of L. plantarum A6 by PCR and presents no similarity with any genes in the genomes of L. plantarum WCFS1, ST-II or JDM1, and consequently cannot account for the amylolytic property of L. plantarum A6.

108


Fig. 3. Transcriptional expression of the starch metabolism related genes in L. plantarum A6 inoculated on slurry. Measurements were performed on each sample, taken at hourly intervals (from 0 h to 18 h). Results are expressed individually from three independent experiments named A, B and C. Grey boxes represent an absence of detection of the transcript coding for the corresponding gene. White boxes represent a positive detection of the transcript coding for the corresponding genes. On the right of each figure is represented the range of Cq from each experiment.

Our work focused on L. plantarum because of its importance in the fermentation of plants – particularly in the fermentation of cereals and cassava – in developing countries. However Petrov et al. (2008) investigated four genes in a phylogenetically distant lactic acid bacterium, Lactococcus lactis subsp. lactis B84 isolated in rye sourdough. The four genes: amyL, amyY, apu and glgP coding for cytoplasmic and extracellular alpha-amylases, amylopullulanase and glycogen phosphorylase, respectively, are involved in starch degradation. When the strain was grown in a modified MRS-starch medium, the two genes coding for alpha-amylases (amyl and amyY) were expressed, whereas apu and glgP were not, strengthening the hypothesis of the central role of alpha-amylase genes in starch degradation (Petrov et al., 2008). The efficiency of the A6 strain in fermenting starchy materials appears to be mostly due to extracellular alpha-amylase. Glucose, maltose and maltooligosaccharides (mainly G3 and G4) were consistently produced. However, the kinetic patterns differed with the growth conditions. In MRS-starch, degradation products only appeared transiently. The starch in this liquid medium was at a very low concentration and was rapidly consumed compared to that in the pearl millet slurry. The kinetic patterns of carbohydrates and enzymes in the culture medium were of limited use in describing the real behaviour of the ALAB in the food matrix but, under controlled conditions, confirmed the synthesis of enzymes whose pattern also depended on growth conditions. High

alpha-amylase activity was observed in both incubation media from 6 h to the end of the fermentation, with higher activity in the MRSstarch than in the pearl millet slurry. Amylase activity in the A6 strain grown in the MRS-starch medium was as expected (Giraud et al., 1991). In the fermented pearl millet slurries, this activity was higher in the whole matrix than in the liquid phase obtained after centrifugation. This suggests that the enzyme in the matrix is difficult to recover, probably because it is trapped in the gel formed by the gelatinized starch and other macropolymers. Another possible reason is that the alpha-amylase could be bound to the starch (Rodriguez-Sanoja et al., 2005). Less frequent detection of the transcripts of the six genes was found in the food matrix than in MRS-starch. This could be due to the longer duration of the extraction step of total RNA from the food matrix than from MRS-starch culture. Indeed, we had to extract the microbial fraction from the food matrix prior to RNA extraction as it was essential to obtain enough bacterial RNA to perform the experiments. This is consistent with the results of Ampe et al. (1998), which showed the importance of this step on gelatinized starchy food made from maize (Ampe et al., 1998). Very few studies have analysed in situ expression of LAB genes in food matrixes, especially starchy ones. Recently, in a study of two African fermented cereal foods, the in situ expression of an alpha-

400

40

350

35

300

30

250

25

200

20

150

15

100

10

50

5

0

0 0

2

4

6

8

10

12

14

16

18

neopullulanse activity (mmol of reducing sugar produced/min/L of MRS-starch)

Ceralpha unit of alpha-amylase/Lof MRS-starch


time (h) alpha-amylase

neopullulanase

Fig. 4. Alpha-amylase and neopullulanase activities of L. plantarum A6 inoculated on MRSstarch. Results are expressed as the mean ± SEM of three independent experiments.

Ceralpha unit of alpah-amylase/kg of slurry

350

300

30

250

20

200

150

10

100

50

0

0

2

4

6

8

10

12

14

16

18

0

alpha-glucosidase activity (mmol glucose formed/min/kg of slurry)

40

400

neopullulanse activity (mmol of reducing sugar produced/min/kg of slurry)

amylase gene by an amylolytic strain L. plantarum ULAG11 grown in a millet slurry was shown (Oguntoyinbo and Narbad, 2012). Direct comparison with our data was difficult as these authors reported results obtained 12 h and 24 h after inoculation and used a different gene expression method (relative gene expression). Nevertheless, they also measured alpha-amylase mRNA production in the food matrix, which peaked 12 h after inoculation. This is consistent with the results of our

time (h) alpha-amylase

alpha-glucosidase

neopullulanase

Fig. 5. Alpha-amylase, alpha-glucosidase and neopullulanase activities of L. plantarum A6 inoculated on pearl millet slurry. Results are expressed as the mean ± SEM of three independent experiments.

109

experiments, as a peak in expression was observed 13 h after inoculation. In a recent study on the metatranscriptome of wheat and spelt sourdoughs, Weckx et al. (2011) showed that the alpha-amylase gene (the name of the gene was not specified) was moderately expressed in situ by L. plantarum in wheat sourdoughs and only just over the expression threshold in spelt sourdough (Weckx et al., 2011). Together, these data and ours suggest that the expression of the amylase genes varies depending on the starchy matrix. However, unlike in fermented foods made with tropical cereals, ALAB do not play a major role in fermentation in wheat sourdough. The second most active enzyme was neopullulanase, which is able to hydrolyse pullulan to produce panose (6-alpha-D-glucosylmaltose). In the MRS-starch medium, neopullulanase activity and transcripts were observed during approximately the same period of time. During the fermentation of the pearl millet slurry, transcription of the gene was more variable than in MRS-starch. Detection of amylase and neopullulanase activities, in addition to transcript analysis, confirmed that the corresponding genes are functional. The fragment amplified by real time PCR closely resembled a gene coding for a maltogenic alpha-amylase in L. plantarum WCFS1 also called glucan 1,4-alpha-maltohydrolase in the L. plantarum JDM1 and ST-III strains (Table 2). The NCBI database analysis of this enzyme's conserved domain showed that, like alphaamylase, this protein presents a conserved domain on the alphaamylase catalytic domain, in addition to an N-terminal Early set domain associated with the catalytic domain of cyclomaltodextrinase and pullulan-degrading enzymes. Cyclomaltodextrinase, maltogenic amylase, and neopullulanase are all capable of hydrolysing two or all three types of substrates: cyclomaltodextrins, pullulan, and starch, and these enzymes are nearly indistinguishable from each other (Lee et al., 2002) making it difficult to attribute a specific role to the neopullulanase produced by the A6 strain in the metabolism of starch. Indeed, there is a lack of information on the neopullulanase produced by LAB. However, in other Firmicutes like Bacillus subtilis, and Bacillus (Geobacillus) stearothermophilus, neopullulanase has also been reported to be able to hydrolyse the alpha-(1-6)-glucosidic linkages, and to catalyse both alpha-(1-4) and alpha-(1-6) trans-glycosylations from maltotriose to produce oligosaccharides (Takata et al., 1992). Although the liquefying action of the A6 strain and other amylolytic LAB has been attributed to alpha-amylase activity (Nguyen et al., 2007; Songre-Ouattara et al., 2009), the contribution of neopullulanase activity to the modification of the texture of gruels fermented by amylolytic LAB remains to be determined. Understanding the contribution of possible trans-glycosylation activity to generate oligosaccharides with putative prebiotic activity would also be of great interest. Two other enzymes could be involved in starch hydrolysis. Amylopectin phosphorylase, coded by the glgP gene, acts on starch or amylose to release D-glucose-1-phosphate, while alpha-glucosidase, coded by agl, hydrolyses the terminal non-reducing (1-4)linked alpha-D-glucose residue with the release of D-glucose. It is likely that these enzymes do not play a major role in in situ starch hydrolysis since in the slurries their activity was very low or close to the detection limit. Consistently, frequency of detection of transcripts coding for glgP and agl was much lower than that of amyA and dexC. Maltose phosphorylase liberates D-glucose and G1P from maltose. Both alpha-glucosidase and maltose phosphorylase are key enzymes for the metabolism of starch hydrolysis products, through glucose and G1P generation, which enter the Embden–Meyerhof pathway. However, in the A6 strain, only the activity of alpha-glucosidase was detectable in the slurry, though at a low level. Surprisingly, maltose phosphorylase activity was near the detection limit despite a high detection level of transcripts of malP gene in both growth media (MRS-starch, and pearl millet slurry). Alphaglucosidase has been shown to play an important role in the generation of glucose from maltose in the amylolytic LAB L. fermentum Ogi E1 isolated from ogi, an African maize based fermented food (Calderon Santoyo et al., 2003). Maltose-phosphorylase was reported to be the main enzyme involved in glucose generation in LAB from rye and

110


Table 2 BLAST search for alpha-amylase genes encountered in different Lactobacillus species. Bacterial strain

Gene

Predicted function

Accession number

Identity

Reference strain

Publication

L. plantarum WCFS1 L. plantarum A6 L. plantarum JDM1 L. plantarum ST-III L. plantarum WCFS1 L. plantarum JDM1 L. plantarum ST-III L. plantarum A6 L. manihotivorans LMG 18010 T L. amylovorus CIP 102989 L. amylovorus NRRL B-45 L. amylovorus NRRL B-4540 Lactobacillus sp. 'OND 32 L. plantarum L137 L. plantarum L137 plasmid pLTK13 L. plantarum WCFS1 L. plantarum JDM1 L. plantarum ST-III

malZ a-amy amy1 amy1 malS amy2 amy2 No name No name No name amyA amyA amyA apuA No name No name No name No name

Maltodextrin glucosidase Alpha-amylase Alpha amylase Alpha amylase Alpha-amylase Alpha-amylase Alpha-amylase Alpha-amylase Alpha amylase precursor Alpha-amylase Alpha-amylase Alpha-amylase Alpha-amylase Amylopullulanase Amylopullulanase Maltogenic alpha-amylase Glucan 1,4-alpha-maltohydrolase Glucan 1,4-alpha-maltohydrolase

AL935263.2 FR874168.1 CP001617.1 CP002222.1 AL935263.2 CP001617.1 CP002222.1 U62095.1 AF126051 U62096.1 EF419426.1 X80271 AF031369.1 AB369265.1 AB450918.1 AL935263.2 CP001617.1 CP002222.1

100% 98% 99% 99% 100% 99% 100% 100% 99% 99% 99% 99% 98% 100% 100% 100% 99% 100%

L. plantarum WCFS1 L. plantarum WCFS1 L. plantarum WCFS1 L. plantarum WCFS1 L. plantarum WCFS1 L. plantarum WCFS1 L. plantarum WCFS1 L. plantarum A6 L. plantarum A6 L. plantarum A6 L. plantarum A6 L. plantarum A6 L. plantarum A6 L. plantarum L137 L. plantarum L137 L. plantarum WCFS1 L. plantarum WCFS1 L. plantarum WCFS1

Kleerebezem et al. (2003) Turpin et al. (2011) Zhang et al. (2009) Wang et al. (2011) Kleerebezem et al. (2003) Zhang et al. (2009) Wang et al. (2011) Giraud and Cuny (1997) Morlon-Guyot et al. (2001) Giraud and Cuny (1997) Mathiesen et al. (2008) Fitzsimons et al. (1994) NCBI Kim et al. (2008) NCBI Kleerebezem et al. (2003) Zhang et al. (2009) Wang et al. (2011)

The table lists the results of blastn analysis with minimum length N 200 bp.

wheat sourdoughs (De Vuyst and Neysens, 2005; Gobbetti et al., 2005; Vrancken et al., 2008). In contrast, a low transcriptional level has been reported in sourdough using microarray analysis of maltose phosphorylase metatranscriptomes of several samples (Weckx et al., 2011). Genetic screening of a collection of LAB isolates and of the metagenome of the African pearl millet fermented gruel, ben-saalga, showed that agl and malP genes were widely distributed, agl being more frequent than malP. In tropical fermentation of cereals, unit operations in the process play an important role in selecting growth conditions (e.g. by decreasing the availability of free sugars when the kernels are soaking or during other unit operations), consequently selecting the LAB with the best metabolic pathways to adapt to these conditions. The wide distribution of these two genes might reflect this adaptability. Notwithstanding, a complete view of starch metabolism could also be acquired by a thorough investigation of the expression of the genes involved in the oligosaccharide transport systems in these tropical cereal-based fermented foods. This should help link the extracellular activity of alpha-amylase to the use of short maltodextrins in the bacteria (Ganzle and Follador, 2012). All these data gave us a deeper insight into the genetic basis of the amylolytic activity of lactic acid bacteria and revealed the potential role of neopullulanase, which merits further investigation. More striking, the dynamic view of gene expression during starch fermentation suggests that the set of genes we investigated are not expressed continuously but transiently. This gene transcription kinetics draws attention to the entire mechanism of regulation of these genes. Transcriptomic analysis and the construction of mutants to investigate this question would advance our understanding of this mechanism of gene regulation. Acknowledgements Christèle Humblot was the recipient of a grant from the French Society of Nutrition. Williams Turpin acknowledges a PhD grant from the French Ministry of Education and Research. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijfoodmicro.2014.05.016. References Agati, V., Guyot, J.P., Morlon-Guyot, J., Talamond, P., Hounhouigan, D.J., 1998. Isolation and characterization of new amylolytic strains of Lactobacillus fermentum from fermented maize doughs (mawè and ogi) from Benin. J. Appl. Microbiol. 85, 512–520.

Amdekar, S., Dwivedi, D., Roy, P., Kushwah, S., Singh, V., 2009. Probiotics: multifarious oral vaccine against infectious traumas. FEMS Immunol. Med. Microbiol. 58, 299–306. Ampe, F., ben Omar, N., Guyot, J.P., 1998. Recovery of total microbial RNA from lactic acid fermented foods with a high starch content. Lett. Appl. Microbiol. 27, 270–274. Bergmeyer, H.U., Graβl, M., Walter, H.E., 1988. Biochemical reagents for general use: enzymes. In: Bergmeyer, H.U., Bergmeyer, J., Graβl, M. (Eds.), Methods of Enzymatic Analysis. Verlagsgesellschaft, Weinheim, pp. 126–328. Bookwalter, G.N., Peplinski, A.J., Pfeifer, V.F., 1968. Using a Bostwick consistometer to measure consistencies of processed corn meals and their CSM blends. Cereal Sci. Today 13, 407–410. Bustin, S.A., Benes, V., Garson, J.A., Hellemans, J., Huggett, J., Kubista, M., Mueller, R., Nolan, T., Pfaffl, M.W., Shipley, G.L., Vandesompele, J., Wittwer, C.T., 2009. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin. Chem. 55, 611–622. Calderon Santoyo, M., Loiseau, G., Rodriguez Sanoja, R., Guyot, J.P., 2003. Study of starch fermentation at low pH by Lactobacillus fermentum Ogi E1 reveals uncoupling between growth and alpha-amylase production at pH 4.0. Int. J. Food Microbiol. 80, 77–87. Chagnaud, P., Machinis, K., Coutte, L.A., Marecat, A., Mercenier, A., 2001. Rapid PCR-based procedure to identify lactic acid bacteria: application to six common Lactobacillus species. J. Microbiol. Methods 44, 139–148. de Vries, M.C., Vaughan, E.E., Kleerebezem, M., de Vos, W.M., 2006. Lactobacillus plantarum — survival, functional and potential probiotic properties in the human intestinal tract. Int. Dairy J. 16, 1018–1028. De Vuyst, L., Neysens, P., 2005. The sourdough microflora: biodiversity and metabolic interactions. Trends Food Sci. Technol. 1,6, 43–56. Diaz-Ruiz, G., Guyot, J.P., Ruiz-Teran, F., Morlon-Guyot, J., Wacher, C., 2003. Microbial and physiological characterization of weakly amylolytic but fast-growing lactic acid bacteria: a functional role in supporting microbial diversity in pozol, a Mexican fermented maize beverage. Appl. Environ. Microbiol. 69, 4367–4374. Fitzsimons, A., Hols, P., Jore, J., Leer, R.J., O'Connell, M., Delcour, J., 1994. Development of an amylolytic Lactobacillus plantarum silage strain expressing the Lactobacillus amylovorus alpha-amylase gene. Appl. Environ. Microbiol. 60, 3529–3535. Ganzle, M.G., Follador, R., 2012. Metabolism of oligosaccharides and starch in lactobacilli: a review. Front. Microbiol. 3, 340. Giraud, E., Cuny, G., 1997. Molecular characterization of the alpha-amylase genes of Lactobacillus plantarum A6 and Lactobacillus amylovorus reveals an unusual 3′ end structure with direct tandem repeats and suggests a common evolutionary origin. Gene 198, 149–157. Giraud, C., Brauman, A., Keleke, S., Lelong, B., Raimbault, M., 1991. Isolation and physiological study of an amylolytic strain of Lactobacillus plantarum. Appl. Environ. Microbiol. 36, 379–383. Giraud, E., Champailler, A., Raimbault, M., 1994. Degradation of raw starch by a wild amylolytic strain of Lactobacillus plantarum. Appl. Environ. Microbiol. 60, 4319–4323. Gobbetti, M., De Angelis, M., Corsetti, A., Di Cagno, R., 2005. Biochemistry and physiology of sourdough lactic acid bacteria. Trends Food Sci. Technol. 16, 57–69. Haydersah, J., Chevallier, I., Rochette, I., Mouquet-Rivier, C., Picq, C., Marianne-Pepin, T., Icard-Verniere, C., Guyot, J.P., 2012. Fermentation by amylolytic lactic acid bacteria and consequences for starch digestibility of plantain, breadfruit, and sweet potato flours. J. Food Sci. 77, M466–M472. Humblot, C., Guyot, J.P., 2009. Pyrosequencing of tagged 16S rRNA gene amplicons for rapid deciphering of the microbiomes of fermented foods such as pearl millet slurries. Appl. Environ. Microbiol. 75, 4354–4361. Kim, J.H., Sunako, M., Ono, H., Murooka, Y., Fukusaki, E., Yamashita, M., 2008. Characterization of gene encoding amylopullulanase from plant-originated lactic acid bacterium, Lactobacillus plantarum L137. J. Biosci. Bioeng. 106, 449–459. Kim, J.H., Sunako, M., Ono, H., Murooka, Y., Fukusaki, E., Yamashita, M., 2009. Characterization of the C-terminal truncated form of amylopullulanase from Lactobacillus plantarum L137. J. Biosci. Bioeng. 107, 124–129.

C. Humblot et al. / International Journal of Food Microbiology 185 (2014) 103–111 Kleerebezem, M., Boekhorst, J., van Kranenburg, R., Molenaar, D., Kuipers, O.P., Leer, R., Tarchini, R., Peters, S.A., Sandbrink, H.M., Fiers, M.W., Stiekema, W., Lankhorst, R.M., Bron, P.A., Hoffer, S.M., Groot, M.N., Kerkhoven, R., de Vries, M., Ursing, B., de Vos, W.M., Siezen, R.J., 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc. Natl. Acad. Sci. U. S. A. 100, 1990–1995. Lee, H.S., Kim, M.S., Cho, H.S., Kim, J.I., Kim, T.J., Choi, J.H., Park, C., Lee, H.S., Oh, B.H., Park, K.H., 2002. Cyclomaltodextrinase, neopullulanase, and maltogenic amylase are nearly indistinguishable from each other. J. Biol. Chem. 277, 21891–21897. Lei, V., Jakobsen, M., 2004. Microbiological characterization and probiotic potential of koko and koko sour water, African spontaneously fermented millet porridge and drink. J. Appl. Microbiol. 96, 384–397. Mathiesen, G., Sveen, A., Piard, J.C., Axelsson, L., Eijsink, V.G., 2008. Heterologous protein secretion by Lactobacillus plantarum using homologous signal peptides. J. Appl. Microbiol. 105, 215–226. Morlon-Guyot, J., Mucciolo-Roux, F., Rodriguez Sanoja, R., Guyot, J.P., 2001. Characterization of the L. manihotivorans alpha-amylase gene. DNA Seq. 12, 27–37. Mouquet, C., Greffeuille, V., Treche, S., 2006. Characterization of the consistency of gruels consumed by infants in developing countries: assessment of the Bostwick consistometer and comparison with viscosity measurements and sensory perception. Int. J. Food Sci. Nutr. 57, 459–469. Muyzer, G., de Waal, E.C., Uitterlinden, A.G., 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol. 59, 695–700. Nguyen, T.T.T., Loiseau, G., Icard-Vernière, C., Rochette, I., Trèche, S., Guyot, J.P., 2007. Effect of fermentation by amylolytic lactic acid bacteria, in process combinations, on characteristics of rice/soybean slurries: a new method for preparing high energy density complementary foods for young children. Food Chem. 100, 623–631. Oguntoyinbo, F.A., Narbad, A., 2012. Molecular characterization of lactic acid bacteria and in situ amylase expression during traditional fermentation of cereal foods. Food Microbiol. 31, 254–262. Petrov, K., Urshev, Z., Petrova, P., 2008. L+-lactic acid production from starch by a novel amylolytic Lactococcus lactis subsp. lactis B84. Food Microbiol. 25, 550–557. Plumed-Ferrer, C., Koistinen, K.M., Tolonen, T.L., Lehesranta, S.J., Karenlampi, S.O., Makimattila, E., Joutsjoki, V., Virtanen, V., von Wright, A., 2008. Comparative study of sugar fermentation and protein expression patterns of two Lactobacillus plantarum strains grown in three different media. Appl. Environ. Microbiol. 74, 5349–5358. Rodriguez Sanoja, R., Morlon-Guyot, J., Jore, J., Pintado, J., Juge, N., Guyot, J.P., 2000. Comparative characterization of complete and truncated forms of Lactobacillus amylovorus alpha-amylase and role of the C-terminal direct repeats in raw-starch binding. Appl. Environ. Microbiol. 66, 3350–3356. Rodriguez-Sanoja, R., Ruiz, B., Guyot, J.P., Sanchez, S., 2005. Starch-binding domain affects catalysis in two Lactobacillus alpha-amylases. Appl. Environ. Microbiol. 71, 297–302.

111

Songré-Ouattara, L.T., Mouquet-Rivier, C., Vernière, C., Humblot, C., Diawara, B., Guyot, J.P., 2008. Enzyme activities of lactic acid bacteria from a pearl millet fermented gruel (ben-saalga) of functional interest in nutrition. Int. J. Food Microbiol. 128, 395–400. Songre-Ouattara, L.T., Mouquet-Rivier, C., Icard-Verniere, C., Rochette, I., Diawara, B., Guyot, J.P., 2009. Potential of amylolytic lactic acid bacteria to replace the use of malt for partial starch hydrolysis to produce African fermented pearl millet gruel fortified with groundnut. Int. J. Food Microbiol. 130, 258–264. Songre-Ouattara, L.T., Mouquet-Rivier, C., Humblot, C., Rochette, I., Diawara, B., Guyot, J.P., 2010. Ability of selected lactic acid bacteria to ferment a pearl millet–soybean slurry to produce gruels for complementary foods for young children. J. Food Sci. 75, M261–M269. Stolz, P., Hammes, W.P., Vogel, R.F., 1996. Maltose-phosphorylase and hexokinase activity in lactobacilli from traditionally prepared sourdoughs. Adv. Food Sci. 18, 1–6. Takata, H., Kuriki, T., Okada, S., Takesada, Y., Iizuka, M., Minamiura, N., Imanaka, T., 1992. Action of neopullulanase — neopullulanase catalyzes both hydrolysis and transglycosylation at alpha-(1-]4)-glucosidic and alpha-(1-]6)-glucosidic linkages. J. Biol. Chem. 267, 18447–18452. Tou, E.H., Guyot, J.P., Mouquet-Rivier, C., Rochette, I., Counil, E., Traore, A.S., Treche, S., 2006. Study through surveys and fermentation kinetics of the traditional processing of pearl millet (Pennisetum glaucum) into ben-saalga, a fermented gruel from Burkina Faso. Int. J. Food Microbiol. 106, 52–60. Tou, E.H., Mouquet-Rivier, C., Rochette, I., Traoré, A.S., Trèche, S., Guyot, J.P., 2007. Effect of different process combinations on the fermentation kinetics, microflora and energy density of ben-saalga, a fermented gruel from Burkina Faso. Food Chem. 100, 935–943. Turpin, W., Humblot, C., Guyot, J.P., 2011. Genetic screening of functional properties of lactic acid bacteria in a fermented pearl millet slurry and in the metagenome of fermented starchy foods. Appl. Environ. Microbiol. 77, 8722–8734. Turpin, W., Humblot, C., Noordine, M.L., Thomas, M., Guyot, J.P., 2012. Lactobacillaceae and cell adhesion: genomic and functional screening. PLoS ONE 7, 1–14. Vrancken, G., Rimaux, T., De Vuyst, L., Leroy, F., 2008. Kinetic analysis of growth and sugar consumption by Lactobacillus fermentum IMDO 130101 reveals adaptation to the acidic sourdough ecosystem. Int. J. Food Microbiol. 128, 58–66. Wang, Y., Chen, C., Ai, L., Zhou, F., Zhou, Z., Wang, L., Zhang, H., Chen, W., Guo, B., 2011. Complete genome sequence of the probiotic Lactobacillus plantarum ST-III. J. Bacteriol. 193, 313–314. Weckx, S., Allemeersch, J., Van der Meulen, R., Vrancken, G., Huys, G., Vandamme, P., Van Hummelen, P., De Vuyst, L., 2011. Metatranscriptome analysis for insight into wholeecosystem gene expression during spontaneous wheat and spelt sourdough fermentations. Appl. Environ. Microbiol. 77, 618–626. Zhang, Z.Y., Liu, C., Zhu, Y.Z., Zhong, Y., Zhu, Y.Q., Zheng, H.J., Zhao, G.P., Wang, S.Y., Guo, X. K., 2009. Complete genome sequence of Lactobacillus plantarum JDM1. J. Bacteriol. 191, 5020–5021.