Novel route of tannic acid biotransformation and their effect on major biopolymer synthesis in Azotobacter sp. SSB81



Bikas R. Pati, HOD, Department of Microbiology, Vidyasagar University, Midnapore 721 102, West Bengal, India. E-mail:



To examine tannic acid (TA) utilization capacity by nitrogen-fixing bacteria, Azotobacter sp. SSB81, and identify the intermediate products during biotransformation. Another aim of this work is to investigate the effects of TA on major biopolymers like extracellular polysaccharide (EPS) and polyhydroxybutyrate (PHB) synthesis.

Methods and Results

Tannic acid utilization and tolerance capacity of the strain was determined according to CLSI method. Intermediate products were identified using high-performance liquid chromatography, LC-MS/MS and 1H NMR analysis. Intermediates were quantified by multiple reactions monitoring using LC-MS/MS. The strain was able to tolerate a high level of TA and utilized through enzymatic system. Growth of Azotobacter in TA-supplemented medium was characterized by an extended lag phase and decreased growth rate. Presence of TA catalytic enzymes as tannase, polyphenol oxidase (PPO) and phenol decarboxylase was confirmed in cell lysate using their specific substrates. PPO activity was more prominent in TA-supplemented mineral medium after 48 h of growth when gallic to ellagic acid (EA) reversible reaction was remarkable. Phase contrast and scanning electron microscopic analysis revealed elongated and irregular size of Azotobacter cells in response to TA. 1H NMR analysis indicated that TA was transformed into gallic acid (GA), EA and pyrogallol. Biopolymer (EPS and PHB) production was decreased several folds in the presence of TA compared with cells grown in only glucose medium.


This is the first evidence on the biotransformation of TA by Azotobacter and also elevated level of EA production from gallotannins. Azotobacter has developed the mechanism to utilize TA for their carbon and energy source.

Significance and Impact of the Study

The widespread occurrence and exploitation of Azotobacter sp. strain SSB81 in agricultural and forest soil have an additional advantage to utilize the soil-accumulated TA and detoxifies the allelopathic effect of constant accumulated TA in soil.


Tannins are plant-derived phenolic acid accumulated in soil, affect nutrient cycle by hindering decomposition rates, complexing proteins and other macromolecules such as starch, cellulose and minerals, inducing toxicity to microbial populations and inhibiting enzyme activities (Lekha and Lonsane 1997; Aguilar and Gutierrez-Sanchez 2001a; Baptist et al. 2008). The increased amounts of tannic acid (TA) in soil have detrimental effects on plant growth process that is directly affecting the vegetation, soil characteristics, cropping systems and yield of production (Kuiters and Sarink 1987). There is continuing interest to clean up the environmental contaminants in agricultural soil and restore the soil productivity. In the last few decades, major focuses have been made on the recycling of organic wastes and polyphenols in agricultural, industrial or municipal in origin and it is evident that microbial transformations are the most efficient process to recover. The biotransformation and detannification of TA to GA or pyrogallol by microbial tannase have been studied from many years for industrial perspectives only.

Although tannins are highly toxic to organisms, some microbes are able to develop various resistant mechanisms to degrade them into oligomeric tannins such as gallic or ellagic acids (EA), and pyrogallol. Several bacteria, Bacillus sp., Staphylococcus sp., Corynebacterium sp., Citrobacter sp. and Klebsiella sp. Citrobacter freundii (Deschamps et al. 1980; Kumar et al. 1999; Mohapatra et al. 2007), and most of the fungal species, Aspergillus sp., Fusarium sp., Penicillium sp., Sporotrichum sp., Rhizoctonia sp., Cylindrocarpon sp. and Trichoderma sp. (Deschamps et al. 1980; Makkar et al. 1994), isolated from the tannery effluent are able to degrade gallotannins into GA oligomer (Kumar et al. 1999). Still there is no report on the use of agriculturally important organism like Azotobacter for TA transformation; though, TA is the second most abundant aromatics soil contaminants.

Azotobacter sp. are well known for their utility as a successful biofertilizer and able to grow in the presence of different phenolic acids, including p-hydroxybenzoic, m-hydroxybenzoic, vanillic, p-coumaric, syringic, cis- and trans-ferrulic and other unidentified aromatic acids (Wu et al. 1987). Moreno et al. (1999) also noticed that some phenolic compounds, such as p-hydroxybenzoic acid or protocatechuic acid, at higher concentrations over the range commonly found in natural soils supported the growth and nitrogen fixation of Azotobacter vinelandii. Azotobacter has the ability to produce huge amount extracellular polysaccharide (EPS) and polyhydroxybutyrate (PHB), which bring an additional advantage in sustainable agriculture development (Gauri et al. 2012). The synthesis of biopolymer in Azotobacter provides protection from unfavourable conditions. PHB is also the source of energy by the activation of the PHB depolymerase enzymes in resting period. Both are synthesized from different pathway, and PHB is generally produced in large amounts of carbon source and nitrogen, phosphorus or oxygen limitation. The synthesis and incorporation of different monomers is depending upon the suitable substrate that can be converted into the desired hydroxyacyl-CoA through metabolic reactions in the bacterial cell. EPS is synthesized from the central sugar metabolites into the alginate precursor GDP-mannuronic acid and exceed to several steps with complex enzymatic process (Lin and Hassid 1966). However, TA utilization by Azotobacter for energy source and reduction in their amount in soil have not been considered yet.

The original purpose of this investigation was to identify the intermediates of TA biotransformation by Azotobacter strain able to use as the sole carbon source and energy metabolism. The major enzymes as tannase, polyphenol oxidase (PPO) and decarboxylase are studied to describe the mechanism. Regulations of biopolymers (EPS and PHB) synthesis in Azotobacter strain are also evaluated in the presence of TA.



Azotobacter sp. strain SSB81 (Gauri et al. 2009) was grown and maintained in nitrogen-free Burk's medium supplemented with glucose. Exponentially growing cells were used as inoculum (approx. 106 cells) for all the experiments.

The effect of different phenolic acid on bacterial growth

The phenolic acid tolerance capacity of Azotobacter strain SSB81 was studied by following microtiter plate dilution method according to Clinical and Laboratory Standards Institute (CLSI 2007) guideline, similar to antibacterial assay. The concentration of TA (Himedia, India) used in this assay ranged from 1 to 64 mg ml−1. Two sets of growth medium were used in this study, TA-supplemented nitrogen-free glucose medium (GTA) and nitrogen-free mineral medium (MTA). A range of TAs concentrations were made in sterile 96-well microtitre plates with medium, final volume was 250 μl per well, and inoculum load was 106 cells per well. Finally, microtiter plates were incubated at 30°C for 36 h, and bacterial growth was measured at 620 nm using a Multiscan Spectrum spectrophotometer (model 1500; Thermo Scientific, Nyon, Switzerland).

Media and growth condition for biotransformation

The N-free Burk's medium (glucose 20 g; K2HPO4 0.2 g; K2SO4 0.1 g; MgSO4 0.2 g; NaCl 0.2 g; CaCO3 5 g; Na-molybdate 0.01 g; water 1000 ml) with glucose and without glucose was used. TA was supplemented with GTA and MTA medium filter sterilization (0·22-μmol l−1 filter; Millipore, Billerica, MA, USA) at 1 g l−1 concentration. In MTA medium, tannic acid was the only carbon sources for their growth and metabolism, while in GTA medium, glucose was used as primary carbon source with TA. The pH of the medium was monitored and adjusted to 7·4 using 0·1 N sterile NaOH. After inoculation, the medium was incubated at 30°C in a shaker with gentle agitation (150 rpm) to maintain aerobic conditions. Only glucose medium (no TA) inoculated with Azotobacter SSB81 was served as positive control, and uninoculated medium containing phenolic acid was served as negative control. To avoid interference in measuring optical density, an uninoculated control was used in each case.

Extraction of phenolic intermediates

The culture medium was harvested at different time intervals to determine the bacterial growth and transformation products. In brief, cultures were centrifuged and filtered through a 0·22-μm syringe filter (Millipore). Bacteria-free supernatants (2 ml) were then extracted thrice repeatedly with equal volume of ethyl acetate and pooled them. Again the remaining TA in aqueous layer was extracted with equal volume of ethyl acetate after acidifying (pH 3) with glacial acetic acid (Merck, Whitehouse Station, NJ, USA) (Chauhan and Jain 2000). The neutral and acidic extracts were pooled together and evaporated to dryness in rotary evaporator in a bath at 42°C. The dried material was resuspended in methanol for further analysis.

High-performance liquid chromatography analysis

High-performance liquid chromatography (HPLC) is the mostly used chromatographic techniques for the separation of phenolic acids (Mandal and Dey 2008). Here, HPLC was used to detect the phenolic acid and it derivatives after biotransformation as well as enzyme assay reaction. HPLC analysis was performed in a Agilent 1100 series model system, equipped with a binary pump, diode array detector (DAD), online vacuum degasser and Chemstation software (Agilent Technologies, Santa Clara, CA, USA). The method was developed after brief modification in the previously reported method by Misan et al. (2011). The extracted samples were injected (20 μl) into the HPLC system, and separation was performed in reverse-phase Zorbax Eclipse XDB-C18 column (4·6 × 250 mm i.d., 5 μm particle size; Agilent Technologies) with varying the proportion of solvent A (methanol) to solvent B [1% TFA in water (v/v)] as follows: initial 10% A, 0–5 min; 10–35% A, 5–35 min and final 35% A, 35–50 min. The total run time and postrunning time were 50 and 20 min, respectively. The column temperature was 25°C. The spectra were acquired at 254, 280 and 310 nm with a bandwidth of 4 nm, and with reference wavelength/bandwidth of 360:100 nm.

Liquid chromatography and tandem mass spectrometry (LC-MS/MS)

LC-MS/MS analyses were performed using Waters 2695 separation module coupled with QuattroMicroAPI mass spectrometer (Waters, Milford, MA, USA). The liquid chromatographic system consisted of a quaternary pump, online vacuum degasser, autosampler and thermostatic column compartment, connected in line to a photodiode array detector (Waters 2998) before the mass spectrometer. Data acquisition and analysis were carried out in waters Mass Lynx 4.1 software (Waters). The LCMS method was developed with acetonitrile as described by Krizman et al. (2007). Sample (10 μl) was injected into the LC system by autosampler, and separation was performed on a XTerra MS C18 reversed-phase column (2·1 × 100 mm i.d., 2·5 µm particle size; Waters). The mobile phase consisted of 0·1% aqueous formic acid (A) acetonitrile (B) and with a gradient elution of 10–30% B in 0–20 min and 30–40% B in 20–30 min. The flow rate was 0·3 ml min−1, and column temperature was maintained at 25°C. All the compounds were detected within the range of 230–360 nm in a PDA detector. The LC-eluted samples were introduced into the electrospray ionization (ESI) source in a postcolumn splitting ratio of 3 : 1. An electrospray source with negative ionization mode (source block temperature 130°C, desolvation temperature 300°C, capillary voltage 3 kV, cone voltage 30 V) was used for mass analysis. The desolvation and cone gas were 450 and 80 l h−1, respectively. The data were recorded in the MS scanning mode with the scan range of 100–1990 (m/z), the scan time was 0·5 s, and interscan delay time was 0·1 s.

Multiple reaction monitoring-based quantification in LC-MS/MS

Multiple reaction monitoring (MRM) using a triple quadrupole mass spectrometer is a powerful technique for rapid and precise quantification of small molecules (Sanchez-Patan et al. 2011). In MRM experiments, two mass analysers are used as static mass filters, to monitor a particular fragment ion of a selected precursor ion. The selectivity resulting from the two filtering stages combined with the high duty cycle results in quantitative analyses with supreme sensitivity. The MRM method deals with transition of parent and product specific for each compound. Method development starts with the optimization of collision gas for MS/MS fragmentation by autotune wizard where samples were infused with the flow rate of 10 μl min−1. During the infusion, the MS signal in the full scan was optimized. The standard curve was prepared against selected MRM method by running minimum five different concentration of each compound. The concentration of analytes was determined automatically by QuanLynx software (Waters) against calibration curve. Here, GA and EA were quantified by MRM with recovery ±2%.

Nuclear magnetic resonance

1H NMR (200 MHz) spectra were recorded on a Bruker AC 200 MHz spectrometer. Chemical shifts are reported in parts per million from tetramethylsilane with the solvent resonance as the internal standard (Deuterated water, 4·86 ppm). Data are reported as follows: chemical shifts, multiplicity (s = singlet, d = doublet, t = triplet and m = multiplet) and coupling constant (hertz).

Enzymatic assay

Extraction protocol

The crude enzyme was extracted by following the method of Zeida et al. (1998). Enzymes were extracted at 4°C in 50 mmol l−1 phosphate buffer (pH 6·5) containing protease inhibitor and 1 mmol l−1 dithiothreitol. Cells were harvested from TA-supplemented culture broth (1L) after centrifugation for 30 min at 11200 g. The harvested cells were washed twice with same extracted buffer and resuspended in 40 ml of phosphate buffer followed by ultrasonication in ice (19 kHz; Insonator model 201M; Kubota, Tokyo, Japan). The disrupted cell suspension was centrifuged at 11200 g for 20 min at 4°C, to separate the cell debris. The supernatant containing the soluble proteins was filtered aseptically using sterile filters of 0·2 μm pore size (Millipore). The protein concentration was determined using the Bradford protein assay (Bradford 1976).

Tannase assay

Tannase activity of crude extract was determined by the method of Mondal et al. (2001). Enzyme solution (0·05 ml) was incubated with 0·3 ml of 1·0% (w/v) TA, in 0·2 mol l−1 acetate buffer (pH 5·5) at 40°C for 10 min, and then, the reaction was stopped by the addition of 2·0 ml bovine serum albumin (BSA) (1 mg ml−1), which precipitated the remaining TA. A control reaction was performed parallel with heat-denatured enzyme. The reaction tubes were then centrifuged (4480 g, 10 min), and the precipitate was dissolved in 2·0 ml of SDS-triethanolamine (1% w/v SDS in 5% v/v triethanolamine) solution. The absorbance was recorded at 530 nm (UV–vis spectrophotometer; Thermo Scientific) after the addition of 1·0 ml of FeCl3 (0·13 mol l−1) solution. The specific extinction coefficient of TA at 530 nm was 0·577 (Mondal et al. 2001). Using this coefficient, one unit of tannase activity is defined as the amount of enzyme required to hydrolyse 1·0 μmol of ester linkage of TA in 1 min under specified conditions.

PPO assay

Polyphenol oxidase oxidizes tyrosine to dihydroxyphenyl alanine which in turns is oxidized to o-quinone followed by Worthington Enzyme Manual (Worthington and Worthington 2011). In brief, 1 ml of 0·001 mol l−1 l-tyrosine and 0·5 mol l−1 phosphate buffer (pH 6·5) were taken in a cuvette and mixed well after the addition of 0·9 ml water. The reaction mixture was then oxygenated by bubbling oxygen into cuvette for 5 min. The cuvette was kept to the spectrometer, and absorbance was recorded for 4–5 min to achieve temperature equilibration. Finally, 0·1 ml of crude enzyme extract was added into the assay solution, and absorbance was recorded for 10 min. The rate of increase is proportional to enzyme concentration and linear during a period of 5–10 min after an initial lag. One unit causes a change in absorbance at 280 nm of 0·001 per minute at 25°C, pH 6·5 under the specified assay condition. The enzyme activity was calculated using the following equation: units/mg = ΔA280/min × 1000/mg enzyme in reaction. For in vitro process, product formation was tested using gallic and EA as substrate in separate reaction, and the products were confirmed further in LC-MS analysis.

Gallate decarboxylase assay

The phenol decarboxylase assay was performed following the method described by O'Donovan and Brooker (2001). The crude enzyme was prepared from TA medium, and GA was used as substrate in this assay. The enzyme extracts (100 μl) were added to 1 ml of 0·5 mol l−1 phosphate buffer (pH 6·5) containing 2 mmol l−1 DTT and 10 mmol l−1 MgCl in a reaction tube and warmed at 37°C for 5 min. Gallate decarboxylase activity was initiated by adding 0·5 ml (0·5% w/v) GA dissolved in phosphate buffer. The reaction was monitored at 300 nm over a period of 15 min, and the product was also analysed by LC-MS. The heat-denatured enzyme extract was used as control. Activity was expressed as mmol GA decarboxylated per minute.

Extraction and estimation of PHB

PHB of Azotobacter was extracted and estimated following the method of Law and Slepecky (1961). The harvested cells were treated with 10% SDS at 100°C for 20 min followed by centrifugation. The pellet was collected, dried and extracted with 50 volume of chloroform at 60°C for 1 h. The non-PHB cell debris was removed by filtration, and the dissolved PHB was separated from chloroform by evaporation, washed twice with methanol and dried at 60–70°C. The PHB content was determined in percentage as the % ratio of PHB content to the cell mass and also estimated by chemical assay. In brief, PHB was converted to crotonic acid by treating with concentrated sulphuric acid, and the product was quantified spectrophotometrically. The polymer sample in chloroform is transferred to a clean test tube, and 10 ml of concentrated H2SO4 is added to the tube, immediately capped with a glass marble and heated for 10 min at 100°C in a water bath. After that, the solution is allowed to cool and the absorbance was measured at 235 nm against a sulphuric acid blank. A standard curve was established with PHB concentrations ranging from 5 to 50 μg ml−1 (Dekwer and Hempel 1999).

PHB detection and quantification by fluorescence-activated cell sorter

Nile red-specific staining and fluorescence-activated cell sorter (FACS) analysis were performed by following the method of Tyo et al. (2006). The reported method was modified for the standardization of staining protocol and membrane permeability. The cells were harvested by centrifugation at 4000 rpm and then cooled at 4°C. Cells were resuspended in 0·5 ml of 8% (w/v) sucrose TSE buffer (10 mmol l−1 Tris–HCl of pH 7·4, 2·5 mmol l−1 Na-EDTA and sucrose) to increase permeability and incubated for 5 min in 4°C. The cells were again centrifuged, and 1 ml of MgCl2 (1 mmol l−1) was used to resuspended the cells. Finally, 5 μl of Nile red (1 mg ml−1) (Sigma-Aldrich, USA) solution in DMSO was added to the cell suspension. After 20-min incubation in the dark place, flow cytometry analysis was performed. This method was optimized for rapid estimation of PHB from large number of samples by correlating the geometric mean with the estimated value of spectrometric quantification. Data were quantified with Cell Quest Pro software attached to FACS Calibur (BD, San Jose, CA, USA).

Estimation of EPS

Cultures were centrifuged (10 000 g for 10 min), and ammonium acetate (1 mol l−1) was added to the supernatant. EPS was precipitated in the supernatant by the addition of double volume of isopropanol. The precipitate was then dissolved in deionized water. Yields of EPS were estimated following the phenol sulfuric acid assay (Dubois et al. 1956) against freshly prepared nitrogen-free basal glucose medium as control. EPS was quantified after preparing the standard curve of glucose in culture medium.

Phase contrast microscopy

Cells were grown in different medium with their respective conditions as described earlier. Twenty microlitres of bacterial cultures from each condition was spotted onto polylysine-coated slide and viewed by phase contrast microscopy. Images were captured by Image pro discovery software (Olympus, Shinjuku, Tokyo, Japan).

Scanning electron microscopy

Cultures from different growth conditions were harvested separately and washed with 1× PBS for several times, fixed with osmium tetra oxide after dehydration with acetone gradient. Finally, 10 μl of culture solution was spotted onto the lysine-coated glass cover slip as drop caste method. The fixed cells were dried and kept on desiccator until use. Samples were then fixed onto a graphite stub and kept in an auto sputter coater (E5200; Bio-Rad, Shinagawa-ku, Tokyo, Japan) under low vacuum for gold coating up to 120 s. Surface morphology was studied using a scanning electron microscope (SEM), model SUPRATM 40, with an accelerated voltage 15–20 kV. (Carl-Zeiss, Oberkochen, Germany)


TA tolerance by the strain

The minimum inhibitory concentration of Azotobacter strain SSB81 was tested against TA fortified in two different media like GTA and MTA. The used media were supplemented with various concentration of TA, and MIC values were found to be 8 and 4 mg ml−1, respectively. Simultaneously, the observed MIC values were again confirmed with colony forming unit in plate count method against each concentration of TA. Growth of Azotobacter SSB81 was optimized in the presence of different concentrations of TA, and observed 2 and 1 mg ml−1 concentrations are not hindering greatly of bacterial growth in GTA and MTA medium, respectively. To compare their intermediate products, biotransformation rate and biopolymer production, 1 mg ml−1 concentration of TA was used throughout the study for both GTA and MTA media.

Detection of TA biotransformation intermediates

Tannic acid biotransformation by the strain was determined in both GTA and MTA media. The intermediate transformation products were extracted from different time intervals of their growth. The extracted intermediates and substrate TA were run into LC separately, subsequently identified with coupled mass spectrometry analysis of each observed peak in LC (Fig. 1). Four major peaks were observed from LC analysis after the growth period of 24–120 h. Two spectra of each set are represented in Fig. 1. Mass spectrometry analysis of LC peaks 1, 2, 3 and 4 showed the molecular mass of m/z 169, m/z 125, m/z 301 and m/z 939 corresponding to the actual mass of GA, pyrogallol (PY), EA and residual TA (pentagalloyl glucose), respectively (Fig. 1). The mass spectrometric signature of commercial tannin is represented in Fig. 1 and described details earlier (Rodriguez et al. 2008). The identified compounds are listed in Table 1. We have also confirmed the intermediates by UV–vis analysis using diode array detector coupled with LC. Spectra showed their unique signature at 272, 273·5, 252·8 and 278 nm for LC peaks 1, 2, 3 and 4, respectively (Fig. 1). Hence, it is definite proved the presence of GA, EA and PY intermediates after biotransformation of TA by Azotobacter sp. strain SSB81.

Table 1. The MS and MS/MS pattern of commercial tannic acid (TA) and their intermediate
Sl. No.MS (m/z) [M-H]MS/MS (m/z) [MH]Identified compounds
1.169·06125, 79Gallic acid
2.320·97169Digalic acid
3.331·02169, 125Galloyl glucose
4.469·11331, 169Methyl-(digalloyl-glucoside)
5.483·10331, 271, 169Digalloyl glucose
6.634·95483, 465, 313Trigalloyl glucose
7.787·99, 392·99 [M-2H]2−635, 617, 465, 313Tetragalloyl glucose
8.939·05787, 769, 617Pentagalloyl glucose
9.1091·26, 545·12 [M-2H]2−939, 770, 617Hexagalloyl glucose
10.1243·171091, 939, 920Heptagalloyl glucose
11.1396·12, 696·84 [M-2H]2−1243, 1091Octagalloyl glucose
12.1700·02, 849·17 [M-2H]2−1091, 169Tannic acid
13.301·04257·3Ellagic acid
Figure 1.

Detection and identification of intermediates from tannic acid (TA) biotransformation by Azotobacter strain SSB81 using LC-MS/MS. LC chromatograms profile of commercial TA (a); SSB81 grown for 48 h in TA-supplemented MTA medium (b); 96 h in TA-supplemented MTA medium (c); 48 h in TA-supplemented GTA medium (d) and 96 h in TA-supplemented GTA medium (e). LC profile indicating the presence of four major peaks that varied at different time interval of growth phase. MS analysis of selected peaks in LC profile indicating the presence of gallic acid (GA) (m/z 169) in peak 1 (g); pyrogallol (m/z 125) in peak 2 (h); ellagic acid (EA) (m/z 301) in peak 3 (i) and residual TA/TA in peak 4 (f). UV–vis spectral profile also confirmed the presence of GA in peak 1 (U1); pyrogallol in peak 2 (U2); EA in peak 3 (U3) and residual TA in peak 4 (U4). GTA, nitrogen-free glucose medium supplemented with tannic acid; MTA, nitrogen-free mineral medium supplemented with tannic acid.

Quantification of GA and EA

LC-MS/MS based MRM is the most sensitive quantification method. Quantification of GA and EA was made by selecting transition the precursor and product ions. For two pairs of GA transition (169·1–124·9/78·69), the optimum collision energy was found 12 and 21 V, respectively, whereas the collision energy needed for EA transition (301·1–257·3/229·1) was 18 and 23, respectively. Calibration curves were prepared by the method of external standard using five different starting concentrations (20–100 μg ml−1) at 10 different calibration levels each. Weighted (1/x2) least squares regression analysis was applied to obtain the equation regression lines and correlation coefficients (r2). The correlation coefficient for GA and EA was found to be 0·9911 and 0·9923, respectively. Both EA and GA have been changed significantly throughout the growth of Azotobacter. In the presence of TA-supplemented GTA and MTA medium, no significant change was observed of pyrogallol in LC peaks. Attempt was made to quantify the EA and GA with different time intervals of growth. It was observed that EA was constitutively higher in MTA medium than GTA medium, interestingly EA production was started from their growth in MTA medium, but it is observed significantly after 24-h growth in GTA medium (Fig. 2). In MTA medium, the GA production was maximum at 24 h of growth and after that certainly falls down and remains unchanged.

Figure 2.

Gallic acid (GA) and ellagic acid (EA) production by Azotobacter strain SSB81. GA (□) and EA (■) production by SSB81 strain that was grown in GTA medium. GA (○) and EA (●) production by SSB81 strain that was grown in MTA medium. Data are the mean of triplicates ±SE. GTA, nitrogen-free glucose medium supplemented with tannic acid; MTA, nitrogen-free mineral medium supplemented with tannic acid.

Chemical shift analysis by 1H NMR

The transformed products were analysed by 1H NMR and compared with authentic samples. The 1H-NMR spectra of the biotransformation of TA at regular interval are presented in Fig. S1. The 1H-NMR showed that initial biotransformation of TA contains signal in the region of 7·08–6·83 and 6·79–6·50 ppm (aromatic part) predominantly. After 48-h reaction, 1H-NMR showed that the signal present in the region of 7·08–6·83 missed as well as signal nature at 6·79–6·83 ppm and also changed to 6·78–6·68 ppm, indicating the production of GA matched to standard commercial compound. New signal in 4·70–4·51 ppm region that also generated with solvent signal matches with standard EA. Another new signal formed in the 3·62–3·56 ppm region, indicating that some aliphatic compounds formed during transformation. 1H-NMR of 72-h reaction showed that as the reaction time increases, there is a nice change in aromatic region. Signals in the region of 7·08–6·83 and 6·79–6·50 ppm (aromatic part) were very small, almost absent, and the signal in the 4·70–4·51 ppm region along with the solvent signal matches the standard that increases with time. New signals were appeared and corresponding to aliphatic region 3·72–3·58 and 3·35–3·22 ppm, whereas another constant signal was observed at 2·43–2·47 ppm region. This information indicates that TA transformation was quite obvious in Azotobacter and metabolized for their sole carbon source through the production of gallic and ellagic acid.

Enzymatic activity

To identify the presence of catalytic enzymes responsible for tannin biotransformation, tannase activity was determined (0·89 ± 0·12 U ml−1) from cell free extract. Phenol decarboxylase and phenol oxidase activities were also detected and found to be 1·89 ± 0·14 U mg−1 and 4·6 ± 0·33 U mg−1, respectively.

Effect of TA of bacterial morphology

Examination of Azotobacter SSB81 using phase contrast and SEM revealed their significant morphological change when grown in both TA-supplemented GTA and MTA media. Cells that were grown without TA only N-free glucose medium showed homogeneous oval-shaped structure, occurring in single or in a pair. In TA-supplemented GTA medium, the presence of abnormal size and shape was increased. In MTA medium, most of the cells were elongated, and size was increased up to double of their size in only glucose medium (Fig. 3).

Figure 3.

Scanning electron and phase contrast micrographs of Azotobacter SSB81 after overnight growth. SSB81 was grown in glucose medium (a, d), in GTA medium (b, f) and in MTA medium (c, f). Cells were washed in 1× PBS buffer and fixed in glutaraldehyde solution as described in 'Methods' section. The viewed magnification of phase contrast microscopy was 10× × 40×. GTA, nitrogen-free glucose medium supplemented with tannic acid; MTA, nitrogen-free mineral medium supplemented with tannic acid.

EPS production

The growth and EPS synthesis were monitored with different time intervals in MTA and GTA media, and only glucose medium was considered as positive control. The cell biomass was rapidly increases with time (24–72 h) in GTA (1·21 mg ml−1) and glucose medium (2·31 mg ml−1), whereas the biomass of MTA medium was increases very slowly (0·52 mg ml−1) as compared with glucose and GTA media (Table 2). The EPS production trends are very similar in case of MTA and GTA media. The EPS production and growth rate were increased with incubation time for MTA and GTA media, but in glucose medium, the EPS production (1·44 g l−1) was increased up to 48 h and after that remained unchanged.

Table 2. Comparative table for growth, EPS and PHB production in different phenolic acid-supplemented media. The abbreviations are MTA, mineral tannic acid medium (without glucose); GTA, glucose tannic acid medium; glucose, only N-free Burk's medium. Data are the mean of triplicates ±SE
Growth mediumEPS and PHB production at different incubation periods
24 h48 h72 h
EPS g l−1PHB % of CDWBiomassEPS g l−1PHB % of CDWBiomassEPS g l−1PHB % of CDWBiomass
  1. a

    Statistical evaluation carried out using a Welch two-sample t-test, giving a confidence level of 95%, P < 0·01.

  2. EPS, extracellular polysaccharide; PHB, polyhydroxybutyrate.

MTA0·48 ± 0·09a1·06 ± 0·110·3 ± 0·040·82 ± 0·09a1·62 ± 0·220·39 ± 0·060·89 ± 0·07a1·76 ± 0·130·52 ± 0·02
GTA0·65 ± 0·05a8·65 ± 0·190·72 ± 0·090·79 ± 0·06a9·56 ± 0·411·05 ± 0·121·02 ± 0·05a9·71 ± 0·881·21 ± 0·14
Glucose1·21 ± 0·077·89 ± 0·240·84 ± 0·071·44 ± 0·0935 ± 2·261·48 ± 0·111·32 ± 0·0859·5 ± 2·02·31 ± 0·17

PHB synthesis

PHB was quantified from geometric mean of flow cytometry analysis that is standardized from estimated value of spectrometric quantification. The bacteria grown in glucose medium were found to be best for PHB production where they can accumulate up to 59·44% of the total cell dry weight in 72 h (Table 2). In GTA medium, the PHB production rate was nearly constant throughout growth, whereas very low PHB production was observed in MTA medium. At initial growth stage (24 h), the PHB production of GTA medium (8·65) was slightly higher than that in only glucose medium (7·88). Confocal microscopic analysis revealed the presence of PHB accumulation inside the cells grown in glucose medium where Nile red specifically binds to the PHB granules. The stained cells were analysed using FACS Calibur with Cell Quest Pro software for quantification (Fig. 4).

Figure 4.

Flow cytometry analysis of accumulated polyhydroxybutyrate (PHB) in Azotobacter strain SSB81. The histogram plots indicating the PHB accumulation in GTA medium (plot 1), only glucose medium (plot 2), MTA medium (plot 3) and unstained cells (plot 4). Localization of only PHB with Nile red stained (b) and merged with cells (c). The confocal microscopic pictured was captured from 48-h grown cells in glucose medium. GTA, nitrogen-free glucose medium supplemented with tannic acid; MTA, nitrogen-free mineral medium supplemented with tannic acid.


Diazotrophic soil bacteria particularly Azotobacter is extensively used as biofertilizer by agriculturalists, farmers and environmentalists. TA degradation through tannase production by Azotobacter has profound importance to sustain soil health and to improve plant productivity. We have described the biotransformation of TA by a widely used soil diazotrophic bacterium Azotobacter and their regulation of biopolymer synthesis during transformation. Present strain SSB81 is able to tolerate higher level of TA than available free or complex TA in soil (Halvorson and Gonzalez 2008). The biomass and growth rate of Azotobacter are much higher in GTA compared with MTA medium (Table 2). This is due to the presence of simple carbon source as glucose helps Azotobacter to grow rapidly. In MTA medium, extended lag phase was observed because of lack of readily available carbon source. In MTA medium, first enzyme was synthesized by the organism for breaking down of TA, and latter this degradative product was utilized by the organism for their growth. GA production was started from the initial growth of SSB81 in MTA medium, and further, it was metabolized to pyrogallol very slowly. Interestingly, GA was rapidly converted to EA in MTA medium after 24-h growth. In GTA medium, little amount of GA production was found and almost remained unchanged, but EA production started slowly after 48 h (Fig. 2). EA production in GTA medium was very low in comparison with MTA medium (Fig. 1) because of the delayed activation of TA degradation enzyme system. Earlier, it was reported that glucose supplement in the medium having the TA content of 2·5% showed a strong catabolic repression of tannase synthesis (Aguilar et al. 2001b). Several studies have conducted and reported on EA production from ellagotannins (Aguilera-Carbo et al. 2008). This is the first report about EA production from gallotannins.

Pyrogallol production also observed from GA after gallate decarboxylase activity, and this pyrogallol may enter into the energy metabolism via pyruvate in Azotobacter sp. as previously reported by Groseclose and Ribbons (1981). The presence of GA and pyrogallol throughout their growth period also supports the growth of this bacterium, as pyrogallol was catalysed very slowly. GA converts into EA rapidly in MTA medium may be due to the increased phenol oxidase activity after 24-h growth. The amount of GA was nearly constant after 72 h supports the possible reversible reaction of EA to GA conversion. It is clear from the in vitro enzymatic reaction using different substrate that TA was converted into GA and pyrogallol when TA was used as substrate. Further, EA and pyrogallol were observed when GA was substrate, and GA and pyrogallol identified from EA as substrate (Fig. S2). The crude enzyme extract revealed the conversion of EA to GA and vice versa. Tannase first catalyses the hydrolysis of ester and depside linkages in TA releasing glucose and GA (Mondal and Pati 2000). PPO plays an important role in the degradation of tannins and accumulation of EA from GA by contributing to oxidative breakdown. This report is in agreement with earlier report by Shi et al. (2005). Recently, Herter et al. (2011) confirmed the presence of phenol oxidase (PO) in Azotobacter, which showed distinct similarities to fungal PO. The nuclear magnetic resonance (NMR) result confirmed the significant and prominent shift of aromatic ring to aliphatic intermediates with increasing the incubation time. TA biotransformation is also in good agreement with the gradual disappearance of the aromatic region, and the increase in the peak number in aliphatic region indicates the formation of aldehyde derivatives and other methyl esters (3·72–3·58/3·35–3·22) as well as pyruvic acid in 2·43–2·47 ppm region (Fig. S1). A similar kind of findings was reported for resorcinol catabolism in A. vinelandii, where resorcinol first converted into pyrogallol and further degraded into pyruvate and aldehydes (Groseclose and Ribbons 1981).

Observed biomass demonstrated that the growth of Azotobacter was inhibited initially in both the media used. The presence of TA in the medium interferes with initial activation of metabolic cascade and extends the lag phage of Azotobacter for GTA and MTA media. The morphological studies (Fig. 3) also support the delayed growth of this bacterium. Elongated sizes of MTA medium growing cells were much higher than control (N-free glucose medium) and GTA medium growing cells. These data also demonstrate that Azotobacter is initially sensitive to tannins and required much time for adaptation by enzyme induction. Similar kind of observation was made against Streptococcus bovis and Streptococcus gallolyticus by O'Donovan and Brooker (2001).

The biopolymers of Azotobacter have great interest in industry like food industry, agro-industry and pharmaceutical sectors, as well as in agriculture because of their biodegradability and the potentiality to produce renewable carbon sources (Gauri et al. 2012). Azotobacter sp. can accumulate high amount of PHB during late log phase of their growth when cultivated on several rich carbon sources, sucrose (Rehm 2009). In our observation, PHB accumulation was maximum in only glucose medium, whereas PHB accumulation was decreased several fold in GTA and MTA medium. In GTA medium, initially higher amount of PHB accumulation was observed because of the increased concentration of free available glucose molecule. It is indicating that after the degradation of TA in MTA medium, TA is immediately utilized for their energy and lack of excess carbon in medium. The PHB accumulation generally occurs at stationary phase, whereas in GTA or MTA medium, the organism did not enter into the stationary phase up to 120 h, might be the another reason for low PHB accumulation during TA transformation. Further, EPS production was also tested and found very lower amount in TA-supplemented medium than that in only glucose medium. The abundance of NADH is a trigger for PHB and EPS formation (Anderson and Dawes 1990), and citrate synthase and isocitrate dehydrogenase activities are allosterically inhibited by high NADH/NAD+ ratio. Therefore, it seems that acetyl-CoA is continuous fed into the TCA cycle and not converted to acetoacetyl-CoA unfavouring the formation of PHB synthesis. Hence, TA might be involved in the central carbon metabolisms including tricarboxylic acid cycle and respiratory chain to control the higher energy demand of Azotobacter, which reduces the PHB accumulation. The acetyl-CoA was also not entering into the GDP-mannuronic acid pathway and inhibits EPS production. It might be predicted that the acetyl-CoA liberated from the aromatic ring cleavage of pyrogallol helps to fulfil the high energy demands needed for the rapid respiration of Azotobacter SSB81.

Present observation revealed that Azotobacter strain SSB81 respond to TA with an extended lag phase and slow growth rate. Azotobacter is able to initiate an adaptation response by the synthesis of TA catalytic enzymes as tannase, PPO and gallate decarboxylase for utilizing TA as their sole carbon source and energy metabolism via proposed catabolic pathway (Fig. 5). The major intermediates are GA and EA, and those are involved in a reversible reaction and utilized as their necessity for their energy. Thus, the widespread occurrence and utilization of Azotobacter in agricultural have another selective advantage to utilize the soil-accumulated TA as their carbon source and detoxify the allelopathic effect of constant accumulated TA in soil.

Figure 5.

Schematic presentation of tannic acid biotransformation by Azotobacter strain SSB81.