Dr J. Moreno, Unidad de Microbiología, Departamento de Biología Aplicada, Escuela Politécnica Superior, Universidad de Almería, 04120 Almería, Spain.
Azotobacter vinelandii was cultured in chemically defined, nitrogen-free media supplemented with either 4-hydroxyphenylacetic, 4-hydroxybenzoic or protocatechuic acids at different concentrations. Under these conditions, biomass, exopolysaccharide production and consumption of the carbon sources were investigated. Results obtained throughout this study showed that 4-hydroxyphenylacetic acid yielded the highest growth levels measured as biomass, and exopolysaccharide production, independently of the concentration of the carbon source tested. 4-Hydroxybenzoic acid also supported appreciable growth and exopolysaccharide recovery by A. vinelandii. Protocatechuic acid, however, only allowed a very small production of biomass and exopolysaccharide by the strain investigated. Under given conditions, more than 26% of the carbon source supplied was converted to exopolysaccharide in cultures of A. vinelandii.
Azotobacter spp. are free-living, nitrogen-fixing, obligate aerobes which have been isolated from tropical and temperate regions ( Tchan 1984). Although traditionally they have been cultured with simple, oxidizable substrates (i.e. sugars) as carbon and energy sources, it is well established that these micro-organisms have the ability to oxidize many aromatic compounds ( Winogradsky 1935; Hardisson et al. 1969 ). Simple phenolic compounds, such as hydroxy and methoxy benzoic and cinnamic acids, are commonly formed in decaying plant residues, soils and roots ( Whitehead et al. 1983 ). They may also be synthesized by soil micro-organisms ( Flaig 1971) or formed as intermediates during humification ( Haider et al. 1975 ). As carbohydrates can be easily metabolized by many soil micro-organisms, it has been postulated that metabolism of phenolics may provide alternative carbon sources for some diazotrophs in many carbon (carbohydrates) limited environments ( Chan 1986). In addition, phenolic compounds can be metabolized only by a few members of soil micro-organisms, Azotobacter spp. being prominent among these ( Wu et al. 1987 ; Moreno et al. 1990 ).
On the other hand, Azotobacter spp. are known to produce copious quantities of exopolysaccharides (EPS) ( Gorin & Spencer 1966), commonly manifest as large mucoid colonies when isolated from the natural soil habitat. The physiological functions of extracellular polysaccharides produced by bacteria in nature, have been widely discussed. The ability of a micro-organism to surround itself in a highly hydrated EPS layer may provide it with protection against desiccation and predation by protozoans ( Whitfield 1988) or phage attack ( Sutherland 1972). Cells buried within a polymer matrix may be inaccessible to antimicrobial agents such as antibiotics ( Costerton et al. 1987 ) or affect the penetration of toxic metal ions ( Dudman 1977). They may also protect nitrogenase activity against high oxygen concentration ( Postgate 1974a), enable free-living bacteria to adhere to and colonize solid surfaces where nutrients accumulate ( Costerton et al. 1987 ), or participate in interactions between plants and bacteria ( Leigh & Coplin 1992).
Many polysaccharides produced by bacteria have characteristic properties in rheology and physiological activity different from natural gums and synthetic polymers. They are also susceptible to biodegradation in nature and less harmful to environmental pollution than synthetic polymers. For these reasons, some bacterial polysaccharides are produced on industrial scales and used as raw materials for food processing and medical and industrial preparations ( Sutherland 1990; Okamoto & Kaida 1994).
Exopolysaccharide production by Azotobacter has been traditionally investigated in media supplemented with carbohydrates as carbon sources ( Page & Sadoff 1975; Jarman et al. 1978 ; Horan et al. 1983 ). Under these culture conditions, Azotobacter synthesizes bacterial alginate and other polysaccharides. However, as phenolic acids are known to support growth and nitrogen fixation of Azotobacter in nature ( Moreno et al. 1990 ; Chen et al. 1993 ), the aim of this work was to investigate the production of exopolysaccharides by these bacteria in the presence of some phenolic compounds as sole carbon and energy sources.
Materials and methods
Azotobacter vinelandii CECT 204 was obtained from the Spanish Type Culture Collection (Department of Microbiology, Faculty of Biological Sciences, University of Valencia, Burjasot, Valencia, Spain). These organisms were maintained on slants of nitrogen-free glucose Burk’s medium ( Vela & Rosenthal 1972) and periodically checked for purity. Stock cultures were sub-cultured every 3 months, grown at 30 °C for 48 h and stored at 4 °C. Working cultures were sub-cultured from the stock culture every 3 months and were themselves sub-cultured monthly.
Media, inocula and culture conditions
The media used for the growth of A. vinelandii had the same salt composition as nitrogen-free Burk’s medium ( Vela & Rosenthal 1972). Three carbon sources were assayed throughout this study: 4-hydroxybenzoic acid, protocatechuic acid and 4-hydroxyphenylacetic acid, at concentrations of 10, 25 and 50 mmol l−1. Glucose was also used at the same concentrations in some experiments. Micro-organisms from maintenance slants were cultured in nitrogen-free glucose liquid Burk’s medium, and then transferred to nitrogen-free Burk’s liquid media amended with 25 mmol l−1 4-hydroxyphenylacetic acid. After three consecutive transfers, inocula were prepared from this medium. Azotobacter vinelandii cells were incubated at 30 °C under continuous agitation (120 rev min−1) for 48 h, harvested by centrifugation and washed twice in sterile distilled water. Volumes (1 ml) of the suspensions (O.D. 0·5 at 540 nm) were inoculated into 250 ml Erlenmeyer flasks containing 50 ml of medium supplemented with 10, 25 or 50 mmol l−1 of the corresponding carbon source. Optical density measurements were made with a Shimadzu UV-160 A spectrophotometer (Kyoto, Japan). All media, once inoculated, were incubated at 30 °C on a rotary shaker (120 rev min−1). pH was monitored and adjusted at 7·3 as needed. Experiments were performed in triplicate.
4-Hydroxybenzoic acid, 3,4-dihydroxybenzoic (protocatechuic) acid and 4-hydroxyphenylacetic acid, were obtained from Aldrich Chemical Company.
Culture media samples were removed at intervals and each sample was divided into subsamples for the determination of biomass, cell protein, exopolysaccharide concentration and carbon source remaining in solution. All assays were done in triplicate.
For biomass estimation, 1 ml volumes of culture media were harvested by centrifugation, washed twice with sterile distilled water and evaporated to dryness in pre-weighed vials during 3 h at 100 °C. Proteins were routinely determined by the method of Lowry et al. (1951) , using bovine serum albumin (Sigma) as standard. Whole cells were pre-digested in 0·1 n NaOH at 80 °C for 1 h.
Remaining amounts of 4-hydroxybenzoic acid, protocatechuic acid and 4-hydroxyphenylacetic acid in culture media were determined in culture supernatant fluids according to the method described by Marambe & Ando (1992) , with appropriate modifications. A 0·5 ml aliquot of each sample was diluted with 7 ml deionized water with thorough mixing in test tubes. Then, 0·5 ml Folin-Ciocalteau reagent was added to the mixture and the samples were thoroughly mixed again. After standing for 3 min, 1 ml of 20% Na2CO3 (saturated solution) was added and mixed, and 1 ml deionized water was added to make a final volume of 10 ml with continuous mixing. The tubes were allowed to stand for 1 h. The absorbance of the coloured solutions was read at 725 nm in a Shimadzu UV-160 A spectrophotometer. Solutions of pure phenolic acids (4-hydroxybenzoic acid, protocatechuic acid and 4-hydroxyphenylacetic acid) of 10–500 mg l−1 were used as standards.
Exopolysaccharides were extracted with three volumes of isopropanol from the culture supernatant fluid by vigorous shaking, according to the method proposed by Jarman et al. (1978) . After 10 min, precipitated exopolysaccharides were filtered onto pre-dried and pre-weighed GF/A Whatman filter discs (Whatman International Ltd, Kent, UK), and washed with 100 ml isopropanol/water (3 : 1, v/v). The filter disc plus precipitate was dried under vacuum at 45 °C for 24 h. Filters were re-weighed and the concentration of exopolysaccharides in the culture broth was calculated.
Chemical analysis of exopolysaccharides obtained in media supplemented with either phenolic acids or glucose was performed as follows. Total carbohydrate content was determined by the anthrone reagent, using pure glucose as standard ( Herbert et al. 1971 ). Uronic acids content was analysed according to the method described by Blumenkratz & Asboe-Hansen (1973), using pure glucuronic acid as standard. Acetyl content was determined by the method of Hestrin (1949) using acetyl choline HCl as standard.
When A. vinelandii was cultured in chemically-defined media with phenolic acids at different concentrations, 4-hydroxyphenylacetic acid yielded the highest growth levels measured as biomass production, independently of the concentration of carbon source tested. 4-Hydroxybenzoic acid also supported appreciable growth of A. vinelandii, but protocatechuic acid only allowed a very small production of biomass by the strain studied ( Fig. 1).
As previously stated, biomass production was investigated at various concentrations of the different carbon sources tested. When 4-hydroxyphenylacetic or 4-hydroxybenzoic acid were used as carbon source, an increase in their concentration implied an increase in biomass production. Thus, higher concentrations of these phenolic acids yielded higher values of biomass production ( Fig. 1). This was also the pattern for protocatechuic acid. However, as shown in Fig. 1, only small growth levels were reached compared with those obtained with other phenolic acids.
Once inoculated, culture media with phenolic acids added at different concentrations showed maximum levels of growth after 2–3 d. Similar amounts of biomass were maintained in 5-day-old cultures ( Fig. 1).
Measurements of the rate of phenolic acids utilization during growth clearly showed that 4-hydroxybenzoic acid was totally degraded after 3 d by A. vinelandii, independently of the initial concentration supplemented. At low initial concentrations (10 mmol l−1) of 4-hydroxyphenylacetic acid, the degradation rate was very similar to that of 4-hydroxybenzoic acid in A. vinelandii. However, at higher concentrations, utilization of 4-hydroxyphenylacetic acid was much smaller. Thus, appreciable amounts of this compound remained in cultures up to 5 d. Protocatechuic acid was degraded to a very small extent. Thus, no more than 20% was used by the micro-organisms, irrespective of the initial concentration employed ( Fig. 1).
Preliminary experiments were performed in order to determine the incubation time for optimum recovery of exopolysaccharides in A. vinelandii cultures. Thus, culture media samples were removed at intervals and quantitative extractions of EPS were made. According to these results (data not shown), maximum EPS recovery could be obtained in 5-day-old cultures, so this incubation period was selected for EPS extraction and quantification in further experiments. As shown in Figs 2, 4-hydroxyphenylacetic acid gave the optimum EPS recovery in A. vinelandii cultures, irrespective of the initial concentration. At low concentrations (10 mmol l−1) of 4-hydroxybenzoic and protocatechuic acids, EPS production was similar. However, at 25 and 50 mmol l−1 concentrations of 4-hydroxybenzoic acid, EPS recovery was much higher than that obtained with equal concentrations of protocatechuic acid ( Fig. 2).
Irrespective of the phenolic acid tested, a clear correlation could be observed between EPS production and the initial concentration of the carbon source investigated ( Fig. 2).
Table 1 shows specific yields of exopolysaccharide production related to biomass production and carbon source consumption. As shown in this table, specific yields of EPS production (YP/B and YP/C) were higher as the initial concentration of substrates increased. The rate of substrate conversion to exopolysaccharide (QCP) was highest when 4-hydroxyphenylacetic acid was used as carbon source.
Table 1. Yields of exopolysaccharide production in Azotobacter vinelandii CECT 204 cultures supplemented with either p-hydroxybenzoic, protocatechuic or p-hydroxyphenylacetic acids at different concentrations
QC→P, Conversion carbon source to polysaccharide (%).
Production and chemical composition of EPS obtained in media supplemented with glucose were investigated and compared with those obtained in media supplemented with phenolic acids. Results shown in Fig. 2 demonstrate that irrespective of the initial concentration employed, EPS recovery in 4-hydroxyphenylacetic acid-amended media was even higher than that obtained in media supplemented with glucose. However, 4-hydroxybenzoic and protocatechuic acids yielded lower amounts of EPS when compared with EPS production in glucose-amended media. The chemical composition of EPS obtained either in the presence of phenolic acids or glucose was also different ( Table 2). Thus, total carbohydrate content was significantly higher in EPS produced when glucose was used as carbon source. On the other hand, EPS obtained in 4-hydroxyphenylacetic acid-amended media were shown to contain higher amounts of uronic acids and acetyl residues.
Table 2. Chemical composition of exopolysaccharides obtained in cultures supplemented with either glucose or p-hydroxyphenylacetic acid
The ability of Azotobacter species to use aromatic compounds has long been known ( Winogradsky 1935). Phenolic acids included in this study are metabolized through dioxygenases by ring cleavage ( Hardisson et al. 1969 ; Chen et al. 1993 ). Dissimilation of 4-hydroxybenzoic acid and related hydro-aromatic acids is initiated by introducing oxygen as a hydroxyl group. The mono-oxygenase responsible for this initial reaction, 4-hydroxybenzoate hydroxylase, converts 4-hydroxybenzoic acid into protocatechuic acid. Further metabolism is performed by ring ortho-fission through the protocatechuate branch of the β-ketoadipate pathway, yielding the intermediaries of the Krebs’ cycle, succinate and acetyl-CoA ( Hardisson et al. 1969 ). On the other hand, metabolism of 4-hydroxyphenylacetic acid is likely to proceed by complete oxidation of the side (acetyl)-chain giving phenol, which is dissimilated, after conversion to cathecol, by ring meta-fission yielding pyruvate and acetaldehyde as final products ( Dagley 1971). All the enzymes catalysing both the ring ortho- and meta-fission pathways are strictly inducible ( Hardisson et al. 1969 ; Chen et al. 1993 ).
According to the results presented here ( Fig. 1), differences observed in biomass production can be primarily ascribed to the metabolic pathway by which phenolic acids present in culture media are degraded. Data presented in Fig. 1 show that A. vinelandii only degraded 20% of the protocatechuic acid added to the medium after 5 d of growth. However, 4-hydroxybenzoic acid, which is dissimilated by previous conversion to protocatechuic acid, was totally degraded after 3 d. This was surprising, but re-examination of the literature suggested that it might not be abnormal. The oxidative ring cleavage of protocatechuic acid, as well as most of the oxidative steps, is mediated by oxygenases ( Hayaishi 1966). Consequently, molecular oxygen does not simply act as a terminal electron acceptor, but is also a specific substrate for a number of step-reactions. Oxygen is reduced at the substrate level, by-passing completely the respiratory electron-transport system and the machinery of oxidative phosphorylation. Hence, the operation of such pathways rarely leads directly to generation of ATP. Synthesis of ATP is for the most part secondary, associated with the terminal oxidation of the aliphatic end-products through the reactions of the tricarboxylic acid cycle ( Stanier & Ornston 1973). The fact that most of the steps associated with the oxidation of aromatic substrates are mediated by oxygenases has a second important physiological consequence for Azotobacter when grown in nitrogen-free media. Nitrogen fixation is a process that requires high amounts of available ATP ( Postgate 1974b). In addition, according to the hypothesis of respiratory protection of nitrogenase ( Hill et al. 1972 ; Yates & Jones 1974; Robson & Postgate 1980), Azotobacter is able to protect nitrogenase against damage by oxygen only if it is able to increase the cellular respiratory activity concomitantly. In other words, under moderate or high rates of aeration, dinitrogen fixation by Azotobacter is markedly influenced by the rate at which oxygen is reduced by respiratory activity. Another important factor influencing metabolism of aromatic substrates by Azotobacter when grown in nitrogen-free media, is the concentration at which they are present in the culture. Some phenolic acids become toxic to Azotobacter, inhibiting nitrogenase activity when present in high concentrations ( Moreno et al. 1990 ). This could be explained by the fact that the amount of available oxygen is not high enough to support both the activity of dioxygenases and the high respiratory activity required to protect nitrogenase against damage by oxygen. Some reports have indicated good growth levels of Azotobacter on protocatechuic acid at concentrations lower than those used here ( Hardisson et al. 1969 ; Chen et al. 1993 ). In our experiments, protocatechuic acid was used at 10, 25 and 50 mmol l−1. These were also the concentrations used for 4-hydroxybenzoic acid. However, this compound first has to be converted to protocatechuic acid by hydroxylation and then the degradation pathway initiated as previously stated. Under these conditions, it could be said that protocatechuic acid is slowly released from the hydroxylation of 4-hydroxybenzoic acid, avoiding the accumulation of the former compound at concentrations that would be toxic to the micro-organism.
It is well established that in Azotobacter, EPS formation is markedly increased after cessation of growth ( Okabe et al. 1981 ; Horan et al. 1983 ). Under the conditions employed in this study, maximum levels of EPS recovery were obtained in the stationary phase after 120 h of growth. Similar results have been reported in media supplemented with carbohydrates as carbon sources, where optimum amounts of EPS could be recovered after 70–120 h ( Jarman et al. 1978 ; Okabe et al. 1981 ; Horan et al. 1983 ). According to the results presented here, some phenolic acids can support EPS production by Azotobacter in nitrogen-free media. Moreover, EPS recovery from media supplemented with 4-hydroxyphenylacetic acid was even higher than that obtained in media supplemented with glucose ( Fig. 2). The amounts of EPS produced are actually small in comparison with those obtained by others ( Jarman et al. 1978 ; Okabe et al. 1981 ; Horan et al. 1983 ). However, it has to be taken into account that sugars are readily degradable carbon sources, and strains used in those studies are mostly mutants with an increased ability for the production of EPS. In spite of this, in our study, at concentrations of 25 mmol l−1 of 4-hydroxyphenylacetic acid, more than 26% of the carbon source was converted to EPS by A. vinelandii ( Table 1). A yield of 31% alginate from sucrose in Burk’s medium after 110 h has been reported for A. vinelandii NCIB 9068 ( Chen et al. 1985 ). A yield of 5–26% alginate of the amount of carbon source supplied to various strains and mutants of A. vinelandii has been mentioned in the same report.
The different chemical composition of EPS produced with either phenolic acids or glucose, mainly with respect to uronic acids and acetyl content, is noteworthy. Alginate is a heteropolysaccharide composed of guluronic and mannuronic acids and acetyl residues. This polymer is the main polysaccharide produced by A. vinelandii in chemically-defined media ( Jarman et al. 1978 ), although the production of other polysaccharides by this micro-organism has also been reported ( Gorin & Spencer 1966; Vermani et al. 1996 ).
The results presented in Table 2 show a large difference between the content of uronic acids in media containing glucose and p-hydroxyphenylacetic acid. Our data also demonstrated that O-acetyl content was very low in comparison with previously reported data (4–57%) ( Skjåk-Bræk et al. 1986 ; Sutherland 1990). These results could be explained by the highly variable quantitative composition of alginates, which mainly depends on the culture conditions and the strain used ( Sutherland 1990).
Phenolic acids are carbon compounds available from a variety of sources, including soil and residues of vegetable origin, from which they are slowly but constantly released. Results presented here indicated that some of these substrates can support growth and nitrogen fixation by A. vinelandii. In addition, it can be said that these compounds have potential for the production of EPS on a large scale.