The lack of a nitrogen source and/or the C/N ratio affects the molecular weight of alginate and its productivity in submerged cultures of Azotobacter vinelandii

Abstract

The effect of a nitrogen (N) source on growth of the N-fixing bacteria Azotobacter vinelandii was evaluated in terms of the production of alginate and its macromolecular characteristics [mean molecular weight (MMW) and viscosity] in the presence of an organic or inorganic N-source or under N-fixing conditions, in submerged shake flask cultures. The effect of the C/N ratio was also investigated in the best alginate-producing nitrogen sources. The best N-sources were peptone, tryptone, and yeast extract, yielding a maximal final alginate concentration of 4.0 ± 0.4 g/L. Under N-fixing conditions, only 1.7 ± 0.2 g/L alginate was obtained. The highest MMW was obtained in cultures grown with peptone (1,520 ± 110 kDa), and cultures grown with yeast extract, tryptone, ammonium acetate, and ammonium sulfate were not significantly different, showing values between 1,400 and 1,100 kDa. On the other hand, a lower MMW was obtained under N-fixing conditions (625 ± 110 kDa). Higher alginate productivity was obtained using a C/N ratio of 14:1 for the best inorganic N-source, but when using the best organic N-source, no significant differences were observed by increasing the amount of nitrogen at a C/N ratio of 29:1. The lower alginate production under N-fixing conditions, along with the lower MMW, led us to propose that A. vinelandii sacrifices production of the biopolymer by establishing other mechanisms that can protect the polymerase complex—findings that are consistent with the current existing literature.

Introduction

Alginates are linear copolymers of linked β-d mannuronic acid, and its C-5 epimer, α-l-guluronic acid, in variable amounts (Gacesa 1998; Sabra et al. 2001; Galindo et al. 2007). Alginates are used widely as thickeners, stabilizers, gelling agents, and emulsifiers in the textile, papermaking, and pharmaceutical industries (Gacesa 1988, 1998; Galindo et al. 2007). As an indigestible polysaccharide, alginate is typically used as a source of dietary fiber (Brownlee et al. 2005). Commercial alginates are currently obtained by extraction from brown seaweeds such as Laminaria digitata, Laminaria hyperborea, and Macrocystis pyrifera (Gacesa 1988). Several bacteria, such as the nitrogen-fixing aerobic bacteria Azotobacter vinelandii, and the opportunistic pathogen Pseudomonas aeruginosa, produce alginate (Rehm and Valla 1997; Gacesa 1998; Galindo et al. 2007). Only alginates synthesized by A. vinelandii show block copolymer structures similar to those of algal alginate (Galindo et al. 2007). The commercial value of this biopolymer is generally low, and its manufacture and reliability depend on natural (i.e., climatic) or non-natural (e.g., contamination) conditions (Chéze-Lange et al. 2002). Nevertheless, bacterial alginate production may be enhanced in order to yield a high value and high purity product to be used in the food and pharmaceutical industries (Sabra et al. 2001). Currently, alginate productivity in submerged bacterial cultures has not reached values high enough to be of commercial interest, but a continuous research effort is aimed at gaining a better understanding of the metabolic production pathways, biological and physiological function, regulation of formation and composition, and optimization of production of alginate (Sabra et al. 2001; Galindo et al. 2007).

Azotobacter vinelandii has been reported to grow under different culture strategies, such as shake flask, batch, fed-batch, and in continuous bioreactors, and the effect of several culture variables are under study (a complete review was published recently by Galindo et al. 2007). Some of these studies were carried out with the addition of an organic N-source (Peña et al. 2000; Trujillo-Roldán et al. 2001, 2003a, 2003b, 2004; Peña et al. 2007, 2008); others reports used a N-inorganic source (Clementi et al. 1995; Clementi 1997; Diaz-Barrera et al. 2007), while others used N-fixing conditions (Sabra et al. 2000; Diaz-Barrera et al. 2009). Despite all this accumulated knowledge, key elements are missing from these investigations, e.g., comparing cultures grown with and without N-source, using strains with high respiratory capacity like A. vinelandii, and assessing the ability to grow and produce alginate by analyzing molecular characteristics such as alginate mean molecular weight (MMW) and viscosity. The biosynthesis of alginate by A. vinelandii depends on the carbon and nitrogen sources, but results are contradictory, which could be due mainly to the strain used and the composition and concentration of the culture medium components (Clementi et al. 1995; Clementi 1997; Sabra et al. 2001). Oppenheim and Marcus (1970) reported the effect on the internal membranous network of growing A. vinelandii with and without a N-source, observing that cells grown under nitrogen-fixing conditions were smaller than cells grown in any nitrogen source, and had the lowest specific growth rate. The A. vinelandii cultures resulting from a nitrogen-free source must be understood as cultures capable of converting atmospheric nitrogen into NH₄⁺. Moreover, the nitrogen-fixation reaction is very expensive in terms of biological energy equivalents, requiring large amounts of both reducing power and high-energy phosphate (ATP) (Halbleib and Ludden 2000). Brivonese and Sutherland (1989) reported poor alginate production in cultures grown with no nitrogen source, and that there was no improvement upon addition of nitrogen sources, such as nitrate or glutamate, but up to 7.0 g/L was obtained when a phosphate- and nitrogen-rich medium (PNR) was used (Brivonese and Sutherland 1989). In other models producing exopolysaccharides (EPS), such as A. pullulans, yields and chemical composition of EPS, in terms of overall pullulan content and the ratios of maltotetraose/maltotriose substitutions, are affected by changes in nitrogen source and its concentration (Orr et al. 2009). The aim of the present study was to determine the best correlation between nitrogen source and C/N ratio in terms of biomass growth, alginate production, the macromolecular characteristics of the polymer produced, and carbon source consumption in submerged cultures.

Materials and methods

Microorganism and culture conditions

Azotobacter vinelandii ATCC-9046 was used throughout this research. The strain was maintained by monthly subculture on Burk’s agar slopes, and stored at 4°C (Peña et al. 1997, 2000). Pre-cultures were grown at 200 rpm and 29°C in 250 mL shake flasks containing 50 mL modified Burk’s medium with the following composition (in g/L): sucrose 20; yeast extract 3, K₂HPO₄ 0.66, KH₂PO₄ 0.16, MOPS 1.42, CaSO₄ 0.05, NaCl 0.2, MgSO₄⋅7H₂O 0.2, Na₂MoO₄⋅2H₂O 0.0029, FeSO₄⋅7H₂O 0.027 (all reagents from Sigma, St. Louis, MO, with the exception of yeast extract from BD-Difco, Franklin Lakes, NJ). The pH was adjusted to 7.2 with a concentrated NaOH solution. To avoid precipitation during autoclaving, the FeSO₄⋅7H₂O and Na₂MoO₄⋅2H₂O solutions were separated from the other medium components during sterilization (Peña et al. 1997). In order to avoid unknown components from the exhausted culture medium, washed cells were used to inoculate shake flasks; exhausted broth components from the inoculum are known to play important regulatory roles in alginate biosynthesis, and in the determination of its molecular weight (Trujillo-Roldán et al. 2003b).

Analytical determinations

Biomass and alginate concentration were determined gravimetrically as described by Peña et al. 1997. Sucrose was hydrolyzed chemically using HCl (2 N), and then assayed for reducing power with DNS reagent as described by Miller 1959 and Peña et al. 1997. The viscosity of the culture broth was measured in a cone/plate viscosimeter (Wells-Brookfield LVT, series 82198; Brookfield Engineering Laboratories, Stoughton, MA), the MMW, polydispersity index (PI), and molecular weight distribution (MWD) were estimated by gel-permeation chromatography with a serial set of ultrahydrogel columns (UG 500 and linear; Waters, Milford, MA), using a conventional HPLC system with a differential refractometer detector (Waters, 410) (Peña et al. 2000; Trujillo-Roldán et al. 2004). The eluant was 0.1 M NaNO₃ at 35°C at a flow rate of 0.9 ml/min. Pullulans from Aureobasidium pullulans (from 5,800 to 1,600,000 Da) were used as standards (Peña et al. 2000).

Culture conditions and reproducibility

Cultures were grown at 200 rpm and 29°C in conventional 250 mL Erlenmeyer shake flasks containing 50 mL modified Burk’s medium containing the corresponding nitrogen source. All cultures were grown in several parallel flasks, three of which were regularly withdrawn and analyzed. Experiments were made at least in triplicate. For all figures, the mean value of at least three independent cultures is shown, and the standard deviation among replicas is presented as error bars.

Evaluation of C/N ratio and comparison with diazotrophic cultures

A C/N ratio of 29/1 was selected in order to match the previously reported C/N ratio of 20/3 g_sucrose/g_{yeast extract} (Peña et al. 2000; Trujillo-Roldán et al. 2001, 2003b, 2004; Peña et al. 2007). In all experiments the concentration of sucrose was kept constant (20 g/L), and the nitrogen source was changed in order to obtain the C/N ratio. Table 1 shows the percentage of carbon and nitrogen of the selected sources (peptone, tryptone, yeast extract, and casein from BD-Difco; sucrose, ammonium sulfate, ammonium acetate, and ammonium nitrate from Sigma). Table 1 shows the concentration of each nitrogen and carbon source used to obtain the required C/N ratio. Additionally, four C/N ratios (81/1, 29/1, 19/1, and 14/1) were evaluated in cultures grown with peptone and ammonium acetate, and these were compared with diazotrophic cultures; sucrose was kept constant at 20 g/L, and nitrogen sources were used as presented in Table 2.

Table 1 Carbon and nitrogen weight percentage per source and concentration used for a C/N mass ratio of 29/1

Table 2 Media composition for screening C and N sources as shown in Fig. 3

Results and discussion

Figure 1 shows the biomass concentrations, remaining reducing sugars, and final alginate concentrations in cultures grown for 120 h using a C/N ratio of 29/1 (as shown in Table 1). The 29/1 ratio is equivalent to the previously reported 20/3 g_sucrose/g_{yeast extract} (Peña et al. 1997, 2000, 2007; Trujillo-Roldán et al. 2001, 2003b). As shown in Fig. 1, no significant differences in final alginate concentration were observed when using peptone (4.0 ± 0.4 g/L), yeast extract (3.9 ± 0.3 g/L), and tryptone (3.9 ± 0.4 g/L). However, a lower final concentration was obtained in diazotrophic cultures (1.7 ± 0.2 g/L), and in cultures using inorganic sources (Fig. 1a). The highest biomass concentration was obtained using the same organic N-sources: peptone with 7.2 ± 0.4 g/L, yeast extract with 6.7 ± 0.3 g/L, and tryptone with 6.1 ± 0.3 g/L. On the other hand, cultures grown using ammonium acetate resulted in final biomass concentrations of 4.9 ± 1.1 g/L, similar to those in diazotrophic cultures (5.2 ± 0.3 g/L). In cultures grown using ammonium sulfate and ammonium nitrate, the biomass was lower (1.7 ± 0.5 and 1.4 ± 0.4 g/L, respectively). However, the highest product/biomass yields were found in cultures grown using ammonium sulfate (0.72 g_alginate/g_biomass), followed by cultures grown with tryptone (0.64 g_alginate/g_biomass), ammonium acetate and yeast extract (0.58 g_alginate/g_biomass); these observations lead us to believe that it is possible to design processes that could potentially increase the final biomass, increasing alginate productivity, e.g., fed batch cultures. The lowest yields were obtained in cultures with no addition of an N-source (0.33 g_alginate/g_biomass), showing that, under N-fixing conditions, A. vinelandii moves its metabolism to other processes, such as those involved in the protection, and the action, of the nitrogenase complex. Additionally, the yield in terms of substrate consumption for these N-fixing cultures (0.09 g_alginate/g_sucrose) was lower than those obtained from cultures grown in peptone, yeast extract, tryptone (0.20 g_alginate/g_sucrose), and ammonium acetate (0.18 g_alginate/g_sucrose).

Fig. 1

Comparison of final biomass and final alginate concentration (a), residual sugars (b), viscosity (c), mean molecular weight (MMW) and polydispersity index (PI) (d), in 120 h cultures using the same C/N ratio of 29/1 with various N-sources

Consumption of the carbon source is reduced significantly in cultures grown with inorganic N-sources (Fig. 1b), where between 30 and 50% of the carbon source remained unconsumed at the end of 120 h. These data concur with those published by Diaz-Barrera et al. (2007), reporting alginate concentrations between 0.5 and 1.5 g/L, and sucrose consumption of lower than 50%; said cultures were grown in a bioreactor using the same strain and similar ammonium acetate concentrations (1.0 g/L).

On the other hand, a high quality alginate can be understood as a polymer having high MMW (normally up to 1,000 kDa), and capable of generating high viscosity solutions (Sabra et al. 2001; Trujillo-Roldán et al. 2004; Galindo et al. 2007). Higher viscosity was achieved in cultures grown in yeast extract (0.092 ± 0.008 Pa.s), peptone (0.079 ± 0.007 Pa.s), and tryptone (0.072 ± 0.003 Pa.s), compared to casein, diazotrophic or inorganic N-sources (Fig. 1c). In terms of the MMW, the highest value was obtained in cultures grown with peptone (1520 ± 110 kDa); cultures grown with yeast extract, tryptone, ammonium acetate, and ammonium sulfate did not differ significantly, having MMWs between 1,400 and 1,100 kDa. The lowest MMW was obtained in cultures where nitrogen fixation occurred (625 ± 110 kDa). Therefore, differences in culture viscosity were assumed to be due to the MMWs, along with concentration, and probably the ratio of l-guluronic acid and d-mannuronic acid (G/M) and degree of acetylation in the biopolymer (Peña et al. 2006).

Due to the large differences between cultures grown with and without N-sources, a comparison was made between them, using yeast extract as an N-source. Figure 2 shows the kinetics of the growing cultures in terms of biomass, alginate production, viscosity of the culture medium, alginate MMW, PI, and sucrose consumption. In both cases, no lag phase was observed, and a similar maximal biomass concentration was obtained (Fig. 2a) for cultures both with and without a nitrogen source (5.9 ± 0.2 and 5.5 ± 0.6 g/L). Nevertheless, the growth of A. vinelandii in diazotrophic conditions showed a higher specific growth rate (μ = 0.11 h⁻¹) than the culture with yeast as a nitrogen source (μ = 0.06 h⁻¹). Oppenheim and Marcus (1970) reported a different behavior; they found that, under nitrogen fixation conditions, A. vinelandii had a specific growth rate lower than that of cultures grown with casaminoacids, NH₄Cl or NaNO₃. On the other hand, there was a noticeable decrease in biomass concentration at the end of the growing period (Fig. 2a), most probably caused by the intracellular consumption of poly-beta-hydroxybutyrate (PHB) (Peña et al. 1997; Pyla et al. 2009). Page and Cornish (1993) reported a decrease in biomass dry weight and PHB consumption when nutritional constraints occur. In this work, a decrease in biomass dry weight was seen in cultures grown in the absence of a N-source. Moreover, Peña et al. (1997) showed PHB consumption in baffled flasks in the first 12 h of culture, using as inoculum cells from conventional flasks whose cells had already accumulated PHB. Pyla et al. (2009) reported a decline in the concentration of PHB, by unit of biomass dry weight, in the stationary growth phase of a wild type strain of A. vinelandii.

Fig. 2

Biomass growth (a), alginate production (b), residual sugars (c), broth viscosity (d), MMW (e), and PI (f) of diazotrophic cultures (■) and cultures with yeast extract as N-source (•)

Under N-fixing conditions, the production of alginate by A. vinelandii is remarkably low (Fig. 2b), producing a maximal alginate concentration of 2.1 ± 0.1 g/L, compared to cultures using yeast extract (4.3 ± 0.5 g/L). In diazotrophic cultures, alginate production was partially growth-associated, and a large fraction of the total alginate synthesis took place in the stationary growth phase. In cultures where a nitrogen source is present, A. vinelandii produced the highest amount of polysaccharide in the exponential growth phase (Fig. 2). Alginate yield and the productivity in terms of biomass for cultures grown in yeast extract (1.14 g_alg/g_biom and 0.023 g_alg/g_biomh, respectively) were higher than in diazotropic cultures (0.31 g_alg/g_biom and 0.006 g_alg/g_biomh, respectively). Data from cultures grown with N-sources were of the same order of magnitude as those previously reported in shake flasks (Peña et al. 1997) and bioreactors (Peña et al. 2000; Trujillo-Roldán et al. 2001, 2003a, 2003b, 2004). Data from diazotrophic cultures were similar to those reported by Sabra et al. (2000) where a different strain of A. vinelandii (DSMZ 93-541b) was grown in continuous bioreactors, but lower in final concentration than those reported in batch cultures (Sabry et al. 1996). Brivonese and Sutherland (1989) reported a similar tendency of poor alginate production, which was not improved by the addition of N-sources (nitrate or glutamate), whereas up to 7.0 g/L was obtained when a phosphate- and nitrogen-rich medium (PNR) was used. Moreover, Oelze (2000) reported that both activity and formation of the respiratory system of A. vinelandii, are controlled by the C/N ratio in ammonium-limited chemostats, but respiratory activity in cultures with no nitrogen has not been reported (Oelze 2000).

The viscosity of the culture broth was also influenced by the medium conditions: the highest viscosity of culture broth was obtained using yeast extract (0.105 ± 0.0052 Pa.s), after 84 h of fermentation, whereas in diazotrophic cultures the viscosity obtained was as much as 0.058 ± 0.012 Pa.s (Fig. 2d). The molecular characteristics of the polysaccharides obtained under the two sets of culture conditions varied considerably (Fig. 2e). The alginate produced using nitrogen sources increases the MMW up to 1,180 kDa during the first 48 h, before decreasing to 990 kDa by the action of alginate lyases, as reported previously (Peña et al. 1997, 2000; Trujillo-Roldán et al. 2003a, 2003b). In contrast, in the product obtained from diazotrophic cultures, the MMW degraded rapidly from 1,400 kDa at the beginning of the culture, to 550 kDa. As can be seen in Fig. 2f, the PI behaves in concordance with the MMW. As alginate is degraded, the PI increases as a function of the formation of low molecular weight polymers (Trujillo-Roldán et al. 2003b, 2004).

The high MMW (with low PI) from cultures grown in yeast extract in comparison with cultures grown diazotrophically, and the differences in yields and productivities described above, can be explained by the very expensive nature of the reaction, in terms of biological energy equivalents, to convert N₂ into NH₄ ⁺ (Halbleib and Ludden 2000). Moreover, the protection of the nitrogenase in diazotrophic cultures by a high respiration rate, and the specificity of formation of an alginate capsule, are also important metabolic processes that help determine differences in alginate productivity and MMW (Sabra et al. 2000). Finally, to a certain extent, the differences in alginate MMW might be also associated with dissimilarities in specific growth rates. Priego-Jimenéz et al. (2005) demonstrated that, in fed-batch cultures grown at low specific growth rates, A. vinelandii produces alginates with higher MMW (and lower PI) than in fed-batch cultures at high specific growth rates. Such results support the idea of the existence of a strong relationship between A. vinelandii growth rate and alginate molecular characteristics, at least with the alginate produced in the exponential phase of growth.

In 1997, using the same culture medium and A. vinelandii strain, Peña et al. (1997) reported that alginate production in conventional flasks was partially growth-associated and that an important fraction of the total alginate synthesis took place in the stationary growth phase (Peña et al. 1997). Moreover, these authors reported a maximum alginate concentration (obtained after 72 h of cultivation) of 4.5 g/L, a viscosity of 0.52 Pa.s, and a MMW of 1.98 x 10⁶ kDa with a PI of 1.50 (Peña et al. 1997). However, productive differences between the cultures of Peña et al. (1997) and the cultures grown in this work could be due to the dimensions of the Erlenmeyer shake flasks, volumes, and the inoculation procedure. In Peña et al. (1997) shake flasks were 500 mL Erlenmeyer flasks containing 100 mL medium, inoculated with 10% pre-culture (about 0.1–0.3 g/L, dry weight). In this work, we used 250 mL shake flasks containing 50 mL culture medium, inoculating washed cells (inoculum cell concentration of 0.1–0.3 g/L, dry weight). As Peña et al. (2007) reported, depending on the filling volume of medium in the 500 mL shake flasks, a characteristic maximum oxygen transfer rate (OTR_max) and curve shape were observed; consumption power during alginate production also changes as a function of the filling volume, modifying alginate molecular weight and productivity (Peña et al. 2007). Thus, it is not surprising that, when comparing our results to those previously reported (added to the fact that we inoculated washed cells), alginate production was highly growth-associated and its molecular characteristics were also different, despite using the same strain of A. vinelandii (ATCC-9046).

For all C/N values evaluated, the quantity of the initial carbon source (i.e., sucrose at 20 g/L) in the medium was held constant (Table 2), and the initial N-content was varied using peptone (Fig. 3a), and ammonium acetate (Fig. 3b) to obtain C/N ratios of 81/1, 29/1, 19/1, and 14/1 (Table 2). As can be seen in Fig. 3, for an increase in any added nitrogen source (peptone or ammonium acetate), an increase in biomass and alginate production was observed. However, in the case of peptone, no differences in alginate production were seen at ratios of 29/1 to 14/1, showing a saturation tendency in the amount of nitrogen added. A similar profile of increasing alginate concentration was found for the viscosity of the culture medium, i.e., by decreasing the C/N ratio an increase of the viscosity of the culture medium was observed (data not shown). The results obtained in this study correspond to those reported by Cho et al. (2001), who reported that a low C/N ratio (17/1) yields higher values of A. vinelandii biomass and higher production of polyhydroxyalkanoates (Cho et al. 2001). In another model, Bhattacharya and Dubey (1997) reported for recombinant cultures of E. coli that the highest specific growth rate and biomass concentration and productivity of the heterologous gene were achieved at a C/N ratio of 15/1; values above that showed lower growth and productivity (Bhattacharya and Dubey 1997). Finally, by decreasing the C/N ratio, biomass and alginate production increases, and a decrease in residual sucrose is observed. This implies that consumption of the carbon source is proportional to the increase in peptone concentration in the culture medium.

Fig. 3

Final biomass (▲), alginate (•), and reducing sugars (■) concentration in 120 h cultures using the different C/N ratios for peptone (a), and ammonium acetate (b) as organic and inorganic N-sources, respectively

Our results suggest that the culture medium, and especially the nitrogen source, play important roles in the production and quality of alginate produced by A. vinelandii. In addition to the observations reported in this manuscript, the use of innovative culture strategies might produce higher yields of alginate. However, the molecular mechanisms that limit the production of the polymer have yet to be elucidated. Trujillo-Roldán et al. (2003b) suggest that small amounts of alginate and/or alginate-lyase (or possibly other unknown components synthesized by the bacteria) may serve as signals that modify production of the polymer. One possible explanation is a mechanism of action that may involve regulatory systems present in A. vinelandii, and that has been shown to regulate the biosynthesis of alginate and growth of the bacteria (Castañeda et al. 2000, 2001). This model proposes that the bacteria detect as yet unidentified environmental signals, and activate transcriptional regulators by a mechanism of phosphorylation, which in turn, acts to deregulate genes. In A. vinelandii, deregulation of these genes results, for example, in depression of the transcription of algD; AlgD is a key enzyme in the production of alginate (Castañeda et al. 2000, 2001). Moreover, it is possible that such deregulation directly affects the production of alginate and its molecular weight, due to variations in the metabolic flux and/or enzymatic activities (e.g., AlgD, AlgL, Alg8, Alg44, AlgJ, AlgEs, among others) involved in biosynthesis and post-modification of alginate. However, proteomic or metabolomic analysis evaluating the effect of different components of culture medium and/or culture conditions are needed to identify probable components of the culture medium, such as proteins or oligosaccharides, that could reveal which of these mechanisms serve as regulators of alginate production and growth of A. vinelandii.