- Email: [email protected]
Conversion of the nitrogen content in liquid manure into biomass and polyglutamic acid by a newly isolated strain of Bacillus licheniformis
FEMS Microbiology Letters 218 (2003) 39^45
www.fems-microbiology.org
Conversion of the nitrogen content in liquid manure into biomass and polyglutamic acid by a newly isolated strain of Bacillus licheniformis Astrid Hoppensack, Fred Bernd Oppermann-Sanio, Alexander Steinbu«chel
Institut fu«r Mikrobiologie der Westfa«lischen Wilhelms-Universita«t Mu«nster, CorrensstraMe 3, 48149 Mu«nster, Germany Received 6 August 2002; accepted 17 October 2002 First published online 29 November 2002
Abstract Extensive spreading of liquid manure onto agricultural fields causes eutrophication of ground and surface water and also pollution of the atmosphere due to the high ammonium nitrogen content. A poly(Q-glutamic acid) (PGA)-producing strain of Bacillus licheniformis was isolated in this study and investigated for its ability to reduce the ammonium nitrogen by converting ammonium into biomass and PGA as depot forms of nitrogen. In batch cultivations swine manure and an optimized mineral salts medium were used for PGA production. For example the cultivation of B. licheniformis strain S2 in liquid manure, which was modified by adding of 18 g citrate and 80 g glycerol l31 and exhibited a carbon to nitrogen ratio of 15.5:1, led to severe reduction of the ammonium content from 2.83 to 0.1 g l31 and to the production of 0.16 g PGA and 7.5 g cell dry mass l31 within 410 h. Approximately 28% (w/w) of the total nitrogen was converted into cellular biomass, whereas 0.1% (w/w) was used for the production of PGA. In addition, approximately 33% (w/v) of the original ammonium was lost by stripping. ; 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Liquid manure ; Biomass ; Polyamide; Bacillus; Nitrogen
1. Introduction Intensi¢ed animal breeding and growth or generation of products from animals have caused serious environmental problems in some geographic regions during the last two decades (European Communities website, http://europa. eu.int). Extensive spreading of liquid manure for example arising from large-scale pig fattening on agricultural ¢elds has resulted in pollution of ground water and the atmosphere due to the high concentrations of ammonium in this waste [1]. On the other hand, liquid manure provides an excellent source of nutrients for crops, if managed properly (European Communities website, http://europa. eu.int). Conversion of the nutrients into microbial biomass with liquid manure as substrate and conversion of the ammonium nitrogen into nitrogen-containing biopolymers such
* Corresponding author. Tel. : +49 (251) 8339821; Fax : +49 (251) 8338388. E-mail address : [email protected] (A. Steinbu«chel).
as peptidoglycan, nucleic acids, proteins and polyamino acids (PAAs) would provide transient depots for ammonium and also other nutrients occurring in swine waste. This microbial biomass and the biopolymers could exhibit a fertilizing function for a prolonged period during decay. Moreover, knowledge obtained on the handling of manure from pigs may be also applicable to manures from other animal sources. Poly(Q-D-glutamic acid) (PGA) is a naturally occurring polyanionic, water-soluble extracellular polyamide [2] with potential applications as a biodegradable thickener, humectant, sustained-release material or drug carrier in food, cosmetics or medicine. It is also an important component of some Asian foods such as Natto [3]. It is one of the few naturally occurring polyamides which are not synthesized by ribosomal protein biosynthesis [4]. Due to the remarkably resistance of PGA against proteolytic attack, degradation of PGA by a non-adapted micro£ora is expected to be very slow. This should result in a continuously low rate of release of free ammonium. In this study, we investigated the conversion of nitrogen into biomass and PGA by batch fermentation of a newly
0378-1097 / 02 / $22.00 ; 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII : S 0 3 7 8 - 1 0 9 7 ( 0 2 ) 0 1 1 3 2 - 1
FEMSLE 10772 20-1-03
40
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
isolated PGA-producing strain of Bacillus licheniformis in liquid manure.
2. Materials and methods 2.1. Growth of bacteria Isolate S2 was cultivated in medium E [5] without glutamate as well as in particle-free liquid manure and in particle-free liquid manure supplemented with di¡erent carbon sources. The chemical composition of the liquid manure was determined at the Landwirtschaftliche Untersuchungs- und Forschungsanstalt (Oldenburg, Germany) and was as follows: 0.46% (w/w) nitrogen, 0.32% (w/w) ammonium N, 0.29% (w/w) P2 O5 , 0.26% (w/w) K2 O, 0.09% (w/w) MgO, 10 mg Cu kg31 , 1.9% (w/w) C, and 4.6% (w/w) dry matter. The liquid manure as received was stored at 330‡C, and it was centrifuged for 30 min at 3000Ug and autoclaved to obtain almost particle-free and sterile liquid manure before use. Carbon sources were sterilized separately by autoclaving before they were added to the manure. All cultures were inoculated with cells obtained from a 3-day-old preculture in liquid manure containing 2% glucose, and the cells were washed twice with sterile saline and concentrated three-fold before inoculation. Batch cultivations were carried out at 30‡C and at a stirring speed of 120 rpm in 2-l conical £asks equipped with three ba¥es containing 200 ml medium. Growth was monitored photometrically at 600 nm.
Glucose was determined photometrically according to the method described by Bergmeyer et al. [8]. For photometric quanti¢cation of glycerol and citrate, respective analysis kits (#148270, #139076) purchased from Boehringer Mannheim were used. Photometric determination of nitrogen was carried out using the nitrogen determination kit (#83) from Macherey-Nagel. Concentrations of NHþ 4 were determined as ammonia (NH3 ) with a type 15 230 3000 gas-sensitive electrode (Mettler Toledo, Steinbach, Germany; detection limit: 1036 mol l31 ) according to the instructions provided by the manufacturer. The amount of isolated PAAs was analyzed after precolumn derivatization of PAA hydrolysates with orthophthaldialdehyde reagent on a reversed-phase high-performance liquid chromatography (HPLC) column as recommended by the manufacturer (Merck, Darmstadt, Germany). Hydrolysis was carried out by adding 1 ml of 6 N HCl to 1 mg of lyophilized PAAs and incubating the mixture for 8 h at 100‡C. 2.5. Nucleotide sequence accession number The 16S rDNA gene nucleotide sequence of isolate strain S2 was submitted to the EMBL nucleotide sequence database and is available under accession number AY052767.
3. Results 3.1. Isolation and taxonomic classi¢cation of the new isolate strain S2
2.2. Analysis of 16S rDNA Extraction of chromosomal DNA was carried out by the method of Ausubel et al. [6]. The 16S rDNA gene (rDNA) was ampli¢ed using oligonucleotide primers as described previously [7]. The nucleotide sequence was determined by DNA sequencing according to the chain termination method using the 4000L automatic sequencer LiCor (MWG-Biotech, Ebersberg, Germany). The 16S rDNA sequence was aligned with published sequences from the NCBI database using the Clustal X software. 2.3. Isolation of PGA and cell dry mass analysis PGA was isolated and puri¢ed by the method reported previously [2]. For analysis of cell dry mass, cells were harvested (60 min, 3000Ug, 4‡C) and washed once with 0.9 N NaCl. The obtained pellet was lyophilized, and the weight was determined. 2.4. Analytical methods In the cell-free supernatants the concentrations of glucose, glycerol, citrate, and ammonium were analyzed at di¡erent times during the time course of the incubation.
The microorganism used in this study was strain S2 which was isolated from a salty sea sludge sample from the island of Malta. Approximately 1 ml of the salty sea sludge sample was used to inoculate 20 ml of Luria^Bertani medium containing 12.5% NaCl in order to obtain PGA-producing bacteria. The culture was incubated aerobically at 30‡C for 3 days. Aliquots of this culture were plated on solid medium E, and a visible viscose colony was identi¢ed as a potential plasmid-free producer of PGA, which consists of 2.2% L-glutamate and 97.8% D-glutamate. The morphological and physiological characteristics of strain S2 are summarized in Table 1. For phylogenetic a⁄liation of this strain the complete 16S rDNA gene was ampli¢ed by polymerase chain reaction and its nucleotide sequence revealed as described in Section 2. Comparison of the obtained sequence comprising 1547 nucleotides with 16S rDNA sequences available from the NCBI database revealed the greatest similarity to the corresponding sequence of B. licheniformis DSM13 (99.4%). The pro¢le of the cellular fatty acids of strain S2 is typical of the Bacillus subtilis group, to which B. licheniformis also belongs (15:0 iso, 22.45% ; 15:0 aniso, 42.59% ; 16:0 iso, 4.49%, 16:0 aniso, 5.36%; 17:0 iso, 8.75%; 17:0 aniso,
FEMSLE 10772 20-1-03
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
16.37%; deviation for B. licheniformis : 0.522; deviation for B. subtilis: 0.572). Furthermore, the 16S rDNA sequence of strain S2 and the other members of the B. subtilis group as well as Bacillus anthracis were aligned, and a phylogenetic tree was constructed based on the neighborjoining method using the Clustal X program (version 1.8; Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. and Higgins, D.G., 1997). A bootstrap analysis was carried out, and the tree (Fig. 1) indicates that strain S2 and B. licheniformis DSM13 are the most closely related strains. Strain S2 will therefore be tentatively referred to as B. licheniformis strain S2.
41
B. vallismortis DSM 11031
B. amyloliquefaciens ATCC23350
B. atrophaeus JCM 9070
B. licheniformis strain S2 681 1000
B. licheniformis DSM 13 710
B. pumilus NCDO 1766
3.2. Growth kinetics of B. licheniformis strain S2
690
984
B. anthracis 1392
To enable a continuous measurement of growth parameters, almost particle-free liquid manure was used in this study. Because B. licheniformis strain S2 was not able to grow on particle-free liquid manure without an additional carbon source (Fig. 2D), the in£uence of 2% (w/v) glucose (Fig. 2C) or of 8% (w/v) glycerol plus 1.87% (w/v) citrate (Fig. 2B) on growth and PGA production was investigated in batch cultivations. With respect to reducing the concentration of nitrogen in liquid manure, we investigated if PGA production occurred in medium without glutamate as precursor, because glutamate contains nitrogen and is therefore not a suitable co-substrate in liquid manure. In medium E containing no glutamate, the speci¢c growth rate (W) of B. licheniformis strain S2 was 0.12 h31 , and the concentrations of PGA were 0.26 g l31 in the middle of the exponential growth phase and 0.18 g l31 at the end
Table 1 Morphological and physiological characteristics of B. licheniformis strain S2 Cell form Cell size Spores Sporangium swollen Gram reaction Catalase Oxidase Anaerobic growth VP reaction pH in VP medium Temperature for growth at 50‡C at 55‡C in medium pH 5.7 in NaCl 2% in NaCl 5%
rods 0.7^0.9U2.0^3.5 Wm +, oval 3 positive + 3 + + 5.4
Acid from D-Glucose L-Arabinose D-Xylose D-Mannitol D-Fructose Hydrolysis of Starch Gelatin Casein Tween 20
+ 3 + + +
in NaCl 10% Lysozyme medium Esculin hydrolysis Gas from glucose Nitrate reduction
+ 3 + 3 +
Utilization of Citrate Propionate Tyrosine hydrolase Phenylalanine deaminase Resistant against Ampicillin Chloramphenicol Tetracycline Kanamycin
+ + + + + + + + +
+ + 3 3
+ + 3 3
B. mojavensis IFO 15718 744
B. subtilis NCDO 1769
0.01
Fig. 1. Bootstrap neighbor-joining phylogenetic tree of B. licheniformis strain S2 and related bacteria of the B. subtilis group and B. anthracis based on 16S rRNA sequence comparisons. The scale bar indicates 1% sequence dissimilarity ; bootstrap values (1000 replicates) are mentioned at the nodes. Accession numbers for the reference sequences used are: B. amyloliquefaciens (X60605) ; B. anthracis (X55059); B. atrophaeus (AB021181); B. licheniformis (X68416) ; B. mojavensis (AB021191); B. pumilus (X60637) ; B. subtilis (X60646); B. vallismortis (AB021198).
of the incubation after 80 h. When this strain was incubated in swine manure, a relatively long lag phase was needed for adaptation. Slower growth (W = 0.1 h31 ) and less PGA (0.16 g l31 at the end of the exponential growth phase and 0.02 g l31 at the end of incubation) were obtained in liquid manure supplemented with glycerol and citrate. Faster growth (W = 0.20 h31 ) was achieved in liquid manure containing glucose. At the beginning of the stationary growth phase the PGA concentration was 0.014 g l31 , whereas after a further 10 h incubation it was 0.26 g l31 and after a further 26 h incubation no PGA was detected. The highest cell densities (10.7 g cell dry mass l31 ) were obtained in modi¢ed medium E and in liquid manure with glucose, whereas in liquid manure supplemented with glycerol and citrate the cell density was 7.5 g l31 . In contrast to growth in modi¢ed medium E, B. licheniformis strain S2 needed an extremely long lag phase of approximately 220 h and 340 h for adaptation in liquid manure supplemented with glucose or with glycerol plus citrate. 3.3. Diminishing of ammonium During incubation of B. licheniformis strain S2 in modi¢ed medium E, the ammonium concentration decreased
FEMSLE 10772 20-1-03
42
100
5
80
4
20
12
100
10
10
80
8
8
6
6
4
4
2
2
12
14
6 4 4
5 1
2
-1
-1
)
)
) 40
2
10
pH
8
6
3
Citrate (g l
pH
10
60
NH (g l 3
2
10
-1
-1
)
8
5 20
1
0
0
60
40
20
2 0
0
0
0
0
20
40
60
80
0
0
100
320
340
Incubation time (h)
FEMSLE 10772 20-1-03
5
12
0
0 300
360
380
400
420
Incubation time (h)
16
C
25
5
20
4
12
10
D
2
6 4 4
1
2
10
8
)
8 3
6 6
2
4 4
5
1
0
0
Optical density at 600 nm
8
6
15
10
pH
pH
NH (g l 3
-1
)
10
-1
8 3
NH (g l 3
12
Optical density at 600 nm
10
4
-1 Glucose (g l )
14
2
2
2 0
0
0 160
180
200
220
240
Incubation time (h)
260
280
0
0 0
100
200
300
400
500
Incubation time (h)
Fig. 2. Incubation of B. licheniformis strain S2 in modi¢ed medium E (A), liquid manure with 8% (w/v) glycerol plus 1.87% (w/v) citrate (B), liquid manure with 2% glucose (C) and liquid manure without co-substrate (D). The cultivations were carried out at 30‡C in 2-l conical £asks equipped with three ba¥es containing 200 ml of the corresponding medium. Symbols : F, optical density; E, citrate ; S, glucose; O, pH; b, NH3 ; a, glycerol. At times indicated by arrows, 50 ml of the culture was removed for isolation of PGA.
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
Citrate (g l
) -1 NH (g l 3
3
15
Glycerol (g l
12
15
Optical density at 600 nm
10
4
B
)
16
A
-1
12
Glycerol (g l
20
Optical density at 600 nm
5
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
from 2.4 to 0.14 g l31 . Approximately 1.8% of the reduced ammonium (94.1%) was lost by stripping, whereas 98.2% was assimilated by the cells after 80 h. When liquid manure was used as medium for growth, the portion of stripped ammonium was much higher than in modi¢ed medium E, regardless of whether co-substrates were added or not. Incubation of B. licheniformis strain S2 in liquid manure containing glycerol and citrate led to a decrease of ammonium from 2.8 to 0.11 g l31 , which corresponded to a total loss of 96.1%. From this 34.4% was due to stripping and 65.6% to bacterial growth. In contrast to this, cultivation of this strain in liquid manure supplemented with glucose resulted in less diminishing of ammonium than in liquid manure containing glycerol plus citrate or in modi¢ed medium E. It accounted for only 64.4% of the ammonium; from this 52% was due to stripping and 48% was assimilated during growth of B. licheniformis strain S2. Without co-substrate no signi¢cant bacterial growth occurred, and the decrease of the ammonium concentration (31.8%) was predominantly due to stripping. 3.4. Conversion of nitrogen into biomass and PGA As shown in Table 2, a signi¢cant portion of the total nitrogen of the employed media was converted into cellular biomass, but only a minor portion was converted into extracellular PGA. The concentration of total nitrogen in modi¢ed medium E was 1.9 g l31 at the beginning of cultivation, and after an incubation period of 80 h 0.75 g l31 nitrogen was detected as constituents of the resulting biomass, whereas 1.0 g l31 nitrogen was found in the supernatant (Fig. 3). However, the synthesized PGA contained only 0.018 g l31 nitrogen. After incubation in liquid manure containing glycerol plus citrate, 0.70 g l31 nitrogen of the total nitrogen content of 2.43 g l31 was converted into biomass, whereas only 0.002 g l31 nitrogen was detected in the synthesized PGA. When B. licheniformis strain S2 was grown on liquid manure with glucose as co-substrate, 0.7 g l31 of the total nitrogen of 2.8 g l31 was converted into biomass after 270 h incubation time, but no nitrogen occurred as PGA.
43
Fig. 3. Balance of the nitrogen after cultivation of B. licheniformis strain S2 in modi¢ed medium E, liquid manure with 8% (w/v) glycerol plus 1.87% (w/v) citrate, liquid manure with 2% glucose and liquid manure without co-substrate. LM: liquid manure ; white: stripped nitrogen ; vertical hatching: ammonium nitrogen in the supernatant, cross-hatching: other nitrogen in the supernatant, black : nitrogen in PGA; horizontal hatching: nitrogen in biomass. Concentrations of nitrogen: Medium E (modi¢ed) : 100% = 1.9 g l31 ; LM+citrate/glycerol: 100% = 2.5 g l31 ; LM+2% glucose : 100% = 2.75 g l31 ; LM: 100% = 2.77 g l31 .
4. Discussion One aim of this study was the conversion of ammonium nitrogen occurring in liquid manure into depot forms such as PAA or bacterial cell mass. Since di¡erent PGA-producing Bacillus strains, which are available in culture collections, were recently tested to produce PGA on swine manure [9], we were interested in new wild-type organisms to contribute more e¡ectively to the reduction of ammonium nitrogen of liquid manure. As a result a new PGAproducing strain of B. licheniformis was isolated from a saline. Strains of B. licheniformis are non-pathogenic to
Table 2 Cultivation of B. licheniformis strain S2 in medium E and particle-free liquid manure supplemented with di¡erent carbon sources Medium
Cell density and sedimentable compounds (g cell dry mass l31 )
PGA (g l31 ) late exp.
late stat.
Medium E (modi¢ed) Medium E (modi¢ed) (sterile control) Liquid manure+citrate/glycerol Liquid manure+citrate/glycerol (sterile control) Liquid manure+2% glucose Liquid manure 2% glucose (sterile control) Liquid manure Liquid manure (sterile control)
10.7 3 9.0 1.5* 12.3 1.6* 1.8 2.0*
0.3 3 0.2 3 3 3 3 3
018 3 0.02 3 3 3 3 3
Cell density and PGA content were determined as described in Section 2. exp., exponential growth phase; stat., stationary growth phase; 3, below detection limit ; *solid particle remaining in swine manure after incubation period.
FEMSLE 10772 20-1-03
44
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
humans and are widely distributed in the environment [10]. Typical habitats of this Gram-positive bacterium are soils, especially nutrient-poor soils such as moorland and deserts [10]. The highest PGA yield was achieved after cultivating B. licheniformis strain S2 in original medium E. With a view to application in agriculture and costs the manure should not be modi¢ed by addition of huge amounts of co-substrate. Therefore, we investigated whether the new isolate is able to diminish the ammonium concentration by producing cellular biopolymers and PGA under less optimal conditions related to agricultural practice. The ability of B. licheniformis strain S2 to grow rapidly in mineral salts medium with ammonium as sole nitrogen source indicated that ammonium is a suitable nitrogen source for this bacterium which is rapidly metabolized and allowed high growth rates. At high concentrations of NH3 (generally s 1 mM) ammonium is directly incorporated into L-glutamate by glutamate dehydrogenase [11]. This situation also occurs in Bacillus [12,13]. Therefore, the high concentration of ammonium in liquid manure (approximately 160 mM) in combination with an excess of carbon should enable a signi¢cant conversion of the ammonium nitrogen into amino acids and PGA. From the di¡erent growth experiments of this study, it can be concluded that a signi¢cant fraction of the ammonium is converted into biomass and to a lesser extent also into PGA. At carbon to nitrogen ratios of 20.5:1 and 15.5:1 as occurring in modi¢ed medium E and liquid manure containing citrate and glycerol, growth of B. licheniformis S2 diminished the total ammonium concentration by 92% or 63% respectively. After incubation a signi¢cant fraction of the total nitrogen that did not occur as ammonium was found in the supernatant. Therefore, it seems probable that in addition to the production of biomass and PGA a signi¢cant portion of the ammonium was used to produce extracellular nitrogen-containing compounds such as proteins, short peptides or amino acids. HPLC analysis of the cell-free supernatants of each culture (data not shown) revealed that after incubation of B. licheniformis strain S2 in modi¢ed medium E or in liquid manure supplemented with citrate and glycerol, compounds containing primary amino groups were synthesized in signi¢cant amounts. These compounds could not be exactly identi¢ed yet. Only when the modi¢ed medium E was used, extracellular proteins (1 g l31 ) were detected at signi¢cant concentrations. When the carbon to nitrogen ratio was only 2.5:1 (liquid manure with 2% glucose), the decrease of ammonium and the production of extracellular nitrogen-containing compounds was signi¢cantly lower than in liquid manure containing citrate and glycerol and exhibiting a carbon to nitrogen ratio of 15.5:1. With respect to this observation, it can be postulated that the production of extracellular nitrogen-containing compounds occurred only if carbon sources were provided in excess.
A higher proportion of ammonium was lost during aerobic treatment of swine waste (approximately 30%) in comparison to cultivation in modi¢ed medium E (1.7%) ; this was caused by non-biological processes (stripping) due to the higher pH of liquid manure as compared to the pH occurring in medium E. The process of ammonium stripping by increasing the pH is a method which is applied to remove the ammonium nitrogen as gaseous ammonia from swine waste [14,15]. However, in order to minimize the pollution of the environment due to emission of ammonia and to optimize the consumption of ammonium by bacterial growth, the pH of swine manure should be decreased before inoculation with the microorganism. With a view to application in agriculture further optimization must be done including medium speci¢cations, growth conditions, growth rate and polymer production rate. Only then will it be possible to determine the economic feasibility of such a process.
Acknowledgements This study was supported by a grant from Bayer AG (Leverkusen, Germany) which is gratefully acknowledged. The salty sample from Malta was kindly provided by Rainer Plaggenborg.
References [1] Archer, J.R. and Nicholson, R.J. (1992) Liquid wastes from farm animal enterprises. In: Farm Animals and the Environment (Phillips, C. and Piggins, D., Eds.), pp. 325^343. CAB International, Wallingford. [2] Troy, F.A. (1973) Chemistry and biosynthesis of the poly(Q-D-glutamyl) capsule in Bacillus licheniformis. I. Properties of the membranemediated biosynthesis reaction. J. Biol. Chem. 248, 306^315. [3] Hara, T. and Ueda, S. (1982) Regulation of polyglutamate production in Bacillus subtilis (natto): transformation of high Q-D-PGA productivity. Agric. Biol. Chem. 46, 2275^2281. [4] Oppermann-Sanio, F.B. and Steinbu«chel, A. (2002) Occurrence, functions and biosynthesis of polyamides in microorganism and biotechnical function. Naturwissenschaften 89, 11^22. [5] Leonard, C.G., Housewright, R.D. and Thorne, C.B. (1958) E¡ects of metallic ions on glutamyl polypeptide synthesis by Bacillus subtilis. J. Bacteriol. 76, 499^503. [6] Ausubel, F.M., Brent, R. Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. and Struhl, K. (1995) Current Protocols in Molecular Biology. John Wiley and Sons, New York. [7] Rainey, F.A., Ward-Rainey, N.L., Janssen, P.H., Hippe, H. and Stackebrandt, E. (1996) The genus Nocardiopsis represents a phylogenetically coherent taxon and a distinct actinomycete lineage: proposal of Nocardiaceae fam. nov. Int. J. Bacteriol. 46, 1088^1092. [8] Bergmeyer, H.U. (1974) Methoden der enzymatischen Analyse, 3. Au£age, pp. 1163^1168. Verlag Chemie, Weinheim. [9] Po«tter, M., Oppermann-Sanio, F.B. and Steinbu«chel, A. (2001) Cultivation of bacteria producing polyamino acids with liquid manure as carbon and nitrogen source. Appl. Environ. Microbiol. 67, 617^622. [10] De Boer, A.S., Priest, F. and Diderichsen, B. (1994) On the industrial use of Bacillus licheniformis : a review. Appl. Microbiol. Biotechnol. 40, 595^598.
FEMSLE 10772 20-1-03
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45 [11] Lengeler, J.W., Drews, G. and Schlegel, H.G. (1999) Biology of the Prokaryotes. Thieme-Verlag, Stuttgart. [12] Schreier, H., Smith, T.M. and Bernloh, R.W. (1982) Regulation of nitrogen catabolic enzymes in Bacillus spp. J. Bacteriol. 151, 971^975. [13] Kim, C.H. and Hollocher, T.C. (1982) 13 N isotope studies on the pathway of ammonia assimilation in Bacillus megaterium and Escherichia coli. J. Bacteriol. 151, 358^366.
45
[14] Beaudet, R., Gagnon, C., Bisaillon, J.G. and Ishaque, M. (1990) Microbiological aspects of aerobic thermophilic treatment of swine waste. Appl. Environ. Microbiol. 56, 971^976. [15] Rittstieg, K., Robra, K.H. and Somitsch, W. (2001) Aerobic treatment of a concentrated urea wastewater with simultaneous stripping of ammonium. Appl. Microbiol. Biotechnol. 56, 820^825.
FEMSLE 10772 20-1-03
www.fems-microbiology.org
Conversion of the nitrogen content in liquid manure into biomass and polyglutamic acid by a newly isolated strain of Bacillus licheniformis Astrid Hoppensack, Fred Bernd Oppermann-Sanio, Alexander Steinbu«chel
Institut fu«r Mikrobiologie der Westfa«lischen Wilhelms-Universita«t Mu«nster, CorrensstraMe 3, 48149 Mu«nster, Germany Received 6 August 2002; accepted 17 October 2002 First published online 29 November 2002
Abstract Extensive spreading of liquid manure onto agricultural fields causes eutrophication of ground and surface water and also pollution of the atmosphere due to the high ammonium nitrogen content. A poly(Q-glutamic acid) (PGA)-producing strain of Bacillus licheniformis was isolated in this study and investigated for its ability to reduce the ammonium nitrogen by converting ammonium into biomass and PGA as depot forms of nitrogen. In batch cultivations swine manure and an optimized mineral salts medium were used for PGA production. For example the cultivation of B. licheniformis strain S2 in liquid manure, which was modified by adding of 18 g citrate and 80 g glycerol l31 and exhibited a carbon to nitrogen ratio of 15.5:1, led to severe reduction of the ammonium content from 2.83 to 0.1 g l31 and to the production of 0.16 g PGA and 7.5 g cell dry mass l31 within 410 h. Approximately 28% (w/w) of the total nitrogen was converted into cellular biomass, whereas 0.1% (w/w) was used for the production of PGA. In addition, approximately 33% (w/v) of the original ammonium was lost by stripping. ; 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Liquid manure ; Biomass ; Polyamide; Bacillus; Nitrogen
1. Introduction Intensi¢ed animal breeding and growth or generation of products from animals have caused serious environmental problems in some geographic regions during the last two decades (European Communities website, http://europa. eu.int). Extensive spreading of liquid manure for example arising from large-scale pig fattening on agricultural ¢elds has resulted in pollution of ground water and the atmosphere due to the high concentrations of ammonium in this waste [1]. On the other hand, liquid manure provides an excellent source of nutrients for crops, if managed properly (European Communities website, http://europa. eu.int). Conversion of the nutrients into microbial biomass with liquid manure as substrate and conversion of the ammonium nitrogen into nitrogen-containing biopolymers such
* Corresponding author. Tel. : +49 (251) 8339821; Fax : +49 (251) 8338388. E-mail address : [email protected] (A. Steinbu«chel).
as peptidoglycan, nucleic acids, proteins and polyamino acids (PAAs) would provide transient depots for ammonium and also other nutrients occurring in swine waste. This microbial biomass and the biopolymers could exhibit a fertilizing function for a prolonged period during decay. Moreover, knowledge obtained on the handling of manure from pigs may be also applicable to manures from other animal sources. Poly(Q-D-glutamic acid) (PGA) is a naturally occurring polyanionic, water-soluble extracellular polyamide [2] with potential applications as a biodegradable thickener, humectant, sustained-release material or drug carrier in food, cosmetics or medicine. It is also an important component of some Asian foods such as Natto [3]. It is one of the few naturally occurring polyamides which are not synthesized by ribosomal protein biosynthesis [4]. Due to the remarkably resistance of PGA against proteolytic attack, degradation of PGA by a non-adapted micro£ora is expected to be very slow. This should result in a continuously low rate of release of free ammonium. In this study, we investigated the conversion of nitrogen into biomass and PGA by batch fermentation of a newly
0378-1097 / 02 / $22.00 ; 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII : S 0 3 7 8 - 1 0 9 7 ( 0 2 ) 0 1 1 3 2 - 1
FEMSLE 10772 20-1-03
40
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
isolated PGA-producing strain of Bacillus licheniformis in liquid manure.
2. Materials and methods 2.1. Growth of bacteria Isolate S2 was cultivated in medium E [5] without glutamate as well as in particle-free liquid manure and in particle-free liquid manure supplemented with di¡erent carbon sources. The chemical composition of the liquid manure was determined at the Landwirtschaftliche Untersuchungs- und Forschungsanstalt (Oldenburg, Germany) and was as follows: 0.46% (w/w) nitrogen, 0.32% (w/w) ammonium N, 0.29% (w/w) P2 O5 , 0.26% (w/w) K2 O, 0.09% (w/w) MgO, 10 mg Cu kg31 , 1.9% (w/w) C, and 4.6% (w/w) dry matter. The liquid manure as received was stored at 330‡C, and it was centrifuged for 30 min at 3000Ug and autoclaved to obtain almost particle-free and sterile liquid manure before use. Carbon sources were sterilized separately by autoclaving before they were added to the manure. All cultures were inoculated with cells obtained from a 3-day-old preculture in liquid manure containing 2% glucose, and the cells were washed twice with sterile saline and concentrated three-fold before inoculation. Batch cultivations were carried out at 30‡C and at a stirring speed of 120 rpm in 2-l conical £asks equipped with three ba¥es containing 200 ml medium. Growth was monitored photometrically at 600 nm.
Glucose was determined photometrically according to the method described by Bergmeyer et al. [8]. For photometric quanti¢cation of glycerol and citrate, respective analysis kits (#148270, #139076) purchased from Boehringer Mannheim were used. Photometric determination of nitrogen was carried out using the nitrogen determination kit (#83) from Macherey-Nagel. Concentrations of NHþ 4 were determined as ammonia (NH3 ) with a type 15 230 3000 gas-sensitive electrode (Mettler Toledo, Steinbach, Germany; detection limit: 1036 mol l31 ) according to the instructions provided by the manufacturer. The amount of isolated PAAs was analyzed after precolumn derivatization of PAA hydrolysates with orthophthaldialdehyde reagent on a reversed-phase high-performance liquid chromatography (HPLC) column as recommended by the manufacturer (Merck, Darmstadt, Germany). Hydrolysis was carried out by adding 1 ml of 6 N HCl to 1 mg of lyophilized PAAs and incubating the mixture for 8 h at 100‡C. 2.5. Nucleotide sequence accession number The 16S rDNA gene nucleotide sequence of isolate strain S2 was submitted to the EMBL nucleotide sequence database and is available under accession number AY052767.
3. Results 3.1. Isolation and taxonomic classi¢cation of the new isolate strain S2
2.2. Analysis of 16S rDNA Extraction of chromosomal DNA was carried out by the method of Ausubel et al. [6]. The 16S rDNA gene (rDNA) was ampli¢ed using oligonucleotide primers as described previously [7]. The nucleotide sequence was determined by DNA sequencing according to the chain termination method using the 4000L automatic sequencer LiCor (MWG-Biotech, Ebersberg, Germany). The 16S rDNA sequence was aligned with published sequences from the NCBI database using the Clustal X software. 2.3. Isolation of PGA and cell dry mass analysis PGA was isolated and puri¢ed by the method reported previously [2]. For analysis of cell dry mass, cells were harvested (60 min, 3000Ug, 4‡C) and washed once with 0.9 N NaCl. The obtained pellet was lyophilized, and the weight was determined. 2.4. Analytical methods In the cell-free supernatants the concentrations of glucose, glycerol, citrate, and ammonium were analyzed at di¡erent times during the time course of the incubation.
The microorganism used in this study was strain S2 which was isolated from a salty sea sludge sample from the island of Malta. Approximately 1 ml of the salty sea sludge sample was used to inoculate 20 ml of Luria^Bertani medium containing 12.5% NaCl in order to obtain PGA-producing bacteria. The culture was incubated aerobically at 30‡C for 3 days. Aliquots of this culture were plated on solid medium E, and a visible viscose colony was identi¢ed as a potential plasmid-free producer of PGA, which consists of 2.2% L-glutamate and 97.8% D-glutamate. The morphological and physiological characteristics of strain S2 are summarized in Table 1. For phylogenetic a⁄liation of this strain the complete 16S rDNA gene was ampli¢ed by polymerase chain reaction and its nucleotide sequence revealed as described in Section 2. Comparison of the obtained sequence comprising 1547 nucleotides with 16S rDNA sequences available from the NCBI database revealed the greatest similarity to the corresponding sequence of B. licheniformis DSM13 (99.4%). The pro¢le of the cellular fatty acids of strain S2 is typical of the Bacillus subtilis group, to which B. licheniformis also belongs (15:0 iso, 22.45% ; 15:0 aniso, 42.59% ; 16:0 iso, 4.49%, 16:0 aniso, 5.36%; 17:0 iso, 8.75%; 17:0 aniso,
FEMSLE 10772 20-1-03
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
16.37%; deviation for B. licheniformis : 0.522; deviation for B. subtilis: 0.572). Furthermore, the 16S rDNA sequence of strain S2 and the other members of the B. subtilis group as well as Bacillus anthracis were aligned, and a phylogenetic tree was constructed based on the neighborjoining method using the Clustal X program (version 1.8; Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. and Higgins, D.G., 1997). A bootstrap analysis was carried out, and the tree (Fig. 1) indicates that strain S2 and B. licheniformis DSM13 are the most closely related strains. Strain S2 will therefore be tentatively referred to as B. licheniformis strain S2.
41
B. vallismortis DSM 11031
B. amyloliquefaciens ATCC23350
B. atrophaeus JCM 9070
B. licheniformis strain S2 681 1000
B. licheniformis DSM 13 710
B. pumilus NCDO 1766
3.2. Growth kinetics of B. licheniformis strain S2
690
984
B. anthracis 1392
To enable a continuous measurement of growth parameters, almost particle-free liquid manure was used in this study. Because B. licheniformis strain S2 was not able to grow on particle-free liquid manure without an additional carbon source (Fig. 2D), the in£uence of 2% (w/v) glucose (Fig. 2C) or of 8% (w/v) glycerol plus 1.87% (w/v) citrate (Fig. 2B) on growth and PGA production was investigated in batch cultivations. With respect to reducing the concentration of nitrogen in liquid manure, we investigated if PGA production occurred in medium without glutamate as precursor, because glutamate contains nitrogen and is therefore not a suitable co-substrate in liquid manure. In medium E containing no glutamate, the speci¢c growth rate (W) of B. licheniformis strain S2 was 0.12 h31 , and the concentrations of PGA were 0.26 g l31 in the middle of the exponential growth phase and 0.18 g l31 at the end
Table 1 Morphological and physiological characteristics of B. licheniformis strain S2 Cell form Cell size Spores Sporangium swollen Gram reaction Catalase Oxidase Anaerobic growth VP reaction pH in VP medium Temperature for growth at 50‡C at 55‡C in medium pH 5.7 in NaCl 2% in NaCl 5%
rods 0.7^0.9U2.0^3.5 Wm +, oval 3 positive + 3 + + 5.4
Acid from D-Glucose L-Arabinose D-Xylose D-Mannitol D-Fructose Hydrolysis of Starch Gelatin Casein Tween 20
+ 3 + + +
in NaCl 10% Lysozyme medium Esculin hydrolysis Gas from glucose Nitrate reduction
+ 3 + 3 +
Utilization of Citrate Propionate Tyrosine hydrolase Phenylalanine deaminase Resistant against Ampicillin Chloramphenicol Tetracycline Kanamycin
+ + + + + + + + +
+ + 3 3
+ + 3 3
B. mojavensis IFO 15718 744
B. subtilis NCDO 1769
0.01
Fig. 1. Bootstrap neighbor-joining phylogenetic tree of B. licheniformis strain S2 and related bacteria of the B. subtilis group and B. anthracis based on 16S rRNA sequence comparisons. The scale bar indicates 1% sequence dissimilarity ; bootstrap values (1000 replicates) are mentioned at the nodes. Accession numbers for the reference sequences used are: B. amyloliquefaciens (X60605) ; B. anthracis (X55059); B. atrophaeus (AB021181); B. licheniformis (X68416) ; B. mojavensis (AB021191); B. pumilus (X60637) ; B. subtilis (X60646); B. vallismortis (AB021198).
of the incubation after 80 h. When this strain was incubated in swine manure, a relatively long lag phase was needed for adaptation. Slower growth (W = 0.1 h31 ) and less PGA (0.16 g l31 at the end of the exponential growth phase and 0.02 g l31 at the end of incubation) were obtained in liquid manure supplemented with glycerol and citrate. Faster growth (W = 0.20 h31 ) was achieved in liquid manure containing glucose. At the beginning of the stationary growth phase the PGA concentration was 0.014 g l31 , whereas after a further 10 h incubation it was 0.26 g l31 and after a further 26 h incubation no PGA was detected. The highest cell densities (10.7 g cell dry mass l31 ) were obtained in modi¢ed medium E and in liquid manure with glucose, whereas in liquid manure supplemented with glycerol and citrate the cell density was 7.5 g l31 . In contrast to growth in modi¢ed medium E, B. licheniformis strain S2 needed an extremely long lag phase of approximately 220 h and 340 h for adaptation in liquid manure supplemented with glucose or with glycerol plus citrate. 3.3. Diminishing of ammonium During incubation of B. licheniformis strain S2 in modi¢ed medium E, the ammonium concentration decreased
FEMSLE 10772 20-1-03
42
100
5
80
4
20
12
100
10
10
80
8
8
6
6
4
4
2
2
12
14
6 4 4
5 1
2
-1
-1
)
)
) 40
2
10
pH
8
6
3
Citrate (g l
pH
10
60
NH (g l 3
2
10
-1
-1
)
8
5 20
1
0
0
60
40
20
2 0
0
0
0
0
20
40
60
80
0
0
100
320
340
Incubation time (h)
FEMSLE 10772 20-1-03
5
12
0
0 300
360
380
400
420
Incubation time (h)
16
C
25
5
20
4
12
10
D
2
6 4 4
1
2
10
8
)
8 3
6 6
2
4 4
5
1
0
0
Optical density at 600 nm
8
6
15
10
pH
pH
NH (g l 3
-1
)
10
-1
8 3
NH (g l 3
12
Optical density at 600 nm
10
4
-1 Glucose (g l )
14
2
2
2 0
0
0 160
180
200
220
240
Incubation time (h)
260
280
0
0 0
100
200
300
400
500
Incubation time (h)
Fig. 2. Incubation of B. licheniformis strain S2 in modi¢ed medium E (A), liquid manure with 8% (w/v) glycerol plus 1.87% (w/v) citrate (B), liquid manure with 2% glucose (C) and liquid manure without co-substrate (D). The cultivations were carried out at 30‡C in 2-l conical £asks equipped with three ba¥es containing 200 ml of the corresponding medium. Symbols : F, optical density; E, citrate ; S, glucose; O, pH; b, NH3 ; a, glycerol. At times indicated by arrows, 50 ml of the culture was removed for isolation of PGA.
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
Citrate (g l
) -1 NH (g l 3
3
15
Glycerol (g l
12
15
Optical density at 600 nm
10
4
B
)
16
A
-1
12
Glycerol (g l
20
Optical density at 600 nm
5
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
from 2.4 to 0.14 g l31 . Approximately 1.8% of the reduced ammonium (94.1%) was lost by stripping, whereas 98.2% was assimilated by the cells after 80 h. When liquid manure was used as medium for growth, the portion of stripped ammonium was much higher than in modi¢ed medium E, regardless of whether co-substrates were added or not. Incubation of B. licheniformis strain S2 in liquid manure containing glycerol and citrate led to a decrease of ammonium from 2.8 to 0.11 g l31 , which corresponded to a total loss of 96.1%. From this 34.4% was due to stripping and 65.6% to bacterial growth. In contrast to this, cultivation of this strain in liquid manure supplemented with glucose resulted in less diminishing of ammonium than in liquid manure containing glycerol plus citrate or in modi¢ed medium E. It accounted for only 64.4% of the ammonium; from this 52% was due to stripping and 48% was assimilated during growth of B. licheniformis strain S2. Without co-substrate no signi¢cant bacterial growth occurred, and the decrease of the ammonium concentration (31.8%) was predominantly due to stripping. 3.4. Conversion of nitrogen into biomass and PGA As shown in Table 2, a signi¢cant portion of the total nitrogen of the employed media was converted into cellular biomass, but only a minor portion was converted into extracellular PGA. The concentration of total nitrogen in modi¢ed medium E was 1.9 g l31 at the beginning of cultivation, and after an incubation period of 80 h 0.75 g l31 nitrogen was detected as constituents of the resulting biomass, whereas 1.0 g l31 nitrogen was found in the supernatant (Fig. 3). However, the synthesized PGA contained only 0.018 g l31 nitrogen. After incubation in liquid manure containing glycerol plus citrate, 0.70 g l31 nitrogen of the total nitrogen content of 2.43 g l31 was converted into biomass, whereas only 0.002 g l31 nitrogen was detected in the synthesized PGA. When B. licheniformis strain S2 was grown on liquid manure with glucose as co-substrate, 0.7 g l31 of the total nitrogen of 2.8 g l31 was converted into biomass after 270 h incubation time, but no nitrogen occurred as PGA.
43
Fig. 3. Balance of the nitrogen after cultivation of B. licheniformis strain S2 in modi¢ed medium E, liquid manure with 8% (w/v) glycerol plus 1.87% (w/v) citrate, liquid manure with 2% glucose and liquid manure without co-substrate. LM: liquid manure ; white: stripped nitrogen ; vertical hatching: ammonium nitrogen in the supernatant, cross-hatching: other nitrogen in the supernatant, black : nitrogen in PGA; horizontal hatching: nitrogen in biomass. Concentrations of nitrogen: Medium E (modi¢ed) : 100% = 1.9 g l31 ; LM+citrate/glycerol: 100% = 2.5 g l31 ; LM+2% glucose : 100% = 2.75 g l31 ; LM: 100% = 2.77 g l31 .
4. Discussion One aim of this study was the conversion of ammonium nitrogen occurring in liquid manure into depot forms such as PAA or bacterial cell mass. Since di¡erent PGA-producing Bacillus strains, which are available in culture collections, were recently tested to produce PGA on swine manure [9], we were interested in new wild-type organisms to contribute more e¡ectively to the reduction of ammonium nitrogen of liquid manure. As a result a new PGAproducing strain of B. licheniformis was isolated from a saline. Strains of B. licheniformis are non-pathogenic to
Table 2 Cultivation of B. licheniformis strain S2 in medium E and particle-free liquid manure supplemented with di¡erent carbon sources Medium
Cell density and sedimentable compounds (g cell dry mass l31 )
PGA (g l31 ) late exp.
late stat.
Medium E (modi¢ed) Medium E (modi¢ed) (sterile control) Liquid manure+citrate/glycerol Liquid manure+citrate/glycerol (sterile control) Liquid manure+2% glucose Liquid manure 2% glucose (sterile control) Liquid manure Liquid manure (sterile control)
10.7 3 9.0 1.5* 12.3 1.6* 1.8 2.0*
0.3 3 0.2 3 3 3 3 3
018 3 0.02 3 3 3 3 3
Cell density and PGA content were determined as described in Section 2. exp., exponential growth phase; stat., stationary growth phase; 3, below detection limit ; *solid particle remaining in swine manure after incubation period.
FEMSLE 10772 20-1-03
44
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45
humans and are widely distributed in the environment [10]. Typical habitats of this Gram-positive bacterium are soils, especially nutrient-poor soils such as moorland and deserts [10]. The highest PGA yield was achieved after cultivating B. licheniformis strain S2 in original medium E. With a view to application in agriculture and costs the manure should not be modi¢ed by addition of huge amounts of co-substrate. Therefore, we investigated whether the new isolate is able to diminish the ammonium concentration by producing cellular biopolymers and PGA under less optimal conditions related to agricultural practice. The ability of B. licheniformis strain S2 to grow rapidly in mineral salts medium with ammonium as sole nitrogen source indicated that ammonium is a suitable nitrogen source for this bacterium which is rapidly metabolized and allowed high growth rates. At high concentrations of NH3 (generally s 1 mM) ammonium is directly incorporated into L-glutamate by glutamate dehydrogenase [11]. This situation also occurs in Bacillus [12,13]. Therefore, the high concentration of ammonium in liquid manure (approximately 160 mM) in combination with an excess of carbon should enable a signi¢cant conversion of the ammonium nitrogen into amino acids and PGA. From the di¡erent growth experiments of this study, it can be concluded that a signi¢cant fraction of the ammonium is converted into biomass and to a lesser extent also into PGA. At carbon to nitrogen ratios of 20.5:1 and 15.5:1 as occurring in modi¢ed medium E and liquid manure containing citrate and glycerol, growth of B. licheniformis S2 diminished the total ammonium concentration by 92% or 63% respectively. After incubation a signi¢cant fraction of the total nitrogen that did not occur as ammonium was found in the supernatant. Therefore, it seems probable that in addition to the production of biomass and PGA a signi¢cant portion of the ammonium was used to produce extracellular nitrogen-containing compounds such as proteins, short peptides or amino acids. HPLC analysis of the cell-free supernatants of each culture (data not shown) revealed that after incubation of B. licheniformis strain S2 in modi¢ed medium E or in liquid manure supplemented with citrate and glycerol, compounds containing primary amino groups were synthesized in signi¢cant amounts. These compounds could not be exactly identi¢ed yet. Only when the modi¢ed medium E was used, extracellular proteins (1 g l31 ) were detected at signi¢cant concentrations. When the carbon to nitrogen ratio was only 2.5:1 (liquid manure with 2% glucose), the decrease of ammonium and the production of extracellular nitrogen-containing compounds was signi¢cantly lower than in liquid manure containing citrate and glycerol and exhibiting a carbon to nitrogen ratio of 15.5:1. With respect to this observation, it can be postulated that the production of extracellular nitrogen-containing compounds occurred only if carbon sources were provided in excess.
A higher proportion of ammonium was lost during aerobic treatment of swine waste (approximately 30%) in comparison to cultivation in modi¢ed medium E (1.7%) ; this was caused by non-biological processes (stripping) due to the higher pH of liquid manure as compared to the pH occurring in medium E. The process of ammonium stripping by increasing the pH is a method which is applied to remove the ammonium nitrogen as gaseous ammonia from swine waste [14,15]. However, in order to minimize the pollution of the environment due to emission of ammonia and to optimize the consumption of ammonium by bacterial growth, the pH of swine manure should be decreased before inoculation with the microorganism. With a view to application in agriculture further optimization must be done including medium speci¢cations, growth conditions, growth rate and polymer production rate. Only then will it be possible to determine the economic feasibility of such a process.
Acknowledgements This study was supported by a grant from Bayer AG (Leverkusen, Germany) which is gratefully acknowledged. The salty sample from Malta was kindly provided by Rainer Plaggenborg.
References [1] Archer, J.R. and Nicholson, R.J. (1992) Liquid wastes from farm animal enterprises. In: Farm Animals and the Environment (Phillips, C. and Piggins, D., Eds.), pp. 325^343. CAB International, Wallingford. [2] Troy, F.A. (1973) Chemistry and biosynthesis of the poly(Q-D-glutamyl) capsule in Bacillus licheniformis. I. Properties of the membranemediated biosynthesis reaction. J. Biol. Chem. 248, 306^315. [3] Hara, T. and Ueda, S. (1982) Regulation of polyglutamate production in Bacillus subtilis (natto): transformation of high Q-D-PGA productivity. Agric. Biol. Chem. 46, 2275^2281. [4] Oppermann-Sanio, F.B. and Steinbu«chel, A. (2002) Occurrence, functions and biosynthesis of polyamides in microorganism and biotechnical function. Naturwissenschaften 89, 11^22. [5] Leonard, C.G., Housewright, R.D. and Thorne, C.B. (1958) E¡ects of metallic ions on glutamyl polypeptide synthesis by Bacillus subtilis. J. Bacteriol. 76, 499^503. [6] Ausubel, F.M., Brent, R. Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. and Struhl, K. (1995) Current Protocols in Molecular Biology. John Wiley and Sons, New York. [7] Rainey, F.A., Ward-Rainey, N.L., Janssen, P.H., Hippe, H. and Stackebrandt, E. (1996) The genus Nocardiopsis represents a phylogenetically coherent taxon and a distinct actinomycete lineage: proposal of Nocardiaceae fam. nov. Int. J. Bacteriol. 46, 1088^1092. [8] Bergmeyer, H.U. (1974) Methoden der enzymatischen Analyse, 3. Au£age, pp. 1163^1168. Verlag Chemie, Weinheim. [9] Po«tter, M., Oppermann-Sanio, F.B. and Steinbu«chel, A. (2001) Cultivation of bacteria producing polyamino acids with liquid manure as carbon and nitrogen source. Appl. Environ. Microbiol. 67, 617^622. [10] De Boer, A.S., Priest, F. and Diderichsen, B. (1994) On the industrial use of Bacillus licheniformis : a review. Appl. Microbiol. Biotechnol. 40, 595^598.
FEMSLE 10772 20-1-03
A. Hoppensack et al. / FEMS Microbiology Letters 218 (2003) 39^45 [11] Lengeler, J.W., Drews, G. and Schlegel, H.G. (1999) Biology of the Prokaryotes. Thieme-Verlag, Stuttgart. [12] Schreier, H., Smith, T.M. and Bernloh, R.W. (1982) Regulation of nitrogen catabolic enzymes in Bacillus spp. J. Bacteriol. 151, 971^975. [13] Kim, C.H. and Hollocher, T.C. (1982) 13 N isotope studies on the pathway of ammonia assimilation in Bacillus megaterium and Escherichia coli. J. Bacteriol. 151, 358^366.
45
[14] Beaudet, R., Gagnon, C., Bisaillon, J.G. and Ishaque, M. (1990) Microbiological aspects of aerobic thermophilic treatment of swine waste. Appl. Environ. Microbiol. 56, 971^976. [15] Rittstieg, K., Robra, K.H. and Somitsch, W. (2001) Aerobic treatment of a concentrated urea wastewater with simultaneous stripping of ammonium. Appl. Microbiol. Biotechnol. 56, 820^825.
FEMSLE 10772 20-1-03