Improvement of a synthetic medium for Dictyostelium discoideum

Process Biochemistry 39 (2004) 925–930

Improvement of a synthetic medium for Dictyostelium discoideum Sang-In Han, Karl Friehs, Erwin Flaschel∗ Bielefeld University, Faculty of Technology, D-33594 Bielefeld, Germany Received 31 January 2003; received in revised form 7 April 2003; accepted 6 June 2003

Abstract Dictyostelium discoideum is of considerable interest as an expression system for the production of proteins of high value. The cultivation of this social amoeba is not as easy as that of other common microbial expression systems. Wildtype strains grow on bacteria. Mutant strains growing on axenic media reach cell densities of 1–2 × 107 ml−1 when cultivated in commonly used complex media. A totally synthetic medium formulated by Franke and Kessin (Proc. Natl. Acad. Sci. USA 74 (1977) 2157) has become popular and allows cell densities of about 3 × 107 ml−1 to be obtained. This medium (FM) is being improved mainly on the basis of the analysis of limitations with respect to amino acids. With this improved synthetic medium (SIH) cell densities in the order of 5–6 × 107 ml−1 have been achieved. © 2003 Elsevier Ltd. All rights reserved. Keywords: Dictyostelium discoideum; Axenic media; Synthetic medium; Medium design; Batch cultivation; Shake-flask cultivation

1. Introduction The social amoeba Dictyostelium discoideum is a promising organism for the production of recombinant pharmaceutical proteins requiring post-translational modifications [2–5]. However, it’s cultivation characteristics do not compare well with other common microbial expression systems. Wildtype D. discoideum is adapted to live on soil feeding bacteria. Growth rates are acceptable with doubling times of 3 h, but the presence of bacteria would certainly not be accepted for the production of pharmaceuticals. Mutant strains are known that feed on axenic (soluble) media by pinocytosis. Depending on the composition of axenic media cell densities of 1 – 2 × 107 ml−1 may commonly be expected. With the exception of one medium all these axenic media are of complex composition. The ingredients of these complex axenic media are listed in Table 1. The most popular complex axenic medium seems to be HL-5 owing to problems with the peptone fraction of other media. The authors of this article prefer HL-5C (personal communication of C. Reymond, University of Lausanne, Switzerland) due to reproducible results and even less dependence on specific complex ingredients. The only synthetic medium for D. discoideum was ∗ Corresponding author. Tel.: +49-521-106-5301; fax: +49-521-106-6475. E-mail address: [email protected] (E. Flaschel).

0032-9592/$ – see front matter © 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0032-9592(03)00215-2

developed by Franke and Kessin [1]. Based on the composition of medium HL-5 it is currently widely applied, since higher maximal cell densities up to 3 × 107 ml−1 , may be achieved. The composition of this medium is listed in Table 2 together with the original analytical results for HL-5 of Franke and Kessin [1]. On the basis of measurements of the amino acid consumption a number of concentrations of amino acids have been altered to remove potential bottlenecks in the utilisation of this long list of nutrients. Shake-flask cultures are presented showing the difference between the classical FM medium with the newly developed SIH medium.

2. Materials and methods 2.1. Chemical substances and media components Yeast extract and bacto tryptone were purchased from Difco and casein peptone and d-glucose from Merck (Darmstadt, Germany). Asparagine was obtained from ICN, histidine from Senn Chemicals and other amino acids from Ajinomoto. Vitamins were purchased from Fluka except folic acid which was obtained from Serva. Dihydroxystreptomycin sulphate was purchased from Fluka and Geneticin (G-418) from Serva. All other chemicals were at least of analytical grade.

926

S.-I. Han et al. / Process Biochemistry 39 (2004) 925–930 Table 2 Composition of synthetic media for D. discoideum

Nomenclature

t µ ρ τ

molar concentration (mmol l−1 ) factor for calculating substrate profiles from growth profiles (g l−1 ) factor for calculating product profiles from growth profiles (g l−1 ) apparent Monod constant (g l−1 ) cell (number) density (ml−1 ) saturation parameter of the limiting substrate (–) concentration of the (limiting) substrate (g l−1 ) time (of cultivation) (h) or (d) specific growth rate (h−1 ) mass concentration (g l−1 ) time constant of growth (h)

Indices 0 aa Amm Glc max

referring to initial conditions amino acid ammonia (and/or ammonium ion) glucose maximal-, maximum-

c fSN fPN Ks N s0S S

Substances

Media composition

Common names Literature

HL-5a [1]

FM [1]

SIH This article

Glucose (mmol l−1 ) Amino acids (mmol l−1 ) l-Arginine l-Asparagine l-Aspartic acid l-Cysteine·HCl Glycine l-Glutamic acid l-Histidine l-Isoleucine l-Leucine l-Lysine·HCl l-Methionine l-Phenylalanine l-Proline l-Threonine l-Tyrosine l-Tryptophan l-Valine

> 56

56

56

4.4 – 6.9 0.58 16.6 9.7 1.6 4.3 7.2 5.9 1.9 3.1 > 6.3 4.3 2.5 0.52 5.9

3.3 2.3 – 1.7 12.0 3.4 1.4 4.6 6.9 4.9 2.0 3.0 7.0 4.2 – 1.0 6.0

3.3 2.3 1.1 2.5 12.0 3.7 1.4 4.6 6.9 8.5 2.3 3.3 7.0 4.2 – 1.7 6.0

Vitamins (mg l−1 ) Biotin Cyanocobalamin Folic acid Lipoic acid Riboflavin Thiamine·HCl Calcium pantothenate Pyridoxine·HCl Choline chloride Nicotinic acid p-Aminobenzoic acid

Abbreviations Amm ammonia (and/or ammonium ion) Glc Glucose

2.2. Media preparation The composition of the complex axenic medium HL-5C is given in Table 1. The pH was adjusted to 6.3 prior to autoclaving the medium for 20 min at 121 ◦ C. Water was obtained from a purification device Seralpur Pro 90CN.

Table 1 Composition of common complex axenic media for D. discoideum Ingredients

Media composition (g l−1 )

Common names Literature

A [6]

AX [7]

HL-5 [8]

HL-5C [9]

Yeast extract Proteose peptone Bacto peptone Casein peptone Thiotone Bacto–tryptone Glucose Maltose Na2 HPO4 KH2 PO4 K2 HPO4 ·12H2 O MgSO4 ·7H2 O pH

0.5 – 5 – – – 5 – – 2.25 1.5 0.5 6.3

7.15 14.3 – – – – – 18 0.49 0.49 – – 6.7

5 5 – – 5 – 10 – 0.6 0.34 – – 6.5–6.7

5 5 – 2.5 – 2.5 10 – 0.35 1.2 – – 6.3

0.023 0.005 0.12 – 0.4 0.53 0.59 0.18 30.0 3.7 0.13

0.020 0.005 0.20 0.4 0.5 0.6 – – – – –

0.020 0.005 0.20 0.4 0.5 0.6 – – – – –

Phosphate salts (mmol l−1 ) K2 HPO4 KH2 PO4 NaH2 PO4

5.0 – –

5.0 – –

– 8.82 2.47

Salts (mmol l−1 ) NaOH NaCl NaHCO3 NH4 Cl CaCl2 FeCl3 MgCl2

7.0 6.0 – – 0.31 0.20 0.49

2.0 – 0.2 1.0 0.02 0.10 0.40

– – 0.2 1.0 0.02 0.10 0.40

13 1.8 0.7 0.6 0.08 2.6 8

13 1.8 0.7 0.6 0.08 2.6 8

Trace elements (␮mol l−1 ) Na2 EDTA H3 BO3 CoCl2 CuSO4 (NH4 )6 Mo7 O24 MnCl2 ZnSO4 a

Analysed after hydrolysis.

– – – 6.4 – 1.4 13

S.-I. Han et al. / Process Biochemistry 39 (2004) 925–930

FM and SIH medium were prepared on the basis of four solutions containing either amino acids, vitamins, salts, or trace elements. The composition of these solutions is given in Table 2. They may be stored at −20 ◦ C for convenience. The amino acid solution was prepared four times concentrated, the vitamin solution 20 times concentrated, the salt solution 50 times concentrated, and the trace element solution 10 000 times concentrated. Two hundred and fifty millilitres of the amino acid solution, 50 ml of the vitamin solution, 20 ml of the salt solution and 0.1 ml of the trace element solution together with 10 g glucose, phosphate salts and antibiotics (50 mg dihydroxystreptomycin sulphate and 5 mg geneticin) were filled up to 500 ml with water. The pH was adjusted to 6.5 by adding dilute HCl or NaOH solutions. Finally, the volume of the medium was brought to 1 l by adding water. The medium was sterilised by filtering through a membrane filter system Sartobran 300 (Sartorius, Göttingen, Germany). The sterile medium can be stored at 4 ◦ C for about 4 weeks. 2.3. Strain and cultivation systems Experiments are being performed with the axenic strain AX2-CR II, which was kindly provided by Reymond [10,11]. This strain expresses a fusion protein of the circumsporozoite protein of Plasmodium falciparum under the control of the actin 6-promoter. D. discoideum was cultivated in Erlenmeyer-shake flasks on a rotary shaker with an eccentricity of 25 mm and a rotational frequency of 150 min−1 at a temperature of 21 ◦ C. Unless stated otherwise shake flasks of 500 ml were used filled with 50 ml medium. The cultivation experiments were started with an inoculation cell density of about 105 ml−1 . 2.4. Analysis of the cultivation system Cell density was determined by counting cells either in a Neubauer chamber under a phase-contrast microscope (Nikon Optiphot 2) or with an automatic counter from Schärfe System (CASY® 1). Dilutions for counting were prepared in 0.017 M Soerensen phosphate buffer of pH 6.0. Glucose was determined enzymically using an assay kit (716251) according to the instructions of the manufacturer Böhringer Mannheim (Mannheim, Germany). The ammonia concentration (the total of both species ammonia and ammonium ions) was analysed with an ammonia electrode (WTW, Germany). Samples were prepared by centrifugation (3000 min−1 ) for 5 min and the supernatant was centrifuged again (7000 min−1 ) for 5 min. A centrifuge Eppendorf 5415 D was used. Detection of the ammonia concentration was carried out by mixing 1 ml of the sample with 10 ml water and 0.15 ml 1 M NaOH solution. Amino acid analysis was performed according to the method described by Büntemeyer et al. [12] by means of HPLC on a RP18 column. Prior to the injection the samples were derivatized with o-phthaldialdehyde followed by

927

fluorescence detection. Prior to derivatization the samples were treated with a protein precipitation reagent.

3. Results and discussion D. discoideum has to be adapted for growth on synthetic media. For this purpose cells from shake-flask cultures are collected by centrifugation, washed with fresh medium and re-inoculated into fresh medium at a cell density of about 105 ml−1 . This was repeated by gradually replacing the complex medium HL-5C against synthetic medium FM. Using this strategy the axenic strain AX2-CR II was adapted to FM medium in 3 weeks. The starting point of medium development was FM medium. Typical results of shake-flask cultures are shown in Fig. 1. Shake flasks of different size but the same relative working volume were taken for cultivation. After 7 d a cell density of almost 3 × 107 ml−1 was achieved. This is almost twice the cell density, which can commonly be achieved by cultivation on complex media like HL-5C [13]. Growth was slow and the solid line (Fig. 1) represents the logistic curve, based on the balance equation for batch cultivation with first order growth with respect to a (virtual) limiting substrate [13]: φ=

φ0


 φ0 + (1 − φ0 ) exp − (1 −1φ0 ) τt

(1)

with: φ=

N Nmax

φ0 =

(2a)

N0 Nmax

(2b)

S0 Ks

(2c)

and s0S =

Fig. 1. Cultivation of D. discoideum in shake-flask cultures on FM medium with shake flasks of different size. The working volumes amounted to 10% of the total flask volumes.

928

S.-I. Han et al. / Process Biochemistry 39 (2004) 925–930

with φ0 the relative cell density at the time of inoculation and s0S the saturation parameter of the limiting substrate according to a Monod kinetics and with the characteristic time constant of growth τ=

1

(3)

µmax s0S

with the maximum specific growth rate (µmax ) according to Monod kinetics. The growth curve in Fig. 1 is characterized by the following parameters: Nmax = 2.9 × 107 ml−1 , φ0 = 0.0023, and τ = 17.9 h. This leads to an initial specific growth rate of µ0 = 1/τ = 0.056 h−1 . A semi-logarithmic plot shows a sigmoidal function. Linear regression in the range of cultivation times from 20 to 100 h yields a specific growth rate of 0.064 h−1 —with little significance for the whole curve. These values compare with time constants in the range of 9–12 h for growth on HL-5C [13]. However, slower growth is compensated by much higher cell densities. The higher maximal cell density was accompanied with a much better utilisation of glucose compared to the situation for complex media. Almost 80% of the initial glucose was utilised after 7 d of cultivation. The profile of the glucose concentration (ρGlc ) as given by the broken line in Fig. 1 was modelled after a balance equation directly derived from Eq. (1): ρGlc = ρGlc,0 − fSN

φ0


φ0 + (1 − φ0 ) exp − (1 −1φ0 ) τt

(4)

Ammonia evolved in parallel to the growth curve and it’s final concentration appeared normal. However, for complex media most of the ammonia only evolved during stationary and starvation phases. The ammonia profile as given by the dashed line in Fig. 1 was modelled again after a balance equation directly derived from Eq. (1): ρAmm = ρAmm,0 + fPN

φ0


φ0 + (1 − φ0 ) exp − (1 −1φ0 ) τt

Fig. 2. Profiles of amino acid consumption during cultivation of D. discoideum on FM medium.

On this basis and some intermediate trials the composition of the FM medium was changed gradually. Supplementation of the FM medium with glucose had a detrimental effect on maximal cell density (data not shown). A greater supplementation of biotin, folic acid, and omission of ammonium chloride or ion chloride had no appreciable effect (data not shown). Therefore, the focus was drawn onto the consumption of amino acids during cultivation. The final composition of the improved medium (SIH) is given in Table 2. The changes from the FM to the SIH medium are highlighted by bold numbers. Major changes have been introduced for the concentrations of lysine and tryptophan. In addition, aspartic acid was introduced and the concentration of phosphates was increased. A typical result for the cultivation of D. discoideum on this improved medium is shown in Fig. 4. Cell densities above 5 × 107 ml−1 can be achieved. Glucose consumption was comparable to that obtained with FM medium, whereas the ammonium ion concentration increased substantially. The ammonium ion concentration increased substantially during the early phase of cultivation, when the cell density was still

(5) In conclusion, the growth of D. discoideum on the synthetic FM medium was characterised by a relatively high maximal cell density and a better utilisation of glucose compared to growth on complex media, but with a lower growth rate. Most of the amino acids present in the FM medium can be followed directly by HPLC as discussed in Section 2. Their profiles during a cultivation experiment are gathered in Fig. 2. All amino acids were utilised during growth, but the extent of utilisation was quite different. Fig. 2 shows that lysine was almost completely consumed after 7 d of cultivation. For a clearer picture the relative consumption of the amino acids is presented in Fig. 3. The numbers given represent the remaining fractions of the initial amino acid concentrations in the FM medium.

Fig. 3. Absolute and relative consumption of amino acids during cultivation on FM medium. The data may differ from those given in Table 2 due to medium preparation.

S.-I. Han et al. / Process Biochemistry 39 (2004) 925–930

929

Fig. 4. Shake-flask cultivation of D. discoideum on SIH medium.

quite low. The cultivation process seemed to be faster than with FM medium, but the balance of the batch cultivation still followed the logistic curve showing that the medium was still one of rather low quality not allowing for exponential growth. The profiles of cell density, glucose and ammonia were modelled with the same Eqs. (1), (4) and (5), respectively, as before. The growth curve in Fig. 4 is characterized by the following parameters: Nmax = 5.8 × 107 ml−1 , φ0 = 0.00003, and τ = 9.14 h. This lead to an initial specific growth rate of µ0 = 1/τ = 0.11 h−1 . However, the small inoculum (φ0 ) shows that this is not the whole story and that it is not realistic to compare any single parameters. A semi-logarithmic plot shows a sigmoidal function again. Linear regression in the range of cultivation times of 20–100 h yields a specific growth rate of 0.084 h−1 . Therefore, the novel medium composition may not only lead to higher cell densities, but also to higher growth rates. The amino acid profiles are gathered in Fig. 5 for the growth experiment on SIH medium shown in Fig. 4. Although amino acids have been supplemented, the general utilisation of amino acids increased. Lysine, as a special case, was used completely, although it’s initial concentration has been raised substantially. The utilisation of amino acids

Fig. 6. Absolute and relative consumption of amino acids during growth of D. discoideum on SIH medium. The data may differ from those given in Table 2 due to medium preparation.

may be better deduced from Fig. 6. This was much more even than in the case of cultivation on FM medium. Alanine was produced and accumulated to a concentration of 1.5 g l−1 during cultivation on SIH medium, but only up to 0.5 g l−1 during cultivation on FM medium (data not shown).

4. Conclusions Analysis of the consumption of amino acids during the cultivation of D. discoideum on synthetic medium FM has shown that their utilisation was not evenly distributed. Some supplementation especially of lysine and tryptophan together with the novel addition of aspartic acid led to a more even and better utilisation of the amino acids. This was accompanied by the production of a cell density in excess of 5 × 107 ml−1 . The better utilisation of amino acids was accompanied by the production of a high concentration of ammonia. The crucial step, however, remains the adaptation of the strain to the synthetic medium. This may cost some time and effort, but seems to be rewarding. In conclusion, the improved synthetic medium (called SIH medium) may represent an even better, because more balanced, nutrient mix than the original FM medium. It has now to be shown that the synthetic media FM and SIH are suitable for mass production of D. discoideum in suspension culture as well as in immobilized form [14].

Acknowledgements

Fig. 5. Amino acid consumption during growth of D. discoideum on SIH medium.

The authors are obliged to Heino Büntemeyer, who managed the amino acid analysis. Dictyostelium discoideum AX2-CR II was kindly provided by C. Reymond. The scholarship for San-In Han from the Max Buchner Forschungsstiftung is gratefully acknowledged.

930

S.-I. Han et al. / Process Biochemistry 39 (2004) 925–930

References [1] Franke J, Kessin R. A defined minimal medium for axenic strains of Dictyostelium discoideum. Proc Natl Acad Sci USA 1977;74:2157– 616. [2] Glenn D, Williams KL. Dictyostelium discoideum: its future in biotechnology. Aust J Biotech 1988;1:46–51. [3] Slade EM, Emslie KR, Williams KL. Expression of recombinant glycoproteins in the simple eukaryote Dictyostelium discoideum. Biotechnol Gen Eng Rev 1997;14:1–35. [4] Rai M, Padh H. Expression systems for production of heterologous proteins. Curr Sci 2001;80(9):1121–8. [5] Jung E, Williams KL. The production of recombinant glycoproteins with special reference to simple eukaryotes including Dictyostelium discoideum. Biotechnol Appl Biochem 1997;25:3–8. [6] Sussman M. Cultivation and serial transfer of the slime mould, Dictyostelium discoideum in liquid nutrient medium. J Gen Microbiol 1961;25:375–8. [7] Claviez M, Pagh K, Maruta H, Baltes W, Fisher P, Gerisch G. Electron microscopic mapping of monoclonal antibodies on the tail region of Dictyostelium myosin. EMBO J 1982;1:1017–22.

[8] Cocucci S, Sussman M. RNA in cytoplasmic and fractions of cellular slime mold amebae. J Cell Biol 1970;45:399–407. [9] C.D. Reymond, personal communication. [10] Fasel N, Begdhadi-Rais C, Bernard M, Bron C, Corradin G, Reymond CD. Dictyostelium discoideum as an expression host for the circumsporocoite protein of Plasmodium falciparum. Gene 1992; 111(2):157–63. [11] Reymond CD, Beghadadi-Rais C, Roggero M, Duarte EA, Desponds C, Bernard M, et al. Anchoring of an immunogenic Plasmodium falciparum circumsporocoite protein on the surface of Dictyostelium discoideum. J Biol Chem 1995;270:12941–7. [12] Büntemeyer H, Lütkemeyer D, Lehmann J. Optimization of serumfree processes for antibody production. Cytotechnology 1991;5:57– 67. [13] Stephan M, Beshay U, Friehs K, Flaschel E. Influence of medium composition on growth behaviour of Dictyostelium discoideum for cultivation on axenic media. Process Biochem., in press. [14] Beshay U, Friehs K, Azzam AE-M, Flaschel E. Cultivation of Dictyostelium discoideum in immobilized form by colonization of porous supports. Process Biochem 2003;38:1521–9.