A Flexible System for Cultivation of Methanococcus and Other Formate-Utilizing Methanogens.

Many hydrogenotrophic methanogens use either H2 or formate as the major electron donor to reduce CO2 for methane production. The conventional cultivation of these organisms uses H2 and CO2 as the substrate with frequent replenishment of gas during growth. H2 is explosive and requires an expensive gassing system to handle safely. Formate is as an ideal alternative substrate from the standpoints of both economy and safety but leads to large changes in the culture pH during growth. Here, we report that glycylglycine is an inexpensive and nontoxic buffer suitable for growth of Methanococcus maripaludis and Methanothermococcus okinawensis. This cultivation system is suitable for growth on liquid as well as solid medium in serum bottles. Moreover, it allows cultivation of liter scale cultures without expensive fermentation equipment. This formate cultivation system provides an inexpensive and flexible alternative for the growth of formate-utilizing, hydrogenotrophic methanogens.

1. Introduction

Methanogens are strictly anaerobic microorganisms belonging to the Euryarchaeota. As a large and diverse group, they are distinguished by their capability to obtain most if not all of their energy for growth from methane production or methanogenesis [1]. In general, methanogens only utilize a limited number of substrates for methanogenesis, such as CO2; H2; formate; methyl-group containing compounds such as methylamines, methylsulfides, and methanol; acetate; and a few low molecular weight alcohols. They do not use sugars, amino acids, or most other common organic substrates [2]. Most methanogens are hydrogenotrophs that use H2 as the primary electron donor to reduce CO2 to methane. Many hydrogenotrophic methanogens can also use formate as the major electron donor [2]. As shown in (1), four molecules of sodium formate are oxidized, yielding one molecule of methane and three molecules of CO2.

Because no more than one ATP is formed per mol of CH4 [3], relatively large amounts of formate are required for even modest growth. Growing cells with sodium formate also leads to a significant accumulation of NaOH, which raises the pH of the medium and inhibits growth. For methanococci, the alkaline pH also causes cell lysis and rapid killing [4]. As a result, pH control becomes a critical concern when cultivating methanogens with sodium formate.

One solution is to titrate the rise in pH with formic acid during growth in a fermenter [5]. For growth in culture tubes and plates, the medium pH can also be controlled with a built-in formic acid reservoir [4]. This cultivation system uses a 6 × 55 mm acid reservoir containing 200 μL formic acid to stabilize the medium pH [4]. As the pH increases, the absorption of formic acid from the headspace also increases, maintaining the pH within levels that support growth. Although this method allows good growth on formate-containing medium, its requirement for manual dexterity precludes it from routine use. Using formate as substrate has also been established in a chemostat system. Costa et al. [6] used formate to grow M. maripaludis in chemostat for studying the transcriptional regulation. The sodium formate was added at 0.38 M, while the pH was maintained at 6.95 by automatic addition of 10% (v/v) H2SO4 [6]. The cell density and growth rate achieved with either formate or H2/CO2 were the same during chemostat cultivation [6-9].

During growth of M. maripaludis with formate, formate dehydrogenase (Fdh) is the key enzyme for formate utilization. Fdh is encoded by two sets of genes, fdhA1B1 and fdhA2B2 in M. maripaludis [10]. Lupa et al. [11] found that mutants with deletions in fdhA1 grew poorly on formate only after an extended lag. In contrast, mutants with deletions in fdhA2 grew nearly the same as wild-type. Because of this and other evidence, Fdh1 was proposed to play a major role in fomate utilization [11].

Over the past decade, many genetic methodologies have been developed in M. maripaludis. These include effective selectable genetic markers [12-16], multiple plasmid shuttle vectors [17], high-efficiency transformation [18], direct gene replacement mutagenesis [19], markerless gene deletion systems [20], random mutagenesis [21], in vivo transposon mutagenesis [22, 23], reporter genes technologies [24, 25], and chemostat cultivation [6-9]. Thus, genetic manipulation of M. maripaludis is easy and effective, and these approaches have become powerful tools to study the metabolism and physiology of multiple Methanococcus species. However, the requirement for H2 growth limits the ability of these genetic tools to be widely applied in laboratories that do not have established systems for handling H2 gas.

Here, we report a medium to cultivate the mesophilic, marine species M. maripaludis on formate using glycylglycine buffer as the pH stabilizer. Ordinarily, M. maripaludis is cultivated in aluminum-sealed tubes with 5 mL of medium under H2/CO2 mixture (4 : 1, v/v) at 276 kPa [26]. For comparison, in our formate cultivation system, the pressure is reduced to 104 kPa, allowing use of more inexpensive stoppers. In addition, frequent gas refilling is avoided without greatly sacrificing growth yield. Simple modifications of common glassware also allows liter-scale cultivation using only a gassing station and a vacuum pump. In addition, the solid medium has a high plating efficiency suitable for genetic experiments. With minor adjustment in medium composition, the procedure is also suitable for growth of the extreme thermophile Methanothermococcus okinawensis.

2. Materials and Methods

2.1. Strains, Media, and Growth Conditions

Methanococcus maripaludis strain S2 was obtained from our laboratory collection (Whitman et al.) [27] and cultured at 37°C. Methanothermococcus okinawensis strain IH1 was obtained from Takai et al. and cultured at 62°C [28].

Cultures were grown in H2/CO2 medium (McNA, a minimal medium with 10 mM sodium acetate) or formate medium (McF) reduced with 3 mM cysteine hydrochloride. The 5 mL cultures were grown in 28 mL aluminum-sealed tubes. For McNA, the tubes were pressurized to 276 kPa with H2/CO2 (4 : 1, v/v) and refilled with the same gas every 24 hours after inoculation. Detailed protocols for growth on formate are given in Appendix A. Briefly, McF medium contained 0.4 M sodium formate and was buffered with 0.2 M glycylglycine (pH = 8.0). The medium was first sparged with N2 to remove most of the O2, and 3 mM cysteine chloride was then added. Tubes were pressurized to 103 kPa with N2/CO2 (4 : 1, v/v) before autoclaving. Prior to inoculation, 3 mM sodium sulfide was added as the sulfur source.

The buffers tested were obtained from Sigma Chemical Co. and included (with the counter ion) Tricine/NaOH (N-[Tris(hydroxymethyl)methyl]glycine), Bicine/NaOH (N,N-bis(2-hydroxyethyl)glycine), Tris/HCl (2-amino-2-hydroxymethyl-propane-1,3-diol), glycine/NaOH, and glycylglycine/NaOH. During formate medium preparation, ingredients were added as listed in the appendices, and the organic buffers were added from stock solutions at pH 7. The concentration of NaCl was adjusted depending upon the amount of sodium formate and sodium in the buffer used so that the final concentration of sodium ion was 0.4 M.

The final medium was also tested for plating (Appendix B) and growth of 1.5 L cultures (Appendix C).

2.2. Rapid Protocol for Preparation of Formate Medium

After combining the components of McF medium, cysteine was added and the medium was dispensed into culture tubes on the bench without anaerobic precautions (Appendix D). Without delay, the tubes were sealed with stoppers and aluminum seals. The tubes were then connected to a gassing manifold, and the air was removed by three successive cycles comprising 45 seconds of vacuum followed by 15 seconds of 104 kPa N2: CO2 (4 : 1, v/v). After exchanging the gas, the medium was autoclaved for 20 min with rapid exhaust. For the control medium, the medium was dispensed in an anaerobic chamber as described in Appendix A, and the gas was exchanged for three cycles with N2/CO2 (4 : 1, v/v) prior to autoclaving.

3. Results

3.1. Optimization of the Formate Medium and Growth Conditions

To determine if organic buffers were inhibitory for growth, they were added to the medium during growth of M. maripaludis on H2/CO2. Because the medium was strongly buffered with bicarbonate and CO2, the buffers did not affect the initial pH. Under these conditions, Tricine was strongly inhibitory (Figure 1). While glycine and Bicine had little effect on cell yield, both increased the lag phase at higher concentrations (data not shown). In contrast, Tris and glycylglycine were not inhibitory and resulted in moderate decreases in the lag phase, presumably by maintaining an optimal pH during the early growth phase (data not shown). Therefore, Tricine and Bicine were omitted from further experiments.

Figure 1

Effect of selected buffers on growth. M. maripaludis S2 was grown in McN (H2/CO2) medium with different concentrations of tested buffers. The culture absorbance was determined after one day.

Tris, glycine, and glycylglycine were further tested for their buffering capacity during growth with 200 mM sodium formate. In the presence of 100 mM buffer, the culture reached a maximal absorbance of about 0.4–0.45 after 20 h (Figure 2). During the first two days of incubation at 37°C, all three buffers maintained the medium pH around 7.2–7.6. However, during extended incubations, decreased absorbance and cellular lysis were observed in media buffered with Tris and glycine (Figure 2, data not shown). In contrast, the absorbance of cultures supplemented with glycylglycine remained stable for six days at 37°C (Figure 2). Moreover, in glycylgylcine-buffered medium, the culture absorbance did not change for up to six weeks at room temperature, and it was still possible to transfer stock cultures to fresh medium. Cultures in McF medium were also used to prepare −80°C freezer stocks in 30% (v/v) glycerol [26, 29], and these cultures retained viability for at least five years.

Figure 2

Growth of M. maripaludis S2 with 200 mM formate and 100 mM of Tris, glycine, and glycylglycine buffers. Two kinds of serum bottle stoppers were used. Blue stoppers are thick butyl rubber stoppers (Bellco Glass Inc., Vineland, NJ, cat. number: 2048-11800, 704.82 USD/1000). They are commonly used for H2/CO2 medium. Butyl rubber gray stoppers (Wheaton Science Products, cat. number: W224100-202, 174.2 USD/1000) were also tested for their durability during M. maripaludis cultivation.

To reduce the cost of anaerobic medium preparation, the influence of different types of stoppers on growth was also tested. Cultivation on H2/CO2 is usually performed at 276 kPa in 28 mL aluminum-sealed tubes. For this reason, thick butyl rubber stoppers (Bellco Glass Inc., Vineland, NJ, cat. number: 2048-11800) are commonly used. These stoppers are made to minimize gas leakage and sustain multiple needle stabs during medium preparation, inoculation, and sampling. As an alternative, butyl rubber grey stoppers (Wheaton Science Products, cat. number: W224100-202) are much less expensive although thinner. Although these stoppers cannot maintain high pressure, they might be suitable for growth on formate at lower pressure. As shown in Figure 2, Wheaton stopper-sealed cultures showed comparable growth profiles and stability, especially in medium supplemented with glycylglycine. In contrast, white precipitates were observed in cultures supplemented with Tris and glycine (data not shown). The composition of the medium resembles that of seawater and contains high levels of divalent cations. During autoclaving, the pH of this medium increases due to the reduced solubility of CO2 at high temperatures. Presumably, these precipitates represent phosphate salts that become insoluble at alkaline pH. The precipitates were rarely observed following autoclaving with the thicker stoppers, probably because they retained CO2 better during autoclaving.

In the presence of 100 mM glycylglycine, the growth yields increased with formate concentrations in nonlinear fashion and were maximal at 0.6 M. Growth was inhibited with 1 M sodium formate, presumably due to sodium toxicity (data not shown). At high formate concentrations and 100 mM glycylglycine, cells lysed in the stationary phase, presumably due to alkalinization of the medium. Increasing the glyclyglycine concentration to 200 mM with 0.4 M formate was found to be optimal for batch growth. In this condition, the growth rate was similar to that in H2/CO2 medium. Moreover, the maximum OD600 nm of 1.0 was comparable to 1.4 in H2/CO2 medium (Figure 3). Thus, the cellular yields per mole of electron donor were nearly equivalent. For instance, medium with 0.4 M formate contained about 2 mmol of formate in 5 mL, and the growth yield was about 340 mg dry wt L−1 or 0.85 g dry wt mol−1 of formate. For 5 mL H2/CO2 cultures with 2.7 mmol of H2, the growth yield was about 400 mg dry wt L−1 or 0.74 g dry wt mol−1 of H2.

Figure 3

Growth of M. maripaludis S2 in H2/CO2 (●) and formate medium (○). The inoculum size was 5 × 104 cells per 5 mL of culture. All values were the averages of five cultures.

Good growth was also found on formate medium containing 1.0% (w/v) agar in serum bottles. Details on preparation are given in Appendix B, but it is similar to the protocols described earlier [30, 31]. Similar to growth with H2/CO2 medium, isolated colonies appeared after 3 to 5 days of incubation, and the plating efficiency was 100%.

3.2. A Simple Medium-Scale Cultivation System for M. maripaludis and M. okinawensis with Formate

The modified formate medium was also useful for cultivation of M. maripaludis and M. okinawensis at liter or medium-scale for the preparation of biomass for enzyme and other studies. For this purpose, a simple cultivation system was developed using common lab glassware and equipment (Appendix C). Comprised largely of a 2 L cultivation bottle, a water trap, and a gas trap, each assembly supported growth of 1.5 L of culture. During growth, the exhaust line allowed the CH4 and CO2 formed to escape, the water trap prevented backflow of water into the culture, and the gas trap prevented back diffusion of air into the culture bottle. A protocol was also developed to ensure complete reduction of the medium before inoculation (Appendix C). Although the medium was sparged prior to inoculation, no gassing was required after inoculation, and the system could be easily moved to fume hood, incubator, or some other well-ventilated space. With a 2% inoculum, M. maripaludis S2 grew to about OD600 nm = 0.8 after 15 hours of incubation at 37°C. In the same medium, M. okinawensis IH1 grew to an OD600 nm = 0.6 (Figure 4). However, reduction of the pH of the glycylglycine buffer stock solution to 6.5 reduced the lag phase of M. okinawensis to 12 h at 62°C without reducing the yield (data not shown). For both cultures, the cell yield was around 1 g (wet weight) per L.

Figure 4

Growth of M. maripaludis S2 (●) and M. okinawensis (▲) in the medium-scale culture system. The inoculum was 1010 cells per 1.5 L of culture. All values are the averages of three cultures.

3.3. Rapid Preparation of Medium without an Anaerobic Chamber

An anaerobic chamber is often used for preparing medium for methanogens, but it is expensive and occupies a large amount of laboratory space. To determine if the formate medium could be prepared in laboratories with limited anaerobic equipment, it was prepared aerobically, and the gas was exchanged with a vacuum pump and gas line connected to a simple gassing manifold controlled by a three-way ball valve. The system was constructed from standard compression fittings so that its fabrication required little equipment and no special expertise. It was designed so that ten tubes or serum bottles could be prepared at one time. A vacuum pressure gauge was used to monitor the gas. After dispensing the medium aerobically, gassing/vacuum cycles were performed to remove O2 from the medium (Appendix D). Interestingly, growth in medium with even one gassing/vacuum cycle was nearly the same as in conventionally prepared medium (data not shown). Cultures of M. maripaludis are often tolerant to O2, and growth of log phase cultures is unaffected by O2 partial pressures < 20 kPa according to our experience. Therefore, it was possible that the large size of inoculum may have protected cells from residual O2. To examine the suitability of this method for small inocula, a most probable number (MPN) experiment was performed in medium prepared with three cycles of gas exchange (Table 1). The most probable numbers were 50 and 160 in media prepared by the rapid or standard protocol with an anaerobic chamber, respectively [32]. These high numbers were not significantly different and would only be possible if growth could be initiated by only one or two cells in both media.

Table 1

Most probable number dilution of M. maripaludis S2 in medium prepared by the rapid protocola.

	Inoculum (number of cells)	Positive number	Negative number
Three O2 removal cycles	1000	5	0
	100	5	0
	10	1	4
	1	1	4
	0.1	0	5
Control	1000	5	0
	100	5	0
	10	3	2
	1	1	4
	0.1	0	5

aThree cycles of gas exchange used in preparation of the McF medium as described in Appendix B. The inoculum was serially diluted into 1000, 100, 10, 1, and 0.1 cells. Growth was monitored for 6 days. When the OD600 nm was greater than 0.6, growth was defined as positive. Control medium was prepared in the anaerobic chamber as described in Appendix A.

This protocol was also suitable for preparation of solid medium and plating for isolation of mutants or other clonal cultures (Appendix D). Agar slabs were formed in serum bottles as described in Appendix B. After growth, single colonies were picked with a syringe needle and transferred to broth under a stream of N2 gas.

4. Discussion

The medium and culturing system for methanogens developed here attempted to address multiple concerns. First, the reagents and equipment should be accessible to many research laboratories. The replacement of H2 with formate as the major substrate for methanogenesis removed the need for a H2 handling system, reducing the cost as well as increasing the safety of culturing. The cost of medium preparation can be further reduced by using much less expensive septum stoppers. Moreover, a simple gassing manifold was sufficient, and an anaerobic chamber was not needed. These methods are straightforward and do not require extensive training. At the University of Georgia, this culturing system was widely used by undergraduate students to isolate and cultivate mutants of M. maripaludis. While training is still required, especially for the safe use of syringes and pressurized glassware, many of the elaborate manipulations of the Hungate method [33] are avoided. For many biological investigations, it is often necessary to generate cultures from single cells as well as generate large amounts of biomass. Both of these are often difficult with fastidious anaerobes. The system developed here had a high plating efficiency, and it was possible to develop cultures from only a few cells. Therefore, it is suitable for the isolation of mutants or other genetic experiments. In addition, it was possible to generate sufficient biomass for enzymatic assays and other biochemical analyses. The glycylglycine buffer prevented alkalinization of the medium and allowed the cultures to remain viable for several weeks on the bench. The addition of glycerol allowed maintenance of viable cultures for at least five years at −80°C. Nevertheless, the formate medium allows a similar growth rate and cellular yield as H2/CO2 medium. Moreover, this medium and protocol were adapted by Weimar et al. [34] for a multiwell plate method to screen chemical compound libraries [34]. M. maripaludis was grown in 96-well microtiter plates sealed in an AGS AnaeroGen compact bag (Oxoid) and incubated at 37°C inside an anaerobic chamber containing 5% H2, 5% CO2, and 90% N2 [34]. Therefore, these methods can be readily adapted for a number of experimental approaches.

Table 2

Component	For tubes	For 1-liter bottle
	For 100 mL	For 1000 mL
Glass-distilled water	30 mL	300 mL
Glycylglycine buffer, 1 M, pH = 8.0	20 mL	200 mL
General salt solution	50 mL	500 mL
K2HPO4, 14 g/L	1.0 mL	10 mL
Na acetate·3H2O, 136 g/L	1.0 mL	10 mL
Trace mineral solution [27]	1.0 mL	10 mL
Iron stock solution [35]	0.5 mL	5 mL
Resazurin, 0.1 g/100 mL	0.1 mL	1 mL
Sodium formate (NaCOOH)	2.7 g	27 g
Sodium bicarbonate (NaHCO3)	0.5 g	5.0 g
Casamino acids (for complex medium)	0.5 g	5.0 g
Alanine (optional, 100 mM)	1.0 mL	10 mL

aMedium components are based upon Balch and Wolfe [35], Romesser et al. [36], and Whitman et al. [27].

Table 3

Component	For tubes	For 1-liter bottle
	For 100 mL	For 1000 mL
Glass-distilled water	10 mL	100 mL
Glycylglycine buffer, 1 M, pH = 6.5	40 mL	400 mL
General salt solution	50 mL	500 mL
K2HPO4, 14 g/L	1.0 mL	10 mL
Na acetate·3H2O, 136 g/L	1.0 mL	10 mL
Trace mineral solution	1.0 mL	10 mL
Iron stock solution	0.5 mL	5 mL
Resazurin, 0.1 g/100 mL	0.1 mL	1 mL
Sodium formate (NaCOOH)	2.7 g	27 g
Sodium bicarbonate (NaHCO3)	0.5 g	5.0 g
Casamino acids (for complex medium)	0.5 g	5.0 g

Table 4

Composition	g/L	Medium concentration (mM)
KCl	0.67	4.5
MgCl2·6H2O	5.50	13.5
MgSO4·7H2O	6.90	14.0
NH4Cl	1.00	9.0
CaCl2·2H2O	0.28	0.95

Table 5

Composition	g/L	Medium concentration (μM)
Nitriloacetic acid	1.5	78
MnSO4·2H2O	0.1	5.3
Fe(NH4)2(SO4)2·H2O	0.2	5.1
CoCl2·6H2O	0.1	4.2
ZnSO4·7H2O	0.1	3.5
CuSO4·5H2O	0.01	0.4
NiCl2·6H2O	0.025	1.1
Na2SeO3	0.2	11.6
Na2MoO4·2H2O	0.1	4.1
Na2WO4·2H2O	0.1	3.0