High Substrate Uptake Rates Empower Vibrio natriegens as Production Host for Industrial Biotechnology

ABSTRACT

The productivity of industrial fermentation processes is essentially limited by the biomass-specific substrate consumption rate (q_S) of the applied microbial production system. Since q_S depends on the growth rate (μ), we highlight the potential of the fastest-growing nonpathogenic bacterium, Vibrio natriegens, as a novel candidate for future biotechnological processes. V. natriegens grows rapidly in BHIN complex medium with a μ of up to 4.43 h⁻¹ (doubling time of 9.4 min) as well as in minimal medium supplemented with various industrially relevant substrates. Bioreactor cultivations in minimal medium with glucose showed that V. natriegens possesses an exceptionally high q_S under aerobic (3.90 ± 0.08 g g⁻¹ h⁻¹) and anaerobic (7.81 ± 0.71 g g⁻¹ h⁻¹) conditions. Fermentations with resting cells of genetically engineered V. natriegens under anaerobic conditions yielded an overall volumetric productivity of 0.56 ± 0.10 g alanine liter⁻¹ min⁻¹ (i.e., 34 g liter⁻¹ h⁻¹). These inherent properties render V. natriegens a promising new microbial platform for future industrial fermentation processes operating with high productivity.

IMPORTANCE Low conversion rates are one major challenge to realizing microbial fermentation processes for the production of commodities operating competitively with existing petrochemical approaches. For this reason, we screened for a novel platform organism possessing characteristics superior to those of traditionally employed microbial systems. We identified the fast-growing V. natriegens, which exhibits a versatile metabolism and shows striking growth and conversion rates, as a solid candidate to reach outstanding productivities. Due to these inherent characteristics, V. natriegens can speed up common laboratory routines, is suitable for already existing production procedures, and forms an excellent foundation for engineering next-generation bioprocesses.

INTRODUCTION

High productivity is an important pillar of economic success for the microbial production of commodities (1, 2). Sophisticated methods in genetic and metabolic engineering allow the construction of producer strains, which operate close to maximal theoretical product yields (3–8). By analogy, fermentation processes are optimized and scaled to technical limits to achieve the highest space-time yields. As a consequence, process optimization steadily struggles to further improve the volumetric productivity (Q_P) to eventually lower capital and operating expenditures. Typical Q_P values of industrially employed microbial systems, such as Escherichia coli, Corynebacterium glutamicum, and Saccharomyces cerevisiae, for the fermentative production of chemicals and fuels are below 4 g liter⁻¹ h⁻¹ (9). Few examples of processes with high volumetric productivities in the range of 4 g liter⁻¹ h⁻¹ with glucose as the substrate have been reported for aerobic l-lysine production with engineered Corynebacterium glutamicum or for anaerobic zero-growth production of alanine with genetically tailored E. coli (10, 11). A key determinant for realizing a high Q_P is the biomass-specific productivity (q_P) of the applied microbial system. The q_P is limited by the biomass-specific substrate consumption rate (q_S), as it essentially determines the amount of carbon that can be funneled into the desired product. Under aerobic conditions in minimal medium with glucose, the fast-growing Bacillus subtilis and E. coli show q_S values of 1.80 and 1.90 g g⁻¹ h⁻¹, respectively (12, 13). Some facultative anaerobic organisms respond to anaerobiosis with an increased q_S to provide sufficient ATP for biomass formation and maintenance requirements. This so-called Pasteur effect increases the q_S of E. coli to 3.33 g g⁻¹ h⁻¹ (13). C. glutamicum grows marginally on glucose as the sole carbon source under anaerobic conditions in the absence of nitrate (14, 15); however, resting cells still exhibit a q_S of 0.37 g g⁻¹ h⁻¹ (16). In the presence of high glucose concentrations, aerobically cultivated S. cerevisiae shows the Crabtree effect, with a high q_S of 3.52 g g⁻¹ h⁻¹, which is only slightly lower than its q_S of 3.85 g g⁻¹ h⁻¹ under anaerobic conditions (17, 18).

Since q_S is directly linked to the growth rate (μ), our search identified the fast-growing gammaproteobacterium Vibrio natriegens as a possible novel microbial factory for future applications in industrial biotechnology. V. natriegens (formerly classified as Pseudomonas natriegens and Beneckea natriegens) is a facultatively anaerobic, marine bacterium originally isolated from salt marsh mud (19). Remarkably, V. natriegens is the fastest-growing nonpathogenic organism identified so far, with a minimal generation time of 9.8 min in complex medium (20). Recently, V. natriegens was exploited as a fast-growing host for molecular biology, providing a set of genetic engineering tools for targeted modification of the metabolism (21, 22).

RESULTS

To evaluate the potential of V. natriegens for industrial applications, we performed aerobic cultivations in complex and minimal media. In shaking flasks, V. natriegens grew rapidly on standard complex medium supplemented with 15 g NaCl liter⁻¹ (LBN, 2×YTN, and BHIN media [see Materials and Methods for medium components]), with maximal growth rates of 2.70 ± 0.03 h⁻¹ in BHIN medium between 0.5 and 1.5 h of cultivation (Fig. 1A). At a low cell density in sterile-filtered BHIN medium, the μ was further improved to 3.18 ± 0.12 h⁻¹ (brain heart infusion [BHI] medium from BD Bacto) and 3.54 ± 0.22 h⁻¹ (BHI medium from Oxoid Ltd.). Under the latter condition, we observed maximal differential growth rates of up to 4.43 h⁻¹, which corresponds to a doubling time of 9.4 min (Fig. 1B).

FIG 1.

(A and B) Growth of V. natriegens in different standard complex media (A) and in BHIN medium at a low cell density as shaking-flask cultures (B). (C) Growth rates in VN minimal medium with 10 g glucose liter⁻¹ under different NaCl concentrations in shaking flasks. For each condition, at least three independent experiments were performed. Error bars show standard deviations.

We tested different standard minimal media in shaking flasks supplemented with 10 g glucose liter⁻¹ and found that VN minimal medium (modified CGXII medium) (23) yielded maximal growth rates of 1.66 ± 0.04 h⁻¹ under optimized NaCl concentrations (15 g liter⁻¹) (Fig. 1C). To further analyze relevant growth characteristics, we performed aerobic bioreactor studies. In BHIN complex medium, V. natriegens grew with a μ of 2.34 ± 0.03 h⁻¹, decoupling the biomass within 1 h (Fig. 2A and B). In VN minimal medium containing 10 g glucose liter⁻¹, V. natriegens showed a lag phase of 2 h, followed by exponential growth, with a μ of 1.48 ± 0.06 h⁻¹ to a biomass concentration (c_X) of 4.23 ± 0.08 g cell dry weight (CDW) liter⁻¹ after 5 h (Fig. 2A to C and E). V. natriegens exhibited a biomass yield (Y_X/S) of 0.38 ± 0.03 g g⁻¹, secreted 3.73 ± 0.25 g CO₂ liter⁻¹ and 1.77 ± 0.11 g acetate liter⁻¹ into the medium (Fig. 2E), and possessed an exceptionally high q_S of 3.90 ± 0.08 g g⁻¹ h⁻¹ (Fig. 2D) in the exponential growth phase.

FIG 2.

Bioreactor cultivations of V. natriegens in BHIN complex medium and VN minimal medium with 10 g glucose liter⁻¹ under aerobic and anaerobic conditions. Biomass formation as g cell dry weight (CDW) liter⁻¹, growth rates (μ), biomass yields (Y_X/S), and biomass-specific glucose consumption rates (q_S) are depicted in panels A to D, respectively. Time courses of biomass formation, glucose consumption, and product formation under aerobic and anaerobic conditions are shown in panels E and F. For each condition, three independent experiments were performed. Error bars show standard deviations.

Furthermore, we investigated the capability of V. natriegens to metabolize other industrially relevant carbon sources and performed aerobic shaking-flask cultivations in VN medium with different sugars, alcohols, and organic acids as the sole carbon and energy source (see Fig. S1 in the supplemental material). No biomass formation was observed on xylose, lactose, mannose, cellobiose, ethanol, methanol, and formate, whereas V. natriegens grew moderately on galactose (μ = 0.18 ± 0.01 h⁻¹) and rhamnose (μ = 0.40 ± 0.01 h⁻¹) and well on maltose (μ = 1.22 ± 0.01 h⁻¹), arabinose (μ = 0.83 ± 0.02 h⁻¹), glycerol (μ = 0.86 ± 0.03 h⁻¹), and fructose (μ = 1.51 ± 0.08 h⁻¹) and showed a slightly higher growth rate on sucrose (μ = 1.79 ± 0.02 h⁻¹) than on glucose (μ = 1.68 ± 0.02 h⁻¹). V. natriegens also grew exponentially on the chitin monomers glucosamine (μ = 0.68 ± 0.02 h⁻¹) and N-acetylglucosamine (μ = 1.74 ± 0.01 h⁻¹) and on the organic acids acetate (μ = 0.45 ± 0.03 h⁻¹), malate (μ = 0.85 ± 0.01 h⁻¹), fumarate (μ = 0.99 ± 0.02 h⁻¹), succinate (μ = 1.00 ± 0.02 h⁻¹), and gluconate (μ = 1.51 ± 0.01 h⁻¹) and was able to degrade soluble starch (μ = 0.19 ± 0.03 h⁻¹).

Analysis of the genome sequence (21, 24, 25) identified genes encoding enzymes such as pyruvate formate lyase which are typical for facultative anaerobes performing a type of mixed-acid fermentation. Therefore, we cultivated V. natriegens in VN medium with 10 g glucose liter⁻¹ under anaerobic conditions in a bioreactor. After a lag phase of 2.5 h, V. natriegens showed a μ of 0.92 ± 0.01 h⁻¹ and a Y_X/S of 0.12 ± 0.01 g g⁻¹ (Fig. 2A to C) and produced 1.93 ± 0.15 g acetate liter⁻¹, 0.66 ± 0.05 g succinate liter⁻¹, 3.01 ± 0.22 g formate liter⁻¹, 0.17 ± 0.01 g lactate liter⁻¹, 1.41 ± 0.06 g ethanol liter⁻¹, 0.04 ± 0.00 g valine liter⁻¹, 0.03 ± 0.00 g glutamate liter⁻¹, and 0.31 ± 0.01 g alanine liter⁻¹ (Fig. 2F; Fig. S2). Liquid chromatography-tandem mass spectrometry (LC-MS/MS)-based analysis of the culture supernatants taken after 7 h identified 3.7 ± 0.3 mg liter⁻¹ of the l-valine biosynthetic intermediate 2,3-dihydroxy-3-methylbutanoic acid and 15.5 ± 1.3 mg liter⁻¹ of the leucine derivative 2-hydroxyisocaproic acid. Compared to that of the aerobic environment, the q_S doubled to 7.81 ± 0.71 g g⁻¹ h⁻¹ under anaerobic conditions (Fig. 2D).

To further analyze the suitability for industrial applications, we packed V. natriegens to high cell densities (∼20 g CDW liter⁻¹) in a bioreactor under anaerobic conditions (Fig. 3). In VN medium with 42.5 g glucose liter⁻¹, the cells did not grow; however, they remained metabolically active and consumed the glucose rapidly within 2 h of fermentation. Resting cells of V. natriegens showed a q_S of 1.02 ± 0.05 g g⁻¹ h⁻¹ and secreted the fermentation products acetate, formate, succinate, lactate, ethanol, and alanine with volumetric productivities (Q_P) ranging between 1.7 and 4.3 g liter⁻¹ h⁻¹. The combined Q_P of the secreted compounds amounted to 17.7 ± 0.1 g liter⁻¹ h⁻¹ (Fig. 3).

FIG 3.

Course of product formation and volumetric productivity (Q_P) of the respective products and combined Q_P (Σ) of resting high-cell-density cultures (∼20 g CDW liter⁻¹) of V. natriegens in VN medium with 42.5 g glucose liter⁻¹ under anaerobic conditions. Three independent experiments were performed. Error bars show standard deviations.

With regard to the high production capacity, we engineered V. natriegens for the anaerobic biosynthesis of alanine. Therefore, we established the sacB-based suicide vector system pDM4 (26) to perform allelic exchanges in V. natriegens. Applying this method, we deleted genes annotated as encoding d- and l-lactate dehydrogenase, pyruvate formate lyase, and malate dehydrogenase, progressively, to reduce by-product formation and elevate pyruvate availability for alanine production. To compare performance, the engineered V. natriegens strains and the wild type were cultivated as resting cells in anaerobic test tubes filled with VN minimal medium containing 5 g glucose liter⁻¹ without the addition of bicarbonate. The implemented genetic modifications eliminated the formation of formate, reduced the secretion of succinate, lactate, acetate, and ethanol, and gradually increased the formation of alanine (Fig. 4A; Fig. S3). V. natriegens Δdldh Δlldh Δpfl Δmdh showed a q_S of 0.91 ± 0.15 g g⁻¹ h⁻¹ and efficiently produced alanine from glucose with a Y_P/S of 0.81 ± 0.01 g g⁻¹ (Fig. 4A; Fig. S4), which corresponds to 82% of the theoretical maximum. Zero-growth production with V. natriegens Δdldh Δlldh Δpfl Δmdh at a high cell density yielded an overall Y_P/S of 0.73 ± 0.05 g g⁻¹ and a Q_P of 0.56 ± 0.10 g alanine liter⁻¹ min⁻¹ (i.e., 34 ± 6 g liter⁻¹ h⁻¹) between 0 and 30 min of fermentation, while glucose was exhausted after 45 min (Fig. 4B; Fig. S5).

FIG 4.

(A) Alanine yields (Y_P/S) of V. natriegens wild-type (V.n.) and its engineered derivatives, V. natriegens Δdldh (V.n. Δ), V. natriegens Δdldh Δlldh (V.n. ΔΔ), V. natriegens Δdldh Δlldh Δpfl (V.n. ΔΔΔ), and V. natriegens Δdldh Δlldh Δpfl Δmdh (V.n. ΔΔΔΔ) in VN medium with 5 g glucose liter⁻¹ under anaerobic conditions. (B) Alanine yield (Y_P/S) and volumetric productivity (Q_P) of zero-growth production at a high cell density with V. natriegens Δdldh Δlldh Δpfl Δmdh. At least three independent experiments were performed. Error bars show standard deviations.

DISCUSSION

V. natriegens is the fastest-growing nonpathogenic bacterium isolated so far (20). However, its potential for industrial applications has not been exploited to date. As we show here, V. natriegens fulfills basic requirements for biotechnological applications. This bacterium is considered biologically safe and can easily be cultivated with established laboratory equipment in cheap minimal medium at high growth rates. NaCl at 15 g liter⁻¹ is essentially required for optimal cell proliferation but might create additional challenges for process development. Increasing NaCl fractions might influence ion-exchange approaches in downstream processing for separating the product from the broth and increase the conductibility of the liquid. Since chloride especially fosters corrosion processes on metal surfaces of the bioreactor (27), we have designed a minimal medium that contains chloride only in traces but sufficient sodium to obtain 91% of the growth rate of that in VN medium (data not shown).

Providing glucose as the substrate, the observed growth rate is at least two times higher than those of traditionally employed microbial systems such as E. coli (13, 28), Bacillus subtilis (12, 29), C. glutamicum (30), and yeast (17, 18, 31). Furthermore, V. natriegens is prototrophic, possesses a versatile metabolism, and has the capability to metabolize several relevant substrates, which represents a basic prerequisite for the favored flexible feedstock concept in industrial biotechnology (32). Deduced from the genome sequence, the sugars d-glucose, d-fructose, d-sucrose, and d-maltose are taken up and activated by specific phosphotransferase systems and further metabolized by glycolysis, the pentose phosphate pathway, and the citric acid cycle. In contrast, the presence of gluconate activates the Entner-Doudoroff pathway (33), yielding growth rates as high as those with fructose as the sole carbon source. Recently, tools for genetic engineering have been made accessible (21, 22); these tools allow, on the basis of the available genome sequence (21, 24, 25), the targeted manipulation of the metabolism. V. natriegens yields superior glucose consumption rates under aerobic and anaerobic conditions. The determined q_S values under both conditions outcompete those of typical industrially relevant microbes by a factor of at least 2 for E. coli (13), Pseudomonas putida (34, 35), and yeast (17, 18) to a factor of 20 for C. glutamicum (16). The high q_S values of V. natriegens provide an excellent basis to achieve high overall productivities to realize cost-efficient industrial fermentation processes. The anaerobic environment in particular, where V. natriegens possesses a two-times-higher q_S than that under aerobic conditions, circumvents oxygen transfer limitations that probably occur under unlimited aerobic growth. Accordingly, the applied anaerobic zero-growth production process yielded a cumulative Q_P of about 18 g liter⁻¹ h⁻¹, which is already significantly higher than that of streamlined fermentation processes for the production of succinate, d-lactate, or valine with genetically optimized microbial systems (36–40). V. natriegens natively provides a multifaceted portfolio of promising target compounds, such as organic and amino acids, making this organism an ideal candidate for the directed production of these target molecules with genetically tailored derivatives. As a proof of concept, we engineered V. natriegens for anaerobic alanine production. V. natriegens Δdldh Δlldh Δpfl Δmdh efficiently produced alanine from glucose with a Q_P of about 0.56 g liter⁻¹ min⁻¹ (i.e., 34 g liter⁻¹ h⁻¹), which is about 9 and 13 times higher than the Q_P values of tailored E. coli and C. glutamicum alanine producer strains, respectively (37).

Taken together, the observed characteristics render V. natriegens an attractive host for future biotechnological applications. The potential to reach outstanding overall productivities, which is an essential prerequisite to realize microbial fermentation processes for the production of commodities, enables V. natriegens to operate competitively with existing petrochemical approaches. V. natriegens can speed up common laboratory routines (21), is suitable for already existing production procedures, and forms an excellent foundation to engineer next-generation bioprocesses.

MATERIALS AND METHODS

Bacterial strains and plasmids.

All bacterial strains with the respective genotypes, plasmids, and primers used in this study are listed in Table 1.

TABLE 1.

Strains, plasmids, and oligonucleotides used in this study

Strain, plasmid, or oligonucleotide	Relevant characteristic(s) or sequence (5′→3′)a	Source, reference, or purpose
Strains
E. coli S17-1 λpir	thi pro hsdR hsdM1 recA RP4-2-Tc::Mu-Km::Tn7 λpir	41
V. natriegens	DSM 759	German Collection of Microorganisms and Cell Cultures
V. natriegens Δdldh	V. natriegens with deletion of dldh gene, encoding d-lactate dehydrogenase (EC 1.1.1.28)	This work
V. natriegens Δdldh Δlldh	V. natriegens Δdldh with deletion of lldh gene, encoding l-lactate dehydrogenase (EC 1.1.1.27)	This work
V. natriegens Δdldh Δlldh Δpfl	V. natriegens Δdldh Δlldh with deletion of pfl gene, encoding pyruvate formate lyase (EC 2.3.1.54)	This work
V. natriegens Δdldh Δlldh Δpfl Δmdh	V. natriegens Δdldh Δlldh Δpfl with deletion of mdh gene, encoding malate dehydrogenase (EC 1.1.1.37)	This work
Plasmids
pDM4	sacBR oriT oriR6K Cmr	26
pDM4 Δdldh	pDM4 carrying a dldh deletion construct	This work
pDM4 Δlldh	pDM4 carrying a lldh deletion construct	This work
pDM4 Δpfl	pDM4 carrying a pfl deletion construct	This work
pDM4 Δmdh	pDM4 carrying a mdh deletion construct	This work
Oligonucleotides
dldh1	CAATTTGTGGAATCCCGGGAGAGCTCGATGAGAAGAAAGAGCATGTG	Primer for pDM4 Δdldh construction (SacI)
dldh2	TCGAGAGAGAACCATAAGGAATCGAGGAGGATAG	Primer for pDM4 Δdldh construction
dldh3	CTCGATTCCTTATGGTTCTCTCTCGAAATCATTG	Primer for pDM4 Δdldh construction
dldh4	GCAAAAGCACCGCCGGACATCAGCGCTAGCCTCAAACATCTGCCACCTCTAG	Primer for pDM4 Δdldh construction (NheI)
lldh1	CAGGTTACCCGCATGCAAGATCTATCTAGACCTAAACGGTAGCCTCTGCAG	Primer for pDM4 Δlldh construction (XbaI)
lldh2	CGTTTTAAACTCGTTTTCTCTCCCTGATAATTCTAAAAAATC	Primer for pDM4 Δlldh construction
lldh3	CAGGGAGAGAAAACGAGTTTAAAACGGAAACCC	Primer for pDM4 Δlldh construction
lldh4	GTGTATATCAAGCTTATCGATACCGTCGACCTATAGCCTTTACCACGAAACG	Primer for pDM4 Δlldh construction (SalI)
pfl1	CAGGTTACCCGCATGCAAGATCTATCTAGACTATTGCCTAAGTTGGTTCC	Primer for pDM4 Δpfl construction (XbaI)
pfl2	AGGTAGGTATGTGACTGTCGCGAAATAACGTTATAG	Primer for pDM4 Δpfl construction
pfl3	TTTCGCGACAGTCACATACCTACCTTTTTAGTAGAAAAAAATAC	Primer for pDM4 Δpfl construction
pfl4	GTGTATATCAAGCTTATCGATACCGTCGACCCGTCAAATGGTTGGTACTG	Primer for pDM4 Δpfl construction (SalI)
mdh1	CAGGTTACCCGCATGCAAGATCTATCTAGACGGATTTGAACAACTGCTCTG	Primer for pDM4 Δmdh construction (XbaI)
mdh2	TAAACGCTTAAACGTAGTTCTCCTTGAGAGTATTTTTTTATAAATG	Primer for pDM4 Δmdh construction
mdh3	CAAGGAGAACTACGTTTAAGCGTTTATACGTTTATAAAAAG	Primer for pDM4 Δmdh construction
mdh4	GTGTATATCAAGCTTATCGATACCGTCGACGTAAGTACTTAGGTACCGCATTC	Primer for pDM4 Δmdh construction (SalI)

^aUnderlined sequences indicate recognition sites of the respective restriction endonucleases.

Media and culture conditions.

V. natriegens was stored at −70°C as 30% (vol/vol) glycerol stock and for cultivation was streaked on brain heart infusion (BHI [BD Bacto]; 37 g liter⁻¹) medium containing 15 g NaCl liter⁻¹ (BHIN medium) and 18 g agar liter⁻¹. LB and 2×YT medium (42) contained overall 15 g NaCl liter⁻¹ (LBN and 2×YTN media). To prepare a preculture of V. natriegens, 5 ml BHIN medium in a test tube was inoculated with a single colony and cultivated at 37°C on a rotary shaker at 120 rpm.

For bioreactor and shaking-flask cultivations, cells of an overnight preculture of V. natriegens were washed with 9 g NaCl liter⁻¹ and inoculated to an optical density at 600 nm (OD₆₀₀) of about 0.1 into VN minimal medium composed of the following (liter⁻¹): 5 g (NH₄)₂SO₄, 15 g NaCl, 1 g KH₂PO₄, 1 g K₂HPO₄, 0.25 g MgSO₄, 0.01 g CaCl₂, 16.4 mg FeSO₄·7H₂O, 10 mg MnSO₄·H₂O, 0.3 mg CuSO₄·5H₂O, 1 mg ZnSO₄·7H₂O, and 0.02 mg NiCl₂·6H₂O. The medium was additionally supplied with 21 g 3-(N-morpholino)propanesulfonic acid (MOPS) liter⁻¹ for experiments in shaking flasks. For anaerobic cultivations, the medium also contained 4.2 g NaHCO₃ liter⁻¹, which was introduced as a freshly prepared 10× stock solution after N₂ stripping (see below). Shaking-flask cultivations were performed with an initial pH of 7.5 as 50-ml cultures in 500-ml baffled Erlenmeyer flasks on a rotary shaker at 120 rpm.

Aerobic and anaerobic fermentations were performed at 37°C as 200-ml cultures in glass reactors. The pH was maintained at 7.0 by online measurement using a standard pH probe (Mettler Toledo, Giessen, Germany) and the addition of 2.5 M NaOH. For aerobic cultivations, the dissolved oxygen was measured online using a polarometric oxygen electrode (Mettler Toledo, Giessen, Germany) and adjusted to ≥30% saturation in a cascade by stirring at 200 to 550 rpm with an aeration at 0.25 to 1 volume per volume per minute. For anaerobic fermentations, the residual oxygen was removed by N₂ stripping for 20 min.

For anaerobic organic acid production at a high cell density, a 5-ml BHIN culture grown for 6 h was used to inoculate a second 50-ml BHIN preculture of V. natriegens to an OD₆₀₀ of 0.1, which was cultivated aerobically overnight in 500-ml baffled Erlenmeyer flasks on a rotary shaker at 120 rpm and 37°C. Afterwards, the stationary cells were harvested by centrifugation (Eppendorf 5417 R; Eppendorf, Hamburg, Germany) (5,000 × g, room temperature [RT], 15 min), resuspended in VN medium, and used to inoculate the bioreactor (100-ml culture) containing VN medium supplemented with 4.2 g NaHCO₃ liter⁻¹ and 42.5 g glucose liter⁻¹.

Testing of the engineered V. natriegens strains for anaerobic alanine production was performed in VN medium with 10 g (NH₄)₂SO₄ liter⁻¹ and 5 g glucose liter⁻¹ without NaHCO₃ as 15-ml cultures in airtight closed test tubes that were inoculated to an OD₆₀₀ of 10 with stationary cells of a 50-ml overnight BHIN preculture. The anaerobic test tubes were cultivated for 5 h at 37°C and 120 rpm on a rotary shaker. Anaerobic alanine production in bioreactors at a high cell density was performed as described above for organic acid production with the modification that bicarbonate was omitted, the initial glucose and (NH₄)₂SO₄ concentrations in the VN medium were set to 32 and 20 g liter⁻¹, respectively, and 12.5% (vol/vol) ammonium hydroxide was used for titration. To obtain sufficient biomass for inoculation of the bioreactor, 400-ml BHIN cultures in 2-liter baffled Erlenmeyer flasks were used as a second preculture.

Determination of μ, Y_X/S, and q_S.

The growth rate, μ, was calculated by linear regression of ln(OD₆₀₀) plotted against time (in hours) during the exponential phase. The biomass yield Y_X/S (g g⁻¹) was determined by linear regression of the biomass concentration c_X (g liter⁻¹) plotted against the respective glucose concentration (g liter⁻¹) during the exponential growth phase. The biomass-specific glucose consumption rate q_S (g g⁻¹ h⁻¹) was calculated by the equation q_S = μ/Y_X/S. The standard deviation (SD) for q_S was determined by applying Gaussian error propagation.

Bacterial matings.

Transfer of pDM4 derivatives to V. natriegens was performed by bacterial mating using E. coli S17-1 λpir as the donor strain. A single colony of E. coli S17-1 λpir, carrying the respective pDM4 derivative, and a single colony of V. natriegens were used to inoculate 5 ml 2×YT medium with 15 μg chloramphenicol ml⁻¹ and BHIN medium, respectively, and the test tubes were cultivated for 2 h at 37°C and 120 rpm on a rotary shaker. The donor cells were washed twice with 2×YT medium to remove any leftover antibiotics of the growth medium and mixed with the recipient strain at a ratio of 1:1. The cells were then pelleted together (3 min, 12,000 × g), resuspended in 100 μl LBN medium, spotted on an LBN agar plate, and incubated at RT for 4 to 5 h. For selection of chloramphenicol-resistant transconjugants, cells were scooped from the mating spot, resuspended in 100 μl BHIN medium, plated on VN agar plates containing 1% (wt/vol) glucose and 15 μg chloramphenicol ml⁻¹, and incubated for 24 h at 37°C.

Construction of V. natriegens deletion mutants.

Chromosomal deletion of the dldh, lldh, pfl, and mdh genes encoding d- and l-lactate dehydrogenase, pyruvate formate lyase, and malate dehydrogenase was performed using crossover PCR and the suicide vector pDM4 (26). DNA fragments were generated using the primer pairs dldh1/dldh2 and dldh3/dldh4, lldh1/lldh2 and lldh3/lldh4, pfl1/pfl2 and pfl3/pfl4, and mdh1/mdh2 and mdh3/mdh4. The respective two fragments were purified, mixed in equal amounts, and subjected to crossover PCR using primers dldh1/dldh4, lldh1/lldh4, pfl1/pfl4, and mdh1/mdh4. The resulting fusion products were ligated into XbaI/SalI (for deletion of lldh, pfl, and mdh)- or SacI/NheI (for deletion of dldh)-restricted plasmid pDM4 and transformed into E. coli S17-1 λpir. After isolation and sequencing (GATC Biotech, Konstanz, Germany), the recombinant plasmid was conjugated into the V. natriegens strains. By applying the method described by Milton et al. (26), the intact chromosomal dldh, lldh, pfl, and mdh genes were deleted via homologous recombination (double crossover). To screen for the dldh, lldh, pfl, and mdh mutants, transconjugants from the mating procedure were cultivated in 5 ml LBN medium with 10% (wt/vol) sucrose for 18 h at 30°C and 120 rpm on a rotary shaker, spread on LBN agar plates with 10% (wt/vol) sucrose, and incubated for 24 h at RT. Deletion mutants were identified by colony PCR using primers dldh1/dldh4, lldh1/lldh4, pfl1/pfl4, and mdh1/mdh4.

Analytics.

Biomass formation was either followed by determining the OD₆₀₀ or the cell dry weight (CDW; in g liter⁻¹) at a given point in time. Both techniques were correlated for several independent cultivations in VN medium, resulting in a CDW (g liter⁻¹) equal to OD₆₀₀ × 0.27 (see Fig. S6 in the supplemental material) for exponentially growing cells. V. natriegens cells of a stationary BHIN preculture showed a CDW/OD₆₀₀ ratio of 0.28, and for cells of the anaerobic high-cell-density culture harvested after 2.5 h, a CDW/OD₆₀₀ ratio of 0.33 was determined.

For determination of glucose, ethanol, and organic and amino acid concentrations in the culture fluid, 2 ml of the culture was harvested by centrifugation (12,100 × g, 5 min, RT), and the supernatant was analyzed. Glucose concentrations were determined enzymatically as described by Lamprecht and Heinz (43). In MOPS containing VN minimal medium, ethanol was determined using an ethanol test kit (catalog no. 10176290035; Roche).

HPLC metabolite quantification.

The amino acid concentration was determined by reversed-phase high-performance liquid chromatography (HPLC) using an Agilent 1200 series apparatus (Agilent Technologies) equipped with an Agilent Zorbax Eclipse Plus C₁₈ column (250 by 4.6 mm, 5 μm) protected by an Agilent Zorbax Eclipse Plus C₁₈ guard column (12.5 by 4.6 mm, 5 μm) at 40°C with a flow rate of 1.5 ml min⁻¹. Fluorometric detection (excitation at 340 nm and emission at 450 nm) was carried out after automatic precolumn derivatization with ortho-phthaldialdehyde and 9-fluorenylmethoxycarbonylchloride. The elution buffer consisted of a polar phase (10 mM Na₂HPO₄, 10 mM Na₂B₄O₇, 0.5 mM NaN₃, pH 8.2) and a nonpolar phase (45% [vol/vol] acetonitrile, 45% [vol/vol] methanol, 10% [vol/vol] H₂O). Quantification of the analytes was conducted by using l-ornithine as an internal standard to correct for derivatization variability and by an 8-point calibration curve for each component as an external reference standard.

Organic acid and ethanol concentrations were measured by HPLC using an Agilent 1200 series apparatus equipped with a Rezex ROA organic acid H⁺ (8%) column (300 by 7.8 mm, 8 μm; Phenomenex), protected by a Phenomenex guard column carbo-H (4 by 3.0 mm inside diameter). A protocol for phosphate precipitation was applied to each sample and standard prior to measurement. Thus, 45 μl 4 M NH₃ and 100 μl 1.2 M MgSO₄ were added to a 1,000-μl sample. After 5 min of incubation, the sample was centrifuged for 5 min at 18,000 × g at RT. Five hundred microliters of supernatant was then transferred to 500 μl 0.1 M H₂SO₄. After thorough mixing and 15 min of incubation at RT, the samples were finally centrifuged for 15 min at 18,000 × g at RT. Subsequently, the supernatant was provided for HPLC injection (10-μl injection volume). Separation was performed under isocratic conditions at 50°C (column temperature) for 45 min with 5 mM H₂SO₄ as the mobile phase at a constant flow rate of 0.4 ml min⁻¹. Detection of organic acids and ethanol was achieved with a refractive index detector at 32°C. Quantification of the analytes was conducted using a 7-point calibration curve for each component as an external reference standard.

LC-MS/MS analysis.

To further investigate the product spectrum of V. natriegens, culture supernatants were analyzed on an Agilent 1260 HPLC coupled to an Agilent 6540 accurate-mass LC-quadrupole time-of-flight (Q-TOF)/MS system with ESI Jet Stream technology (Agilent Technologies). Samples were prepared in 60% (vol/vol) acetonitrile and 10 mM ammonium acetate buffer (pH 5.6), resulting in 1:10 dilutions. Five microliters was injected onto a SeQuant ZIC-pHILIC column (150 by 2.1 mm, 5 μm) with a guard column (SeQuant ZIC-pHILIC; 20 by 2.1 mm, 5 μm) at 40°C with a flow rate of 0.2 ml min⁻¹. Mobile phase A was composed of 10% (vol/vol) aqueous buffer solution (10 mM ammonium acetate) and 90% (vol/vol) acetonitrile, and the composition of mobile phase B was 90% (vol/vol) aqueous buffer solution and 10% (vol/vol) acetonitrile, both adjusted to pH 5.6 with acetic acid (44). After gradient elution, the analytes were detected in negative MS mode and negatively targeted MS/MS mode using the following conditions: drying gas flow rate of 8 liters min⁻¹ with a gas temperature of 325°C, a nebulizer with a 40-lb/in² gauge, a sheath gas flow rate of 12 liters min⁻¹, and a sheath gas temperature of 350°C, capillary voltage of 4,000 V, and fragmentor voltage of 140 V. System control, acquisition, and data analysis were performed using MassHunter workstation software (version B.05.00), and peaks were extracted by the “Find by Molecular Feature” algorithm. Initial identification of compounds was performed by an online accurate-mass database search and a comparison of precursor ion fragmentations (MS/MS) (METLIN and MassBank). Applying this method, two prominent signals were identified as 2-hydroxyisocaproic acid and 2,3-dihydroxy-3-methylbutanoic acid. By comparing the retention time, accurate mass, and MS/MS fragmentation with those of commercially available standards of the two compounds (purchased from Sigma-Aldrich), identification was verified. Both compounds were quantified using a 7-point calibration curve for each component as an external reference standard.

TC analysis.

Total inorganic carbon (TIC) and total organic carbon (TOC) contents in bioreactor samples were quantified with a total carbon (TC) analyzer (Multi N/C 2100s; Analytik Jena, Jena, Germany). The apparatus was operated in the TOC mode by using the differential detection method with parallel measurement of TIC and TC to determine the TOC amount of the sample (TOC = TC – TIC) as described previously (45). The instrument was calibrated with a standard solution containing 2,500 mg l-valine liter⁻¹, producing a 9-point carbon calibration range of 32 to 1,280 mg C liter⁻¹ for TC quantification.

To prevent the outgassing of CO₂ in the sample prior to detection, 100 μl of sample was immediately transferred to 2-ml reaction tubes containing 1.886 ml H₂O and 14 μl 5 N potassium hydroxide and deionized water, effectively increasing the basicity and thus shifting the carbonic acid equilibrium toward HCO₃⁻ and CO₃²⁻.