Investigation of Nitrate Respiration in Cupriavidus necator for Application in Life Support System

Abstract

Cupriavidus necator is a well-studied microorganism with potential application in bioregenerative life support systems for single-cell protein and bioplastic production. Most studies have been carried out in autotrophy or heterotrophy, requiring O₂ as the final electron acceptor. In the context of inhabited missions, access to O₂ will primarily be limited to the crew. In this study, we investigated the capacity of C. necator to carry out nitrate respiration as a strategy to limit oxygen supply to the cultures by providing nitrate from another compartment of the Bioregenerative Life Support System (BLSS). Batch bioreactor experiments were carried out to determine the best conditions for nitrate utilization in terms of pH and aeration. Continuous cultures were then performed under two carbon sources (glucose vs. acetic acid) and two substrate limitations (nitrate vs. carbon). The optimal conditions were found to be pH 7.5 under anaerobiosis. They were applied in chemostats, where three steady-states were obtained at a low dilution rate. In all cases, the biomass consisted of a mixture of protein (from 29 ± 1% Cell Dry Weight (CDW) to 39 ± 2% CDW) and polyhydroxybutyrate (from 45 ± 2% CDW to 57 ± 3% CDW), which was found to be a key component for nitrate respiration metabolism. Microaerobic conditions were also tested in batch culture, reporting for the first time aerobic nitrate respiration in C. necator. Under these conditions, growth parameters improved during the nitrate phase; however, the specific growth rate during the nitrite phase was lower than that observed under strictly anaerobic conditions.

1. Introduction

The renewed interest in human space exploration is accompanied by the need to develop new life support systems. These can be defined as complex systems of technologies that work in a network to maintain favourable conditions for human survival. Space agencies around the world are promoting research in this area to enable food production and resource recycling in space [1,2,3]. Therefore, the introduction of biological reactions is needed, and such systems are known as bioregenerative life support systems (BLSSs).

As early as the 1960s, ref. [4] investigated the potential use of Cupriavidus necator (C. necator), formerly known as Hydrogenomonas eutropha, Wautersia eutropha, Alcaligenes eutrophus, and, later, Ralstonia eutropha [5], for space applications, focusing on its ability to convert CO₂, urea, and hydrogen into single-cell protein (SCP). More recently, its integration into a bioregenerative life support system (BLSS) has been further explored [6,7,8,9]. In the last two studies, urea from crew urine served as the nitrogen source, while volatile fatty acids (VFAs) generated during the first stage of anaerobic digestion of organic waste provided the carbon source, either as a mixture (mainly acetate, followed by propionate, valerate, butyrate, iso-valerate, and iso-butyrate) or individually. Laboratory-scale experiments reported protein productivities ranging from 0.046 to 0.132 g·g⁻¹·h⁻¹ (equivalent to 6.2–23 g·day⁻¹) [7].

C. necator is also one of the most efficient and extensively studied bacteria for producing polyhydroxyalkanoates (PHAs) [10,11]. Using anaerobic digestion effluent as a carbon source provides a promising way to integrate waste management with biopolymer production [12,13]. For a space application, in situ production of PHA-derived materials could enable astronauts to manufacture tools or construction components using 3D printing. The incorporation of PHA production into a BLSS has been investigated using Rhodospirillum rubrum [14,15,16], where slow specific growth rates (µ) (h⁻¹) (0.005–0.010 h⁻¹) and relatively low PHA yields (averaging 10% of cell dry weight (CDW), and up to 35% of CDW under optimal conditions) were achieved, consisting of a mix of polyhydroxybutyrate (PHB) and polyhydroxyvalerate (PHV).

The metabolic shift between SCP production and PHA accumulation is generally driven by nutrient imbalance, particularly the carbon-to-nitrogen (C/N) ratio. Typically, C. necator favours SCP production at C/N < 10 and PHA accumulation at C/N > 20 [17]. Notably, significant amounts of PHA have been detected even under conditions optimized for SCP production and high growth rates [7,18].

Most of these process characterizations were performed aerobically. However, for a life support system application in space, access to O₂ will be scarce and primarily limited to the crew. C. necator is a well-studied bacterium whose genome has been sequenced and annotated [19,20]. It is known to have the ability to grow anaerobically using nitrate and nitrite as electron acceptors. Nitrate respiration consists of the total reduction of nitrate to N₂ by four enzymes: a nitrate reductase NADH-dependent (nar), a nitrite reductase (nir), a nitric oxide reductase (nor), and a nitrous oxide reductase (nos) [21]. Few data are available to describe C. necator fermentation using its anoxic metabolism. The authors of [22] were the first to test it in autotrophy and heterotrophy. Nitrate cultures were characterized by a diauxic growth, first with nitrate consumption and nitrite production, and then with nitrite consumption. Reference [23] determined the kinetic parameters of C. necator H16 in autotrophy and confirmed a drastic reduction in the µ, at best, µ_nitrate = 0.021 h⁻¹ and µ_nitrite = 0.008 h⁻¹, compared to aerobic conditions in autotrophy (µ = 0.24 h⁻¹) [24]. These very low specific growth rates represent a major limitation for practical applications in autotrophy. However, studying heterotrophic anaerobic growth has been lacking in the literature. Therefore, the present study focuses on characterizing biomass production, physiology, and metabolic outcomes under nitrate-respiring conditions in order to assess the feasibility and constraints of using nitrate as an alternative electron acceptor.

Interestingly, nitrate and nitrite are key compounds in the natural nitrogen cycle. They are produced by nitrifying microorganisms and consumed by plants [25,26]. A BLSS attempts to reproduce the nitrogen cycle in artificial ecosystems. For example, BLSS designs typically include a compartment that allows ammonium (NH₄⁺) to be converted to nitrate, which is then used for plant growth [2,27]. In this context, aerobic cultivation of C. necator could potentially be replaced by nitrate-respiring cultivation, thereby reducing the overall oxygen demand supplied by plants. It is thus necessary to determine the best conditions for nitrate respiration and then to test its effect on the biomass composition.

Therefore, the aim of this work was to investigate the potential of C. necator as a nitrate utilizer for biomass production. First, optimum pH and aeration conditions were determined for nitrate utilization. Finally, nitrate-limited or carbon-limited continuous culture experiments were performed to characterize the biomass production and composition at steady-state. Two carbon sources (glucose and acetate) were tested in continuous cultures. Acetate was selected because it is the main VFA present in anaerobic digestate.

2. Materials and Methods

2.1. Strain and Culture Media

2.1.1. Strains

Wild-type strains C. necator CECT 4623 (Spanish Type Culture Collection, Universitat de Valencia, Valencia, Spain) and DSM 541 (DSMZ-German Collection of Microorganisms and Cell cultures GmbH, Leibniz Institute, Braunschweig, Germany) were used in the present work. C. necator CECT 4623 is a H16-derived, glucose-utilizing mutant, while C. necator DSM 541 is also derived from the wild-type strain H16 but is not able to synthesize PHA [28,29]. This strain was used to allow for an indicative comparison of growth behaviour in the absence of PHB synthesis. Indeed, although this strain is derived from H16, it was not obtained from CECT 4623, thus preventing the complete exclusion of the influence of other mutations.

2.1.2. Media

Tryptone Soy Broth (TSB, Becton Dickinson, Sparks, MD, USA) was used as a rich medium for the first steps of preculture. To prepare TSB agar plates, 20 g·L⁻¹ of agar was added to the TSB medium. Mineral medium 1 (MM1) was used for precultures and was previously described in [30]. Glucose (9.6 g·L⁻¹) and urea (1.6 g·L⁻¹) were used as carbon and nitrogen sources, respectively. Mineral medium 2 (MM2) was employed for bioreactor cultures and was prepared as described in [31]. For the batch experiments, 1800 mL of MM2 was prepared with glucose (9.6 g·L⁻¹) or fructose (9.6 g·L⁻¹), urea (1.6 g·L⁻¹), and NO₃^- (3 g·L⁻¹) as the carbon (C), nitrogen (N), and energy sources, respectively. Due to the large excess of glucose and urea quantified during Experiment C, their concentrations were then reduced to (4.9 g·L⁻¹) for glucose and (1 g·L⁻¹) for urea throughout all the remaining batch experiments (A, B, D, and E). For chemostat conditions, three independent 20 L tanks were prepared with MM2 containing urea (1 g·L⁻¹) and NO₃⁻ (3 g·L⁻¹). In addition, three 4 L tanks were prepared, one containing glucose (49 g·L⁻¹) and the others containing acetic acid (57 g·L⁻¹).

2.2. Inoculum Preparation

The inoculum was prepared according to [7].

2.3. Bioreactor Cultivation

The 2 L working-volume bioreactor set-up was composed as presented in [7], with some slight modifications. Nitrogen was used to flush the reactor and maintain a pO₂ < 0.5% saturation. The gas analysis allowed the quantification of N₂, CO₂, and O₂ in both the inlet and outlet of the reactor (MicroGC fusion, Inficon, Bad Ragaz, Switzerland). For the continuous-experiment set-up, fresh medium was added by peristaltic pumps (520Du, Watson Marlow, Spirax-Sarco Engineering plc, Cheltenham, UK) from sterile tanks and removed from the reactor with another pump to a waste tank.

2.4. Fermentation Strategy

A series of four batch cultures were performed in a bioreactor to determine the optimal nitrate utilization conditions for C. necator. First, three separate culture runs tested five pH values (7.0, 7.3, 7.5, 8.0, and 8.5) under anaerobic conditions with strain CECT 4623 on MM2 medium with glucose. These five conditions were tested in three cultures because some of them showed no growth or very slow dynamics. Microaeration was then studied again at pH 7.5 with C. necator CECT 4623. Anaerobiosis was performed with a continuous flow of N₂ in the liquid phase of the bioreactor, while microaerobiosis consisted of a continuous, limited flow of air, resulting in pO₂ < 0.5% throughout the culture. Finally, a culture with the PHA-deleted strain DSM 541 was tested anaerobically at pH 7.5 on MM2 with fructose. Figure 1 summarizes the fermentation strategy in each batch.

Under the conditions defined above, two chemostats using the strain CECT 4623 at low dilution rates (0.02 h⁻¹), with glucose or acetic acid as the sole C source and urea as the N source, were studied. After an approximately 50 h batch phase to consume all nitrate and nitrite, a 20 L tank of MM2 and a 4 L bottle containing the carbon source (glucose or acetic acid) were connected to the bioreactor. The flow rate of the 4 L tank pump containing the C-sources was set at 4 mL·h⁻¹, which corresponds to 10% of the total flow rate (40 mL·h⁻¹). This set-up allows the separate control of both nitrate and carbon flow. Therefore, multiple substrate limitations were attempted, specifically nitrate or carbon limitations. For each experimental condition, a steady-state was reached after at least a five-residence time, i.e., about 250 h at this dilution rate. The steady-states were then characterized by at least three samples. Finally, the next condition was set up by connecting new tanks to the bioreactor. Thus, each condition corresponded to a 20 L tank of culture medium 2 and a 4 L bottle of the carbon source, so they were all independent from each other. The experimental dilution rates were calculated by dividing the effluent flow rate by the volume of the liquid phase in the reactor at the end of the culture.

Table 1. Culture conditions tested in this study.

Mode	Batch
Experiment	A	B *	C	D	E
Strain	CECT 4623	CECT 4623	CECT 4623	CECT 4623	DSM 541
pH	7.0/7.3	7.5	8.0/8.5	7.5	7.5
Carbon source	Glucose	Glucose	Glucose	Glucose	Fructose
Nitrogen source	Urea	Urea	Urea	Urea	Urea
Aeration	Anaerobiosis	Anaerobiosis	Anaerobiosis	Microaerobiosis	Anaerobiosis
Mode	Chemostat
D (h−1)	0.02		0.02	0.02
Experiment	C1		C2	C3
Carbon source	Glucose		Acetic acid	Acetic acid
Nitrogen source	Urea		Urea	Urea
Limiting substrate	Nitrate and nitrite		Nitrate and nitrite	Acetic acid
Transient period duration (h)	286		359	505
Steady-state duration (h)	98		66	71

* Experimental condition performed in duplicate.

summarizes the conditions tested in this study for the batch and chemostat phases.

2.5. Analytical Procedure

2.5.1. Biomass Quantification and Characterization

The quantification of the biomass, protein, PHB, PHV, DNA, and RNA contents was performed according to [7]. The results are expressed in %CDW. In addition, flow cytometry was used to monitor cell viability with propidium iodide as described in [32].

2.5.2. Metabolite Quantification

The supernatant was used to quantify substrates and metabolites. Nitrate, nitrite, and acetic acid consumption and possible byproduct formation were quantified by high-pressure ionic chromatography (HPIC) using an ICS-3000 system (Dionex, Sunnyvale, CA, USA) equipped with a conductivity detector. IonPac AS11-HC analytical and AG11 guard columns equipped with an anion self-generating suppressor (ASRS^® 300, 4 mm), a carbonate removal device (CRD 200, 4 mm), a continuously regenerated anion trap column (CR-ATC) and a conductivity detector were used for anion quantification (Thermo Fisher Scientific, Waltham, MA, USA). The mobile phase was KOH solution (EGC II KOH cartridge, Thermo Fisher Scientific, Waltham, MA, USA) at a flow rate of 1 mL·min⁻¹. A concentration gradient was applied: 1 mM for 13 min, increasing to 60 mM over 32 min with two steps at 15 mM and 30 mM for times of 25 min and 35 min, respectively, then 60 mM for 5 min and finally decreasing to 1 mM over 15 min. Glucose was also quantified with HPIC using the ICS-3000 system (Dionex, USA) equipped with an amperometric detector. CarboPac PA1 analytical and CarboPac PA1 guard columns (4 × 250 mm) (4 × 50 mm) (Thermo Fisher Scientific, Waltham, MA, USA), followed by an amperometric detector composed of an electrochemical cell, a reference electrode, and a working electrode (pH–AgCl) (Dionex, USA), were used. The isocratic mobile phase was 200 mM NaOH, purchased concentrated (Fisher Chemical, Leicestershire, UK) and set at 1 mL·min⁻¹. In addition, the ammonium concentration was determined as described in [33]. Chromatographic data were collected and analysed with Chromeleon (version 6.80 SP4 Build 2361) software. External standards (acetate, nitrate, nitrite, glucose, and ammonium) were used from 1.25 mg·L⁻¹ to 20 mg·L⁻¹, at the beginning and the end of the sequence, to ensure the stability throughout the analysis.

Finally, urea was quantified as presented in [7].

2.6. Data Treatment

For the batch experiments, specific growth rates (µ_max) (h⁻¹) were determined as being the slope of Ln(OD) = f(t). The error on the µ_max was calculated as the standard deviation of the slope. Correlations between CDW and optical density measurements were determined for each batch experiment and were used to determine substrate (S) conversion yields to biomass. The state variables were plotted pairwise in a scatter plot within the considered period of time. A linear regression was applied; the conversion yields (Y_nitrate), (Y_nitrite), and (Y_glucose) (g_CDW·g_S⁻¹) correspond to its slope and the error to its standard deviation.

For the chemostat experiments, the CO₂ production rate (rCO₂) (Cmol·L⁻¹·h⁻¹) was calculated as described in [34]. Substrate consumption rates and carbon and nitrogen mass balances were determined by adaptation of the data treatment presented in [7]. Briefly, the dilution rate (D) (h⁻¹) was defined by the following equation (Equation (1)):

$$D={\displaystyle \frac{Q_{out}}{Vl}}$$

(1)

where Q_out is the flow rate at the outlet of the chemostat (L·h⁻¹) and V_l is the volume of the liquid phase in the bioreactor tank (L).

At steady-state, the consumption rates of carbon (rC) and nitrogen (rN = r_urea + rNH₄⁺) or production rates of total biomass (rX), catalytic biomass (rX_C) (biomass subtracted from PHAs), and PHA (rPHA) were determined according to the following equation (Equation (2)):

$$r_E={\displaystyle \frac{Q_{out}·{\left[E\right]}_{out}-Q_{in}·{\left[E\right]}_{in}}{V_l}}$$

(2)

For transitional regimes the following equation (Equation (3)) was used:

$$r_E={\displaystyle \frac{Q_{out}·{\left[E\right]}_{out}-Q_{in}·{\left[E\right]}_{in}}{V_l}}+{\displaystyle \frac{d(Q_{out}·{\left[E\right]}_{out})}{dt}}$$

(3)

where E represents any compound, [E]_in its concentration at the chemostat inlet (mol·L⁻¹), Q_in its inlet flow rate (L·h⁻¹), and [E]_out its outlet concentration (mol·L⁻¹). In this study, the double-feeding strategy used to separate the carbon source from the rest of the culture medium allowed for the distinction between Q_in and Q_out. The units Cmol and Nmol, which represent the amount containing one mole of carbon or nitrogen, were used to characterize the organic compounds (acetate, glucose, biomass, CO₂, and PHB) and the nitrogen-containing compounds (urea, ammonium, and biomass) [35].

The elemental composition of the catalytic biomass was C H_1.69 O_0.45 N_0.25, determined from the culture of C. necator under unlimited growth conditions [7]. The catalytic biomass yields (Ysx_c) in (Cmol_xc·Cmol_S⁻¹) and the total biomass yields (Ysx) (g_CDW·g_S⁻¹) were calculated with the equations below (Equations (4) and (5)); in the latter case, the rates were expressed in (g·L⁻¹·h⁻¹):

$$Y_{S{X}_C}=-{\displaystyle \frac{r_{X_C}}{r_C}}$$

(4)

$$Y_{SX}=-{\displaystyle \frac{r_X}{r_C}}$$

(5)

Finally, the specific rates (q) were determined, and the carbon and nitrogen mass balances were checked as presented in [7].

3. Results

3.1. Determination of the Nitrate Fermentation Conditions

3.1.1. Optimum pH

Three batch cultures (A, B, and C) (Table 1) were performed at different pHs to characterize the growth parameters: the maximum specific growth rate (µ_max) and the conversion yields of nitrate, nitrite, and glucose into biomass (Y_nitrate, Y_nitrite, and Y_glucose). Biomass composition was also quantified in terms of proteins, DNA, RNA, PHB, and PHV over time. Figure 2 shows the kinetic results of the experiments described below, and

Table 2. Growth parameters and average macromolecular composition of C. necator grown in batch cultures with two strains at different pH and aeration conditions.

Strains	Aeration	Experiment	pH	µnitrate	µnitrite	Ynitrate	Ynitrite	Yglucose	Proteins	PHB	DNA	RNA	Sum of the Biomass Components
				(h−1)	(h−1)	(gCDW·gnitrate−1)	(gCDW·gnitrite−1)	(gCDW·gglucose−1)	(%CDW)	(%CDW)	(%CDW)	(%CDW)	(%CDW)
CECT 4623	Anaerobiosis	A	7.0	0.0044 ± 0.0002	/	0.27 ± 0.04	/	/	22 ± 1	29 ± 6	3.7 ± 0.3	1.4 ± 0.2	56 ± 6
			7.3	0.0073 ± 0.0001	0.012	0.31 ± 0.04	/	0.35 ± 0.02	27 ± 1	37 ± 2	3.6 ± 0.3	0.8 ± 0.1	68 ± 7
		B *	7.5	0.036 ± 0.001	0.050 ± 0.003	0.11 ± 0.00	0.26 ± 0.08	0.29 ± 0.01	42 ± 6	28 ± 12	3.7 ± 0.9	1.0 ± 0.2	75 ± 5
		C	8.0	0.012 ± 0.001	0.009	0.14 ± 0.01	/	0.30 ± 0.04	35 ± 4	23 ± 9	2.1 ± 0.3	1.4 ± 0.5	62 ± 9
			8.5	0	/	/	/	/	/	/	/	/	/
	Microaerobiosis	D	7.5	0.104 ± 0.013	0.027	0.24 ± 0.02	0.52 ± 0.00	0.29 ± 0.02	49 ± 5	32 ± 7	3.1 ± 0.7	2.0 ± 0.2	86 ± 3
DSM 541	Anaerobiosis	E	7.5	0.033 ± 0.001	0	0.05 ± 0.01	0	/	/	/	/	/	/

* Experimental condition performed in duplicate.

summarizes the kinetic parameters and biomass composition results. Culture A consisted of two different periods (Figure 2A). The first, performed at pH 7.0 from 0 h to 180 h, showed very slow growth dynamics (µ_nitrate = 0.0044 ± 0.0002 h⁻¹); the pH was then changed to 7.3 for the rest of the experiment. The specific growth rates increased at µ_nitrate = 0.0073 ± 0.0001 h⁻¹ and µ_nitrite = 0.012 h⁻¹ at pH 7.3.

The Y_nitrate was found to be the highest in the study at 0.27 ± 0.04 g_CDW·g_nitrate⁻¹ and 0.31 ± 0.04 g_CDW·g_nitrate⁻¹ at pH 7.0 and 7.3, respectively. The average biomass composition for both pH conditions is shown in Table 2. Interestingly, the total biomass composition (protein + PHB + nucleic acids) was found to be low at the beginning of the cultivation and increased over time (from 56% CDW to 68% CDW). Similarly, PHB content tended to increase over time, while DNA and RNA contents decreased.

At pH 7.5, the growth dynamics were faster and showed a diauxic profile (Figure 2B). After around 20 h of lag phase, cell growth started on nitrate, which was converted into nitrite. Total nitrate consumption was achieved at 65 h, and a second growth phase powered by nitrite consumption was observed until 100 h. The specific growth rates were µ_nitrate = 0.036 ± 0.001 h⁻¹ and µ_nitrite = 0.050 ± 0.003 h⁻¹. In terms of biomass yields, Y_nitrate = 0.11 ± 0.00 g_CDW·g_nitrate⁻¹, Y_nitrite = 0.26 ± 0.08 g_CDW·g_nitrite⁻¹, and Y_glucose = 0.29 ± 0.01 g_CDW·g_glucose⁻¹ were obtained. At the end of the nitrate growth phase, the biomass was composed of 48% of CDW protein, 20% CDW PHB, and 5.1% CDW nucleic acids. At the end of the nitrite growth phase, the protein content decreased to 34% CDW, the PHB content increased to 43% CDW, and the nucleic acids dropped to 3.4% CDW. The average values obtained for all the samples of this experiment are shown in Table 2.

Culture C was also composed of two distinct periods (Figure 2C). The first, at pH 8.5 from 0 h to 60 h, was characterized by an absence of cell growth, an absence of nitrate and nitrite consumption, and by an increase in cell membrane permeability that reached 35% of the total population instead of less than 5% for the other cultures. Thus, the decision to decrease the pH to 8.0 was made. Then, from 60 h to 137 h, a first phase of cell growth with nitrate consumption and nitrite production was observed. This was followed by a second phase of cell growth on nitrite. The two µ_max were calculated: µ_nitrate = 0.012 ± 0.001 h⁻¹ and µ_nitrite = 0.009 h⁻¹. The yields were determined and are presented in Table 2. Protein content increased during nitrate consumption and reached a maximum of 41% CDW, then decreased to 32% CDW at the end of the experiment. Nucleic acids accumulated during the nitrate phase to a maximum of 4.5% CDW before decreasing and stabilizing at 3% CDW for the entire nitrite phase. PHB content increased continuously, from 11% CDW to 31% CDW (Table 2).

In all cases, glucose, urea, and ammonium were quantified in excess, and no PHV was detected at any time. Interestingly, the urea and NH₄⁺ concentrations remained stable over time, with ammonium concentrations not exceeding 11 Nmmol·L⁻¹ regardless of pH. Under anaerobiosis, pH 7.5 clearly led to the highest µ_max on both the nitrate and nitrite phases of diauxic growth, with µ_nitrate < µ_nitrite. This condition was repeated in a final batch culture and showed comparable results in terms of specific growth rates (µ_nitrate = 0.042 ± 0.003 h⁻¹ and µ_nitrite = 0.050 ± 0.000 h⁻¹) and biomass composition. It was therefore chosen as the condition for the next steps of the study.

3.1.2. Anaerobiosis vs. Microaerobiosis

Figure 2D presents the kinetic data for the batch experiment at pH 7.5 under microaerobiosis. The nitrate and nitrite phases were characterized with µ_nitrate = 0.104 ± 0.019 h⁻¹ and µ_nitrite = 0.027 ± 0.001 h⁻¹. Due to the high µ_nitrate, the overall culture time was significantly reduced compared to the same pH conditions under anaerobiosis, from 100 h to 45 h. Glucose, urea, and ammonium, i.e., carbon and nitrogen, were in excess as expected. Unlike previous cultures, NH₄⁺ concentrations increased throughout the cultivation period, reaching 17 Nmmol·L⁻¹ and 43 Nmmol·L⁻¹ at the end of the nitrate and nitrite growth phases, respectively. The biomass yield on glucose (Y_glucose) was 0.29 ± 0.02 g_CDW·g_glucose⁻¹, which is similar to the value obtained under anaerobic conditions. However, the yields on nitrate and nitrite were approximately twice as high as in anaerobiosis. As before, the protein, DNA, and RNA contents decreased over time (from 54% CDW to 46% CDW and from 6% CDW to 4% CDW, respectively) while the PHB content increased (from 24% CDW to 33% CDW). Our results indicate that C. necator can simultaneously utilize oxygen and nitrate as terminal electron acceptors under the tested conditions. To our knowledge, such concurrent oxidative and nitrate respiration has not been explicitly described for C. necator in the literature. Although the underlying regulatory and enzymatic mechanisms were not investigated here, this observation suggests a degree of respiratory flexibility allowing partial substitution of oxygen by nitrate.

Microaerobiosis led to the highest µ_max obtained in this study, for the nitrate growth phase at pH 7.5. Interestingly, microaerobiosis cled to µ_nitrate higher than µ_nitrite, which is the opposite of the anaerobic condition. Since the microaerobic condition had a lower µ_nitrite than the anaerobic condition, it was not retained for the following stages of the study.

3.1.3. PHA Synthetic Metabolism

Finally, the necessity of PHA synthetic metabolism for nitrate respiration was tested. One final batch experiment was conducted with C. necator DSM 541, a strain that cannot synthesize PHA, under anaerobic conditions at a pH of 7.5. The kinetic data are presented in Figure 2E. A growth phase occurred on nitrate with a yield of 0.05 ± 0.01 g_CDW·g_nitrate⁻¹ and nitrite production. After reaching 2 g·L⁻¹ of nitrites, cell growth did not resume. On the contrary, biomass slowly decreased with nitrite concentration. This was accompanied by pyruvate excretion up to 1 g·L⁻¹ and an increase in cell permeability (27%). These results illustrate the necessity of PHA synthesis for the cell to ensure nitrite respiration.

3.2. Continuous Cultures

The conditions established above (pH 7.5 under anaerobiosis with strain CECT 4623) were applied to the chemostat experiments. Three steady-states were obtained: one on glucose and two on acetic acid as the carbon source. The objective was to study the effect of nitrate respiration on biomass composition under nutrient-unbalanced conditions. The dilution rates were 0.023 h⁻¹, 0.021 h⁻¹, and 0.024 h⁻¹ for experiments C1, C2, and C3, respectively. The kinetic data are shown in Figure 3. The nitrogen source was in excess in the form of urea and ammonium during all steady-states (on glucose, the total nitrogen concentration in the culture broth was 14.7 ± 0.4 Nmmol·L⁻¹, and on acetic acid it was 14.8 ± 0.3 Nmmol·L⁻¹ and 20 ± 3 Nmmol·L⁻¹). Urea accounted for the vast majority of excess nitrogen in the culture broth. Ammonium concentrations never exceeded 1 Nmmol·L⁻¹. Similarly, high concentrations of residual carbon were quantified during the first two steady-states (glucose = 0.45 ± 0.03 g·L⁻¹ for C1 and acetic acid = 1.6 ± 0.5 g·L⁻¹ for C2), but neither nitrate nor nitrite was detected, meaning that the reducing power was the limiting substrate in these cases. In contrast, for the final steady-state (C3), nitrate and nitrite were quantified in excess with concentrations of 5.3 ± 0.9 mmol·L⁻¹ and 1.0 ± 0.1 mmol·L⁻¹, respectively. Acetic acid was not detected, indicating that carbon limitation was reached. However, it was difficult to achieve perfect stability even after 500 h, as shown by the evolution of the total nitrogen concentration. Thus, it was the only rate calculated with (Equation (4)).

In addition, the biomass composition in proteins, DNA, RNA, and PHB was determined. The results presented in

Table 3. Growth parameters and average macromolecular composition of C. necator grown in continuous cultures at steady-state with different limiting substrates.

Experiment	D	Carbon Source	Limiting Substrate	CDW	Ysx	Ysxc	Protein	PHA	DNA	RNA
	(h−1)			(g·L−1)	(gCDW·gS−1)	(Cmolxc·cmolS−1)	(%CDW)	(%CDW)	(%CDW)	(%CDW)
C1	0.023	Glucose	Nitrate and nitrite	1.6 ± 0.1	0.38 ± 0.02	0.43 ± 0.03	33 ± 2	57 ± 1	3.9 ± 0.4	1.4 ± 0.2
C2	0.021	Acetic acid		0.94 ± 0.08	0.29 ± 0.03	0.33 ± 0.02	29 ± 1	57 ± 3	3.6 ± 0.3	0.80 ± 0.1
C3	0.024		Acetic acid	0.59 ± 0.02	0.21 ± 0.01	0.24 ± 0.01	39 ± 2	45 ± 2	2.2 ± 0.4	0.40 ± 0.01

correspond to the mean values and their standard deviations for at least three samples collected at steady-state. In all cases, the biomass was composed of a mixture of proteins and PHB. Nitrate and nitrite being used as the limiting substrate resulted in a greater accumulation of PHB than that of the acetic acid-limited chemostat, with 57% CDW and 45% CDW, respectively. As expected with glucose and acetic acid, no PHV was detected. The total nucleic acid contents were found to be 4.5 ± 0.7% CDW for C1, 5.1 ± 1.2% CDW for C2, and 2.6 ± 0.4% CDW for C3.

Using both online and offline data, the following were determined: instantaneous qC/qN ratios; carbon-to-biomass yields (Ysx) and (Ysx_c) (Table 3); carbon (qC)- and nitrogen (qN)-specific consumption rates; CO₂ (qCO₂)- and PHB (qPHB)-specific production rates; and carbon and nitrogen balances (

Table 4. Nutritional ratios, specific rates, and mass balances of the three steady-states.

Experiment	D	C/N	qC/qN	qN	qC	qCO2	qPHB	Carbon Balance	Nitrogen Balance
	(h−1)			(Nmmol·g−1·h−1)	(Cmmol·g−1·h−1)			(%)	(%)
C1	0.023	8.2	23 ± 5	−0.20 ± 0.02	−4.7 ± 0.6	1.9 ± 0.2	1.5 ± 0.1	88 ± 1	108 ± 10
C2	0.021	8.7	29 ± 4	−0.20 ± 0.02	−5.4 ± 0.5	2.7 ± 0.4	1.4 ± 0.2	93 ± 2	99 ± 9
C3	0.024	2.8	6 ± 1	−1.1 ± 0.2	−6.8 ± 0.2	6.0 ± 0.1	0.91 ± 0.07	114 ± 8	79 ± 4

). For all the steady states, the mass balances were satisfactory and closed within 21%, meaning that all substrates and products were accounted for. This gap could be explained by the low quantities of biomass, urea, ammonium, PHB, and CO₂ that were measured.

4. Discussion

4.1. Effects of pH and Aeration on C. necator Anaerobic Growth

A series of five batch cultivations were carried out to determine the optimal kinetic parameters for nitrate utilization by C. necator. Firstly, the anaerobic-growth kinetics started with a consistent lag phase regardless of pH. This is probably explained by the use of aerobically grown precultures [22]. Cells needed time to transition from O₂ to nitrate respiration. Subsequently, as expected under these conditions, diauxic growth was reported, confirming proteome adaptation to nitrite reduction after total nitrate consumption [36]. Under anaerobic conditions, the best µ_max was found at pH = 7.5, with a sharp decrease in µ_max values at higher and lower pHs. These results are consistent with the findings of [22], who determined an optimal pH range from 7.5 to 8.0 in heterotrophy. In terms of yields, the behaviour of C. necator varied. Y_nitrate was found to be lower at pH 7.5. Due to the small number of sampling points during the nitrite growth phase, Y_nitrite could only be calculated at pH 7.5 under anaerobiosis and microaerobiosis. The lower Y_nitrate at pH 7.5 may be explained by increased carbon flux toward PHB (reaching 20%CDW), reducing biomass yields.

Microaerobic conditions yielded the best growth parameters in terms of µ_nitrate and yields, compared to anaerobiosis at the same pH. Only µ_nitrite was found to be lower than in anaerobiosis. According to [22], O₂ represses the formation of the denitrifying system (nitrite reductase), which could explain why, in this case, µ_nitrate was higher than µ_nitrite. Similarly, Pfitzner and Schlegel [22] found that ammonium repressed the expression of nitrate reductase. Thus, the use of urea as a nitrogen source limited the effect of high NH₄⁺ concentrations during nitrate growth. This is supported by the strong regulation of ammonium concentration by cells under anaerobic conditions reported in this study. To our knowledge, this is the first report of simultaneous oxidative and nitrate respiration in C. necator. However, this phenomenon, also called aerobic denitrification, has been well characterized in other denitrifiers [37,38]. A direct link between dissolved oxygen concentration, growth rate, and yields has been established for Pseudomonas stutzeri [39].

4.2. Effects of the pH and Aeration on C. necator Biomass Composition

In batch cultures, the composition of the quantified total biomass (sum of the protein, PHB, DNA, and RNA contents) was influenced by the culture conditions (Table 2). Furthermore, it increased over time during the kinetics, from 46% CDW (Figure 2C) to 81% CDW (Figure 2B). This means that the unquantified cell components, namely carbohydrates and lipids that form cell walls and membranes, reached 40–50% CDW instead of 10–20% CDW, as usually determined in C. necator [40]. One possible explanation for this phenomenon is an increase in cell size over time. An increase in forward-scatter width (FSC-W) and forward-scatter area (FSC-A), which are known to correlate with the size of the cells, was measured by flow cytometry over time for the kinetics (A, B, and C) (Supplementary Materials Figure S1). In the case of bacilli-shaped bacteria, it has been shown that the surface-to-volume ratio decreases with increasing width [41]. Thus, the surface-to-volume ratio decreased over time during our kinetics, which could have led to a decrease in the proportion of the cell surface constituents (lipids and carbohydrates) relative to the protein, PHB, DNA, and RNA present inside the cell. Harris and Theriot (2018) and R. H. Pritchard (1974) have already suggested this idea [41,42], and it has been quantitatively measured in Bacillus megaterium [43]. Cell size is known to be influenced by nutrient availability and cell cycle progression, although work is still needed to characterize all the physiological processes involved in cell size regulation and, notably, the repartition between lipid and carbohydrate accumulation [44,45,46]. In our case, non-optimal pH conditions may have triggered this phenomenon.

Furthermore, PHA synthesis is a key element of nitrate respiration metabolism, and several lines of evidence support this interpretation. First, the biomass composition evolved in the same way in batch culture, regardless of the condition. A decrease in protein and nucleic acid content was measured over time in favour of an increase in PHB. In the chemostat at steady-state, the biomass was also mainly composed of a mixture of protein and PHB. Proteomic analyses comparing aerobic and anaerobic conditions with C. necator H16 suggested a strong link between anaerobiosis and PHB synthesis [36]. Our experiment with the PHA-negative strain DSM 541 demonstrated the necessity of PHB to perform complete nitrate respiration. Because nitrate reduction showed comparable µ_max between strains 4623 and 541 (Table 2), this step still worked even with a lower yield. However, the absence of PHB synthesis resulted in the absence of nitrite reduction and, therefore, no growth. To our knowledge, this is the first time that a direct correlation link between PHB synthesis and anaerobic respiration has been reported in C. necator. Nevertheless, the importance of PHB accumulation to perform nitrate respiration has been shown in anaerobic and aerobic conditions with other microorganisms. Interestingly, ref. [47] reported that electron flux coming from PHA production or consumption powered denitrification in Accumulibacter phosphatis. Similarly, detailed biochemical pathways of Pseudomonas stutzeri under aerobic and anaerobic denitrification were proposed by [39]. In this case, microbial stoichiometry and carbon flux analyses revealed that the carbon source and PHB stock played a critical role in electron balance in nitrate respiration [39,48].

4.3. Consideration for BLSS Integration

Finally, the interest of using anaerobic nitrate respiration with C. necator can be discussed for a spatial application in a life support system. To date, this is one of the few studies describing chemostat experiments under anaerobic nitrate respiration with C. necator. Three different steady-states were obtained, and their biomass compositions were characterized. Only two continuous culture experiments under these conditions have been reported. Both were performed with an ecological biotechnology approach to remove nitrate and nitrite pollutants from wastewater, one in autotrophy [49] and the other in heterotrophy using a complex medium in membrane bioreactors [50]. Therefore, it is difficult to compare those results. Moreover, the biomass consisted of a mixture of protein and PHB, regardless of nutrient limitation and carbon source. The C/N ratio of nutrients has been shown to be a key parameter influencing PHB production [17]. In this case, it was possible to reduce the PHB content from 57% CDW to 45% CDW in favour of protein accumulation by adjusting the C/N ratio in the culture medium (Table 4). Examination of the qC/qN, i.e., the ratio of the specific consumption rates of carbon and nitrogen, shows that only C3 showed a qC/qN below 10, known to favour SCP production. Further optimization of the C/N ratio toward SCP production seems difficult, as it would lead to a reduction in biomass quantity, which was already very low. Continuous experiments showed lower nucleic acid accumulation under anaerobiosis than aerobiosis, approaching the 2% target for food application [51]. This result was confirmed in our batch cultures, where DNA and RNA levels decreased over time, showing less nucleic acid accumulation than under aerobic conditions [40].

Anaerobic nitrate respiration also revealed some challenges compared to oxidative respiration. Reference [36] showed that all intermediate compounds were produced by C. necator, leading to the production of toxic gaseous oxides, but all these gases were consumed by the cells until total reduction into N₂. However, only a genetically engineered strain, deleted for the nitrous oxide reductase (nor), showed N₂O accumulation inside the gas phase of the bioreactor [50]. Furthermore, under anaerobic conditions on glucose, the total biomass yield (Ysx) in chemostat cultivation (0.43 g_CDW·g_glucose⁻¹) was higher than in batch culture (0.29 g_CDW·g_glucose⁻¹), which is encouraging. However, operating chemostats anaerobically rather than aerobically caused a substantial drop in catalytic biomass yields (Ysx_c). This trend was further confirmed in chemostat experiments with C. necator, where total biomass yields were lower: under similar conditions of strain and nutrient limitation, the Ysx_c were 0.46 ± 0.02 Cmol_xc·Cmol_glucose⁻¹ and 0.43 ± 0.02 Cmol_xc·Cmol_{acetic acid}⁻¹ on glucose and acetic acid, respectively [8]. In terms of biomass productivity, values ranging from 0.037 gCDW·L⁻¹·h⁻¹ (C1) to 0.020 gCDW·L⁻¹·h⁻¹ (C2) to 0.014 gCDW·L⁻¹·h⁻¹ (C3) were obtained. In aerobiosis at low dilution rate, biomass productivities of 0.19 gCDW·L⁻¹·h⁻¹ and 0.24 gCDW·L⁻¹·h⁻¹ were obtained on acetic acid and glucose, respectively [8]. In addition, the specific growth rates obtained remain very low compared to aerobic culture. Doubling times were significantly longer, with the best anaerobic growth rate observed in this study being approximately five times lower than published aerobic heterotrophic values (μ = 0.24 h⁻¹) [40]. These results could be expected because anaerobic nitrate respiration yields around 35% to 40% less ATP compared to aerobic respiration [52], which is combined with the drastic inhibitory effect of nitrite accumulation on growth [23]. Nitrate respiration involves a trade-off between reduced oxygen demand and slower growth rates, which would require larger reactor volumes or longer cultivation times. While this limitation is critical for SCP production, where high biomass productivity is required, it may be more acceptable for PHB production, in which polymer accumulation rather than rapid growth is the primary objective. Due to the mix of SCP and PHB, it could be interesting to directly use this biomass as a 3D-printable material. This strategy was proven to reduce production costs by removing the PHB purification step, which would be very advantageous for an application in space [53]. An increase in the compostability and varying mechanical properties was determined with PHB-rich biomass material, depending on the blend ratio between PHA and biomass [54,55].

Given that anaerobic nitrate respiration would significantly reduce biomass production and require larger reactor capacities or longer culture times, the results obtained in this study represent a significant limitation for the practical applications of BLSS. One of our main findings was the identification of these constraints, highlighting the need for further research into different approaches to reducing oxygen demand while maintaining acceptable growth performance, such as aerobic nitrate respiration.

5. Conclusions

In conclusion, C. necator’s nitrate utilization conditions were found to be optimal at a pH of 7.5 under anaerobiosis. These results provide a proof of principle that nitrate can substitute for oxygen as a terminal electron acceptor in C. necator, with growth parameters substantially reduced compared to aerobic conditions. PHB was a key component of this metabolism, and the biomass always consisted of a mixture of protein and PHB. Urea proved to be an ideal nitrogen source under these conditions, allowing the cells to tightly regulate ammonium concentrations. For the first time, simultaneous oxidative and nitrate respiration, also known as aerobic denitrification, was characterized in C. necator and led to improved growth parameters. Further research is needed to optimize this approach, lowering oxygen demand while preserving acceptable growth performance. Future work should quantify N₂ production directly and assess whether the reduced growth rates are offset by O₂ conservation benefits.