Characterization of a designed synthetic autotrophic–heterotrophic consortia for fixing CO₂ without light

Abstract

Microbial interactions are important for metabolism, and they improve metabolic substrate types and metabolic efficiency. To discover microbial combinations with high CO₂ fixation efficiencies, a series of synergistic microbial consortia of increasing diversity and complexity were devised using chemoautotrophic strains, including Ochrobactrum, Stenotrophomonas, Castellaniella, and Sinomicrobium strains, which were isolated from a non-photosynthetic microbial community (NPMC) with CO₂ fixation capacity. Addition of a small inocula of NPMC universally improved the CO₂ fixation efficiencies of the consortia by up to 10-fold, while the CO₂ fixation efficiencies of most multimember consortia were similar to those of single strains. An analysis of the microbial community structure revealed that both autotrophic–autotrophic microbial interactions and autotrophic–heterotrophic microbial interactions occurred in the synthetic microbial consortia. Ochrobactrum and Castellaniella strains were crucial for autotrophic metabolism, while Lysinibacillus and Pseudomonas strains were crucial for heterotrophic metabolism. These devised microbial consortia have potential applications in addressing environmental issues.

---

1. Introduction

CO₂ fixation by microorganisms is considered to be a potential technology for CO₂ capture, utilization, and sequestration, which could be significant to carbon resource exploitation.^1,2 Non-photosynthetic microorganisms (namely chemoautotrophic microorganisms) exhibit a wider range of physiological and ecological traits than photosynthetic microorganisms. As a result, they have a wider range of applications and far-reaching effects on CO₂ fixation, especially in bioreactors and soil environments without light.^3–5

The available energy supply is the most important factor that limits the growth of photosynthetic and non-photosynthetic carbon-fixing microorganisms.^6,7 Energy sources for CO₂ fixation by chemoautotrophic microorganisms include many types of inorganic compounds (namely available electron donors), such as hydrogen, nitrogen (like NH₄⁺ and NO₂⁻), sulfur (like S²⁻, S⁰, S₂O₃²⁻), phosphite, and metal species (like Fe²⁺ and Mn²⁺).⁸ Some of these are pollutants, of which H₂S is an industrial waste.⁹ Thus, CO₂ fixation by non-photosynthetic microorganisms may have dual benefits, including CO₂ fixation and environmentally friendly waste treatment. It has been reported that mixed electron donors that are composed of many reductive inorganic compounds could improve the CO₂ fixation efficiency of a non-photosynthetic microbial community.¹⁰ This effect was only found in mixed microbial communities, whereas it had no effect on the CO₂ fixation efficiency of individual species.¹⁰

Microbial interactions between different strains, such as syntrophy, are common in nature. These interactions can promote metabolism and the metabolic efficiency of an entire microbial community, that is, the availability and efficient use of resources of the entire microbial community will be high and the community can effectively adjust its metabolism according to a change in the environmental conditions.^11–13 The detailed effects of microbial interactions are mainly as follows: providing nutrients for microorganisms, promoting thermodynamic conversion efficiency (keeping the overall reaction exergonic to support bacterial growth by simultaneously carrying out reactions with positive and negative free energy changes), increasing the number of metabolizable substrates, and eliminating toxic microbial substances.¹¹

Based on these microbial interactions, many artificial mixed microbial communities have been devised. A fungal–bacterial community was used to degrade cellulosic biomass to produce isobutanol.¹⁴ A co-culture of engineered yeast and the cellulolytic bacterium Actinotalea fermentans degraded switchgrass, corn stover, and sugarcane bagasse to produce methyl halide.¹⁵ These microbial communities have great potential applications in global health, energy, and environmental issues.¹⁶

In our previous experiments, a mixed non-photosynthetic microbial community (NPMC) with CO₂ fixation capacity, using NO₂⁻, S²⁻, and S₂O₃²⁻ as electron donors, was obtained from oceans near France, Thailand, China, Japan, the South Pole, the North Pole, and the Equator.¹⁷ NPMC possesses richer metabolic pathways and higher CO₂ fixing efficiencies than single strains because of the interactions among many different strains, even including uncultured microorganisms. The maximum CO₂ fixing efficiency of the NPMC under aerobic and anaerobic conditions was approximately 2.65 mmol min⁻¹ g⁻¹ protein and 3.14 mmol min⁻¹ g⁻¹ protein, respectively,¹⁰ which were far higher than the values of the endosymbiotic chemoautotrophs (approximately 50.8 μmol min⁻¹ g⁻¹ protein).¹⁸ Compared with photosynthetic microorganism, the maximum CO₂ fixing efficiency of NPMC was about 128 mg CO₂ L⁻¹ d⁻¹ that was slightly lower than the CO₂ uptake rate of the algae in reactor which could reach 162 mg CO₂ L⁻¹ d⁻¹.¹⁹ Considering that culturing the algae in reactor was conducted using a constant biomass concentration of 0.126 g L⁻¹, while NPMC was cultured using an initial biomass concentration of 0.64 mg C per L. This means that NPMC may own great potential for carbon sequestration. However, enriching microbial biomass, regulating the fermentation and collecting a microbial community that contains numerous strains, like NPMC, are difficult. A simplified microbial community composed of limited members might be a better choice for practical applications. Thus, in this study, multiple chemoautotrophic strains isolated from NPMC were combined in various ways to devise a synergistic microbial consortium with a high CO₂ fixing efficiency. The different responses of the dominant bacteria from different consortia were investigated to further optimize the consortia, and their application in other fields was explored by examining changes in the microbial community structure. This study will provide the foundation for future endeavors to develop better and more robustly engineered synthetic communities that can be used in industrial biotechnology applications to fix and reuse CO₂ efficiently, which was important to carbon resource exploitation.

2. Materials and methods

2.1 Non-photosynthetic CO₂ fixing strains

NPMC capable of fixing CO₂ was a mixed microbial community isolated from the surface seawater of four oceans around the world, including those near Australia, China, Japan, France, Thailand, the Equator, the South Pole, and the North Pole.¹⁷ Five chemoautotrophic strains that were isolated from the NPMC were deposited in the China Center for Industrial Culture Collection (CICC) were Ochrobactrum sp. WH-2 (CICC 23802, GenBank accession number: KM502331), Stenotrophomonas sp. WH-11 (CICC 23803, GenBank accession number: KM975671), Ochrobactrum sp. WH-13 (CICC 23804, GenBank accession number: KM975672), Castellaniella sp. WH-14 (CICC 23805, GenBank accession number: KM975673), and Sinomicrobium oceani WH-15 (CICC 23806, GenBank accession number: KM975674).

2.2 Autotrophic culture

The autotrophic culture medium contained (g L⁻¹): Na₂CO₃, 1.0; NaHCO₃, 1.0; (NH₄)₂SO₄, 5.0; KH₂PO₄, 1.0; K₂HPO₄, 2.0; MgSO₄·7H₂O, 0.2; NaCl, 20; CaCl₂, 0.01; and FeSO₄·7H₂O, 0.01. To the autotrophic culture medium, 2 mL of a trace element solution was added, which contained (mg L⁻¹): Na₂MoO₄·2H₂O, 1.68; H₃BO₃, 0.4; ZnSO₄·7H₂O, 1.0; MnSO₄·5H₂O, 1.0; CuSO₄·5H₂O, 7.0; CoCl₂·6H₂O, 1.0; and NiSO₄·7H₂O, 1.0. A total of 0.46% NaNO₂, 0.50% Na₂S₂O₃, and 1.25% Na₂S were added to the autotrophic culture medium as electron donors to provide adequate energy for the microorganisms in the experiments. Serum bottles (150 mL) containing 40 mL of the medium were prepared and autoclaved for 20 min at 121 °C. The bacterial inoculum concentration was 2.5% (v/v). The serum bottle was sealed with a silicon stopper and filled with 20% (v/v) CO₂ using a syringe (the air : CO₂ ratio was approximately 80 : 20). All of the cultures were incubated at 30 °C, with shaking at 120 rpm, without light for 96 h. The atmosphere within the serum bottles was readjusted to the initial ratios after culturing for 48 h.

2.3 Heterotrophic enrichment culture

A total of 2.5% (v/v) of a microbial culture (either single strain or NPMC) was inoculated into Luria–Bertani medium and incubated at 30 °C, with shaking at 120 rpm, for 48–96 h. Two blank samples were prepared for each heterotrophic enrichment process to examine whether the bacteria were contaminated.

2.4 Estimation of the CO₂ fixing efficiency

The total organic carbon (TOC) value, which reflects the microbial CO₂ fixing efficiency, was analyzed using a Shimadzu TOC-VCPH total organic carbon analyzer (Shimadzu Seisakusho Co., Ltd., Kyoto, Japan). To reduce the impact of inorganic carbon on the analyses, the pH of the sample was adjusted to about 4.0 prior to TOC analysis.

2.5 DNA extraction, PCR amplification of 16S rRNA genes, and pyrosequencing

Bacterial DNA was extracted using a PowerSoil DNA Isolation Kit (MO BIO Laboratories, Inc., Carlsbad, CA, USA) according to the manufacturer's instructions. The V1–V3 regions of the bacterial 16S rRNA genes were amplified using bacterial primers 8F (5′–AGAGTTTGATCCTGGCTCAG–3′) and 533R (5′–TTACCGCGGCTGCTGGCAC–3′).²⁰ The fused forward and backward primers included an 8-nucleotide barcode that was inserted between the Illumina adaptor and the sequencing primer. The barcodes were used to sort multiple samples in a single Illumina MiSeq (Illumina, San Diego, CA, USA) run.

PCR products were purified using a DNA gel extraction kit (Axygen, Hangzhou, China) and quantified using a TBS-380 Mini-Fluorometer (Turner BioSystems, Sunnyvale, CA, USA). Following quantitation, the products from different samples were mixed at equal ratios for pyrosequencing with the Illumina MiSeq platform. All the reads of samples, including consortium-I-N, consortium-II-N, consortium-I-H, and consortium-II-H have been submitted to the National Center for Biotechnology Information (NCBI) under the accession numbers SRR1642850, SRR1642849, SRR1642908, and SRR1642907, respectively.

2.6 Bioinformatics analysis

Data preprocessing was performed mainly using mothur software.²¹ The sequences were clustered into operational taxonomic units (OTUs) at a 97% sequence identity (furthest neighbor method). A similarity analysis, a heatmap genus-level analysis, and a VENN diagram were also conducted using mothur software. The heatmap revealed the distribution of the 50 most abundant bacterial genera among the samples. The relationship among the samples and the genera they contained was determined by Chao distance and the complete clustering method. The heatmap plot depicts the relative percentage of each bacterial genus (variables clustering on the vertical axis) within each sample (horizontal axis clustering). The intensity of the color scale indicates the relative abundance of each genus. Relative abundance was defined as the number of sequences affiliated with that genus divided by the total number of sequences per sample.

2.7 Statistical analysis

The correlations between the variables were assessed by 2-tailed Spearman's correlation test using SPSS version 23 (IBM, Armonk, NY, USA). Statistical significance is defined as P < 0.05.

3. Results and discussion

3.1 CO₂ fixing efficiency of the designed synthetic microbial consortia

The composition of the NPMC that could efficiently fix CO₂ is too complex, and it even contained many uncultured microorganisms. Compared with the NPMC, a devised synergistic microbial consortium composed of limited members could more easily achieve regulation and culture collection. To devise synergistic microbial consortia with high CO₂ fixing efficiencies, multiple chemoautotrophic strains were isolated from the NPMC, and five fast-growing strains were selected for further study. Using these five strains, a series of synergistic microbial consortia of increasing diversity and complexity was devised. The CO₂ fixing efficiencies of the microbial consortia are shown in Fig. 1.

Fig. 1 Comparative carbon fixation efficiency of the synthetic microbial consortia composed of different strains. Individual dashes in each column represent the average carbon fixation efficiency of each strain composition (the original data could be seen in Table S1†).

The maximum carbon fixation efficiency of the synthetic consortia that were composed of two strains was higher than that of the single strains. However, overall, the carbon fixation efficiency of most of the synthetic consortia that were composed of two or more strains (except those composed of the aforementioned five strains) was similar to that of the single strains (Fig. 1). The carbon fixation efficiency was not universally improved by increasing the number of strains in the consortia. This indicated that while there were positive interactions between strains, the number of positive interactions was small. Moreover, these positive microbial interactions were only observed in some of the synthetic consortia that were composed of two strains, and they were not observed in the synthetic consortia that were composed of three or more strains. With increasing members, the interactions could not play a role in the new consortium, or an unknown side effect offset the effect of the interactions. The self-inhibition effect of autotrophic microorganisms on CO₂ fixation²² may be the reason for the failure (or limited role) of positive interactions to increase the carbon fixation efficiency.

Taking into account that the NPMC indeed had a higher carbon fixation efficiency and richer microbial diversity than the consortia, there is certainly some mechanism by which positive interactions could increase the carbon fixation efficiency. The mechanism may involve a single strain or multiple strains. However, the probability of screening these strains from the NPMC may be very low, especially because such a strain(s) may be a symbiotic bacterium or unculturable and, therefore, could not be isolated individually. Thus, the addition of the NPMC at a low inoculum concentration not only enriched the number of bacterial species of the synthetic consortia, but it also retained the major roles of the isolated strains in the synthetic consortia.

The carbon fixation efficiency of the synthetic consortium containing the NPMC was universally higher than that of the synthetic consortium without the NPMC (Fig. 1). Thus, the NPMC played a role in CO₂ fixation by the synthetic consortia. Additionally, the consortium with the highest CO₂ fixation efficiency in 3-member consortia containing the NPMC was named as consortium-I-N (consortium-I containing the NPMC). Consortium-I was composed of Ochrobactrum sp. WH-2, Stenotrophomonas sp. WH-11, and Castellaniella sp. WH-14. The consortium with the highest CO₂ fixation efficiency in 4-member consortia containing the NPMC was named as consortium-II-N (consortium-II containing the NPMC). Consortium-II was composed of Ochrobactrum sp. WH-2, Castellaniella sp. WH-14, Ochrobactrum sp. WH-13, and Sinomicrobium oceani WH-15. The carbon fixation efficiencies of consortium-I-N and consortium-II-N were significantly higher than those of the other synthetic consortia. The carbon fixation efficiencies of consortium-I-N and consortium-II-N were 971 and 366% higher than those of consortium-I and consortium-II, respectively. Thus, the NPMC significantly affected the microbial interactions in consortium-I and consortium-II.

Although the addition of the NPMC increased the carbon fixation efficiency, its underlying mechanism is still unknown. Overall, its role may be mainly divided into two aspects: one was to enhance positive microbial interactions, and the other was to reduce negative microbial interactions. According to the growth rates of consortium-I-N and consortium-II-N, a low inoculum of the NPMC was more likely to reduce negative microbial interactions in consortium-I and consortium-II. CO₂ fixation by autotrophic microorganisms is negatively influenced by organic compounds,²³ including extracellular metabolites.²⁴ The results of the community structure of the NPMC indicated that the NPMC consisted of a variety of heterotrophic, and even obligate heterotrophic, microorganisms.¹⁷ The consumption of organic compounds by heterotrophic microorganisms may improve the CO₂ fixation of autotrophic microorganisms by reducing the inhibition of extracellular metabolites.¹⁷ Thus, a medium including organic carbon source was used to culture the NPMC to increase the ratio of heterotrophic to autotrophic microorganisms, and then these heterotrophic microorganisms were transferred into an inorganic medium, which may decrease the CO₂ fixation efficiency of the NPMC. To distinguish between the original NPMC and the NPMC that was obtained after the aforementioned culturing procedure, the heterotrophic culture was named NPMC-H.

When NPMC-H was cultured in organic medium, it almost completely lost the ability to fix CO₂ (Table 1); however, consortium-I containing NPMC-H (consortium-I-H) and consortium-II containing NPMC-H (consortium-II-H) still exhibited the similar CO₂ fixation efficiencies to consortium-I-N and consortium-II-N. This confirmed that heterotrophic microorganisms were the main factor responsible for improving the CO₂ fixation efficiencies, and that the inhibition of extracellular metabolites was the main reason for the limited carbon fixation efficiency of consortium-I and consortium-II. In addition, the net carbon fixation values (Table 1) of the bacterial components of consortium-I-H and consortium-II-H were both higher than the sum of the theoretical CO₂ fixation efficiency values of the microbial components. If the elevated carbon fixation values (beyond that of the sum of the individual strains) were considered to result from microbial interactions, then these interactions accounted for approximately 33% of the carbon fixation in consortium-I-H and consortium-II-H. Thus, consortium-I-H and consortium-II-H were the ideal synergistic microbial consortia for fixing CO₂.

Table 1 Percentage of net carbon fixation by the members in consortium-I containing NPMC-H (consortium-I-H) and consortium-II containing NPMC-H (consortium-II-H)^a

	Net carbon fixation (mg C per L)	Theoretical percentage of net carbon fixation by the members in consortium (%)
		Consortium-I-H	Consortium-II-H
a The inoculum concentrations of each strain and NPMC-H were about 2 mg C per L. The carbon fixation values resulting from microbial interaction in consortium-I-H and consortium-II-H were the TOC values of consortium-I-H and consortium-II-H minus the theoretical net carbon fixation efficiency of each strain.
Consortium-I-H	27.51 ± 0.14
Consortium-II-H	32.12 ± 0.26
NPMC-H	0	0	0
Ochrobactrum sp. WH-2 + NPMC-H	8.04 ± 0.45	29.23	25.03
Stenotrophomonas sp. WH-11 + NPMC-H	3.19 ± 0.45	11.60	Not contained
Castellaniella sp. WH-14 + NPMC-H	7.3 ± 0.27	26.54	22.73
Ochrobactrum sp. WH-13+ NPMC-H	4.78 ± 0.09	Not contained	14.88
Sinomicrobium oceani WH-15 + NPMC-H	1.56 ± 0.06	Not contained	4.86
Microbial interaction		32.64	32.50

---

A correlation analysis (Table S2 in ESI†) of the data in Fig. 1 was used to determine the role and weight of the microbial members of the consortia. The results showed that the TOC values of the consortia increased significantly with the addition of the NPMC (Spearman correlation = 0.764, P < 0.001). There was no negative or positive correlation between the strains in the consortia and the TOC values (all P values were greater than 0.1) in the absence of NPMC (Table S3 in ESI†). This indicated that the CO₂ fixation efficiencies of the strains were not additive because of the lack of positive interactions. Only the addition of Castellaniella sp. WH-14 improved the TOC of the consortia (Spearman correlation = 0.457, P < 0.001) in the presence of NPMC (Table S4 in ESI†). Thus, the interactions between Castellaniella sp. WH-14 and the other four strains were complicated, and they included negative and positive effects. However, the interactions between Castellaniella sp. WH-14 and the other four strains mainly had positive effects when the consortia contained NPMC. As Castellaniella sp. WH-14 was a member of consortium-I and consortium-II, the results demonstrate that it played an important role in the consortia. However, further studies will be needed to determine the role of Castellaniella sp. WH-14 in the consortium containing NPMC-H, as well as the crucial bacterial species of the NPMC and NPMC-H that engaged in synergistic interactions.

3.2 Microbial community structure analysis of the synthetic microbial consortia

The carbon fixation efficiencies of consortium-I and consortium-II containing the NPMC or NPMC-H were all far higher than those of the individual strains. These indicated that these synthetic microbial consortia not only synergized the carbon fixation capacity of their microbial members, but also led to many microbial interactions. Net carbon fixation by consortium-I-N was similar to that consortium-I-H (about 30 mg C per L). However, net carbon fixation by consortium-II-H was approximately twice that of consortium-II-N. This suggested that the acting bacteria in the NPMC and NPMC-H may differ. Next, to determine and optimize the functional bacteria of the synthetic microbial consortia, high-throughput sequence was used to study the microbial community structure of the synthetic microbial consortia.

The similarity between consortium-I-H and consortium-II-H was higher than that between consortium-I-H (or consortium-II-H) and consortium-I-N (or consortium-II-N) (Fig. 2a), indicating that the microbial community structures of the synthetic microbial consortia may mainly depend on the NPMC or NPMC-H. However, the inoculation concentration of the NPMC or NPMC-H accounted for a small fraction of the synthetic microbial consortia (Fig. 1 caption and Table 1), which implied that the growth of dominant strains in the NPMC and NPMC-H grew very fast and even exceeded that of the strains in consortium-I and consortium-II during carbon fixation by the synthetic microbial consortia. This may be the reason for the high ratios of the NPMC and NPMC-H in the microbial community structures at the end of culturing. In addition, NPMC-H was derived from the NPMC, and it could be assumed that the dominant bacteria that they contributed to consortium-I and consortium-II were the same. However, the similarity between consortium-I-H and consortium-I-N was lower than that between consortium-I-H and consortium-II-H, which was inconsistent with the aforementioned assumption. Thus, the dominant bacteria contributed by NPMC-H and the NPMC differed.

Fig. 2 (a) Similarity analysis of different consortia based on OTUs. The scale bar indicates the number of substitutions per site. (b) Heatmap analysis based on the genus distribution and abundance of samples (the figure shows the 50 most prominent OTUs at the genus level and the original data could be seen in Table S5†).

Fig. 2b shows the differences between the dominant genera in the synthetic microbial consortia. The relative abundances of the yellow and red sections will be mainly discussed, while other colors, such as green and blue, were not included because of their small order of magnitude and proportion in the microbial community structure. Overall, the dominant genera of the consortia containing NPMC-H were Ochrobactrum and Lysinibacillus, while the proportions of other bacteria were small, except that there was one more dominant genus, Pseudomonas, in consortium-II-H. The dominant genera of the consortia containing the NPMC were Ochrobactrum and Castellaniella; thus, Ochrobactrum occupied a dominant position in all of the consortia. The colors in Fig. 2b showed that the relative abundance of Ochrobactrum was about 80% of the total. The characteristics of the other three genera were as follows. In contrast to the type species of the genus Bacillus, the peptidoglycan of Lysinibacillus contains with lysine, aspartic acid, alanine, and glutamic acid.²⁵ Castellaniella is a Gram-negative, facultatively anaerobic, denitrifying genus, and it exhibits good growth after 24 h of incubation on nutrient agar at 25–30 °C.²⁶ Pseudomonas contains a great deal of metabolic diversity, and Pseudomonas species can use many different organic compounds as carbon and energy sources.⁸

Further analysis of the microbial community structures of these four consortia needs to consider the initial compositions of their microbial communities. The inoculation concentration of the NPMC in consortium-I-N was 3%, while the concentrations of the other three bacteria (Stenotrophomonas sp. WH-11, Ochrobactrum sp. WH-2, and Castellaniella sp. WH-14) were all approximately 32%. The inoculation concentration of the NPMC in consortium-II-N was 2%, while the concentrations of the other four bacteria (S. oceani WH-15, Ochrobactrum sp. WH-2, Ochrobactrum sp. WH-13, and Castellaniella sp. WH-14) were approximately 24%. The concentrations of NPMC-H and other bacteria in consortium-I-H and consortium-II-H were all approximately 25 and 20%, respectively. Consequently, Ochrobactrum and Castellaniella were both included in consortium-I and consortium-II, while Stenotrophomonas only existed in consortium-I, and Sinomicrobium only existed in consortium-II. The initial proportion of Castellaniella in the consortia ranged from 20 to 32%, and it decreased to 1% after 96 h of culturing in consortium-I-H and consortium-II-H. Additionally, the proportions of Stenotrophomonas and Sinomicrobium in the consortia were too low to be determined after 96 h of culturing. There are many examples in which the proportion of a strain in multimember syntrophic consortia varied significantly during culturing, and the proportions of crucial bacteria that take part in beneficial microbial interactions may also decrease (possibly to 0) by the end of culturing. As study of specific cross-feeding principles that drive the formation and maintenance of individuals within a mixed population by Mee et al.¹² showed that the proportions of some microbial members in a 13-member syntrophic consortium increased, while others decreased, during culturing. The proportions of many bacteria decreased to 0 after 24 h.¹² Compared with other genera, the proportion of Ochrobactrum in the four consortia increased to 80% from initial proportions of 25–48% after 96 h of culturing.

In summation, the response of bacteria in the NPMC and NPMC-H to consortium-I and consortium-II differed, which resulted in different dominant bacteria in these consortia. Ochrobactrum and Castellaniella were the main carbon fixers in the consortia containing the NPMC. There were many types of non-dominant genera in the consortia that contained the NPMC, which resulted in diverse interactions between the microbial genera, including interactions between the added strains and the autotrophic and heterotrophic bacteria in the NPMC. Ochrobactrum and Lysinibacillus (and Pseudomonas in consortium-II-H) were the main carbon fixers in the consortia containing NPMC-H. There were fewer non-dominant genera in the consortia containing NPMC-H compared with the consortia containing NPMC. Basically, NPMC-H could not fix carbon. This led to fewer interactions between microbial genera. The interactions of the consortia containing NPMC-H mainly occurred among strains of consortium-I and consortium-II and the heterotrophic bacteria in NPMC-H. Net carbon fixation by consortium-II-H was approximately twice that of consortium-II-N, indicating that the interactions between NPMC-H and the strains in consortium-II have enormous CO₂ fixation potential. The aforementioned interactions mainly occurred among the heterotrophic bacteria of NPMC-H. A study of these heterotrophic bacteria may be the key to developing microbial consortia that comprise heterotrophic bacteria that synergistically and efficiently fix CO₂.

3.3 Crucial strains of autotrophic–heterotrophic consortia for efficient CO₂ fixation

All the reads of NPMC-H (initial) and NPMC-H (cultured) have been submitted to the NCBI under the accession numbers SRR1642851 and SRR1642906, respectively. The similarity and overlap of the consortia are shown in Fig. 3a. The numbers of species in NPMC-H (initial), NPMC-H (cultured), consortium-II-H, and consortium-I-H were 2278, 1572, 1530 and 1568, respectively. As the number of initial species of consortium-II-H and consortium-I-H was only 3–4 more than that of NPMC-H (initial), NPMC-H (initial) could be approximately set as the initial species number. This suggested that the microbial community structure varied obviously when the carbon source was changed from organic carbon to CO₂. As a result, the microbial community diversity declined and the number of species decreased by 31–33%. The numbers of species shared between NPMC-H (initial) and NPMC-H (cultured), consortium-II-H, and consortium-I-H were 476, 345, and 369, respectively. That is, the microbial community structures of consortium-II-H and consortium-I-H changed more significantly than that of NPMC-H (cultured). There were 238 species that were common to all four consortia, which was approximately 10% of the initial number of species. This implied that in addition to the carbon source, microbial interactions (the response of bacteria in NPMC-H to the strains of consortium-I and consortium-II) played an important role in changing the microbial community structures. The number of species shared between consortium-II-H and consortium-I-H was 499, about 32–33% of the total. The result showed that the response of NPMC-H to consortium-I and consortium-II differed, that is, the microbial interactions in consortium-II-H and consortium-I-H were distinct.

Fig. 3 (a) VENN diagram of the consortia. (b) Heatmap analysis based on the genus distribution and abundance of the consortia (the original data could be seen in Table S5†).

Lysinibacillus accounted for 99% of the genera of NPMC-H (initial) after culturing with organic carbon, while Pseudomonas accounted for 0.6% (Fig. 3b). This suggested that the two genera consisted of heterotrophic bacteria. When NPMC-H was cultured with CO₂ as the sole carbon source, the TOC value did not change (Table 1). However, the microbial community structure of NPMC-H (cultured) differed significantly compared with that of NPMC-H (initial), that is, the proportion of Lysinibacillus decreased sharply, while the proportion of Ochrobactrum increased, and the proportion of Castellaniella increased to 2% from undetected. Thus, Ochrobactrum and Castellaniella might be related to microbial autotrophic metabolism, and Lysinibacillus might be related to heterotrophic metabolism. Compared with NPMC-H (cultured), the proportion of Lysinibacillus in consortium-I-H declined slightly, while the proportion of Ochrobactrum increased further, and the proportions of other genera were essentially unchanged. The proportion of Lysinibacillus in consortium-II-H decreased slightly, while the proportion of Pseudomonas increased substantially, and the proportions of other genera remained unchanged compared to NPMC-H (cultured).

The above results indicated that synergistic microbial consortium could be obtained using consortium-I, consortium-II, and Lysinibacillus (although consortium-II with Pseudomonas may be better). As NPMC-H was a mixed microbial community, its composition may be unstable and influence its effect. Determining the key heterotrophic strains in NPMC-H, like Lysinibacillus and Pseudomonas, could be favorable to maintain and promote the effect of NPMC-H on improving CO₂ fixation efficiency of autotrophic strains. The roles of Stenotrophomonas in consortium-I and Sinomicrobium in consortium-II may be understood by examining the dynamics of the microbial community structure. The results shown in Fig. 1 revealed that Stenotrophomonas and Sinomicrobium were both necessary for improving carbon fixation efficiency. The results shown in Fig. 3 indicated that they may play important roles in the early stage, and then their concentrations decreased gradually. The previous results suggested that inorganic compounds used as energy sources have great energizing potential.¹⁰ In the case of a sufficient energy source, the carbon fixation capacity of a synergistic microbial consortium may be improved by optimizing microbial community structures of consortium-I and consortium-II containing Lysinibacillus, as well as by increasing total bacteria concentration.

Moreover, as the extracellular metabolites of consortium-I and consortium-II could be used as carbon sources by heterotrophic bacteria, the further exploration of the components and concentrations of these extracellular metabolites may connect microbial CO₂ fixation with high-value fermentation processes, for example, many species of Lysinibacillus and Pseudomonas can produce biosurfactants that can be used to degrade hazardous materials (including alkanes and polycyclic aromatic hydrocarbons).^27,28 Lastly, the addition of autotrophic bacteria to heterotrophic bacteria may promote the autotrophic metabolism of the entire consortium, which is conducive to the application of autotrophic bacteria in soil remediation and for improving soil organic matter.

4. Conclusions

Consortium-I, which was composed of three strains, and consortium-II, which was composed of four strains, were found to be part of a devised synergistic microbial consortium that has CO₂ fixation potential. Addition of the NPMC at a low inoculum concentration stimulated the microbial interactions of consortium-I and consortium-II. Using NPMC-H, which was derived from the NPMC, the CO₂ fixation efficiencies of consortium-I and consortium-II were improved by approximately 900%. Lysinibacillus and Pseudomonas in NPMC-H were the major factors that played key roles in consortia CO₂ fixation by reducing the inhibition of extracellular metabolites from autotrophic microorganisms.

Acknowledgements

This work was supported by National Natural Science Foundation of China under Grant No. 21307093, No. 21577101, and No. 51408588; Shanghai Natural Science Foundation of China No. 16ZR1440000; Shanghai Municipal Education Commission No. 14ZZ091 and No. ZZSD15105; Youth Innovation Fund for Interdisciplinary Research of SARI No. Y526453235.

References

1. S. Judd, L. J. P. van den Broeke, M. Shurair, Y. Kuti and H. Znad, Water Res., 2015, 87, 356–366.
2. S. K. Bagchi and N. Mallick, RSC Adv., 2016, 6, 29889–29898.
3. Y. Shen, RSC Adv., 2014, 4, 49672–49722.
4. X. Wu, T. Ge, H. Yuan, B. Li, H. Zhu, P. Zhou, F. Sui, A. G. O'Donnell and J. Wu, Appl. Microbiol. Biotechnol., 2013, 98, 2309–2319.
5. X.-E. Long, H. Yao, J. Wang, Y. Huang, B. K. Singh and Y.-G. Zhu, Environ. Sci. Technol., 2015, 49, 7152–7160.
6. S. C. Pierobon, J. Riordon, B. Nguyen and D. Sinton, Bioresour. Technol., 2016, 209, 391–396.
7. I. A. Berg, Appl. Environ. Microbiol., 2011, 77, 1925–1936.
8. M. T. Madigan, J. M. Martinko, D. Stahl and D. P. Clark, Brock biology of microorganisms, Pearson Higher Education, New York, 13th edn, 2010.
9. J. B. M. Klok, P. L. F. van den Bosch, C. J. N. Buisman, A. J. M. Stams, K. J. Keesman and A. J. H. Janssen, Environ. Sci. Technol., 2012, 46, 7581–7586.
10. J. Hu, L. Wang, S. Zhang, X. Fu and Y. Le, Environ. Sci. Technol., 2010, 44, 6364–6370.
11. B. E. L. Morris, R. Henneberger, H. Huber and C. Moissl-Eichinger, FEMS Microbiol. Rev., 2013, 37, 384–406.
12. M. T. Mee, J. J. Collins, G. M. Church and H. H. Wang, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, E2149–E2156.
13. K. Brenner, L. You and F. H. Arnold, Trends Biotechnol., 2008, 26, 483–489.
14. J. J. Minty, M. E. Singer, S. A. Scholz, C. H. Bae, J. H. Ahn, C. E. Foster, J. C. Liao and X. N. Lin, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 14592–14597.
15. T. S. Bayer, D. M. Widmaier, K. Temme, E. A. Mirsky, D. V. Santi and C. A. Voigt, J. Am. Chem. Soc., 2009, 131, 6508–6515.
16. P. E. M. Purnick and R. Weiss, Nat. Rev. Mol. Cell Biol., 2009, 10, 410–422.
17. J. Hu, L. Wang, S. Zhang, Y. Le and X. Fu, Appl. Biochem. Biotechnol., 2014, 173, 2307–2320.
18. K. M. Scott and C. M. Cavanaugh, Appl. Environ. Microbiol., 2007, 73, 1174–1179.
19. D. D.-W. Tsai, R. Ramaraj and P. H. Chen, Ecol. Eng., 2016, 92, 106–110.
20. S. M. Huse, L. Dethlefsen, J. A. Huber, D. M. Welch, D. A. Relman and M. L. Sogin, PLoS Genet., 2008, 4, e1000255.
21. P. D. Schloss, S. L. Westcott, T. Ryabin, J. R. Hall, M. Hartmann, E. B. Hollister, R. A. Lesniewski, B. B. Oakley, D. H. Parks, C. J. Robinson, J. W. Sahl, B. Stres, G. G. Thallinger, D. J. Van Horn and C. F. Weber, Appl. Environ. Microbiol., 2009, 75, 7537–7541.
22. S. Mazzoleni, F. Cartenì, G. Bonanomi, M. Senatore, P. Termolino, F. Giannino, G. Incerti, M. Rietkerk, V. Lanzotti and M. L. Chiusano, New Phytol., 2015, 206, 127–132.
23. J. Hu, L. Wang, S. Zhang, Y. Wang and X. Xi, Bioresour. Technol., 2011, 102, 7147–7153.
24. A. Żak and A. Kosakowska, Estuarine, Coastal Shelf Sci., 2015, 167, 113–118.
25. I. Ahmed, A. Yokota, A. Yamazoe and T. Fujiwara, Int. J. Syst. Evol. Microbiol., 2007, 57, 1117–1125.
26. P. Kampfer, K. Denger, A. M. Cook, S. T. Lee, U. Jackel, E. B. M. Denner and H. J. Busse, Int. J. Syst. Evol. Microbiol., 2006, 56, 815–819.
27. M. H. Seo, K. R. Kim and D. K. Oh, Appl. Microbiol. Biotechnol., 2013, 97, 8987–8995.
28. W. Xia, Z. Du, Q. Cui, H. Dong, F. Wang, P. He and Y. Tang, J. Hazard. Mater., 2014, 276, 489–498.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra13118b

---