Mutualistic relationship between Nitrospira and concomitant heterotrophs.

Nitrifying chemoautotrophs support the growth of diverse concomitant heterotrophs in natural or engineered environments by supplying organic compounds. In this study, we aimed to investigate this microbial association, especially (i) to distinguish whether the relationship between nitrifying chemoautotrophs and heterotrophs is commensal or mutualistic, and (ii) to clarify how heterotrophs promote the growth of autotrophic nitrite-oxidizing bacteria (Nitrospira). Pure cultured Nitrospira (Nitrospira sp. ND1) was employed in this study. Heterotrophs growing with metabolic by-products of Nitrospira as a sole carbon source were isolated from several environmental samples and used to test the growth-promoting activity of Nitrospira. Furthermore, liquid chromatography-mass spectrometry analysis was conducted to evaluate how heterotrophs consumed chemical compounds produced by Nitrospira and newly produced during co-cultivation. Notably, Nitrospira growth was stimulated by co-cultivation with some heterotrophs and the addition of spent media of some strains, suggesting that not only heterotrophs but also Nitrospira received benefits from their mutual co-existence. Furthermore, the data suggested that some of the growth-promoting heterotrophs provided as-yet-unidentified growth-promoting factors to Nitrospira. Overall, Nitrospira and heterotrophs thus appear to exhibit a mutualistic relationship. Such mutualistic relationships between autotrophs and heterotrophs would contribute to the stability and diversity of microbial ecosystems.

Introduction

Nitrifying chemoautotrophs (ammonia and nitrite oxidizers) play important roles in the biological nitrogen cycle and wastewater treatment processes. Nitrifying chemoautotrophs constitute primary producers in natural environments as they assimilate inorganic carbon and release organic carbon compounds to microbial ecosystems. In nitrifying microbial ecosystems, concomitant heterotrophs are always observed representing approximately 50% of microbial communities even under conditions lacking organic carbon supply (Rittmann et al., 1994; Okabe et al., 1999; Okabe et al., 2002; Kindaichi et al., 2004; Martiny et al., 2005; Fujitani et al., 2013). The growth of diverse heterotrophs has been shown to be supported by nitrifying chemoautotrophs through the utilization of soluble microbial products derived from metabolic by‐products and decaying biomass (Noguera et al., 1994; Rittmann et al., 2002; Kindaichi et al., 2004; Okabe et al., 2005; Matsumoto et al., 2010). Nevertheless, although the in situ ecophysiological interactions between nitrifying chemoautotrophs and heterotrophs based on cross‐feeding have been well studied using cultivation‐independent approaches, a better understanding of the ecophysiological interactions between nitrifying chemoautotrophs and heterotrophs is expected to clarify the factors that control the efficiency and stability of nitrification.

However, because of the difficulty in cultivating nitrifying microorganisms, analysis of the microbial associations based on pure cultures has been limited. For example, although it is well‐established that the growth of concomitant heterotrophs completely depends on nitrifying chemoautotrophs in the ecosystems, it remains unclear how the nitrifying chemoautotrophs benefit in turn from such dependent heterotrophs.

Notably, some heterotrophs exhibit the potential to promote the growth of ammonia‐oxidizing bacteria or archaea (Keluskar et al., 2013; Sedlacek et al., 2016; Bayer et al., 2019), whereas the association with nitrite‐oxidizing bacteria has not been reported. The key nitrite oxidizers Nitrospira may represent candidate species able to benefit from microbial associations as they are widespread both in the natural ecosystem and engineered environments and play a significant role in the nitrogen cycle (Wagner et al., 1996; Burrell et al., 1998; Hovanec et al., 1998; Juretschko et al., 1998; Bartosch et al., 1999). Nevertheless, despite their importance, a comprehensive understanding of their physiological properties has been limited because they generally resist cultivation, especially into pure cultivation (Nowka et al., 2015; Daims et al., 2016).

In this study, we tested the growth promotion effect of heterotrophs on nitrifying chemoautotrophs, especially Nitrospira, to clarify the microbial association; in particular, to (i) distinguish whether the relationship between nitrifying chemoautotrophs and heterotrophs is commensal or mutualistic, and (ii) clarify how heterotrophs promote the growth of Nitrospira. For this purpose, pure cultured Nitrospira (Nitrospira sp. ND1) belonging to sublineage I was employed in this study (Fujitani et al., 2014).

The effects of co‐cultivation with several selected concomitant heterotrophs grown with the spent (conditioned) medium of Nitrospira as a carbon source on the growth of Nitrospira were analysed by focusing on specific growth rate and lag time. In addition, the spent medium of each heterotroph was applied to pure Nitrospira culture to clarify how growth‐promoting heterotrophs affect the growth of Nitrospira; i.e., to distinguish whether such strains secrete growth‐promoting factors (GPF). Finally, the fate of chemical compounds during the cultivation of heterotrophs with the spent medium of Nitrospira culture were comparatively analysed among the strains using liquid chromatography–mass spectrometry (LC–MS) to evaluate how heterotrophs consumed compounds produced by Nitrospira and produced during the cultivation.

Results and discussion

Isolating and pre‐screening candidate heterotrophs

A total of 84 strains of heterotrophs were isolated from three different samples using the culture media included spent medium from Nitrospira culture (NSM) (Fig. 1). This medium contains 15.9 ± 6.3 mg‐C L−1 (S.D.) of dissolved organic carbon derived from metabolic by‐products of Nitrospira as a sole carbon source. Among the 84 strains, 15 strains that promoted the nitrite oxidation activity of Nitrospira were selected based on the pre‐screening test. The type strain, Pseudomonas putida (NBRC14164) was also employed as strain No.0, as it has been reported to enhance the growth of several autotrophic bacteria such as ammonia and methane‐oxidizing bacteria (Ho et al., 2014; Sedlacek et al., 2016). Finally, a total of 16 strains were employed for further analysis to investigate the effect of heterotrophs on the nitrite oxidation activity of Nitrospira (Table 1).

Fig. 1

Flowchart of procedure for the preparation of culture media used in this study. Step 1: the spent medium from Nitrospira culture (NSM) was prepared by filtration of Nitrospira culture after cultivation with IN‐medium. Step 2: NSM was employed for two types of culture medium; (i) the medium for isolation of heterotrophs was prepared as gellan gum plates (NSM with gellan gum); (ii) NSM‐medium was prepared by adding nitrite to NSM. Step 3: each tested heterotroph was cultivated in NSM. Then, the spent medium from each culture of heterotroph (HSM) was prepared by filtration. Step 4: HSM was employed for two types of culture medium; (i) HSM‐medium was prepared by adding nitrite to HSM; (ii) 10% HSM‐medium was prepared by adding HSM to IN‐medium (HSM was added to IN medium to 10% of the total volume).

Table 1

Test strains used for assays and effect of those strains on the growth of Nitrospira.

Strain No.	Source	Closest cultured relative	Similarity (%)a	Accession No.	Co‐culture			HSM			10% HSM
					r.a.	μ	λ	r.a.	μ	λ	r.a.	μ	λ
0	CC	Pseudomonas putida	100	AP013070	+	+	+	+	±	+	+	+	+
4	EC1	Mycobacterium grossiae	99	CP043474	+	+	+	+	+	+	±	±	±
71	EC2	Shinella fusca	99	KM210268	+	+	±	+	±	+	+	+	+
49	EC2	Pseudomonas putida	100	CP026115	+**	±	±	+	±	+	+	+	+
55	EC2	Ensifer adhaerens	100	KT229738	+**	±	±	+	±	+	+	+	+
1	EC1	Sphingopyxis alaskensis	99	AY509241	±	±	±	+	±	+	±	±	±
14	AS	Mycobacterium dioxanotrophicus	99	CP020809	±	±	±	+	±	±	±	±	±
61	EC2	Alicycliphilus denitrificans	100	AB908107	±	±	±	+	±	+	±	±	±
38	EC2	Gordonia sihwensis	100	MN880097	−	−	±	+	±	±	n.d.	n.d.	n.d.
11	EC1	Bradyrhizobium elkanii	100	LC515847	−	±	−	±	±	±	n.d.	n.d.	n.d.
20	AS	Gordonia austrails	100	MK680170	±	±	±	±	±	±	n.d.	n.d.	n.d.
39	EC2	Shinella fusca	99	KM210268	±	±	±	±	±	±	n.d.	n.d.	n.d.
40	EC2	Gordonia sihwensis	100	MN880097	−	±	−	±	±	±	n.d.	n.d.	n.d.
63	EC2	Sphingobacterium mizutaii	100	MK253331	±	±	±	±	±	±	n.d.	n.d.	n.d.
64	EC2	Paludibaculum fermantans	94	NR134120	±	±	±	±	±	±	n.d.	n.d.	n.d.
74	EC2	Reyranella massiliensis	99	HM048834	−	±	−	±	±	±	n.d.	n.d.	n.d.

a16S rRNA similarity of each strain to the closest known species.

AS, activated sludge; CC, type strain; EC1, enrichment culture I; EC2, enrichment culture II; n.d., no data; r.a., relative activity.

+: positive; ±: no effect; −: negative; +**: positive (large S.D.).

Effect of heterotrophs on the growth activity of Nitrospira

We tested the effect of heterotrophs grown with NSM‐medium (Fig. 1) as the sole carbon source on the nitrite oxidation activity of Nitrospira under two different conditions. These included (i) co‐cultivation with tested heterotrophs in NSM‐medium and (ii) pure cultivation in the spent medium from heterotroph culture (HSM‐medium) derived from each tested heterotroph (Fig. 1).

Figure S1A and C shows representative time‐course changes in nitrites during co‐cultivation with heterotrophs (strains No.0, No.4 and No.71), and pure cultivation in HSM‐media derived from each heterotroph respectively. The results of co‐culture assays indicated that three heterotrophs (No.0, No.4 and No.71) apparently elevated the nitrite oxidation activity of Nitrospira (equivalent to the growth) (Table 1, Figs S1B and S2). On the other hand, 9 strains of heterotrophs (No.0, No.1, No.4, No.14, No.38, No.49, No.55, No.61 and No.71) elicited a positive effect on the growth of Nitrospira through their HSM‐medium (Table 1, Figs S1D and S2).

Figure S3 shows the effect of co‐cultivation with heterotrophs (Fig. S3A) or pure cultivation in HSM on the growth characteristics of Nitrospira (Fig. S3B). The specific growth rate of Nitrospira was distinctly higher in co‐cultivation with three strains, No.0, No.61 and No.71 (p‐values < 0.05, t‐test) than that in pure culture. The lag time was distinctly shortened when Nitrospira was co‐cultivated with strains No.0, No.4, No.49 and No.55 (p < 0.05) compared with that in pure culture. In contrast, the lag time was extended when Nitrospira was co‐cultivated with strains No.11, No.74 and No.40. The specific growth rate of Nitrospira was lower in co‐cultivation with strain No.38 than that in pure cultivation (Table 1, Fig. S3B). The specific growth rate of Nitrospira was distinctly higher when growing in the HSM from strain No.4 (p < 0.05). The relative lag time of Nitrospira was shortened when growing in HSMs from strains No.1, No.4, No.49, No.55, No.0, No.61 and No.71 (p < 0.05).

The results of co‐culture assays indicated that three heterotrophs (No.0, No.4 and No.71) promoted the growth of Nitrospira (Table 1, Fig. S1A), whereas a larger number (9 strains) of heterotrophs (No.0, No.1, No.4, No.14, No.38, No.49, No.55, No.61 and No.71) elicited a positive effect on the growth of Nitrospira through their HSM‐medium (Table 1, Fig. S1D). There can be several reasons for this difference as follows: (i) the competition for dissolved oxygen (Ohashi et al., 1995; Okabe et al., 1996; Nogueira et al., 2002), (ii) the effect of physical contact between the cells on the induction of secondary metabolite (Onaka et al., 2011; 2015) and (iii) predatory bacteria for Nitrospira (Dolinšek et al., 2013).

Effect of low‐level supplementation of Nitrospira

The positive effect of heterotrophs on the growth of Nitrospira through their HSM‐medium suggested that some of the heterotrophs secreted chemical compounds enhancing the growth of Nitrospira. However, the analysis left another possibility that this positive effect might occur by degradation of toxic compounds by heterotrophs. To confirm whether the heterotrophs produce chemical compounds enhancing the growth of Nitrospira, the effect of low‐level supplementation with HSM on Nitrospira growth activity was examined. Nitrospira was cultivated with 10% HSM‐medium composed of inorganic nitrite (IN)‐medium with the addition of HSM from each heterotroph to 10% of the total volume (Fig. 1).

The results showed that the nitrite oxidation activity of Nitrospira in the 10% HSM‐media from strains No.0, No.49, No.55 and No.71 was apparently promoted whereas that from strains No.1, No.4, No.14 and No.61 did not positively affect the nitrite oxidation activity (Table 1, Fig. 2). This indicated that four HSMs from among eight tested strains apparently contained chemical compounds facilitating the growth of Nitrospira. Thus, these 4 strains could be classified as GPF suppliers. Conversely, the lack of growth‐promoting effects from the other four strains (No.1, No.4, No.14 and No.61) was likely due to (i) the addition of 10% HSM being insufficient to promote Nitrospira growth (containing less amount of growth‐promoting chemical compounds), or (ii) the ability of these strains to promote the growth of Nitrospira being a consequence of the reduction (degradation) of growth‐inhibiting compounds through scavenging of the metabolic by‐products derived from Nitrospira, as previously hypothesized (Paerl and Pinckney, 1996; Paerl et al., 2000; Morris et al., 2008; Sher et al., 2011; Ho et al., 2014).

Fig. 2

The effect of cultivation of Nitrospira with 10% HSM‐medium from each heterotroph on nitrite oxidation activity.

A. The time course change in nitrite during cultivation of Nitrospira with 10% HSM‐medium (closed square with dashed line) and with IN‐medium (closed circle). Data from cultivation with HSM‐medium derived from strain No.0, No.49, No.55 and No.71 are shown.

B. Comparison of the nitrite oxidation activity of Nitrospira among cultivation with 10% HSM‐media from each heterotroph, IN‐medium (as a control, CON) and 10% NSM‐medium (as a negative control, NC). The data are expressed as relative values with that with IN‐medium defined as 100%. Nitrite oxidation activity was calculated based on the time required for reduction of nitrite concentration below the standard value, in the same manner as shown in Figs S2 and S3.

C. The comparison of specific growth rate (μ) and lag time (λ) of Nitrospira among cultivation with IN‐medium, 10% NSM‐medium and 10% HSM‐media derived from each heterotroph. The data are expressed as relative values with that in IN‐medium defined as 1.0. Error bars indicated the standard deviation of the biological triplicate. Asterisk indicates a value significantly different from the control (p < 0.05).

Although nitrate, the main metabolic product of Nitrospira inhibits the growth itself when the concentration was over 15 mM (Fig. S4), consistent with the previous observation (Nowka et al., 2015), the growth promotion observed in the present study was not likely due to the reduction of nitrate by heterotrophs. This is because there was no significant difference in the concentration of nitrate between NSMs and HSMs (data not shown).

Chemical species in NSM and HSMs as analysed by LC–MS

As heterotrophs grew with organic substrates released from Nitrospira, it can be expected that heterotrophs have the potential to consume various types of organic compounds in NSM by assimilation or dissimilation. They would also have the potential to release different types of organic compounds into HSMs as metabolic by‐products that would not be contained in NSM.

To address these possibilities, we compared the chemical compounds from 100 to 2000 kDa contained in NSM and HSMs by LC–MS and signpost analysis to evaluate how each heterotroph consumed and produced chemical compounds during the cultivation in NSM (Fig. 3A).

Fig. 3

Chemical profiles of the spent medium from Nitrospira (NSM) and each heterotroph (HSMs) analysed by LC–MS.

A. Hierarchical cluster analyses and heatmap of chemical profiles of NSM and HSMs from each heterotroph. The x‐axis represents the type of spent medium (showing strain number or NSM), whereas the y‐axis represents the chemical compounds in media. Green represents no difference in relative abundance and red represents maximum relative abundance. The red arrow indicates a chemical compound particularly strongly detected in HSM derived from No.49, No.55 and No.71.

B. Principal component analysis plot of chemical profiles. GPF‐suppliers are represented with black closed circles (No.0, No.49, No.55 and No.71); NSM is represented with an open triangle; growth‐promoting heterotrophs excluding GPF‐suppliers (No.1, No.4, No.14 and No.61) are represented with closed black square; other heterotrophs (No.11 and No.38) are represented with grey square diamond.

The analysis results revealed that each HSM exhibited a different chemical profile. Signal intensities of several compounds detected in NSM disappeared or were significantly reduced in each HSM, suggesting that they were used as substrates for the growth of heterotrophs. In particular, some chemical compounds derived from NSM (m/z = 403.13, 546.20, 585.15, 602.18 and 690.23; blue arrows in Fig. 3A) disappeared or exhibited significantly decreased signal intensity in all HSMs, suggesting that these chemical compounds were commonly consumed by all heterotrophs, and were probably utilized as key substrates for the growth of the tested heterotrophs.

Some HSMs contained substantially more compounds than NSM or characteristic compounds that were not detected in NSM, suggesting that heterotrophs produced diverse organic compounds accompanied by the consumption of organic compounds in NSM and consequently their spent media exhibited unique chemical profiles. Among these compounds, several common chemical compounds highly produced by GPF suppliers (No.0, No.49, No.55 and No.71) in particular were identified (Fig. 3A). However, we were unable to identify chemical compounds exclusive to HSMs from GPF suppliers. Notably, the chemical compound (m/z = 519.08; red arrow in Fig. 3A) was present at particularly high levels in HSMs from three strains (No.49, No.55 and No.71), suggesting this compound as a candidate for the GPF for Nitrospira. Although strain No.0 should also be a GPF supplier, it likely produces different chemical compounds as GPFs than those of strains No.49, No.55 and No.71.

Some organic compounds, such as amino acids, pyruvate and glucose can enhance the growth of autotrophic nitrifying organisms (Clark and Schmidt, 1966; Clark and Schmidt, 1967a,b; Pan and Umbreit, 1972; Kim et al., 2016; Bayer et al., 2021), it is likely that the growth‐promoting heterotrophs provided such organic compounds for utilization as metabolic substrates. In addition, it is expected that the heterotrophs provided vitamins, siderophores or signal‐like molecules that affect phenotypic expression including growth activity (Batchelor et al., 1997; Burton et al., 2005; Keluskar et al., 2013).

Principal component analysis of chemical profiles revealed that the strong GPF suppliers, strains No.49, No.55 and No.71, belonged to the same cluster, whereas strain No.0 was positioned far from this cluster. Moreover, strains No.1, No.14, No.38 and No.61 belonged to the same cluster and were not grouped as strong GPF suppliers, with strain No.4 being located apart from any clusters (Fig. 3B). This further suggested that these strains enhanced the growth of Nitrospira differently compared with the GPF suppliers.

Significance of the mutualistic relationship between Nitrospira and heterotrophs

The evidence provided herein that some heterotrophs promote the growth of Nitrospira suggests that the relationship between autotrophs and some heterotrophs is mutualistic rather than commensal. Not only Nitrospira but also other nitrifying microorganisms may, therefore, exhibit a similar mutualistic relationship with various heterotrophs. In addition, the possibility remains that other tested heterotrophs would also have the potential to promote the growth of Nitrospira but not exhibited under the tested condition because some heterotrophs did not grow efficiently. In the present study, we utilized NSM‐medium derived from the supernatant of Nitrospira culture for the cultivation of heterotrophs without the addition of extra carbon sources. This is to clarify whether heterotrophs could exert their potential to promote the growth of nitrifiers under the conditions in natural ecosystems where heterotrophs grow using metabolic by‐products of Nitrospira as a sole carbon source (Fig. 1). Therefore, the results obtained in this study were considered to closely reflect the actual interactions between nitrifying organisms and heterotrophs in natural ecosystems.

In the present study, we revealed that some heterotrophs supplied chemical compounds promoting Nitrospira growth (GPF‐supplier). The results of LC–MS analysis suggested the existence of chemical compounds produced by GPF suppliers. However, these substances have not been completely characterized in the present study. Moreover, the mechanism of growth promotion might be varied among the microbial types in addition to supplying GPF but remained unclarified. Therefore, additional studies are required to obtain a comprehensive understanding of interactions based on these chemical compounds.

As it is impracticable to test every microbial type among the diverse heterotrophs in various environments, it is not clear how such diverse concomitant heterotrophs affect the growth of Nitrospira. However, the results obtained in the present study suggest that heterotrophs positively affecting the growth of Nitrospira are not rare; rather, a considerable proportion of heterotrophs in environments are likely to positively affect the growth of nitrifying chemoautotrophs.

Notably, one of the bottlenecks toward obtaining pure cultures of Nitrospira is the tendency of Nitrospira to form inseparable aggregates with heterotrophs when they are enriched from the environment (Spieck et al., 2006; Lebedeva et al., 2008; Fujitani et al., 2013; Fujitani et al., 2014; Nowka et al., 2015). However, this also suggests that Nitrospira and heterotrophs are strongly associated to promote their growth. Thus, the complex and diverse interactions revealed in this study may partly underlie the resistance of Nitrospira to isolation. Furthermore, these interactions may contribute to the stabilization of nitrifying microbial ecosystems in nature and engineered systems.

Appendix S1. Experimental procedures.

Click here for additional data file.