Abstract
Sulfurimonas species are among the most abundant sulfur-oxidizing bacteria in the marine environment. They are capable of using different electron acceptors, this metabolic flexibility is favorable for their niche adaptation in redoxclines. When oxygen is depleted, most Sulfurimonas spp. (e.g., Sulfurimonas gotlandica) use nitrate (\({{{{{{{\mathrm{NO}}}}}}}}_3^ -\)) as an electron acceptor to oxidize sulfur, including sulfide (HS-), S0 and thiosulfate, for energy production. Candidatus Sulfurimonas marisnigri SoZ1 and Candidatus Sulfurimonas baltica GD2, recently isolated from the redoxclines of the Black Sea and Baltic Sea respectively, have been shown to use manganese dioxide (MnO2) rather than \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) for sulfur oxidation. The use of different electron acceptors is also dependent on differences in the electron transport chains embedded in the cellular membrane, therefore changes in the membrane, including its lipid composition, are expected but are so far unexplored. Here, we used untargeted lipidomic analysis to reveal changes in the composition of the lipidomes of three representative Sulfurimonas species grown using either \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) and MnO2. We found that all Sulfurimonas spp. produce a series of novel phosphatidyldiazoalkyl-diacylglycerol lipids. Ca. Sulfurimonas baltica GD2 adapts its membrane lipid composition depending on the electron acceptors it utilizes for growth and survival. When carrying out MnO2-dependent sulfur oxidation, the novel phosphatidyldiazoalkyl-diacylglycerol headgroup comprises shorter alkyl moieties than when sulfur oxidation is \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\)-dependent. This is the first report of membrane lipid adaptation when an organism is grown with different electron acceptors. We suggest novel diazoalkyl lipids have the potential to be used as a biomarker for different conditions in redox-stratified systems.
Similar content being viewed by others
Introduction
Chemoautotrophic bacteria of the genus Sulfurimonas [class Campylobacterota, according to the NCBI classification, [1, 2] play an important role in sulfur oxidation in sulfidic habitats, such as stratified marine waters, anoxic sediments, and hydrothermal deep-sea vents, as well as in some terrestrial environments [3]. For example, along the Namibian shelf, Lavik, et al. [4] reported that occasionally occurring, large-scale sulfidic water masses were detoxified by oxidation of sulfide (HS-) using nitrate \(({{{{{{{\mathrm{NO}}}}}}}}_3^ - )\) by two groups of bacteria: Sulfurimonas spp. and bacteria falling in the gammaproteobacterial sulfur oxidizer cluster. Within and below the pelagic redoxcline of the Baltic Sea, where a stable redoxcline separates deep anoxic sulfidic water from oxygenated surface water [5], Sulfurimonas spp. accounted for up to 15% of total prokaryotic abundance. Here, Grote, et al. [6] isolated S. gotlandica GD1T, a denitrifying chemolithotrophic sulfide oxidizer. In oxygen deficient systems, sulfur cycling has been shown to be tightly linked to the nitrogen cycle without affecting chemical gradients, so called cryptic cycling [e.g., [7, 8].
In the stratified water columns of both the Baltic Sea and the Black Sea, \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) and nitrite (\({{{{{{{\mathrm{NO}}}}}}}}_2^ -\)) often disappear before the first appearance of HS- [9, 10]. Geochemical water column profiles suggest that extensive cycling of dissolved and particulate manganese (Mn) may quantitatively account for the oxidation of HS-, thereby shuttling oxidative potential of O2 and \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) over several meters distance. This mechanism, as well as a potential biological catalyzation, was proposed earlier because chemoautotrophic HS- oxidation could not be linked to O2 or \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) [10,11,12]. Recently, Henkel, et al. [13] isolated Ca. Sulfurimonas marisnigri SoZ1 from the redoxcline of the central Black Sea. This bacterium can couple the oxidation of reduced sulfur compounds, including HS-, S0 and thiosulfate, to the reduction of MnO2. The biologically catalyzed oxidation of HS- by Ca. S. marisnigri SoZ1 is faster than the abiotic oxidation of HS- with MnO2 [14] and yields \({{{{{{{\mathrm{SO}}}}}}}}_4^{2 - }\) as the prominent end product, while elemental sulfur (S0) accumulates in the abiotic reaction [13, 15, 16]. Combining the cellular abundance of Sulfurimonas spp. in the Black Sea redoxcline with rates of MnO2 dependent HS- oxidation by Ca. S. marisnigri SoZ1 in a reaction diffusion model yielded adequate rates to counterbalance the HS- flux, resulting in the measured HS- concentration profile, in contrast to a pure abiotic reaction [14]. Hence, this process is likely to play a prominent role in HS- oxidation at the chemocline of the Black Sea. Likewise, Ca. Sulfurimonas baltica GD2 was isolated from the redoxcline of the Gotland Deep in the Baltic Sea [17]. These studies revealed the important role of MnO2-dependent Sulfurimonas spp. in biogeochemical cycling in redox-stratified systems [14]. Changes in the electron acceptor are associated with differences in the electron transport chain (ETC), which is embedded in the cellular membrane. Protein complexes of the ETC are expected to interact with and influence the lipid composition of the membrane, which is partially responsible for the fluidity of the membrane [18]. Nevertheless, few studies have addressed potential changes in the membrane lipid composition upon changes in the electron acceptor as a response to maintain the cell homeostasis.
Structurally diverse microbial lipids play important roles as the building blocks of membranes and play a role in energy storage, signaling and modulation of protein activity [19]. Many microorganisms maintain their membrane functionality, permeability, and fluidity during changing environmental conditions by membrane adaptation [20, 21] through the regulation of its lipid composition [e.g., [22,23,24,25]. For instance, when phosphorus is limited, phytoplankton and some bacteria can use non-phosphorus lipids to replace phospholipids for survival [23, 26, 27]. However, the membrane lipid composition of Sulfurimonas spp., and potential differences when using different electron acceptors to oxidize HS-, are unknown. Until now, only a few studies have reported the predominant cellular fatty acids of Sulfurimonas spp. [e.g., [28,29,30], but their intact polar lipid composition, including both the fatty acid side chain and their linked polar headgroups, remains to be studied. Because of their relatively high abundance in redox-stratified waters, Sulfurimonas spp. are likely to contribute substantially to the lipidome of these environments. Recent advances in the field of lipidomics, using non-targeted approaches combined with computational methods, allows for comprehensive lipidome profiling [31,32,33,34].
Here, we examined the lipidome of three Sulfurimonas species, S. gotlandica GD1T and Ca. S. baltica GD2, both isolated from water column of the Baltic Sea, and Ca. S. marisnigri SoZ1 isolated from the water column of the Black Sea, cultured with either \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) or MnO2 as electron acceptor. Our aim was to investigate how Sulfurimonas spp. adjust their membrane lipid composition when using different electron acceptors (\({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) and MnO2), in order to determine potential microbial adaptations of the membrane linked to different metabolisms.
Methods and materials
Cultures
Anoxic medium was prepared with a salinity of 21, 14 and 10 psu for Ca. S. marisnigri SoZ1, Ca. S. baltica GD2 and S. gotlandica GD1T respectively as described before [17] with minor changes. We used 2 L glass bottles closed with butyl rubber stoppers for both MnO2 and \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) conditions with Ca. S. marisnigri SoZ1 and Ca. S. baltica GD2. S. gotlandica was cultivated in 100 mL serum bottles closed with butyl rubber stoppers. We raised the concentration of NH4Cl and Na2HPO4 from 0.02 mM and 0.01 mM to 2.5 mM and 0.2 mM, respectively. These adjustments were made to exclude deficiencies of macro nutrients during growth and should therefore reduce differences in lipid composition of cultures. Cultures were obtained from the internal culture collection of the IOW, which are also available at the German Collection of Microorganisms and Cell Cultures GmbH (DSMZ) and at the Japan Collection of Microorganisms (JCM) with the identifiers JCM 39139 and DSM 111879 (Ca. S. marisnigri SoZ1), JCM 39138 and DSM 111898 (Ca. S. baltica GD2) and JCM16533 and DSM 19862 (S. gotlandica GD1T). Ca. S. marisnigri SoZ1 and Ca. S. baltica GD2 were cultured with either 3 mM thiosulfate (\({{{{{{{\mathrm{S}}}}}}}}_2{{{{{{{\mathrm{O}}}}}}}}_3^{2 - }\)) and 5 mM MnO2 (Mn-reducing conditions) or 5 mM \({{{{{{{\mathrm{S}}}}}}}}_2{{{{{{{\mathrm{O}}}}}}}}_3^{2 - }\) and 10 mM \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) (\({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) reducing conditions). S. gotlandica GD1T was amended with 15 mM \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) and 10 mM \({{{{{{{\mathrm{S}}}}}}}}_2{{{{{{{\mathrm{O}}}}}}}}_3^{2 - }\). S. gotlandica GD1 grows faster and has a substantially higher growth efficiency with nitrate than Ca. S. marisnigri SoZ1 and Ca. S. baltica GD2. This is because S. gotlandica GD1 is a complete denitrifier (\({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) to N2), while Ca. S. marisnigri SoZ1 and Ca S. baltica GD2 can only reduce \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) to \({{{{{{{\mathrm{NO}}}}}}}}_2^ -\) [17]. Since S. gotlandica grows faster than the other strains, the initial \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) and S2O32- concentrations were higher to prolong the exponential growth phase. Concentrations of \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) (and therefore also S2O32-) in cultivations of Ca. S. marisnigri SoZ1 and Ca. S. baltica GD2 are lower to exclude a putative toxification by \({{{{{{{\mathrm{NO}}}}}}}}_2^ -\). Cell numbers were estimated after 8 days of growth and the biomass were harvested at day 12 at 10 °C in the dark. Cellular abundance of Ca. S. marisnigri SoZ1 and Ca. S. baltica GD2 were ~5 × 106 cells mL−1 under Mn-reducing conditions and ~5 × 105 cells mL−1 under \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) reducing conditions (Ca. S. baltica GD2 only). Very few cells were visible under \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) reducing conditions with Ca. S. marisnigri SoZ1. S. gotlandica GD1T reached a cellular abundance of ~4 × 107 cells mL−1 at day 8. Incubation was continued until day 10 at which the black color of Mn-containing incubations became slightly brownish-gray, indicating the precipitation of Ca-rich Mn-carbonates at the end of the exponential growth phase [13, 17]. Cells were harvested in 50 mL centrifugation tubes at 4060 x g for 15 minutes at 10 °C, pooled and stored at −20 °C before freeze drying and shipping to the NIOZ for lipidome analysis.
Lipidome analysis
The cells of three representative Sulfurimonas species (Ca. S. marisnigri SoZ1, Ca. S. baltica GD2 and S. gotlandica GD1T) were collected and freeze dried for lipidome analysis. A detailed description of sample extraction and ultra-high performance liquid chromatography coupled to high-resolution tandem mass spectrometry (UHPLC-HRMS2) analysis and initial data processing is given in Bale, et al. [35]. Briefly, freeze-dried samples were extracted using a modified Bligh-Dyer procedure. They were extracted ultrasonically for 10 min, twice in a mixture of methanol, dichloromethane and phosphate buffer (2:1:0.8, v-v:v) and twice with a mixture of methanol, dichloromethane and aqueous trichloroacetic acid solution (TCA) pH 3 (2:1:0.8, v-v:v). The organic phase was separated by adding additional dichloromethane and buffer to a final solvent ratio of 1:1:0.9 (v:v) and were re-extracted three times with dichloromethane and dried under a stream of N2 gas. The extract was redissolved in a mixture of MeOH:DCM (9:1, v-v) and were filtered through 0.45 µm regenerated cellulose syringe filters (4 mm diameter; Grace Alltech). The extracts were then analyzed using Agilent 1290 Infinity I UHPLC coupled to a Q Exactive Orbitrap MS (Thermo Fisher Scientific, Waltham, MA). The output data files generated by the UHPLC-HRMS2 analyses were further processed using MZmine software [36]. Process steps included mass peak detection, chromatogram building and deconvolution, isotope grouping, ion component alignment and gap filling [34]. The relative abundance of components was obtained after processing and the combined dataset of MS/MS spectra were analyzed using the Feature Based Molecular Networking tool [33] through the Global Natural Product Social Molecular Networking (GNPS) platform [34] to build molecular networks of the detected components in the dataset (https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=f775f785684e4cc1b9e133d7abcc6f00). Details can be found in Ding, et al. [37]. It should be noted that many of the lipids detected in this study have not been described previously and hence authentic standards for absolute quantification are not available. Therefore, the lipid compositions were examined in terms of their peak area response. Thus, the relative peak area does not necessarily reflect the actual relative abundance of the different lipids, however, this method allows for some comparison between the samples analyzed in this study. Due to the extraction and analytical methods, and based on annotation from Ding, et al. [37], most of the ion components from the molecular network we generated were lipids and contaminants, thus we used the term “lipidome” for parts of the results and discussion where the lipids are discussed.
Results and discussion
Lipidome of three representative Sulfurimonas species
Three representative Sulfurimonas species were cultured with thiosulfate using \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) or MnO2 as electron acceptors. Thiosulfate was used instead of HS- as it is a more convenient electron donor, being non-toxic even in higher concentrations, and non-reactive with the MnO2 [13]. Cells were harvested after 12 days of growth, which corresponds to the late exponential or early stationary phase [17]. The growth stages of Ca. S. marisnigri SoZ1 and Ca. S. baltica GD2 were comparable due to the reliable change in color of the growth media in the late exponential growth phase, and the fact that the time needed to reach final cell densities with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) does not differ from that with MnO2 based on our experience with these strains. Similar to the previous studies, growth yields of Ca. S. marisnigri SoZ1 and Ca. S. baltica GD2 cultured with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\)were substantially lower compared to cultivation with MnO2 as terminal electron acceptor [17]. The biomass yield of Ca. S. marisnigri SoZ1 grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) was too low for a precise lipidomic analysis, therefore it is excluded for the following analysis. S. gotlandica GD1T is unable to grow with MnO2 [17]. Since only Ca. S. baltica GD2 could be cultured with both \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) and MnO2 conditions, this is the only species that could be examined in terms of its membrane adaptation to a change in electron acceptor. The results of Ca. S. marisnigi SoZ1 grown with MnO2 and S. gotlandica GD1 grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) provide complementary information about their lipid specificities and are thus interesting for comparison.
Recently, we established a workflow that provides a substantially expanded view of the molecular composition of the microbial lipidome in environmental settings [37]. Here we used the same workflow for our culture lipidome analysis. The complete dataset produced by analysis with high performance liquid chromatography coupled to high resolution tandem mass spectrometry (HPLC–HRMS2) of the Bligh-Dyer lipid extracts of the cultures contained 9936 unique ion components, of which 4282 ion components (43%) occurred in structure-similarity groupings in the molecular network, while 5654 ion components occurred as singletons (i.e., components without molecular relatives). A search in the Global Natural Products Social Molecular Networking (GNPS) library [34] resulted in only 167 spectral annotations (<2%). Other than ca. 20 contaminants (e.g., plasticizers), the majority of these annotations were those of well-known glycerol-based lipids: phosphatidylethanolamines (PE-DAGs), phosphatidyl glycerols (PG-DAGs) and triayclglycerols (TAGs). However, the vast majority of ion components was left unannotated. Lipidome annotation remains a bottleneck in marine microbiology lipidomic studies because public databases in this field are poorly populated [35, 37].
The major lipid classes detected in the Sulfurimonas spp. were PE/PG-DAGs, diacylglycerols (DAGs) and acyl ether glycerols (AEGs) without a polar head group, ornithine lipids, isoprenoidal quinones, long-chain fatty acids and some unknown lipids (Fig. 1A). The distribution of major lipid classes, based on the peak intensity, were similar between Ca. S. baltica GD2, grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) and with MnO2 (Fig. 1B). PE/PG-DAGs were dominant among all the lipids, accounting for >80% of total lipids. AEGs were the second most abundant lipid class, accounting for nearly 8% of total lipids. Isoprenoid quinones and some unknown lipids accounted for 0.5–3% of total lipids. PE/PG-DAGs were also the most abundant group of lipids in Ca. S. marisnigi SoZ1 grown with MnO2, comprising 62% of the total lipids, followed by 15% AEGs and 11% unknown lipids. In contrast to Ca. S. baltica GD2 and Ca. S. marisnigi SoZ1, PE/PG-DAGs comprised only 39% of total lipids in S. gotlandica GD1T grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\). Ornithine lipids were almost exclusively produced by S. gotlandica GD1T grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) (Fig. S2), accounting for 40% of the total lipids. Ornithine lipids are phosphorus-free amino intact polar lipids. which are relatively common in bacteria, but are absent in eukaryotes and archaea [38]. About 50% of sequenced bacteria are suggested to be able to synthesize ornithine lipids under certain growth conditions [24, 39].
The lipidome differences between Ca. S. marisnigri SoZ1 and S. gotlandica GD1 may be attributed to membrane differences between the species or to factors such as differences in the growth media. Differences in the lipidome of Ca. S. baltica GD2, depending on whether they were grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) or MnO2, are not evident from the total lipid distribution, so we subsequently examined the compositional differences within specific lipid classes.
PE-DAGs and PG-DAGs are the major components of most bacterial membranes [40, 41] and this also holds for the investigated Sulfurimonas species (Fig. 2B). A subnetwork of diverse PE- and PG-DAGs (Fig. 2A) was shown in the molecular network. The acyl moieties (i.e., derived from the esterified fatty acids) of these abundant PEs and PGs had 0–3 double bond equivalents (DBEs) and contained 26–36 carbon atoms (i.e., the sum of the two acyl chains). The majority (i.e., 50–70%) of the PE- and PG-DAGs had 30–32 total acyl carbons with 0 to 2 DBEs. The most dominant individual PE- and PG-associated fatty acids were C14, C15 and C16, saturated and monounsaturated fatty acids, in agreement with a previous study that reported that the dominant fatty acid of S. gotlandica GD1T was C16:1 (66% of total), with smaller amounts of C18:1 and C16:0 fatty acids [30]. Our detailed analyses revealed a minority of smaller PE- and PG-DAGs, containing 26–29 total acyl carbons and with 0–1 DBEs. In the only Sulfurimonas species that was cultivatable under both \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) and MnO2 conditions, i.e., Ca. S. baltica GD2, the proportion of these smaller PE- and PG-DAGs (26–29 acyl carbons, 0–1 DBEs) was 5.5% of total PE/PG-DAGs under \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) conditions and 17.0% under MnO2 conditions, suggesting a phospholipid acyl chain adaption in Ca. S. baltica GD2 depending on the electron acceptor used.
Novel lipids found in the Sulfurimonas species
A subnetwork (>60 components; Fig. 3A, B) entirely populated by unknown lipids (labeled “unknown1” in Fig. 1) revealed a clear division between Ca. S. baltica GD2 grown with either \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) or with MnO2 as the electron acceptor. The composition of this unknown lipid network was also noticeably different between Ca. S. marisnigi SoZ1 grown with MnO2 and S. gotlandica GD1 grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\). One of the central and relatively abundant nodes, with a lipid structure occurring in all Sulfurimonas spp. grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) or with MnO2, represents a compound with a molecular ion at m/z 771.5657 (as per the molecular network, cf. Fig. 3) eluting at 17.0–17.5 min (Fig. S3). This component was assigned an elemental composition (EC) of C42H80O8N2P+ (Δmmu = 1.00; the difference in millimass; [calculated mass – observed mass] x 1000). Upon MS2 fragmentation (Fig. 3C) a dominant fragment ion at m/z 225.1000 with an EC of C7H18O4N2P+ (Δmmu = 0.13) was formed, likely representing the polar headgroup, and a relatively minor and complimentary fragment at m/z 547.4704 (EC C35H63O4+; Δmmu = −1.59) representing the diacyl glycerol core containing the esterified fatty acids with a combined total number of carbon atoms of 32 and 2 DBEs. A single fragment ion at m/z 237.2208 with EC of C16H29O+ suggests the diacyl glycerol is composed of two C16:1 fatty acids. An apparent loss of H3PO4 from the main headgroup ion at m/z 225.1000 results in a fragment ion at m/z 127.1234, (EC C7H15N2+; Δmmu = 0.42) and a further loss of N2H2, suggesting the two nitrogen are adjacent to each other in the molecule, results in a fragment ion at m/z 97.1012 with an EC of \({{{{{{{\mathrm{C}}}}}}}}_7{{{{{{{\mathrm{H}}}}}}}}_{13}^ +\) (Δmmu = −0.52). We, therefore, tentatively identify this lipid as a C16:1/C16:1 phosphatidyldiacylglycerol bound to a diazoheptane moiety. The dominance of the fragment ion representing the headgroup in the MS2 spectrum is very similar to the MS2 behavior of polar headgroups with a quaternary ammonium (i.e., phosphocholine headgroups) and we, therefore, suggest the diazoheptane moiety is bound not via the terminal N, but to the adjacent N resulting in a quaternary ion. The extracted ion chromatogram of m/z 771.5667 (as per the molecular network, cf. Fig. 3) revealed several additional peaks at 18.2–18.5 min (Fig. S3) in the Sulfurimonas species grown with MnO2 with identical m/z and assigned EC. Interestingly, their fragmentation mass spectrum showed a similar pattern as described for the peak at 17.0–17.5 min (Fig. 3C), however the fragment ion representing the headgroup was now observed at m/z 197.0687 (Δmmu = −0.39) and the fragment ion related to the carbon chain in the hydrazine moiety was observed at m/z 69 with an EC of \({{{{{{{\mathrm{C}}}}}}}}_5{{{{{{{\mathrm{H}}}}}}}}_9^ +\). The fragment ion representing the diacyl core was now observed at m/z 575.5038 (C37H67O4+). In addition to a fragment indicating a C16:1 fatty acid, an additional fragment was observed at m/z 265 representing a C18:1 fatty acid. We therefore assigned this lipid as a phosphatidyldiazopentane diacylglycerol (16:1/18:1). Further inspection of the subnetwork revealed a series of diazoalkyl lipids varying both in the carbon number of the alkylhydrazine headgroup moiety from 5 to 11 and in their acyl carbon atoms from 34 to 36, leading to their tentative assignment as phosphatidyldiazoalkyl-diacylglycerols (hereafter, PDA-DAGs). The structural identification here is tentative, further structural elucidation methods (e.g., nuclear magnetic resonance [NMR] spectroscopy) would be required to confirm the proposed structures and elucidate the position of the various functional groups.
The PDA-DAGs with a short C5 alkyl chain in the headgroup were dominant in the two cultures that used MnO2 as the electron acceptor: Ca. S. marisnigri (90% of total PDA-DAGs) and Ca. S. baltica (74% of total PDA-DAGs) (Fig. 3B). In contrast, Ca. S. baltica GD2, grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\), produced much less C5 PDA-DAGs (27% of total PDA-DAGs), and mostly C7 PDA-DAGs, (73%) suggesting a membrane lipid adaptation depending on electron acceptors of this Sulfurimonas species. S. gotlandica GD1, unable to grow with MnO2 as electron acceptor, produced predominantly PDA-DAGs with a headgroup containing a longer alkyl chain (C7 and C11; 49% and 51% of total PDA-DAGs, respectively), while C5 PDA-DAGs were almost absent (0.1%). The PDA-DAGs composition of Ca. S. marisnigi SoZ1 grown with MnO2 and on S. gotlandica GD1 grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) were potentially attributed to their species specificities.
A search using the Mass Spectrometry Search Tool (MASST) of over 1000 public LC-MS datasets available in the GNPS platform database (including lipidomes, metabolomes etc.) was performed to see whether lipids structurally similar to the PDA-DAGs detected in this study of Sulfurimonas cultures had been detected in other organisms or environments [34, 37]. However, none of the databases showed matches to these unique lipids, suggesting that either it was due to a database bias or these novel components are potentially indicative of this taxonomic group, because of their specific metabolism or the environmental adaptation to their niche (redoxcline). Indeed, natural products containing a diazo group are rare [42,43,44] and their presence in the headgroups of microbial membrane lipids has not been reported before to the best of our knowledge. Although their structure remains tentative, one could speculate about their potential biosynthesis. There are enzymes reported that convert an amine group into a diazo functional group [42]. Primary alkyl amines can be enzymatically produced from fatty acids [45] or by decarboxylation of aliphatic amino acids [46]. The only truly enigmatic biochemical reaction would be the coupling of the diazo functional group with the phosphate moiety of the intact polar lipids.
Potential causes for membrane lipid adaptation with different electron acceptors
Our results revealed marked changes in the composition of (i) the short alkyl chain of the headgroup of PDA-DAG lipids and (ii) the acyl moieties of the PE- and PG-DAGs of Ca. S. baltica GD2 depending on the available terminal electron acceptor. It is well established from culture studies that temperature, pressure, pH, growth rate and nutrients are key factors in regulating the fatty acid and headgroup composition of membrane lipids of microorganisms [47,48,49,50]. In order to maintain membrane fluidity and permeability at the cultured conditions where pressure is elevated and temperature is low, microorganisms modify their fatty acid composition by increasing the amount of unsaturation and decreasing the chain lengths [21, 51, 52]. Such phenomena were also observed in the natural environment such as the ocean and global soils [53,54,55]. However, in studies of the eastern subtropical South Pacific, it has been shown that the chain length and degree of unsaturation can be an inherent biochemical property of a lipid class, rather than a function of temperature or pressure [56]. The sequential utilization of terminal electron acceptors is a function of potential energy yield based on thermodynamics which results in the mostly vertically progressive depletion of electron acceptors from O2 to \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\), Mn(IV), Fe(III) and \({{{{{{{\mathrm{SO}}}}}}}}_4^{2 - }\) [57]. Our results suggest that Ca. S. baltica GD2 produces shorter chain length lipids when using MnO2 instead of \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) for energy production. This may indicate that Ca. S. baltica GD2 modify their membrane permeability by changing their lipid composition to achieve electron exchange under MnO2 conditions or due to the different membrane proteins associated with lipids mediating these reactions. To our knowledge, this is the first report of changes in the microbial membrane lipid composition in response to different electron acceptors. Further studies are required to screen for the presence of PDA-DAGs in the natural environment and to determine the biological mechanisms underlying their presence in Sulfurimonas.
Conclusion
This study showed that the membrane lipid composition of a Sulfurimonas species, Ca. S. baltica GD2, depends on the electron acceptor used during culturing, either \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) or MnO2. Differences in the lipid composition of other two species, Ca. S. marisnigi SoZ1 grown with MnO2 and on S. gotlandica GD1 grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) were likely attributed to their species specificities. A range of novel diazoalkyl phospholipids were tentatively identified by using molecular network and high-resolution mass spectrometry. We observed that when carrying out MnO2-dependent sulfur oxidation, Ca. S. baltica GD2 possesses shorter acyl (fatty acid) chain lengths in its PE- and PG-DAGs and shorter alkyl chains in its phosphatidyldiazoalkyl-diacylglycerols lipid polar headgroups, compared to the ones produced when grown with \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\). These chain length modifications may be utilized by the cells to maintain membrane homeostasis when using MnO2 instead of \({{{{{{{\mathrm{NO}}}}}}}}_3^ -\) for energy production. Advances in untargeted lipidomic approaches and the development of computational tools will enable the more-efficient characterization of microbial lipidome and the search for unknowns, leading to a better understanding of their functions and metabolism under different environmental conditions.
Data availability
The HPLC-MS/MS datasets generated during the current study are available from the corresponding author on reasonable request.
References
Waite DW, Vanwonterghem I, Rinke C, Parks DH, Zhang Y, Takai K, et al. Comparative genomic analysis of the class Epsilonproteobacteria and proposed reclassification to Epsilonbacteraeota (phyl. nov.). Front Microbiol. 2017;8:682.
Waite DW, Vanwonterghem I, Rinke C, Parks DH, Zhang Y, Takai K, et al. Addendum: Comparative genomic analysis of the Class Epsilonproteobacteria and proposed reclassification to Epsilonbacteraeota (phyl. nov.). Front Microbiol. 2018;9:772.
Han Y, Perner M. The globally widespread genus Sulfurimonas: versatile energy metabolisms and adaptations to redox clines. Front Microbiol. 2015;6:989.
Lavik G, Stührmann T, Brüchert V, Van der Plas A, Mohrholz V, Lam P, et al. Detoxification of sulphidic African shelf waters by blooming chemolithotrophs. Nature. 2009;457:581–4.
Grote J, Labrenz M, Pfeiffer B, Jost G, Jürgens K. Quantitative distributions of Epsilonproteobacteria and a Sulfurimonas subgroup in pelagic redoxclines of the central Baltic Sea. Appl Environ Microbiol. 2007;73:7155–61.
Grote J, Schott T, Bruckner CG, Glöckner FO, Jost G, Teeling H, et al. Genome and physiology of a model Epsilonproteobacterium responsible for sulfide detoxification in marine oxygen depletion zones. Proc Natl Acad Sci. 2012;109:506–10.
Johnston DT, Gill BC, Masterson A, Beirne E, Casciotti KL, Knapp AN, et al. Placing an upper limit on cryptic marine sulphur cycling. Nature. 2014;513:530–3.
Walsh DA, Zaikova E, Howes CG, Song YC, Wright JJ, Tringe SG, et al. Metagenome of a versatile chemolithoautotroph from expanding oceanic dead zones. Science. 2009;326:578–82.
Wakeham SG, Amann R, Freeman KH, Hopmans EC, Jørgensen BB, Putnam IF, et al. Microbial ecology of the stratified water column of the Black Sea as revealed by a comprehensive biomarker study. Organic Geochem. 2007;38:2070–97.
Jost G, Martens-Habbena W, Pollehne F, Schnetger B, Labrenz M. Anaerobic sulfur oxidation in the absence of nitrate dominates microbial chemoautotrophy beneath the pelagic chemocline of the eastern Gotland Basin, Baltic Sea. FEMS Microbiol Ecol. 2009;71:226–36.
Taylor GT, Iabichella M, Ho T-Y, Scranton MI, Thunell RC, Muller-Karger F, et al. Chemoautotrophy in the redox transition zone of the Cariaco Basin: A significant midwater source of organic carbon production. Limnology Oceanogr. 2001;46:148–63.
Grote J, Jost G, Labrenz M, Herndl Gerhard J, Jürgens K. Epsilonproteobacteria represent the major portion of chemoautotrophic bacteria in sulfidic waters of pelagic redoxclines of the Baltic and Black Seas. Appl Environ Microbiol. 2008;74:7546–51.
Henkel JV, Dellwig O, Pollehne F, Herlemann DPR, Leipe T, Schulz-Vogt HN. A bacterial isolate from the Black Sea oxidizes sulfide with manganese(IV) oxide. Proc Natl Acad Sci. 2019;116:12153.
Henkel JV, Schulz-Vogt HN, Dellwig O, Pollehne F, Schott T, Meeske C, et al. Biological manganese-dependent sulfide oxidation impacts elemental gradients in redox-stratified systems: indications from the Black Sea water column. The ISME J. 2022;16:1523–33.
Herszage J, dos Santos Afonso M, Luther GW. Oxidation of cysteine and glutathione by soluble polymeric MnO2. Environ Sci Technol. 2003;37:3332–8.
Yao W, Millero FJ. Oxidation of hydrogen sulfide by hydrous Fe(III) oxides in seawater. Marine Chem. 1996;52:1–16.
Henkel JV, Vogts A, Werner J, Neu TR, Spröer C, Bunk B, et al. Candidatus Sulfurimonas marisnigri sp. nov. and Candidatus Sulfurimonas baltica sp. nov., thiotrophic manganese oxide reducing chemolithoautotrophs of the class Campylobacteria isolated from the pelagic redoxclines of the Black Sea and the Baltic Sea. System Appl Microbiol. 2021;44:126155.
Kiebish MA, Han X, Cheng H, Lunceford A, Clarke CF, Moon H, et al. Lipidomic analysis and electron transport chain activities in C57BL/6J mouse brain mitochondria. J Neurochem. 2008;106:299–312.
Brügger B. Lipidomics: Analysis of the lipid composition of cells and subcellular organelles by electrospray ionization mass spectrometry. Ann Rev Biochem. 2014;83:79–98.
Siliakus MF, van der Oost J, Kengen SWM. Adaptations of archaeal and bacterial membranes to variations in temperature, pH and pressure. Extremophiles. 2017;21:651–70.
Sinensky M. Homeoviscous adaptation—A homeostatic process that regulates the viscosity of membrane lipids in Escherichia coli. Proc Natl Acad Sci. 1974;71:522–5.
Popendorf KJ, Tanaka T, Pujo-Pay M, Lagaria A, Courties C, Conan P, et al. Gradients in intact polar diacylglycerolipids across the Mediterranean Sea are related to phosphate availability. Biogeosciences. 2011;8:3733–45.
Martin P, Van Mooy BAS, Heithoff A, Dyhrman ST. Phosphorus supply drives rapid turnover of membrane phospholipids in the diatom Thalassiosira pseudonana. The ISME J. 2011;5:1057–60.
Geiger O, González-Silva N, López-Lara IM, Sohlenkamp C. Amino acid-containing membrane lipids in bacteria. Progr Lipid Res. 2010;49:46–60.
Benning C, Huang ZH, Gage DA. Accumulation of a novel glycolipid and a betaine lipid in cells of Rhodobacter sphaeroides grown under phosphate limitation. Arch Biochem Biophys. 1995;317:103–11.
Van Mooy BAS, Fredricks HF, Pedler BE, Dyhrman ST, Karl DM, Koblížek M, et al. Phytoplankton in the ocean use non-phosphorus lipids in response to phosphorus scarcity. Nature. 2009;458:69.
Sebastián M, Smith AF, González JM, Fredricks HF, Van Mooy B, Koblížek M, et al. Lipid remodelling is a widespread strategy in marine heterotrophic bacteria upon phosphorus deficiency. ISME J. 2016;10:968–78.
Wang S, Jiang L, Hu Q, Cui L, Zhu B, Fu X, et al. Characterization of Sulfurimonas hydrogeniphila sp. nov., a Novel bacterium predominant in deep-sea hydrothermal vents and comparative genomic analyses of the Genus Sulfurimonas. Front Microbiol. 2021;12:626705.
Wang S, Jiang L, Liu X, Yang S, Shao Z. Sulfurimonas xiamenensis sp. nov. and Sulfurimonas lithotrophica sp. nov., hydrogen- and sulfur-oxidizing chemolithoautotrophs within the Epsilonproteobacteria isolated from coastal sediments, and an emended description of the genus Sulfurimonas. Int J System Evolut Microbiol. 2020;70:2657–63.
Labrenz M, Grote J, Mammitzsch K, Boschker HTS, Laue M, Jost G, et al. Sulfurimonas gotlandica sp. nov., a chemoautotrophic and psychrotolerant epsilonproteobacterium isolated from a pelagic redoxcline, and an emended description of the genus Sulfurimonas. Int J Syst Evolut Microbiol. 2013;63:4141–8.
Law KP, Zhang CL. Current progress and future trends in mass spectrometry-based archaeal lipidomics. Organic Geochem. 2019;134:45–61.
Collins JR, Edwards BR, Fredricks HF, Van Mooy BAS. LOBSTAHS: an adduct-based lipidomics strategy for discovery and identification of oxidative stress biomarkers. Anal Chem. 2016;88:7154–62.
Nothias L-F, Petras D, Schmid R, Dührkop K, Rainer J, Sarvepalli A, et al. Feature-based molecular networking in the GNPS analysis environment. Nat Methods. 2020;17:905–8.
Wang M, Carver JJ, Phelan VV, Sanchez LM, Garg N, Peng Y, et al. Sharing and community curation of mass spectrometry data with Global Natural Products Social Molecular Networking. Nat Biotechnol. 2016;34:828–37.
Bale NJ, Ding S, Hopmans EC, Arts MGI, Villanueva L, Boschman C, et al. Lipidomics of environmental microbial communities. I: Visualization of component distributions using untargeted analysis of high-resolution mass spectrometry data. Front Microbiol. 2021;12:659302.
Pluskal T, Castillo S, Villar-Briones A, Orešič M. MZmine 2: Modular framework for processing, visualizing, and analyzing mass spectrometry-based molecular profile data. BMC Bioinformatics. 2010;11:395.
Ding S, Bale NJ, Hopmans EC, Villanueva L, Arts MGI, Schouten S, et al. Lipidomics of environmental microbial communities. II: Characterization using molecular networking and information theory. Front Microbiol. 2021;12:659315.
Vences-Guzmán MÁ, Geiger O, Sohlenkamp C. Ornithine lipids and their structural modifications: from A to E and beyond. FEMS Microbiol Lett. 2012;335:1–10.
Sohlenkamp C, Geiger O. Bacterial membrane lipids: diversity in structures and pathways. FEMS Microbiol Rev. 2016;40:133–59.
Raetz C, Dowhan W. Biosynthesis and function of phospholipids in Escherichia coli. J Biol Chem. 1990;265:1235–8.
Gill C. Effect of growth temperature on the lipids of Pseudomonas fluorescens. Microbiology. 1975;89:293–8.
Nawrat CC, Moody CJ. Natural products containing a diazo group. Nat Product Rep. 2011;28:1426–44.
Sugai Y, Katsuyama Y, Ohnishi Y. A nitrous acid biosynthetic pathway for diazo group formation in bacteria. Nat Chem Biology. 2016;12:73–75.
Chen L, Deng Z, Zhao C. Nitrogen–nitrogen bond formation reactions involved in natural product biosynthesis. ACS Chem Biology. 2021;16:559–70.
Anderson R, Huang Y. Fatty acids are precursors of alkylamines in Deinococcus radiodurans. J Bacteriol. 1992;174:7168–73.
Kim DI, Chae TU, Kim HU, Jang WD, Lee SY. Microbial production of multiple short-chain primary amines via retrobiosynthesis. Nat Commun. 2021;12:173.
Wada H, Gombos Z, Murata N. Contribution of membrane lipids to the ability of the photosynthetic machinery to tolerate temperature stress. Proc Natl Acad Sci. 1994;91:4273–7.
DeLong EF, Yayanos AA. Adaptation of the membrane lipids of a deep-Sea bacterium to changes in hydrostatic pressure. Science. 1985;228:1101–3.
Chwastek G, Surma MA, Rizk S, Grosser D, Lavrynenko O, Rucińska M, et al. Principles of membrane adaptation revealed through environmentally induced bacterial lipidome remodeling. Cell Reports. 2020;32:108165.
Vinçon-Laugier A, Cravo-Laureau C, Mitteau I, Grossi V. Temperature-dependent alkyl glycerol ether lipid composition of mesophilic and thermophilic sulfate-reducing bacteria. Front Microbiol. 2017;8:1532.
Oshima M, Miyagawa A. Comparative studies on the fatty acid composition of moderately and extremely thermophilic bacteria. Lipids. 1974;9:476–80.
Sinensky M. Temperature control of phospholipid biosynthesis in Escherichia coli. J Bacteriol. 1971;106:449–55.
Fang J, Barcelona MJ, Nogi Y, Kato C. Biochemical implications and geochemical significance of novel phospholipids of the extremely barophilic bacteria from the Marianas Trench at 11,000m. Deep Sea Res Part I: Oceanogr Res Pap. 2000;47:1173–82.
Weijers JWH, Schouten S, van den Donker JC, Hopmans EC, Sinninghe, Damsté JS. Environmental controls on bacterial tetraether membrane lipid distribution in soils. Geochimica et Cosmochimica Acta. 2007;71:703–13.
Holm Henry C, Fredricks Helen F, Bent Shavonna M, Lowenstein Daniel P, Ossolinski Justin E, Becker Kevin W, et al. Global ocean lipidomes show a universal relationship between temperature and lipid unsaturation. Science. 2022;376:1487–91.
Van Mooy BAS, Fredricks HF. Bacterial and eukaryotic intact polar lipids in the eastern subtropical South Pacific: Water-column distribution, planktonic sources, and fatty acid composition. Geochimica et Cosmochimica Acta. 2010;74:6499–516.
Jørgensen BB, Findlay AJ, Pellerin A. The biogeochemical sulfur cycle of marine sediments. Front Microbiol. 2019;10:849.
Smoot ME, Ono K, Ruscheinski J, Wang P-L, Ideker T. Cytoscape 2.8: new features for data integration and network visualization. Bioinformatics. 2010;27:431–2.
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13:2498–504.
Acknowledgements
Two anonymous referees are thanked for helpful comments and S. Schouten, P. Riekenberg and M.T.J. van der Meer for helpful discussions. We acknowledge A. Abdala, D. Dorhout, J. Ossebaar, M. Verweij, A. Mets for technical support. JSSD received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement no. 694569—MICROLIPIDS) and from a Spinoza award from NWO.
Author information
Authors and Affiliations
Contributions
SD, JVH, NB, LV, ECH and JSSD conceived the concept. SD carried out the data analyses. JVH cultured organisms. MK carried out the lipid extraction and ECH ran MS experiments. SD, ECH, NB and JSSD performed the lipid identification. SD wrote the initial manuscript with the input of all authors. JSSD and LV supervised the study. All authors edited and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Ding, S., Henkel, J.V., Hopmans, E.C. et al. Changes in the membrane lipid composition of a Sulfurimonas species depend on the electron acceptor used for sulfur oxidation. ISME COMMUN. 2, 121 (2022). https://doi.org/10.1038/s43705-022-00207-3
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1038/s43705-022-00207-3
This article is cited by
-
Selective lipid recruitment by an archaeal DPANN symbiont from its host
Nature Communications (2024)