Metaproteomics of a gutless marine worm and its symbiotic microbial community reveal unusual pathways for carbon and energy use

Abstract

Low nutrient and energy availability has led to the evolution of numerous strategies for overcoming these limitations, of which symbiotic associations represent a key mechanism. Particularly striking are the associations between chemosynthetic bacteria and marine animals that thrive in nutrient-poor environments such as the deep sea because the symbionts allow their hosts to grow on inorganic energy and carbon sources such as sulfide and CO(2). Remarkably little is known about the physiological strategies that enable chemosynthetic symbioses to colonize oligotrophic environments. In this study, we used metaproteomics and metabolomics to investigate the intricate network of metabolic interactions in the chemosynthetic association between Olavius algarvensis, a gutless marine worm, and its bacterial symbionts. We propose previously undescribed pathways for coping with energy and nutrient limitation, some of which may be widespread in both free-living and symbiotic bacteria. These pathways include (i) a pathway for symbiont assimilation of the host waste products acetate, propionate, succinate and malate; (ii) the potential use of carbon monoxide as an energy source, a substrate previously not known to play a role in marine invertebrate symbioses; (iii) the potential use of hydrogen as an energy source; (iv) the strong expression of high-affinity uptake transporters; and (v) as yet undescribed energy-efficient steps in CO(2) fixation and sulfate reduction. The high expression of proteins involved in pathways for energy and carbon uptake and conservation in the O. algarvensis symbiosis indicates that the oligotrophic nature of its environment exerted a strong selective pressure in shaping these associations.

Fig. 1.

Positive effect of symbiont enrichment using density-gradient centrifugation. Enrichment considerably increased the number of identified proteins from a given symbiont compared with analyses of whole worms. Average protein numbers were calculated for 2D LC-MS/MS experiments (SI Appendix, Table S1). n = 2 for γ1-symbiont enrichments; n = 2 for δ1-symbiont enrichments; and n = 4 for whole-worm samples. Both metagenomic and proteomic binning information was used for assignment of proteins to a symbiont.

Fig. 2.

Overview of symbiotic metabolism based on metaproteomic and metabolomic analyses. (A) Live O. algarvensis specimen. (B) Light micrograph of a cross section through O. algarvensis. The region containing the symbionts is highlighted in blue. (C) Metabolic reconstruction of symbiont and host pathways. The δ1- and δ4-symbionts are shown as a single cell, because most metabolic pathways were identified in the δ1-symbiont and only a small fraction of the same pathways were identified in the δ4-symbiont because of the low coverage of its metaproteome. 3-HPB, partial 3-hydroxypropionate bicycle; CM, cell material; CODH, carbon monoxide dehydrogenase (aerobic or anaerobic type); NiRes, nitrate respiration; OxRes, oxygen respiration; PHA, polyhydroxyalkanoate granule; S⁰, elemental sulfur; S_red, reduced sulfur compounds; SulOx, sulfur oxidation; Unk. TEA, unknown terminal electron acceptor.

Fig. 3.

Modified version of the 3-HPB in the γ1-symbiont. Reactions not needed for the assimilation of propionate and acetate are shown in the gray box; reaction 1 also can play a role in fatty acid metabolism. (1) Acetyl-CoA carboxylase (2004223475); (2) malonyl-CoA reductase; (3) propionyl-CoA synthase; (4) propionyl-CoA carboxylase (2004223080); (5) methylmalonyl-CoA epimerase (RASTannot_91923); (6) methylmalonyl-CoA mutase (RASTannot_20798); (7) succinyl-CoA:(S)-malate-CoA transferase (RASTannot_529, RASTannot_48547); (8) succinate dehydrogenase (2004223104, 2004223105); (9) fumarate hydratase (2004223692); (10 a,b,c) (S)-malyl-CoA/β-methylmalyl-CoA/(S)-citramalyl-CoA (MMC) lyase (RASTannot_91504); (11) mesaconyl-C1-CoA hydratase (β-methylmalyl-CoA dehydratase) (2004222675); (12) mesaconyl-CoA C1-C4 CoA transferase (RASTannot_38616); (13) mesaconyl-C4-CoA hydratase [(S)-citramalyl-CoA dehydratase] (RASTannot_6738).

Fig. 4.

Suggested role of H⁺-PPase in the δ1-symbiont. Energy is conserved through the use of a membrane-bound proton-translocating pyrophosphatase instead of a cytosolic pyrophosphatase. PPi is produced by abundantly expressed enzymes, which catalyze the initial steps of sulfate reduction, propionate oxidation, and acetate oxidation. Red numbers show the stoichiometry.

Fig. 5.

Comparison of the classical Calvin cycle with a proposed version that is more energy efficient. (A) The text book version of the Calvin cycle. (B) The more energy-efficient version of the Calvin cycle in the γ-symbionts through the use of PPi-dependent trifunctional 6-phosphofructokinase/sedoheptulose-1,7-bisphosphatase/phosphoribulokinase (green) and a proton-translocating pyrophosphatase/proton-translocating pyrophosphate synthase (H⁺-PPase/H⁺-PPi synthase) (red). The main differences between the cycles are highlighted in yellow. CM, cell membrane; DHAP, dihydroxyacetone phosphate; GAP, d-glyceraldehyde-3-phosphate; PPi, inorganic pyrophosphate; Sh-7-P, d-sedoheptulose-7-phosphate. (C) Overview of genes that are replaced by the trifunctional PPi-dependent enzyme in different organisms. (D) Colocalized H⁺-PPase/PPi-PFK genes in the γ-symbionts and other symbiotic and free-living bacteria.

Fig. P1.

Overview of symbiotic metabolism based on metaproteomic and metabolomic analyses. (A) Live O. algarvensis specimen. (B) Light micrograph of a cross section through O. algarvensis. The region containing the symbionts is highlighted in blue. (C) Schematic diagram of possible energy and carbon sources for the symbiosis. External sources of energy and carbon could include carbon monoxide, hydrogen, and carbon dioxide; internal sources could include reduced sulfur compounds and host waste products, such as acetate and glycine betaine. S_ox, oxidized sulfur compounds; S_red, reduced sulfur compounds.