Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Dimethylsulfoniopropionate concentration in coral reef invertebrates varies according to species assemblages


Dimethylsulfoniopropionate (DMSP) is a key compound in the marine sulfur cycle, and is produced in large quantities in coral reefs. In addition to Symbiodiniaceae, corals and associated bacteria have recently been shown to play a role in DMSP metabolism. Numerous ecological studies have focused on DMSP concentrations in corals, which led to the hypothesis that increases in DMSP levels might be a general response to stress. Here we used multiple species assemblages of three common Indo-Pacific holobionts, the scleractinian corals Pocillopora damicornis and Acropora cytherea, and the giant clam Tridacna maxima and examined the DMSP concentrations associated with each species within different assemblages and thermal conditions. Results showed that the concentration of DMSP in A. cytherea and T. maxima is modulated according to the complexity of species assemblages. To determine the potential importance of symbiotic dinoflagellates in DMSP production, we then explored the relative abundance of Symbiodiniaceae clades in relation to DMSP levels using metabarcoding, and found no significant correlation between these factors. Finally, this study also revealed the existence of homologs involved in DMSP production in giant clams, suggesting for the first time that, like corals, they may also contribute to DMSP production. Taken together, our results demonstrated that corals and giant clams play important roles in the sulfur cycle. Because DMSP production varies in response to specific species-environment interactions, this study offers new perspectives for future global sulfur cycling research.


Coral reefs have been described as dimethylsulfoniopropionate (DMSP) hotspots1,2. This compound is an important metabolite that plays a central role in the marine sulfur cycle3. It is involved in numerous cellular and ecological processes. Among several known functions, DMSP possesses antioxidant properties4,5 as evidenced by cellular increase of DMSP under CO2 depletion6. Dimethylsulfoniopropionate has other protective physiological functions, notably serving as an osmolyte and cryoprotectant in marine algae7,8. Acting as a signalling molecule, DMSP is also involved in antiviral defense mechanisms and sulfide detoxification9,10,11. Numerous studies have recently drawn attention to variations of DMSP concentration in organisms subjected to environmental changes. Marine algae and/or coral studies have demonstrated that DMSP concentration changes with light intensity and salinity, as well as in response to oxidative stressors6,12,13,14. DMSP is therefore believed to be involved in organisms’ stress response, especially in coral species14,15. Dimethylsulfoniopropionate has also been extensively studied for its role in climate regulation16, because it can be converted into dimethylsulfide (DMS), a trace gas that is a source of reduced sulfur and which plays a role in cloud formation17,18. Consequently, precise methodologies such as nuclear magnetic resonance (NMR) spectroscopy have been developed for accurate detection and quantification of DMSP16,19.

Dimethylsulfoniopropionate is found only in certain terrestrial or marine organisms. It has been recorded in graminoids, rhodophytes, mussels and also in benthic flatworms20,21,22,23. However, only a few species are able to produce DMSP, and the majority of organisms are believed to accumulate this molecule through their diets or phototrophic symbionts24. To date, pathways of DMSP biosynthesis have been characterized in higher plants22,25, marine algae such as Symbiodiniaceae26,27 and more recently in corals, with methionine as a common precursor15. Higher plants and marine algae use two different DMSP biosynthetic pathways, while corals harbor an algal-like pathway15,26. Similarly to marine algae, coral species such as Acropora sp. likely encode the enzymatic machinery required for biosynthesis of DMSP (specific aminotransferase, reductase, methyltransferase and decarboxylase enzymes)3,15. Even though the specific biosynthesis pathway has not yet been studied in other marine organisms, DMSP has been found in other coral taxa (e.g. Pocilloporidae, Poritidae) and sessile organisms such as giant clams and anemones20,28. The concentrations of DMSP were higher in corals (Acropora sp., Heliopora sp., Pavona sp.) and giant clams (Tridacna maxima and Tridacna squamosa) than in other animals20,28. Dinoflagellates of the family Symbiodiniaceae account for most of the DMSP production in coral reefs24,29. This family is currently classified into nine divergent lineages, hereafter referred to as clades A to I30, of which seven have recently been elevated to genus level31, with clades A-D commonly found in symbiosis with scleractinian corals and giant clams32. Symbiodiniaceae can possess a DMSP-lyase activity leading to the conversion of DMSP into DMS that differs among clades33. Thus, the presence or proportion of different Symbiodiniaceae clades within a host may influence the variation of DMSP concentration often observed between coral species34. Additionally, the differences observed in coral studies could also be the result of other symbionts hosted by corals. Certain bacteria possess crucial genes for DMSP degradation such as homologs of the dmdA gene that are involved in the demethylation pathway or of the dddD gene known to be involved in the degradation of DMSP into DMS35,36,37,38. These homologs have been found in Proteobacteria that are common in coral species. Additionally, the occurrence of bacteria involved in DMSP production in coral is possible39, further highlighting the complexity of DMSP-cycling processes in benthic holobionts. Most of these bacteria have been recently described in microbiomes of giant clams40. Further, DMSP production in heterotrophic bacteria has also been recently highlighted41.

It is well known that coral symbionts such as dinoflagellates (Symbiodiniaceae) and bacteria vary according to environmental conditions42,43,44,45,46, and are therefore assumed to play a key role in their host fitness47. In addition to symbionts, interaction between holobionts could also impact the host fitness. While it is well known that marine sessile organisms interact, either directly or indirectly48,49,50, only a few studies have underlined the role of benthic species assemblages on coral reef dynamics51,52, as well as on giant clams’ fitness40. To better understand how benthic species assemblages could influence their respective fitness, we artificially combined three coral reef builders, including two coral species (Pocillopora damicornis; Linneaus, 1758 and Acropora cytherea; Dana, 1846) and one giant clam species (Tridacna maxima; Röding, 1798), and measured DMSP concentration in each species using NMR spectroscopy. Metabarcoding was used to characterize the Symbiodiniaceae composition within the three studied species and determine the putative role of Symbiodiniaceae in DMSP production. We also examined whether giant clams contribute to DMSP production by screening our previously described transcriptomes of T. maxima53 for enzymes involved in the DMSP biosynthesis. Finally, because DMSP is a stress biomarker, the experiments were performed either at lagoon temperature or under thermal stress.

Material and Methods

Coral and giant clam collection and experimental design

The coral and giant clam samples as well as the experimental design used in this study were described previously52. Briefly, coral species of A. cytherea and P. damicornis were collected in Moorea lagoon, French Polynesia (17°30′S, 149°50′W, Linareva fringing reef)54. For each species, 4 colonies were sampled and nubbinized into 45 small fragments. Giant clams of the species T. maxima were purchased from a French Polynesian nursery on Reao Island (18°28′S, 136°25′W; N°Tahiti: 139 519). Corals nubbins and giant clams were reared in a common garden. A CITES permit was obtained to allow specimen exports (CITES – FR1698700087 – E).

Experiments were conducted in open-circuit aquariums (20 L/h), using distinct assemblages of either one species P. damicornis (P, n = 4 aquariums) or T. maxima (T, n = 4 aquariums), two species P. damicornis + A. cytherea (PA, n = 4 aquariums) or A. cytherea + T. maxima (AT, n = 4 aquariums), or three species P. damicornis + A. cytherea + T. maxima (PAT, n = 4 aquariums; Figure S1a). Three nubbins per coral colony (n = 3 nubbins x 4 colonies per species) and 12 giant clams were used when required in an assemblage. Coral fragments (5–8 cm in height and 2 cm in diameter) and giant clams (4–8 cm in size) were spaced 5 to 10 cm apart to avoid contact. Assemblages were either under thermal stress (32 °C, S, n = 10 aquariums) or at lagoon temperature (27 °C, L, n = 10 aquariums). After 12 days of acclimation at 27 °C (time 0), half of the aquariums (n = 10) were placed under thermal stress by increasing the temperature by 1 °C per day until it reached 32 °C on day 17 (Fig. S1b). At each sampling time (day 12 and 17), half of the sampled giant clam mantles (4 cm2) and 80% of the sampled coral fragments were immediately snap-frozen in liquid nitrogen and stored at −80 °C until further analysis. The remaining portions of the sampled giant clams and corals were stored in 70% ethanol for DNA analysis. Seawater temperature data were recorded every 10 minutes with Temperature/Light Data Loggers (P/N U22–001, Onset, Bourne, Massachusetts; or Ruskin, Ottawa, Canada; Fig. S1b). Seawater temperature was controlled with the Biotherm pro system (Hobby, Stukenbrock, Germany). Health status checks on the coral nubbins were carried out using the standardized coral coloration scale as a proxy of symbiont density and chlorophyll a content variation (Coral Health Chart,,56, and through daily visual observation for giant clams.

Metabolite extraction for dimethylsulfoniopropionate quantification

Coral samples were lyophilized overnight, and 1 to 1.5 g of powder was obtained for extraction. The extraction protocol was adapted from Tapiolas and collaborators19. In total, 51 A. cytherea and 54 P. damicornis samples were extracted with 3 mL of HPLC-grade methanol (CH3OH). After 5 minutes of sonication (35 kHz Transsonic 950/H, Elma, Germany) at room temperature, the extracts were shaken for 3 hours. A second extraction with an additional 2 mL of CH3OH was performed using 5 minutes of sonication and 5 minutes of shaking. The two extracts were then pooled and dried using a vacuum centrifuge drier (Genevac EC-2 plus, Genevac, UK) before being stored at −20 °C until NMR analysis. Samples were lyophilized overnight and weighed to measure the quantity of extracts.

Giant clam samples (n = 42) were lyophilized overnight and 0.1 to 0.5 g of powder was obtained for extraction. Protocol extraction was adapted from Mohamadi57. Solid-liquid extractions were performed using 2 mL of H20 and 0.5 mL of HPLC-grade methanol (CH3OH) prior to vortexing. After adding 1.5 mL of CH3OH and 2 mL of dichloromethane (CH2Cl2), the solutions were mixed using a vortex and sonicated for 10 minutes. The solid-liquid extraction was carried out three times and the three extracts were pooled and centrifuged at 2050 g for 20 min in order to separate the polar and apolar phases. The apolar extracts were pooled as per the polar extracts and lyophilized. In this study, only the hydroalcoholic (polar) extracts of the giant clam samples were analyzed.

All extracts were resolubilized in 500 μL of deuterium oxide (D2O), vortex mixed and transferred into a 5 mm NMR tube, then analyzed immediately by 1H-NMR spectroscopy.

Nuclear magnetic resonance data acquisition and dimethylsulfoniopropionate quantification

All 1H spectra were recorded on a Bruker Avance 1 spectrometer (Bruker, Germany) at 300 MHz in the same conditions with 128 transients; chemical shifts were reported in ppm from tetramethylsilane (TMS). Absolute Area was determined by NMR analysis on the topspin 2.1 Bruker software.

Two external calibration curves of DMSP were performed under the same conditions as the samples with 128 transients (Figure S1c,d). Integration zone A (3.41 ppm to 3.34 ppm) was used for A. cytherea and P. damicornis samples because there were no overlapping signals. Integration zone B was used for T. maxima samples because of overlapping signals on zone A and no overlapping signal on zone B (Fig. S1e–h). The concentrations of DMSP calculated with the calibration curve (in mg per mL) were normalized with the mass ratio corresponding to the dry weight of the extract. Analysis of variance (ANOVA) and ad hoc pairwise comparisons (Tukey test) were performed using R software58.

DNA extraction; PCR amplification and sequencing of Symbiodiniaceae communities

Coral DNA extractions (n = 2 to 9 per assemblage) were performed following a Cethyl Trimethylammonium Bromide (CTAB) based protocol59. Symbiodiniaceae diversity was studied using the internal transcribed spacer 2 (ITS2) marker. The ITS2 region of the nuclear ribosomal array was amplified using the forward primer “its-Dino” (5′-GTGAATTGCAGAACTCCGTG-3′) and the reverse primer “ITS2-Rev2” (5′-GCCTCCGCTTACTTATATGCT-3′)60,61. IlluminaTM overhang adaptors were included in the primers as described in Kozich et al.62. PCR amplification and sequencing were performed as described in Guibert et al.52.

Symbiodiniaceae composition analysis of A. cytherea and P. damicornis samples was adapted from the Arif et al. (2014) pipeline in mothur v1.3963. An ITS2 database64 was used to annotate the sequences via Basic Local Alignment Search Tool (BLASTN). The relative abundance of Symbiodiniaceae subclades was represented using R (Reshape2 and ggplot2 packages).

Symbiodiniaceae composition analysis of T. maxima samples was processed using the same methodology as described in Guibert et al.40.

Identification of candidate genes

A previous predicted peptide data set of T. maxima metatranscriptome53 was blasted against a custom database at a threshold of 10−3. This database contained molluscan and cnidarian sequences extracted from the Protein database of the National Center for Biotechnology Information (NCBI) filtered with the DMSP enzyme names or EC numbers and from the DMSP candidate genes previously identified in Acropora millepora15 (Table S2, sheet 2). The selected sequences were blasted against the whole NCBI Protein database and any sequence with a top hit with a plant, algae or bacterial taxonomy was considered a false positive and removed (Table S2, sheet 1).

A differential gene expression analysis was then performed from previous RNAseq results of T. maxima under thermal stress at 32 °C for 5 days (Bioproject accession number: PRJNA309928). Transcript quantification of the above selected sequences was performed with RNA-Seq by Expectation Maximization (RSEM)65 across 14 samples of the Bioproject (8 biological samples: B0Tx and BT0 as the control group and 6 BTx biological samples as the Heating group53). The differential expression between the 2 groups was performed with the edgeRun package66 with 50,000 iterations.


Analyses of dimethylsulfoniopropionate concentrations according to thermal stress, species and assemblages

Dimethylsulfoniopropionate concentration was evaluated by 1H-NMR spectroscopy and standardized with the mass ratio in the three species: P. damicornis, A. cytherea and T. maxima (Table S3). An ANOVA test (Table 1) revealed significant differences among assemblages (p = 1.01e-11) or species (p < 2e-16) with a significant interaction among them (p = 3.90e-08). No significant differences between assemblages and temperatures were found. Thus, the DMSP data were pooled by assemblages for each species.

Table 1 Results of ANOVA test on DMSP concentration (mass ratio) by assemblage, species and temperature. Df: degrees of freedom; SumSq: Sum of squares; Mean Sq: mean sum of squares, F value: F statistic.

In corals, the concentrations of DMSP were from 5 to 10-fold lower in P. damicornis than in A. cytherea (Tukey test, p < 0.001), with means ranging between 0.65 and 0.87% (% = mg DMSP/100 mg extract) for P. damicornis and between 3.43 and 8.20% for A. cytherea (Fig. 1). In P. damicornis, while mean DMSP concentrations increased with the increasing complexity of assemblages, no significant differences were found (Fig. 1A). However, in A. cytherea, the DMSP concentration increased for the two-species assemblages (PA and AT) but decreased in PAT assemblages. A significant difference was found between AT and the other assemblages (A, p < 0.0001; PA, p < 0.05; PAT, p < 0.001; Fig. 1B).

Figure 1

Concentration of dimethylsulfoniopropionate (Mass ratio: mg/mg %) of Pocillopora damicornis (A) and Acropora cytherea (B) by assemblage. Assemblages: P: P. damicornis (nA = 25); A: A. cytherea (nB = 12); PA: P. damicornis and A. cytherea (nA = 16, nB = 16); AT: A. cytherea and T. maxima (nB = 8); PAT: P. damicornis, A. cytherea and T. maxima (nA = 13, nB = 15). Letters in lower case indicate significant differences between means (Tukey, p < 0.05). Photographs: Isis Guibert.

Interestingly, in T. maxima, mean DMSP concentrations decreased when exclusively associated with A. cytherea (from 1.97% to 0.98%) but increased in the PAT assemblage (3.22%, Fig. 2). Tridacna maxima exhibited a significantly higher DMSP concentration in the PAT assemblages than in the AT assemblages (p < 0.5).

Figure 2

Concentration of dimethylsulfoniopropionate (Mass ratio: mg/mg %) of Tridacna maxima by assemblage. Assemblages: T: T. maxima (n = 19); AT: A. cytherea and T. maxima (n = 9); PAT: P. damicornis, A. cytherea and T. maxima (n = 14). Letters in lower case indicate significant differences between means (Tukey, p < 0,05). Photograph: Isis Guibert.

Symbiodiniaceae composition by assemblage

The relative abundances of Symbiodiniaceae genotypes associated with the different assemblages and thermal stress conditions were measured at the clade and subclade levels (Fig. 3 and Table S4). Among the three Symbiodiniaceae clades (A, C and D) detected in the samples, clade D was the dominant lineage in P. damicornis nubbins (> 90%; Fig. 3a). Two D subclades were systematically detected in all nubbins. Subclade D17 represented more than half of the relative abundance (58–60%) and subclade D2 ranged from 39 to 44%. Clades A and C were only detected as background clades with subclades A1, A4, A6 and C66 sporadically present in the nubbins, either alone or associated, regardless of temperature levels and/or assemblages.

Figure 3

Relative abundance of Symbiodiniaceae subclades in Pocillopora damicornis (A) and Acropora cytherea (B). All samples were collected at day 17 except those with a 0 in their name that were collected at day 12. PAT: P. damicornis, A. cytherea and T. maxima; AT: A. cytherea and T. maxima; A: A. cytherea; P: P. damicornis; 0: time 0 - day 12; (1–3): number of the experiment; L: lagoon temperature; S: thermal stress; (1–4): number of sample. Photograph: Isis Guibert.

Three clades (A, C and D) were also detected in A. cytherea (Fig. 3b), although clade A largely dominated in all nubbins (n = 38), 24 of which harbored only this clade. Subclade A1 was dominant in the Symbiodiniaceae composition of all A. cytherea nubbins, with the presence of A2 subclade only once, and A6, 5 times, with both subclades occurring at a background level (<1%). Clade C was detected in 9 nubbins at the background level (<0.2%) except for one nubbin (A01.3 composed of 59% A and 41% C). Two C subclades, C3 and C3K were found concomitantly (in 4 nubbins) or separately (in 2 and 3 nubbins, respectively). Finally, clade D (subclade D10) was found in 5 nubbins, two of which contained D10 above the background level (7% and 14%). No significant correlation between Symbiodiniaceae clade/subclade and host assemblages or temperature was found. Moreover, no variation of nubbin color, was recorded according to temperature levels and/or assemblages meaning that there was no drastic variation in Symbiodiniaceae densities.

Identification of candidate genes for dimethylsulfoniopropionate biosynthesis in Tridacna maxima

The blasting of the predicted peptide T. maxima dataset against a DMSP enzyme custom database allowed the selection of 475 predicted peptides potentially involved in pathways of DMSP biosynthesis (Table S2 and S5, sheet 2). After functional clustering, all identified peptides with potential roles in DMSP synthesis were found to exhibit at least 1 homolog per step. Among them, 69 were significantly (False Discovery Rate <0.05) up-regulated during a thermal stress event (Table S5, sheet 1). More precisely, 29 peptides were identified as homologs to those of the coral pathway: 6 homologs of AT1 and AT5 for transamination step, 16 for the reduction step, 6 homologs of REDOX2,3,4,6 and 10, for the methylation step of METHYL 1 and 2, and one homolog of DECARB3 for the decarboxylation step (Fig. 4).

Figure 4

Putative pathway of dimethylsulfoniopropionate biosynthesis in Tridacna maxima. ID numbers of T. maxima peptides are indicated for each step.


Dimethylsulfoniopropionate concentrations differ according to species and assemblage

Dimethylsulfoniopropionate concentrations varied among the three invertebrate species analyzed. In A. cytherea, DMSP concentrations were 5 to 10-fold higher than in P. damicornis. A previous study performed on Pocilloporidae and Acroporidae, measured DMSP concentrations of, respectively, 0.333 nmol/mm2 and 2.473 nmol/mm2 in P. damicornis and A. millepora19, which is consistent with our results. Also, similar DMSP concentrations were found in three different species from the Pocilloporidae family, while they were highly variable in Acroporidae19. As we used different NMR quantification approaches for corals and giant clams, we were not able to directly compare concentrations between them. Nevertheless, beyond differences between species, our results revealed that the DMSP concentrations might vary in a species according to its close environment, namely interspecies assemblages.

In corals, increases in DMSP have been associated with direct sunlight, thermal stress, air exposure or oxidative stressors3,5,14. Measurements of DMSP concentration in Acropora species (A. aspera, A. tenuis and A. millepora) after 467 and 53 days at 31 °C revealed an increase of DMSP concentration in bleached corals. Here we show that after 2 days at >31 °C the DMSP production, in still unbleached A. Cytherea, P. damicornis and T. maxima, is not sufficiently enhanced to detect a significant difference in DMSP concentrations between stressed and unstressed corals and giant clams. This indicates that at least three days of temperature stress and/or the concurrent loss/gain of a certain quantity of Symbiodiniaceae/bacteria might be required for enhancing DMSP production. However, significant variations of DMSP concentrations were observed among the three species. These variations occurred in different environmental contexts. For A. cytherea, DMSP concentration increased significantly in AT assemblages but not in PA and PAT assemblages, whereas for T. maxima, it increased in PAT assemblages but decreased in association with A. cytherea. These previous results suggest that DMSP concentration in coral reef holobionts is dependent on neighboring species, possibly acting as a stressor14 or an activator of the sulfur metabolism. For P. damicornis, the association with A. cytherea alone or with T. maxima does not lead to a highly significant increase in DMSP. For A. cytherea, an increase of DMSP is observed in the AT assemblage. However, when P. damicornis is also present in the assemblage (PA or PAT), this effect disappeared. Therefore, the presence of P. damicornis prevents the clams effect’, highlighting possible yet uncharacterized interspecies communication responses. It is also known that, depending on the species, the variation in DMSP is not always observed under the same environmental factor. For example, a decrease in salinity leads to a decrease of DMSP in A. millepora, but such an effect had not previously been observed in Stylophora pistillata and P. damicornis68. In the present study, contrary to A. cytherea, the DMSP concentration in giant clams was increased in PAT assemblages rather than in AT assemblages. For giant clams, the presence of the two coral species seems to have a stronger effect than the sole presence of A. cytherea. In the case of DMSP as an indicator of stress, we have previously shown that giant clam mortality occurred in presence of A. cytherea, in both AT and PAT assemblages40. However, giant clam DMSP increases occurred only in PAT assemblages and we did not find any correlation between health status and DMSP levels in giant clams. Thus, the “health status” (i.e. declining clams) does not appear to systematically lead to a significant change in DMSP level. Taken together, our results suggest that DMSP concentration in the holobionts is influenced by their neighboring species, modifying the metabolism of the sulfur pathway.

Dimethylsulfoniopropionate production

The variations in DMSP concentrations among invertebrates harboring Symbiodiniaceae have been essentially attributed to the types and densities of their symbiotic dinoflagellates20. Significant DMSP variation also exists between Symbiodiniaceae clades5,20,33,67. Consequently, the types and densities of Symbiodiniaceae possibly impact the DMSP concentrations in holobionts, by translocating DMSP to their hosts69,70. Nevertheless, DMSP concentrations and the types of clades in our samples were not correlated. While the DMSP concentration of A. cytherea was high in AT assemblages and low in PAT assemblages, the relative abundance of clade A and D were similar in both assemblages. If Symbiodiniaceae are indeed responsible for the DMSP concentrations, we have to assume that their density should be higher in A. cytherea nubbins in AT than in other assemblages. Moreover, as DMSP concentrations in P. damicornis nubbins are between 5 and 10-fold lower than in A. cytherea nubbins, and since clade D has been described as a lower DMSP producer than clade C67, we can also hypothesize that clade D is a lower DMSP producer than clade A, a permanent clade in A. cytherea. Another non-exclusive alternative hypothesis for explaining the difference in DMSP production between A. cytherea and P. damicornis relies on a difference in Symbiodiniaceae densities between these two coral species, as observed in our previous work, A. cytherea harboring more Symbiodiniaceae than P. damicornis in Moorea59. Regarding giant clams, as for corals, Symbiodiniaceae types are not sufficient to account for DMSP concentrations variations between assemblages. At the individual level, giant clams in the same assemblage, with either 99% of subclade C66 or 99% of subclade A6 did not exhibit any difference in DMSP concentration40. Thus, even if the Symbiodiniaceae are well known DMSP producers, our results suggested that Symbiodiniaceae genotypes did not account for the DMSP variations observed in our study.

In addition to Symbiodiniaceae, corals and giant clams harbor numerous other associated organisms, including various families of bacteria. Raina and collaborators37 reported that more than 65% of the bacteria genera playing a role in DMSP/DMS metabolism can be found in association with coral species. The bacterial communities associated within the giant clams of this experiment have already been characterized40. By overlapping the data from both studies, we demonstrated that several bacterial genera involved in the degradation of DMSP or DMS in corals were also associated with T. maxima (Fig. 5). This suggests that the microbial communities of giant clams are also structured by the presence of DMSP and could indicate that bacteria influence the composition and variation in DMSP concentration between assemblages. However, as for Symbiodiniaceae, no link was found between bacterial composition and assemblages in T. maxima. Altogether our results suggested that the variation in DMSP concentration observed here did not result from Symbiodiniaceae or bacterial genera present in the sample, but might result from differences in their relative densities.

Figure 5

Marine bacterial genera degrading DMSP or DMS in the water column, associated or not with corals and/or giant clams. DMSP-degrading (green circle) and DMS-degrading (brown circle) found in the water column, bacterial genera found in corals (blue oval) and in giant clams (green oval) (adapted from37,40; and further developed from71,72,73,74,75,76,77).

Recent studies have highlighted the existence of DMSP biosynthesis pathways in animals, and in corals notably3,15. The increase in DMSP concentration measured in some assemblages might also be due to an enhancement in DMSP biosynthesis by the hosts, including giant clams. Of relevance, we showed for the first time that, similarly to corals, the giant clam T. maxima possesses algal gene homologs putatively involved in DMSP biosynthesis pathways. Candidate genes were not only homologous to DMSP alga-like pathway genes (top blast hit), but were also overexpressed during a thermal stress event. As at least one candidate gene for each step of the alga-like pathway has been found, we assume that the whole pathway is present in T. maxima. Future studies on complete holobiont communities and gene expression by assemblage will be required to decipher the involvement of each holobiont partner in DMSP biosynthesis.

This study explored the influence of interspecies assemblages on DMSP concentrations in corals and giant clams. We showed that DMSP concentration is higher in A. cytherea when associated with T. maxima and higher in T. maxima when they are in a three species assemblage. Because the balance between Symbiodiniaceae genera was maintained regardless of host assemblages, our results also show that microbial communities cannot solely account for the DMSP differences observed within and among assemblages. Further studies are required to decipher if symbiont densities and/or the host alga-like DMSP biosynthesis pathway contribute to the differences in DMSP production between assemblages.

Data availability

Supplementary data for this article can be found at Aquatic Sciences Website. All data are available upon request.


  1. 1.

    Broadbent, A. D., Jones, G. B. & Jones, R. J. DMSP in corals and benthic algae from the Great Barrier Reef. Estuar. Coast. Shelf Sci. 55, 547–555 (2002).

    ADS  CAS  Article  Google Scholar 

  2. 2.

    Hill, R. W., Dacey, J. W. H. & Krupp, D. A. Dimethylsulfoniopropionate in reef corals. Bull. Mar. Sci. 57, 489–494 (1995).

    Google Scholar 

  3. 3.

    Raina, J. B. et al. DMSP biosynthesis by an animal and its role in coral thermal stress response. Nature 502, 677–80 (2013).

    ADS  CAS  PubMed  Article  Google Scholar 

  4. 4.

    Downs, C. A. et al. Oxidative stress and seasonal coral bleaching. Free Radic. Biol. Med. 33, 533–543 (2002).

    CAS  PubMed  Article  Google Scholar 

  5. 5.

    Deschaseaux, E. S. M. et al. Effects of environmental factors on dimethylated sulfur compounds and their potential role in the antioxidant system of the coral holobiont. Limnol. Oceanogr. 59, 758–768 (2014).

    ADS  CAS  Article  Google Scholar 

  6. 6.

    Sunda, W., Kieber, D. J., Kiene, R. P. & Huntsman, S. An antioxidant function for DMSP and DMS in marine algae. Nature 418, 317–320 (2002).

    ADS  CAS  PubMed  Article  Google Scholar 

  7. 7.

    Kirst, G. O. et al. Dimethylsulfoniopropionate (DMSP) in ice-algae and its possible biological role. Mar. Chem. 35, 381–388 (1991).

    CAS  Article  Google Scholar 

  8. 8.

    Trevena, A. J., Jones, G. B., Wright, S. W. & Van Den Enden, R. L. Profiles of DMSP, algal pigments, nutrients and salinity in pack ice from eastern Antarctica. J. Sea Res. 43, 265–273 (2000).

    ADS  CAS  Article  Google Scholar 

  9. 9.

    Evans, C., Malin, G., Wilson, W. H. & Liss, P. S. Infectious titers of Emiliania huxleyi virus 86 are reduced by exposure to millimolar dimethyl sulfide and acrylic acid. Limnol. Oceanogr. 51, 2468–2471 (2006).

    ADS  CAS  Article  Google Scholar 

  10. 10.

    Havill, D. C., Ingold, A. & Pearson, J. Sulphide tolerance in coastal halophytes. Vegetatio 62, 279–285 (1985).

    Article  Google Scholar 

  11. 11.

    Seymour, J. R., Simó, R., Ahmed, T. & Stocker, R. Chemoattraction todimethylsulfoniopropionate throughout the marine microbial food web. Science 329, 342–346 (2010).

    ADS  CAS  PubMed  Article  Google Scholar 

  12. 12.

    Van Rijssel, M. & Gieskes, W. W. C. Temperature, light, and the dimethylsulfoniopropionate (DMSP) content of Emiliania huxleyi (Prymnesiophyceae). J. Sea Res. 48, 17–27 (2002).

    Article  Google Scholar 

  13. 13.

    Van Alstyne, K. L., Gifford, S. A., Dohman, J. M. & Savedo, M. M. Effects of environmental changes, tissue types and reproduction on the emissions of dimethyl sulfide from seaweeds that form green tides. Environ. Chem. 13, 220–230 (2016).

    Article  CAS  Google Scholar 

  14. 14.

    Yost, D. M., Jones, R. J. & Mitchelmore, C. L. Alterations in dimethylsulfoniopropionate (DMSP) levels in the coral Montastraea franksi in response to copper exposure. Aquat. Toxicol. 98, 367–373 (2010).

    CAS  PubMed  Article  Google Scholar 

  15. 15.

    Aguilar, C. et al. Transcriptomic analysis of the response of Acropora millepora to hypo-osmotic stress provides insights into DMSP biosynthesis by corals. BMC Genomics 18, 612–626 (2017).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  16. 16.

    Rap, A. et al. Natural aerosol direct and indirect radiative effects. Geophys. Res. Lett. 40, 3297–3301 (2013).

    ADS  CAS  Article  Google Scholar 

  17. 17.

    Nunes, J. et al. Two intertidal, non-calcifying macroalgae (Palmaria palmata and Saccharina latissima) show complex and variable responses to short-term CO2 acidification. ICES J. Mar. Sci. 73, 887–896 (2016).

    Article  Google Scholar 

  18. 18.

    Simó, R. Production of atmospheric sulfur by oceanic plankton: Biogeochemical, ecological and evolutionary links. Trends Ecol. Evol. 16, 287–294 (2001).

    PubMed  Article  Google Scholar 

  19. 19.

    Tapiolas, D. M., Raina, J. B., Lutz, A., Willis, B. L. & Motti, C. A. Direct measurement of dimethylsulfoniopropionate (DMSP) in reef-building corals using quantitative nuclear magnetic resonance (qNMR). spectroscopy. J. Exp. Mar. Bio. Ecol. 443, 85–89 (2013).

    CAS  Article  Google Scholar 

  20. 20.

    Van Alstyne, K. L., Schupp, P. & Slattery, M. The distribution of dimethylsulfoniopropionate in tropical Pacific coral reef invertebrates. Coral Reefs 25, 321–327 (2006).

    ADS  Article  Google Scholar 

  21. 21.

    Becker, S. et al. The biology of an Antarctic rhodophyte, Palmaria decipiens: recent advances. Antarct. Sci. 23, 419–430 (2011).

    ADS  Article  Google Scholar 

  22. 22.

    Kocsis, M. G. et al. Dimethylsulfoniopropionate biosynthesis in Spartina alterniflora. Evidence that S-methylmethionine and dimethylsulfoniopropylamine are intermediates. Plant Physiol. 117, 273–281 (1998).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  23. 23.

    White, B. A., Lansing, E., Hill, R. W. & Dacey, J. W. H. Accumulation of Dimethylsulfonioprionate in Geukensia demissa Depends on trophic Interactions. Nature 189, 235–236 (1995).

    CAS  Google Scholar 

  24. 24.

    Van Alstyne, K. L., Dominique, V. J. & Muller-Parker, G. Is dimethylsulfoniopropionate (DMSP) produced by the symbionts or the host in an anemone-zooxanthella symbiosis? Coral Reefs 28, 167–176 (2009).

    ADS  Article  Google Scholar 

  25. 25.

    Hanson, A. D., Rivoal, J., Paquet, L. & Cage, D. A. Biosynthesis of 3-Dimethylsulfoniopropionate in Wollastonia biflora (L.) DC. Plant Physiol. 105, 103–110 (1994).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  26. 26.

    Stefels, J. Physiological aspects of the production and conversion of DMSP in marine algae and higher plants. J. Sea Res. 43, 183–197 (2000).

    ADS  CAS  Article  Google Scholar 

  27. 27.

    Keller, M. D., Bellows, W. K. & Guillard, R. R. L. Dimethyl sulfide production in marine phytoplankton. Biog. sulfur Environ. 393, 167–182 (1989).

    CAS  Article  Google Scholar 

  28. 28.

    Hill, R. W., Dacey, J. W. H. & Edward, A. Dimethylsulfoniopropionate in giant clams (Tridacnidae). Biol. Bull. 199, 108–115 (2000).

    CAS  PubMed  Article  Google Scholar 

  29. 29.

    Steinke, M., Brading, P., Kerrison, P., Warner, M. E. & Suggett, D. J. Concentrations of dimethylsulfoniopropionate and dimethyl sulfide are strain-specific in symbiotic dinoflagellates (Symbiodinium sp., dinophyceae). J. Phycol. 47, 775–783 (2011).

    PubMed  Article  Google Scholar 

  30. 30.

    Pochon, X. & Gates, R. D. A new Symbiodinium clade (Dinophyceae) from soritid foraminifera in Hawai’i. Mol. Phylogenet. Evol. 56, 492–497 (2010).

    CAS  PubMed  Article  Google Scholar 

  31. 31.

    LaJeunesse, T. C. et al. Systematic revision of Symbiodiniaceae highlights the antiquity and diversity of coral endosymbionts. Curr. Biol. 28, 1–11 (2018).

    Article  CAS  Google Scholar 

  32. 32.

    Neo, M. L., Eckman, W., Vicentuan, K., Teo, S. L. M. & Todd, P. A. The ecological significance of giant clams in coral reef ecosystems. Biol. Conserv. 181, 111–123 (2015).

    Article  Google Scholar 

  33. 33.

    Yost, D. M. & Mitchelmore, C. L. Dimethylsulfoniopropionate (DMSP) lyase activity in different strains of the symbiotic alga Symbiodinium microadriaticum. Mar. Ecol. Prog. Ser. 386, 61–70 (2009).

    ADS  CAS  Article  Google Scholar 

  34. 34.

    Van Alstyne, K. L. & Puglisi, M. P. DMSP in marine macroalgae and macroinvertebrates: Distribution, function, and ecological impacts. Aquat. Sci. 69, 394–402 (2007).

    Article  CAS  Google Scholar 

  35. 35.

    Howard, E. C. et al. Bacterial Taxa That Limit Sulfur Flux from the Ocean. Science. 314, 649–652 (2006).

    ADS  CAS  PubMed  Article  Google Scholar 

  36. 36.

    Littman, R. A., Willis, B. L., Pfeffer, C. & Bourne, D. G. Diversities of coral-associated bacteria differ with location, but not species, for three acroporid corals on the Great Barrier Reef. FEMS Microbiol. Ecol. 68, 152–163 (2009).

    CAS  PubMed  Article  Google Scholar 

  37. 37.

    Raina, J. B., Dinsdale, E. A., Willis, B. L. & Bourne, D. G. Do the organic sulfur compounds DMSP and DMS drive coral microbial associations? Trends Microbiol. 18, 101–108 (2010).

    CAS  PubMed  Article  Google Scholar 

  38. 38.

    Todd, J. D. et al. Structural and regulatory genes required to make the gas dimethyl sulfide in bacteria. Science. 315, 666–669 (2007).

    ADS  CAS  PubMed  Article  Google Scholar 

  39. 39.

    Raina, J. B., Tapiolas, D., Willis, B. L. & Bourne, D. G. Coral-associated bacteria and their role in the biogeochemical cycling of sulfur. Appl. Environ. Microbiol. 75, 3492–3501 (2009).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  40. 40.

    Guibert, I., Pochon, X., Lecellier, G. & Berteaux-lecellier, V. Metabarcoding reveals distinct microbiotypes in the giant clam Tridacna maxima. Microbiome 8, 57 (2020).

  41. 41.

    Curson, A. R. J. et al. Dimethylsulphoniopropionate biosynthesis in marine bacteria and identification of the key gene in this process. Nat. Microbiol. 2, 1–35 (2017).

    Article  CAS  Google Scholar 

  42. 42.

    Silverstein, R. N., Cunning, R. & Baker, A. C. Change in algal symbiont communities after bleaching, not prior heat exposure, increases heat tolerance of reef corals. Glob. Chang. Biol. 21, 236–249 (2015).

    ADS  PubMed  Article  Google Scholar 

  43. 43.

    Cunning, R., Yost, D. M., Guarinello, M. L., Putnam, H. M. & Gates, R. D. Variability of Symbiodinium communities in waters, sediments, and corals of thermally distinct reef pools in American Samoa. PLoS One 10, 1–17 (2015).

    Article  CAS  Google Scholar 

  44. 44.

    Ziegler, M., Seneca, F. O., Yum, L. K., Palumbi, S. R. & Voolstra, C. R. Bacterial community dynamics are linked to patterns of coral heat tolerance. Nat. Commun. 8, 14213 (2017).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  45. 45.

    Sharp, K. H., Pratte, Z. A., Kerwin, A. H., Rotjan, R. D. & Stewart, F. J. Season, but not symbiont state, drives microbiome structure in the temperate coral Astrangia poculata. Microbiome 5, 120 (2017).

    PubMed  PubMed Central  Article  Google Scholar 

  46. 46.

    Camp, E. F. et al. Corals exhibit distinct patterns of microbial reorganisation to thrive in an extreme inshore environment. Coral Reefs 3, 1–16 (2020).

  47. 47.

    Rosenberg, E. & Zilber-Rosenberg, I. Symbiosis and development: The hologenome concept. Birth Defects Res. Part C - Embryo Today Rev. 93, 56–66 (2011).

    CAS  Article  Google Scholar 

  48. 48.

    Chornesky, E. Induced development of sweeper tentacles on the reef coral Agaricia agaricites: A response to direct competition. Biol. Bull. 165, 569–581 (1983).

    PubMed  Article  Google Scholar 

  49. 49.

    Connell, J. H. et al. A long-term study od competition and diversity of corals. Ecol. Monogr. 74, 179–210 (2004).

    Article  Google Scholar 

  50. 50.

    Lang, J. Interspecific aggression by scleractinian corals. 2. why the race is not only to the swift. Bull. Mar. Sci. 23, 260–279 (1973).

    Google Scholar 

  51. 51.

    Adjeroud, M., Briand, M. J., Kayal, M. & Dumas, P. Coral assemblages in Tonga: Spatial patterns, replenishment capacities, and implications for conservation strategies. Environ. Monit. Assess. 185, 5763–5773 (2013).

    PubMed  Article  Google Scholar 

  52. 52.

    Guibert, I. et al. Differential effects of coral-giant clam assemblages on biofouling formation. Sci. Rep. 9, 1–12 (2019).

    CAS  Article  Google Scholar 

  53. 53.

    Dubousquet, V. et al. Changes in fatty acid composition in the giant clam Tridacna maxima in response to thermal stress. Biol. Open 5, 1400–1407 (2016).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  54. 54.

    Rouzé, H., Lecellier, G., Langlade, M., Planes, S. & Berteaux-Lecellier, V. Fringing reefs exposed to different levels of eutrophication and sedimentation can support the same benthic communities. Mar. Pollut. Bull. 92, 212–221 (2015).

    PubMed  Article  CAS  Google Scholar 

  55. 55.

    Siebeck, U. E., Marshall, N. J., Klüter, A. & Hoegh-Guldberg, O. Monitoring coral bleaching using a colour reference card. Coral Reefs 25, 453–460 (2006).

    ADS  Article  Google Scholar 

  56. 56.

    Siebeck, U. E., Logan, D. & Marshall, N. J. CoralWatch - a flexible coral bleaching monitoring tool for you and your group. Proc. 11th Int. Coral Reef Symp. 549–553 (2008).

  57. 57.

    Mohamadi, F. La métabolomique appliquée à l’étude de l’impact de stress environnementaux sur les coraux scléractiniaires (2014).

  58. 58.

    R core team. A language and environment for statistical computing (2014).

  59. 59.

    Rouzé, H. et al. An updated assessment of Symbiodinium spp. that associate with common scleractinian corals from Moorea (French Polynesia) reveals high diversity among background symbionts and a novel finding of clade B. PeerJ 5, 1–22 (2017).

    Article  CAS  Google Scholar 

  60. 60.

    Pochon, X., Pawlowski, J., Zaninetti, L. & Rowan, R. High genetic diversity and relative specificity among Symbiodinium-like endosymbiotic dinoflagellates in soritid foraminiferans. Mar. Biol. 139, 1069–1078 (2001).

    Article  Google Scholar 

  61. 61.

    Stat, M., Pochon, X., Cowie, R. O. M. & Gates, R. D. Specificity in communities of Symbiodinium in corals from Johnston Atoll. Mar. Ecol. Prog. Ser. 386, 83–96 (2009).

    ADS  CAS  Article  Google Scholar 

  62. 62.

    Kozich, J. J., Westcott, S. L., Baxter, N. T., Highlander, S. K. & Schloss, P. D. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the miseq illumina sequencing platform. Appl. Environ. Microbiol. 79, 5112–5120 (2013).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  63. 63.

    Schloss, P. D. et al. Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 75, 7537–7541 (2009).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  64. 64.

    Arif, C. et al. Assessing Symbiodinium diversity in scleractinian corals via next-generation sequencing-based genotyping of the ITS2 rDNA region. Mol. Ecol. 23, 4418–4433 (2014).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  65. 65.

    Li, B. & Dewey, C. N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12, 323 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  66. 66.

    Dimont, E., Shi, J., Kirchner, R. & Hide, W. edgeRun: An R package for sensitive, functionally relevant differential expression discovery using an unconditional exact test. Bioinformatics 31, 2589–90 (2015).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  67. 67.

    Deschaseaux, E. S. M. et al. Comparative response of DMS and DMSP concentrations in Symbiodinium clades C1 and D1 under thermal stress. J. Exp. Mar. Bio. Ecol. 459, 181–189 (2014).

    CAS  Article  Google Scholar 

  68. 68.

    Gardner, S. G. et al. Dimethylsulfoniopropionate, superoxide dismutase and glutathione as stress response indicators in three corals under short-term hyposalinity stress. Proc. R. Soc. B Biol. Sci. 283, 1–9 (2016).

    Article  CAS  Google Scholar 

  69. 69.

    Hill, R. W., Dacey, J. W., Hill, S. D., Edward, A. & Hicks, W. A. Dimethylsulfoniopropionate in six species of giant clams and the evolution of dimethylsulfide after death. Can. J. Fish. Aquat. Sci. 61, 758–764 (2004).

    CAS  Article  Google Scholar 

  70. 70.

    Yost, D. M. & Mitchelmore, C. L. Determination of total and particulate dimethylsulfoniopropionate (DMSP) concentrations in four scleractinian coral species: A comparison of methods. J. Exp. Mar. Bio. Ecol. 395, 72–79 (2010).

    CAS  Article  Google Scholar 

  71. 71.

    Alex, A. & Antunes, A. Genus-wide comparison of Pseudovibrio bacterial genomes reveal diverse adaptations to different marine invertebrate hosts. PLoS One 13, 1–25 (2018).

    Article  CAS  Google Scholar 

  72. 72.

    Zeng, Y. X., Qiao, Z. Y., Yu, Y., Li, H. R. & Luo, W. Diversity of bacterial dimethylsulfoniopropionate degradation genes in surface seawater of Arctic Kongsfjorden. Sci. Rep. 6, 1–10 (2016).

    Article  CAS  Google Scholar 

  73. 73.

    González, J. M. et al. Silicibacter pomeroyi sp. nov. and Roseovarius nubinhibens sp. nov., dimethylsulfoniopropionate-demethylating bacteria from marine environments. Int. J. Syst. Evol. Microbiol. 53, 1261–1269 (2003).

    PubMed  Article  CAS  Google Scholar 

  74. 74.

    Rossbach, S., Cardenas, A., Perna, G., Duarte, C. M. & Voolstra, C. R. Tissue-specific microbiomes of the Red Sea giant clam Tridacna maxima highlight differential abundance of Endozoicomonadaceae. Front. Microbiol. 10 (2019).

  75. 75.

    Shiu, J. H. et al. A newly designed primer revealed high phylogenetic diversity of endozoicomonas in coral reefs. Microbes Environ. 33, 172–185 (2018).

    PubMed  PubMed Central  Article  Google Scholar 

  76. 76.

    Schäfer, H. Isolation of Methylophaga spp. from marine dimethylsulfide-degrading enrichment cultures and identification of polypeptides induced during growth on dimethylsulfide. Appl. Environ. Microbiol. 73, 2580–2591 (2007).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  77. 77.

    Cai, L. et al. Season structures prokaryotic partners but not algal symbionts in subtropical hard corals. Appl. Microbiol. Biotechnol. 102, 4963–4973 (2018).

    CAS  PubMed  Article  Google Scholar 

Download references


This work was supported by LabexCorail “SynTox”. I. Guibert was a fellow of Sorbonne University – Doctoral School 129. We acknowledge the InterContinental Resort & Spa Moorea and the Moorea Dolphin Center for providing a coral garden protected area. We also thank Franck Lerouvreur, Pascal Ung, Ewen Morin and all the students from CRIOBE who helped us in the field. We are much indebted to Roy Benyon for English improvement of the manuscript.

Author information




Conceived and designed the experiments: I.G., V.B., P.R., G.L., Performed the experiments: I.G., V.D., V.B., Collected the samples: I.G., Processed the samples: I.G., F.B., G.L., I.B., Generated the data: F.B., V.D., X.P., G.L., Analysed the data: I.G., G.L., Wrote the manuscript: I.G., V.B., G.L., I.B., X.P.

Corresponding author

Correspondence to Isis Guibert.

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.

Summary statement Various assemblages of corals and giant clams modulate dimethylsulfoniopropionate concentration, a key compound in the marine sulfur cycle.

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

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Guibert, I., Bourdreux, F., Bonnard, I. et al. Dimethylsulfoniopropionate concentration in coral reef invertebrates varies according to species assemblages. Sci Rep 10, 9922 (2020).

Download citation


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing