Metabolism of 2,3-dihydroxypropane-1-sulfonate by marine bacteria

Article abstract of DOI:10.1039/c7ob00357a

Both enantiomers of the sulfoquinovose breakdown product 2,3-dihydroxypropane-1-sulfonate, an important sulfur metabolite produced by marine algae, were synthesised in a ³⁴S-labelled form and used in feeding experiments with marine bacteria. The labelling was efficiently incorporated into the sulfur-containing antibiotic tropodithietic acid and sulfur volatiles by the algal symbiont Phaeobacter inhibens, but not into sulfur volatiles released by marine bacteria associated with crustaceans. The ecological implications and the relevance of these findings for the global sulfur cycle are discussed.

Full text of DOI:10.1039/c7ob00357a

Organic &
Biomolecular Chemistry
View Article Online
View Journal
COMMUNICATION
Metabolism of 2,3-dihydroxypropane-1-sulfonate
by marine bacteria†
Cite this: DOI: 10.1039/c7ob00357a
b
Ersin Celik,a Michael Maczka,b Nils Bergen,c Thorsten Brinkhoff,c Stefan Schulz
Received 14th February 2017,
Accepted 15th March 2017
a
and Jeroen S. Dickschat
*
DOI: 10.1039/c7ob00357a
rsc.li/obc
Both enantiomers of the sulfoquinovose breakdown product
2,3-dihydroxypropane-1-sulfonate, an important sulfur metabolite
produced by marine algae, were synthesised in a 34S-labelled form
and used in feeding experiments with marine bacteria. The label-
ling was efficiently incorporated into the sulfur-containing anti-
biotic tropodithietic acid and sulfur volatiles by the algal symbiont
Phaeobacter inhibens, but not into sulfur volatiles released by
marine bacteria associated with crustaceans. The ecological impli-
cations and the relevance of these findings for the global sulfur
cycle are discussed.
Sulfoquinovose (1) is a sugar that occurs in sulfoquinovosyl
diacylglycerol (SQDG) sulfolipids that were discovered more
than 50 years ago1 and are today known to generally occur in
higher plants and algae, and most photosynthetic bacteria.2–4
The estimated annual production of 1010 t (ref. 5) makes 1
besides methionine, cysteine, glutathione and dimethyl-
sulfoniopropionate (DMSP)6 one of the most abundant organic
sulfur compounds in nature. The bacterial degradation of 1 to
2,3-dihydroxypropane-1-sulfonate (6, DHPS) was initially
suggested to proceed via glycolysis,7 but recently a specific
catabolon with similar intermediates as in glycolysis was dis-
covered (Scheme 1A).8 The absolute configurations of the
pathway intermediates were not specified in the original
report, but assuming that all stereocentres in the catabolites of
1 are retained, the pathway will end at (S)-6. This metabolite is
also a major constituent in the cytosol and massively excreted
by the diatom Thalassiosira pseudonana,9 an organism that
Scheme 1 Sulfoquinovose (1) and DHPS catabolism. (A) Degradation of
1 to DHPS (S)-6 by E. coli, (B) conversion of (S)-6 to (R)-8, and (C)
alternative catabolic pathways for the degradation of (R)-8 to sulfite.
Enzymes shown in bold are encoded in P. inhibens DSM 17395.
aKekulé-Institut für Organische Chemie und Biochemie, Rheinische Friedrich-
Wilhelms-Universität Bonn, Gerhard-Domagk-Straße 1, 53121 Bonn, Germany.
E-mail: dickschat@uni-bonn.de; Tel: +49 228 735797
bInstitut für Organische Chemie, TU Braunschweig, Hagenring 30, 38106
Braunschweig, Germany
cInstitut für Chemie und Biologie des Meeres, Universität Oldenburg, Carl-von-
Ossietzky-Straße 9-11, 26129 Oldenburg, Germany
lives in symbiosis with bacteria from the Roseobacter group by
providing DHPS, while the alga depends on exogenous vitamin
B12 excreted by the bacteria.10 The expression of bacterial
genes for uptake and catabolism of DHPS is strongly upregu-
lated in cocultures of the diatom and its bacterial symbionts.9
Therefore, DHPS is a potential source of carbon and sulfur for
†Electronic supplementary information (ESI) available: Synthetic procedures,
isolation and identification of new bacterial strains, analytical data from feeding
experiments with (34S)DHPS, information about genes for DHPS catabolism in
P. inhibens, and NMR spectra of synthetic compounds. See DOI: 10.1039/
c7ob00357a
This journal is © The Royal Society of Chemistry 2017
Org. Biomol. Chem.

View Article Online
Communication
Organic & Biomolecular Chemistry
marine bacteria that are capable of DHPS uptake and and dimethyl trisulfide by P. inhibens,22,29,30 which is consist-
catabolism.
ent with an efficient degradation via the DMSP demethylation
DHPS degradation makes use of the enzymes encoded in pathway that ends at methanethiol.31 As an extension of this
the hps gene cluster11,12 that spans genes for the transcrip- work the use of DHPS by P. inhibens as a sulfur source for sec-
tional regulator HpsR, the major facilitator superfamily trans- ondary metabolites is addressed here.
porter HpsU, the dehydrogenases HpsO, HpsP and HpsN, and
For the investigation of biosynthetic pathways feeding
the (R)-sulfolactate sulfo-lyase SuyAB.13,14 The end product of experiments with isotopically labelled compounds are
the sulfoquinovose degradation (S)-6 may be epimerised by useful.32 Therefore, racemic (34S)DHPS (6) was synthesised
HpsO and HpsP via 2-oxo-3-hydroxypropane-1-sulfonate (7) to starting from solketal (17) that was converted into the mesylate
(R)-6 (Scheme 1B), followed by HpsN oxidation to 3-sulfolactate 18 (Scheme 3). Nucleophilic substitution with K34SCN31
(8) with an unknown absolute configuration,12 however SuyAB yielded the thiocyanate 19 that was reduced to the thiol 20
requires the substrate (R)-8 that reflects likely the enantiomer with LiAlH4. Its oxidation with H2O2 proceeded with simul-
provided by HpsN. Besides the cleavage to pyruvate (11) and taneous loss of the protecting group to yield (34S)DHPS with
SO32− by SuyAB13,14 a few other pathways for the further degra- 24% yield via four steps.
dation of (R)-8 are known (Scheme 1C), including oxidation to
P. inhibens DSM 17395 liquid cultures release the sulfur
sulfopyruvate (9) by SlcD,15 transamination to L-cysteate (12) by volatile dimethyl disulfide (DMDS), its oxidation product
an unidentified enzyme16 and desulfonation to 11 by the PLP- S-methyl methanethiosulfonate and dimethyl trisulfide
dependent CuyA.17 Decarboxylation to sulfoacetaldehyde (10) (DMTS), and small amounts of dimethyl tetrasulfide (Fig. 1 of
by ComDE followed by desulfonation by Xsc yields acetyl phos- the ESI†). Synthetic (rac)-(34S)DHPS was fed to a P. inhibens
phate (13) that enters the C2 pool and is used as a carbon liquid culture and the volatiles emitted by this strain were
source.15,18,19 The sulfite may simply be exported by the trapped on a charcoal filter by the use of a closed-loop strip-
specific exporter TauE,20 but could also be a source for other ping apparatus (CLSA).33 Filter extracts with CH2Cl2 were ana-
sulfur metabolites. We show here by isotopic labelling experi- lysed by GC/MS, showing the incorporation of 34S-labelling
ments that the DHPS sulfur is efficiently used for the bio- into dimethyl trisulfide and dimethyl tetrasulfide, but not into
synthesis of secondary metabolites and sulfur volatiles specifi- dimethyl disulfide and S-methyl methanethiosulfonate (Fig. 1
cally by marine bacteria associated with algae.
A
good candidate organism for our investigations is
Phaeobacter inhibens DSM 17395, a marine bacterium of the
Roseobacter group (family Rhodobacteraceae) that lives in sym-
biosis with marine algae and that is known to produce various
sulfur metabolites.21,22 In the symbiotic relationship with the
microalga Emiliania huxleyi the compatible solute dimethyl-
sulfoniopropionate (DMSP)23 is supplied as a carbon and
sulfur source to the bacterium,24 and, in return, P. inhibens
produces the growth promoter phenylacetic acid and the
proton antiporter25 tropodithietic acid (TDA, 16) that has anti-
biotic activity21 and protects the haptophyte from bacterial
infections.21 The biosynthesis of TDA depends on the TDA bio-
synthetic gene cluster26 and a biosynthetic model for the in-
corporation of sulfur into TDA proceeds via S-thiocysteine (15)
that can arise from cystine (14) via cleavage by PatB
(Scheme 2),27,28 an enzyme that has homology to cystathionine
β-lyases. This model was supported by the incorporation of
Scheme 3 Synthesis of (34S)DHPS.
2−
labelling from 34SO4 and, with much higher rates, from (34S)
cysteine and (34S2)cystine.27 We have also previously reported
about the incorporation of labelling from (methyl-2H6)DMSP
and (34S)DMSP into sulfur volatiles such as dimethyl disulfide
Fig. 1 EI mass spectra of (A) dimethyl trisulfide and (B) labelled DMTS
obtained after feeding of (rac)-(34S)DHPS.
Scheme 2 TDA and its biosynthesis from cystine via S-thiocysteine.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Organic & Biomolecular Chemistry
Communication
and Fig. 1 of the ESI†). Careful interpretation of the fragmenta- lator (black), the sulfoacetaldehyde acetyltransferase Xsc, for
tion pattern located the incorporated 34S-labelling at the the sulfite exporter TauE (green),20 and for an acetate kinase
central (non-methylated) sulfur(s) of DMTS (55% incorporation and a phosphate acetyltransferase (blue) for export and further
rate for the central sulfur) and dimethyl tetrasulfide (Fig. 1 of conversion of the Xsc products sulfite and acetyl phosphate
the ESI†) by an increase of those fragment ions that contain (13) are found.
one of the central sulfurs (m/z = 79 and m/z = 111), but not of
the fragment ion representing a MeS group (m/z = 47).
This genetic information demonstrates that P. inhibens can
convert both enantiomers of DHPS via the pathways shown in
To investigate which of the two enantiomers of 6 serves as a Scheme 1, yielding sulfite from 9 via degradation by ComDE
precursor for DMTS both (R)- and (S)-(34S)DHPS were and Xsc and possibly also via transamination and cleavage by
synthesised via the route shown in Scheme 3 from the com- CuyA. Sulfite may either be exported, but can also be used for
mercially available pure enantiomers of 17. Both compounds the biosynthesis of sulfur metabolites by reduction to sulfide
were obtained in high enantiomeric purity (Fig. 2 of the ESI†). via the sulfite reductase that is encoded in P. inhibens (acces-
Feeding of (R)- and (S)-(34S)DHPS to P. inhibens resulted in the sion number WP_014880398) and incorporation into cysteine
incorporation into the central sulfur atom of DMTS for both that was previously shown to be a good precursor both for
enantiomers (Fig. 3 of the ESI†), with incorporation rates of sulfur volatiles and TDA.27,29
10% from (R)-(34S)DHPS and 30% from (S)-(34S)DHPS. This
The link between the uptake of the algal sulfoquinovose
finding is explainable by the formation of hydrogen sulfide catabolite DHPS and TDA biosynthesis in P. inhibens is of eco-
from DHPS as a source of the central sulfur of DMTS, while logical importance, because P. inhibens colonises marine algae
the methylated sulfurs may directly arise from methionine such as Emiliania huxleyi and Ulva australis and protects the
degradation to methanethiol. The reaction of two units of plant host by excreting its antibiotic TDA.34 Recent research by
methanethiol with one equivalent of hydrogen sulfide under the Moran group has shown that DHPS excreted by marine
oxidative conditions may then result in DMTS. A clear in- algae is eagerly taken up by their bacterial symbionts. We have
corporation of labelling was also observed in both sulfur unraveled here additional functions of DHPS in the symbiotic
atoms of TDA as indicated by increased ions at [M + 2]+ and interaction between algae and bacteria, because the sulfur in
[M + 4]+ from both enantiomers of DHPS (9% incorporation this compound is used to produce the sulfur containing anti-
rate from (S)-(34S)DHPS and 24% from (R)-(34S)DHPS, Fig. 4 of biotic TDA for algal protection. To further investigate the speci-
the ESI†). These data indicate that both enantiomers of DHPS ficity of our findings five other marine bacteria, isolated from
are taken up by P. inhibens and can be used as a sulfur source the carapace of the crustacean Cancer pagurus and affiliated
for sulfur volatiles and TDA.
with the family Rhodobacteraceae (Table 1 of the ESI†), were
A detailed analysis of the genome of P. inhibens reveals the investigated for the incorporation of labelling from (rac)-(34S)
presence of several genes for the degradation of DHPS (Fig. 2 DHPS into DMTS, a sulfur volatile consistently found in the
and Table 2 of the ESI†). This includes the hps gene headspace extracts of all bacterial isolates. As a positive
cluster composed of genes for the regulator HpsR (black), for a control, (34S)cysteine was fed in the second series of experi-
tripartite transport system (green) presumably for DHPS ments. In all cases no incorporation of the sulfur labelling
uptake, for the dehydrogenases, HpsNOP, and directly from (rac)-(34S)DHPS could be observed, while the labelling
adjacent genes for a universal stress protein (purple), for the from (34S)cysteine was efficiently taken up into the central
sulfopyruvate decarboxylase ComDE, and the sulfolactate sulfur of DMTS (50–90% incorporation rate, for strain CP127
dehydrogenase SlcD. Upstream of this cluster a standalone an additional 15% incorporation was observed into the MeS
gene for the PLP-dependent L-cysteate sulfolyase CuyA and, groups, Fig. 5 of the ESI†). These observations reflect a highly
further upstream, a cluster of genes for a transcriptional regu- specific adaption of marine bacteria such as P. inhibens to
their algal symbionts, while other marine bacteria associated
with crustaceans lack the capacity for the degradation of
DHPS, a metabolite that is not produced by animals. Because
of the large amounts of DHPS produced in the oceans and the
high abundance of this compound especially in phytoplankton
blooms9 the conversion of DHPS into sulfur volatiles by bac-
teria associated with algae has ample effects and is likely of
similar importance for the global sulfur cycle as the well inves-
tigated bacterial DMSP degradation to volatile sulfur
compounds.31,35–37
Fig. 2 Genes for DHPS degradation in P. inhibens DSM 17395. Gray:
directly involved in DHPS degradation, green: transporters, black: tran-
scriptional regulators, purple: universal stress protein, blue: further con-
Conclusions
P. inhibens and some other bacteria of the Roseobacter group
version of acetyl phosphate. Numbers above arrows indicate nucleotide
positions of the borders of the gene clusters.
are well known for strong interactions with their hosts.24
This journal is © The Royal Society of Chemistry 2017
Org. Biomol. Chem.

View Article Online
Communication
Organic & Biomolecular Chemistry
Specific adaptations are also reflected by the results of our 15 K. Denger, J. Mayer, M. Buhmann, S. Weinitschke,
study and are in accordance with the isolation source. Several
T. H. M. Smits and A. M. Cook, J. Bacteriol., 2009, 191,
strains of P. inhibens that were obtained from different hosts
5648.
were shown to be metabolically extremely versatile and capable 16 C. A. R. M. Mikosch, K. Denger, E.-M. Schäfer and
of living with various eukaryotic organisms including algae in A. M. Cook, Microbiology, 1999, 145, 1153.
previous work.38 Besides the algal sulfur metabolite DMSP, we 17 K. Denger, T. H. M. Smits and A. M. Cook, Biochem. J.,
have shown here that the sulfur metabolite DHPS, an algal 2006, 394, 657.
breakdown product of sulfoquinovose, is taken up and used 18 R. H. White, Biochemistry, 1986, 25, 5304.
for the biosynthesis of the sulfur containing antibiotic TDA 19 J. Ruff, K. Denger and A. M. Cook, Biochem. J., 2003, 369,
and sulfur volatiles by P. inhibens. The new strains from the
275.
carapace of C. pagurus appear not to have the potential to use 20 S. Weinitschke, K. Denger, A. M. Cook and T. H. M. Smits,
DHPS which underpins the specificity of the adaption of
Microbiology, 2007, 153, 3055.
P. inhibens to its algal symbiont.
21 K. Kintaka, H. Ono, S. Tsubotani, S. Harada and
H. Okazaki, J. Antibiot., 1984, 37, 1294.
22 J. S. Dickschat, C. Zell and N. L. Brock, ChemBioChem,
2010, 11, 417.
23 J. S. Dickschat, P. Rabe and C. A. Citron, Org. Biomol.
Chem., 2015, 13, 1954.
24 M. R. Seyedsayamdost, R. J. Case, R. Colter and J. Clardy,
Nat. Chem., 2011, 3, 331.
Acknowledgements
This work was funded by the DFG (SFB TR51 “Roseobacter”).
25 M. Z. Wilson, R. Wang, Z. Gitai and M. R. Seyedsayamdost,
Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 1630.
26 H. Geng, J. B. Bruhn, K. F. Nielsen, L. Gram and R. Belas,
Appl. Environ. Microbiol., 2008, 74, 1535.
27 N. L. Brock, A. Nikolay and J. S. Dickschat, Chem. Commun.,
2014, 50, 5487.
Notes and references
1 A. A. Benson, Adv. Lipid Res., 1963, 1, 387.
2 G. Datz and G. Döhler, Z. Naturforsch., C: J. Biosci., 1981,
36, 856.
3 D. A. Gage, Z.-H. Huang and C. Benning, Lipids, 1992, 27,
632.
28 R. Wang, E. Gallant and M. R. Seyedsayamdost, mBio, 2016,
7, e02118.
4 N. Sato, S. Kobayashi, M. Aoki, T. Umemura, I. Kobayashi
and M. Tsuzuki, Biochem. Biophys. Res. Commun., 2016,
470, 123.
5 J. N. Harwood and R. G. Nicholls, Biochem. Soc. Trans.,
1979, 7, 440.
29 N. L. Brock, M. Menke, T. A. Klapschinski and
J. S. Dickschat, Org. Biomol. Chem., 2014, 12, 4318.
30 N. L. Brock, C. A. Citron, C. Zell, M. Berger, I. Wagner-
Döbler, J. Petersen, T. Brinkhoff, M. Simon and
J. S. Dickschat, Beilstein J. Org. Chem., 2013, 9, 942.
31 C. R. Reisch, M. J. Stoudemayer, V. A. Varaljay, I. J. Amster,
M. A. Moran and W. B. Whitman, Nature, 2011, 473, 208.
32 J. Rinkel and J. S. Dickschat, Beilstein J. Org. Chem., 2015,
11, 2493.
33 S. Schulz, J. Fuhlendorff and H. Reichenbach, Tetrahedron,
2004, 60, 3863.
34 D. Rao, J. S. Webb, C. Holmström, R. Case, A. Low,
P. Steinberg and S. Kjelleberg, Appl. Environ. Microbiol.,
2007, 73, 7844.
35 J. D. Todd, R. Rogers, Y. G. Li, M. Wexler, P. L. Bond,
L. Sun, A. R. J. Curson, G. Malin, M. Steinke and
A. W. B. Johnston, Science, 2007, 315, 666.
36 A. R. J. Curson, J. D. Todd, M. J. Sullivan and
A. W. B. Johnston, Nat. Rev. Microbiol., 2011, 9, 849.
37 A. W. B. Johnston, R. T. Green and J. D. Todd, Curr. Opin.
Chem. Biol., 2016, 31, 58.
38 S. Thole, D. Kalhoefer, S. Voget, M. Berger, T. Engelhardt,
H. Liesegang, A. Wollherr, S. Kjelleberg, R. Daniel,
M. Simon, T. Thomas and T. Brinkhoff, ISME J., 2012, 6,
2229.
6 J. S. Dickschat, P. Rabe and C. A. Citron, Org. Biomol.
Chem., 2015, 13, 1954.
7 A. B. Roy, M. J. E. Hewlins, A. J. Ellis, J. L. Harwood and
G. F. White, Appl. Environ. Microbiol., 2003, 69, 6434.
8 K. Denger, M. Weiss, A.-K. Felux, A. Schneider, C. Mayer,
D. Spiteller, T. Huhn, A. M. Cook and D. Schleheck, Nature,
2014, 507, 114.
9 B. P. Durham, S. Sharma, H. Luo, C. B. Smith, S. A. Amin,
S. J. Bender, S. P. Dearth, B. A. S. van Mooy,
S. R. Campagna, E. B. Kujawinski, E. V. Armbrust and
M. A. Moran, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 453.
10 M. T. Croft, A. D. Lawrence, E. Raux-Deery, M. J. Warren
and A. G. Smith, Nature, 2005, 438, 90.
11 K. Denger, T. Huhn, K. Hollemeyer, D. Schleheck and
A. M. Cook, FEMS Microbiol. Lett., 2012, 328, 39.
12 J. Mayer, T. Huhn, M. Habeck, K. Denger, K. Hollemeyer
and A. M. Cook, Microbiology, 2010, 156, 1556.
13 U. Rein, R. Gueta, K. Denger, J. Ruff, K. Hollemeyer and
A. M. Cook, Microbiology, 2005, 151, 737.
14 K. Denger and A. M. Cook, Microbiology, 2010, 156, 967.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2017