© 2005 Nature Publishing Group
Methanotrophic symbionts provide carbon for
photosynthesis in peat bogs
Ashna A. Raghoebarsing
1
, Alfons J. P. Smolders
2
, Markus C. Schmid
1
, W. Irene C. Rijpstra
4
,
Mieke Wolters-Arts
3
, Jan Derksen
3
, Mike S. M. Jetten
1
, Stefan Schouten
4
, Jaap S. Sinninghe Damste
´
4
,
Leon P. M. Lamers
2
, Jan G. M. Roelofs
2
, Huub J. M. Op den Camp
1
& Marc Strous
1
Wetlands are the largest natural source of atmospher ic methane
1
,
the second most important greenhouse gas
2
. Methane flux to the
atmosphere depends strongly on the climate
3
; however, by far the
largest part of the methane formed in wetland ecosystems
is recycled and does not reach the atmosphere
4,5
. The biogeo-
chemical controls on the efficient oxidation of methane are still
poorly understood. Here we show that submerged Sphagnum
mosses, the dominant plants in some of these habitats, consume
methane throug h symbiosis with partly endophytic methano-
trophic bacteria, leading to hig hly effective in situ methane
recycling. Molecular probes revealed the presence of the bacteria
in the hyaline cells of the plant and on stem leaves. Incubation with
13
C-methane showed rapid in situ oxidation by these bacteria to
carbon dioxide, which was subsequently fixed by Sphagnum,as
shown by incorporation of
13
C-methane into plant sterols. In this
way, methane acts as a significant (10–15%) carbon source for
Sphagnum. The symbiosis explains both the efficient recycling of
methane and the high organic carbon burial in these wetland
ecosystems.
Peat bogs alternate between law ns and pools. Lawns are dominated
by species that grow up to several decimetres above the water
table. Pools are dominated by aquatic species, such as Sphagnum
cuspidatum, that form layers of liv ing plants below the water table.
We investigated the m ethane-oxidizing activity of submerged
S. cuspidatum from peat bog pools at different field locations in the
Netherlands, and compared it to the activity of S. magellanicum and
S. papillosum growing in lawns. The potential methane-oxidizing
activit y was substantially higher in the submerged mosses (Fig. 1). In
control experiments with bog water, methane was not oxidized,
indicating that the methanotrophic bacteria were mainly present on
or in the living Sphagnum tissue.
The identity and location of these methanotrophs was determined
in a molecular approach. Total genomic DNA from washed Sphagnum
plants was isolated and bacterial 16S ribosomal RNA genes were
amplified, c loned into Escherichia coli, sequenced and analysed
phylogenetically. One of the 16S rRNA gene sequences of the clone
library was affiliated to a cluster of type II methanotrophs that
contained acidophilic methanotrophs isolated from Sphagnum bogs,
such as Methylocella palustris (identity 93%)
6
and Methylocapsa
acidiphila (identity 93%)
7
.
The full 16S rRNA gene sequence was used to design two specific
oligonucleotide probes for fluorescence in situ hybridization (FISH).
FISH was combined with serial sectioning of the stems and the stem
leaves of multiple individuals of submerged S. cuspidatum.The
methanotrophic bacterium targeted by the probes was the dominant
methanotroph in S. cuspidatum sections, accounting for over 75% of
LETTERS
Figure 1 | Methane oxidation potential of different parts of submerged and
non-submerged Sphagnum mosses as a measure of methanotrophs
associated.
Error bars indicate standard deviations of at least six
independent experiments.
1
Department of Microbiology,
2
Department of Aquatic Ecology and Environmental Biology, and
3
Department of Plant Cell Biology, Radboud University Nijmegen, Toernooiveld 1,
6525 ED Nijmegen, The Netherlands.
4
Royal Netherlands Institute for Sea Research (NIOZ), Department of Marine Biogeochemistry and Toxicology, PO Box 59, 1790 AB Den
Burg, The Netherlands.
Vol 436|25 August 2005|doi:10.1038/nature03802
1153
© 2005 Nature Publishing Group
all
a
-Proteobacteria. Application of general probes showed that the
a
-Proteobacteria themselves made up 80% of all detected bacteria,
indicating that the new methanotroph was indeed the dominant
bacterium in S. cuspidatum sections.
g
-Proteobacteria (including
type I methanotrophs) were virtually absent.
In S. cuspidatum stems, clusters of the new methanotroph were
present in the hyaline cells of the outer cortex (Fig. 2a–c; in total
10
6
–10
7
methanotrophs per individual plant, total length of stem
,40 cm). Hyaline cells are dead, water-filled cells that contain pores
by which solutes (and bacteria) can move in or out
8
. The presence of
clusters indicated that this bacterium was actively grow ing inside the
hyaline cells. The bacterial clusters consisted of 5–25 individual
coccoid cells lying closely together in a random arrangement. On
the stem leaves, the same probes hybridized with bacteria occurring
as dense, geometric clusters tightly bound to the living plant cells
(Fig. 2d, e, 10
5
–10
6
methanotrophs per individual plant). Differences
in the morphology of micro-colonies have been observed to depend
on environmental conditions for other microorganisms
9
. On the
basis of the measured in vitro methane-oxidizing capacity of
S. cuspidatum (,20
m
mol per g dry weight per day; Fig. 1) and the
number of methanotrophs per plant, an activity in the order of
l–4 fmol methane cell
21
h
21
was estimated for the associated metha-
notrophs. This is significantly higher than the in vitro methane
oxidation rates reported for acidophilic methanotrophs (,0.3 fmol
methane cell
21
h
21
) (ref. 6), indicating that the actual numbers of
methanotrophs per S. cuspidatum indiv idual might still be
underestimated.
Because FISH analysis had shown that the new methanotroph was
the only bacterium occurring in the characteristic geometric clusters,
it was possible to identify and inspect this bacterium with trans-
mission electron microscopy (TEM). The TEM and FISH results
were consistent with respect to the localization of the methanotroph.
TEM also showed that this bacterium did not contain any intra-
cytoplasmic membranes. The absence of intracytoplasmic mem-
branes was noted previously for the phylogenetically related type II
methanotroph M. palust ris
6
. Otherwise, intracytoplasmic mem-
branes are a characteristic feature of methanotrophic bacteria.
The predominance of type II methanotrophs was further sub-
stantiated by the presence of bishomohopanoic acid in Sphagnum
lipid extracts after periodic acid treatment. This compound was
previously shown to form after periodic acid treatment from the C
35
hopanetetrol derivatives, membrane rigidifiers produced by metha-
notrophic bacteria
10
. The natural
13
C contents of this compound
(
d
13
C ¼ 239.8‰) were substantially depleted relative to Sphagnum
cell material and enriched compared to that of methane ( Table 1), in
accordance with its origin from serine-cycle (type II) methano-
trophic bacteria
11
. Using this methanotrophic biological marker we
were able to determine whether the methanotrophs associated with
Sphagnum were actively growing . After incubating Sphagnum
with
13
C-labelled methane for 5 days, isotopic analysis showed that
13
C-labelled methane was incorporated into this lipid in substantial
amounts; nearly 50% of this lipid was synthesized from the labelled
methane, indicating that the methanotrophic population h ad
doubled over the course of the experiment.
The observed tight association of methanotrophic bacteria with
S. cuspidatum would enable the efficient recycling into living mosses
Figure 2 | In situ detection of the new methanotroph in S. cuspidatum with
fluorescently labelled rRNA-targeted oligonucleotide probes.
a, Cryo-
scanning electron micrograph of a stem cross-section. S, stem leaf; I, outer
cortex; II, internal cylinder; III, inner pith. Scale bar, 100
m
m.
b, c, Epifluorescence micrographs of the new methanotroph (purple or
pink cells) in the outer cortex of Sphagnum stems, after a double
hybridization with the specific probe S-*-18ALF-1437-a-A-18 and the
general probe EUB
21
. Scale bars, 10 and 5
m
m. d, Dense, geometric
clusters of the same bacterium on a stem leaf, after a triple
hybridization with the specific probe S-*-18ALF-1437-a-A-18, the
general probe EUB and probe Alf968 (S-Sc-aProt-0968-a-A-18
(5
0
-GGTAAGGTTCTGCGCGTT-3
0
), specific for
a
-Proteobacteria). Scale
bar, 5
m
m. e, Transmission electron micrograph of a geometric cluster
closely attached to a stem leaf. Scale bar, 1
m
m.
Table 1 | Methane and CO
2
concentrations and
d
13
C values in the
Mariapeel bog pool
CH
4
CO
2
Plants*
Sediment gas composition (%) 52 48 –
Bulk water concentration (
m
M) 50 ^ 20 160 ^ 30 –
d
13
C(‰) 256 214.5 226.5
*The
d
13
C values of growing (226‰) and decaying (227‰) S. cuspidatum were almost
identical.
LETTERS NATURE|Vol 436|25 August 2005
1154
© 2005 Nature Publishing Group
of both oxygen (derived from photosynthesis) and methane (derived
from decaying plants), according to the following set of equations:
CH
4
oxidation : CH
4
þ 2O
2
¼ CO
2
þ 2H
2
O ð1Þ
CO
2
fixation : 2CO
2
þ 2H
2
O ¼ 2CH
2
O þ 2O
2
ð2Þ
Balance : CH
4
þ CO
2
¼ 2CH
2
O ð3Þ
To provide experimental evidence for this scenario, the potential
contribution of methane to carbon fixation by S. cuspidatum was
investigated under conditions relevant to the field. Multiple batches
of individuals of S. cuspidatum were incubated with
13
C-labelled
methane in the presence of unlabelled carbon dioxide. As a control
experiment, only
13
C-labelled carbon dioxide was supplied. Both
compounds were added to a final concentration of 0.2 mM, close to
the in situ concentrations (Table 1). Over 5 days, incorporation of the
label by S. cuspidatum was determined via the
13
C incorporation into
sitosterol, a Sphagnum-specific sterol (Fig. 3). Methane was assimi-
lated into the sitosterol pool at a rate of 0.20 ^ 0.03
m
g C per g dry
weight per day, compared to 1.4 ^ 0.1
m
g C per g dry weight per day
for carbon dioxide. Thus, in the presence of carbon dioxide, at
near in situ concentrations, the capacity of methane incorporation
by S. cuspidatum was ,15% of the carbon dioxide assimilation
capacity.
The natural carbon isotope abundances of Sphagnum mosses in
the field (
d
13
C 226.5‰; Table 1) are consistent with our estimate
that 15% of the carbon fixed by Sphagnum derives from isotopically
depleted methane (that is, 256‰; Table 1). S. cuspidatum fixes
carbon dioxide via the Calvin cycle, and is able to fractionate strongly
against
13
C (up to 29‰) at high carbon dioxide concentrations
(.2 mM)
12,13
. However, unlike vascular (semi-)aquatic plants such
as rice, S. cuspidatum does not have aerenchyma
8
that facilitate the
transport of atmospheric carbon dioxide. Therefore, at lower carbon
dioxide concentrations, carbon assimilation by S. cuspidatum is
limited by mass transfer, and carbon fractionation has been reported
to decrease to at most 4‰ (refs 12, 13). Because the average carbon
dioxide concentration in the field was approximately 0.16 mM, a
range of 4–10‰ was used as a conservative estimate for carbon
fractionation by S. cuspidatum in the field
12,13
. With this assumption,
the data from Table 1 and a simple isotopic mass balance (see
Methods), we calculate d that methane contribut ed on average
between 5% and 20% to the carbon fixed by S. cuspidatum in the
field, in good agreement with the labelling results. It is likely that
variation in local conditions (water depth, exposu re to wind,
temperature, light availability, rates of methane ebullition compared
to diffusion/advection) will affect the relative contributio n of
methane to the carbon uptake of Sphagnum mosses in space and
time. This will also be determined by the location of the symbiotic
metha notrophs in the plants, both in the direct vicinity of the
photosynthetically active cells and in the more remote hyaline cells
of the stems.
Our results show that methane is a significant and as yet over-
looked supplement to the carbon intake of submerged S. cuspidatum
in peat bogs. Peat bogs in the Northern Hemisphere store up to one-
third of the carbon sequestered in soils globally
14
. This is surprising
considering that the primary production is limited by the nutrient
delivery through rain water and the limited delivery of carbon
dioxide to the acidic waters of these ecosystems
5
. The efficient
recycling of peat decomposition products (such as methane) as
demonstrated here may mechanistically explain the paradox of
peatlands as ecosystems with apparent low primary productivity
combined with hig h carbon burial.
METHODS
In situ conditions. In the Mariapeel nature reserve (the Netherlands: 518 24
0
90
00
N; 58 54
0
90
00
E),
d
13
C values of Sphagnum mosses and material from the
decaying peat were determined on freeze-dried homogenized mate rial as
described previously
15
. Concentrations and
d
13
C values of dissolved carbon
dioxide and methane were measured as described previously
16
.
Methane oxidation. Potential methane oxidation of different parts of Sphagnum
were measured by incubating 6 g of thoroughly washed Sphagnum in 100 ml
infusion flasks sealed w ith airtight rubber stoppers. To prevent mass transport
limitations, no additional water was added to the experiments. To each flask 1 ml
of pure methane was added and methane consumption was measured every 6 h
over 2 days. Methane oxidation rates were calculated by linear regression.
Tenfold concentrated water samples (10
6
bacterial cells ml
21
) from the bog
were used as controls and showed no methane oxidation. Samples were collected
in the Netherlands from seven lawn locations (S. magellanicum, S. papillosum)
and six bog pools (S. cuspidatum), one of these being the Mariapeel. Methane
was measured on an HP 5890 gas chromatograph equipped with a flame
ionization detector and a Porapak Q column (80/100 mesh).
16S rRNA gene sequence analysis, FISH and electron microscopy. Total
genomic DNA from S. cuspidatum plants containing methanotrophs, isolated
with combined methods
17
, was used as template for PCR amplification of 16S
rRNA genes. PCR was performed with general bacterial primers
18
using a T
gradient thermal cycler (Biometra), and a clone librar y was made as described
previously
18
. Based on the obtained 16S rRNA gene sequences, two new
oligonucleotide probes S-*-18ALF-0218-a-A-18 (5
0
-GGGCCGATCCCCC
GGCGA-3
0
) and S-*-18ALF-1437-a-A-18 (5
0
-CTTGCGGTTAACAGAACG-3
0
)
were designed using the ARB program package
19
. Apart from these species-
specific probes we used group-specific probes described previously
20,21
. Fresh
S. cuspidatum stems sectioned with a scalpel (section thickness 0.1 ^ 0.05 mm)
were used for FISH as described previously
22
. Formamide concentrations used in
the FISH experiments varied between 10% and 20%. No signal was obtained at
these formamide concentrations when testing the specificity of the probes with
Beijerinckia indica ssp. indica (DSM 1715), which has the fewest mismatches of
all reference organisms to the designed probes. Electron microscopy (TEM/
SEM) was performed on stems and stem leaves following published proto-
cols
6,22
.
Methane incorporation measurements. S. cuspidatum, col lected from the
Mariapeel nature reserve, was washed with demineralized water and incubated
in 250-ml serum bottles in 5 g wet weight aliquots with 150 ml medium as
described previously
8
.
13
C- or
12
C-CH
4
or CO
2
were added to final concen-
trations of 200
m
M as specified in the text. The bottles were shaken at
150 r.p.m. at ambient conditions and sacrificed for lipid analysis at days 0,
1, 3 and 5. Lipids were ultrasonically extracted and analysed by gas
chromatography/mass spectrometry and isotope ratio gas chromatography
mass spectrometry as described
23
. Hopanes were analysed by treatment of the
lipid fr action with periodic acid and sodium borohydride as described
previously
10,24
.
Isotopic mass balancing. The measured
d
13
C values of S. cuspidatum in the field
(226‰, see Table 1) resulted from assimilation of dissolved carbon dioxide
(214.5‰), respired methane (256‰) and fractionation against
13
C during
(mass-transfer-limited) carbon dioxide fixation
12,13
(7 ^ 3‰). The following
equation describes this relationship quantitatively (where a denotes the fraction
of plant carbon derived from methane and Ep denotes the fractionation during
Figure 3 | Incorporation of
13
C label in biological markers for Sphagnum
(circles) and methanotrophic bacteria (squares).
Filled circles/solid lines
show the results for labelled methane (99%
13
C) in the presence of
unlabelled carbon dioxide; open symbols/dashed lines show the results for
labelled carbon dioxide (4%
13
C).
NATURE|Vol 436|25 August 2005 LETTERS
1155
© 2005 Nature Publishing Group
fixation):
d
13
C ðSphagnumÞ¼a
d
13
C ðrespired methaneÞ
þð1 2 aÞ
d
13
C ðcarbon dioxideÞ 2 Ep
ð4Þ
Because all
d
13
C values from equation (4) were known experimentally, it could be
derived that the contribution of methane to Sphagnum carbon (a) was between
0.05 and 0.2 (equivalent to 5–20%).
Received 23 December 2004; accepted 9 May 2005.
1. Hein, R., Crutzen, P. J. & Heimann, M. An inverse modeling approach to
investigate the global atmospheric methane cycle. Global Biogeochem. Cycles 11,
43–-76 (1997).
2. Rodhe, H. A comparison of the contribution of various gases to the greenhouse
effect. Science 248, 1217–-1219 (1990).
3. Smith, L. C. et al. Siberian peatlands a net carbon sink and global methane
source since the early Holocene. Science 303, 353–-356 (2004).
4. Dedysh, S. N. et al. Isolation of acidophilic methane-oxidizing bacteria from
northern peat wetlands. Science 282, 281–-284 (1998).
5. Lamers, L. P. M., Farhoush, C., Van Groenendael, J. M. & Roelofs, J. G. M.
Calcareous groundwater raises bogs; the concept of ombrotrophy revisited.
J. Ecol. 87, 639–-648 (1999).
6. Dedysh, S. N. et al. Methylocella palustris gen. nov., sp. nov., a new methane-
oxidizing acidophilic bacterium from peat bogs, representing a novel subtype
of serine-pathway methanotrophs. Int. J. Syst. Evol. Microbiol. 50, 955–-969
(2000).
7. Dedysh, S. N. et al. Methylocapsa acidophila gen. nov., sp. nov., a novel
methane-oxidizing and dinitrogen fixing acidophilic bacterium from Sphagnum
bog. Int. J. Syst. Evol. Microbiol. 52, 251–-261 (2002).
8. Rydin, H. & Clymo, R. S. Transport of carbon and phosphorus compounds
about Sphagnum. Proc. R. Soc. Lond. 237, 63–-84 (1989).
9. Yao, R., Macario, A. J. L. & Conway de Macario, E. Immunochemical
differences among Methanosarcina mazei S-6 morphologic forms. J. Bacteriol.
174, 4683–-4688 (1992).
10. Rohmer, M., Bisseret, P. & Neunlist, S. in Biological Markers in Sediments and
Petroleum (eds Moldowan, J. M., Albrecht, P. & Philp, R. P.) 1–-17 (Prentice Hall,
London, 1992).
11. Jahnke, L. L., Summons, R. E., Hope, J. M. & Des Marais, D. J. Carbon isotopic
fractionation in lipids from methanotrophic bacteria II: The effects of
physiology and environmental parameters on the biosynthesis and isotopic
signatures of biomarkers. Geochim. Cosmochim. Acta 63, 79–-93 (1999).
12. Keeley, J. E. & Sandquist, D. R. Carbon: freshwater plants. Plant Cell Environ. 15,
1021–-1035 (1992).
13. Smolders, A. J. P., Tomassen, H. B. M., van Mullekom, M., Lamers, L. P. M. &
Roelofs, J. G. M. Mechanisms involved in the re-establishment of
Sphagnum-dominated vegetation in rewetted bog remnants. Wetlands Ecol.
Manag. 11, 403–-418 (2003).
14. Post, W. M., Emanuel, W. R., Zinke, P. J. & Strangenberger, A. G. Soil carbon
pools and world life zones. Nature 298, 156 (1982).
15. Marguillier, S., van der Velde, G., Dehairs, F., Hemminga, M. A. & Rajagopal, S.
Trophic relationship in an interlinked mangrove-seagrass ecosystem as traced
by
d
13
C and
d
15
N. Mar. Ecol. Prog. Ser. 151, 115–-121 (1997).
16. Smolders, A. J. P., Tomassen, H. B. M., Lamers, L. P. M., Lomans, B. P. &
Roelofs, J. G. M. Peat bog restoration by floating raft formation: the effects of
groundwater and peat quality. J. Appl. Ecol. 39, 391–-401 (2002).
17. Lomans, B. P. et al. Microbial populations involved in cycling of dimethyl sulfide
and methanethiol in freshwater sediments. Appl. Environ. Microbiol. 67,
1044–-1051 (2001).
18. Juretschko, S. et al. Combined molecular and conventional analyses of nitrifying
bacterium diversity in activated sludge: Nitrosococcus mobilis and Nitrospira-like
bacteria as dominant populations. Appl. Environ. Microbiol. 64, 3042–-3051
(1998).
19. Ludwig, W. et al. ARB: A software environment for sequence data. Nucleic
Acids Res. 32, 1363–-1371 (2004).
20. Amann, R. I. et al. Combination of 16S rRNA-targeted oligonucleotide probes
with flow cytometry for analyzing mixed microbial populations. Appl. Environ.
Microbiol. 56, 1919–-1925 (1990).
21. Daims, H., Bruhl, A., Amann, R., Schleifer, K. H. & Wagner, M. The domain-
specific probe EUB338 is insufficient for the detection of all Bacteria:
development and evaluation of a more comprehensive probe set. Syst. Appl.
Microbiol. 22, 434–-444 (1999).
22. Wolters-Arts, M. et al. Water-conducting properties of lipids during pollen
hydration. Plant Cell Environ. 25, 513–-519 (2002).
23. Schouten, S. et al. Biosynthetic effects on the stable carbon isotopic
compositions of algal lipids: Implications for deciphering the carbon isotopic
biomarker record. Geochim. Cosmochim. Acta 62, 1397–-1406 (1998).
24. Sinninghe Damste
´
,J.S.et al. The occurrence of hopanoids in planctomycetes:
implications for the sedimentary biomarker record. Organic Geochim. 35,
561–-566 (2004).
Acknowledgements We thank K. T. van de Pas-Schoonen, A. Pol, H. P. M. Geurts,
J. Eygensteyn, M. van Mullekom, J. Berk, H. Tomassen and M. M. A. van Herpen
for technical support. Part of this study was supported by the Dutch Ministry of
Agriculture, Nature Management and Food quality (Research Program
‘Overlevingsplan Bos en Natuur’).
Author Information The 16S rRNA gene sequences were deposited at GenBank
under accession number AY163571. Reprints and permissions information is
available at npg.nature.com/reprintsandpermissions. The authors declare no
competing financial interests. Correspondence and requests for materials should
be addressed to A.J.P.S. (a.smolders@science.ru.nl) or J.S.S.D.
(damste@nioz.nl).
LETTERS NATURE|Vol 436|25 August 2005
1156
