Direct and reversible hydrogenation of CO2to formate by a bacterial carbon dioxide reductase

Article abstract of DOI:10.1126/science.1244758

Storage and transportation of hydrogen is a major obstacle for its use as a fuel. An increasingly considered alternative for the direct handling of hydrogen is to use carbon dioxide (CO₂) as an intermediate storage material. However, CO₂ is thermodynamically stable, and developed chemical catalysts often require high temperatures, pressures, and/or additives for high catalytic rates. Here, we present the discovery of a bacterial hydrogen-dependent carbon dioxide reductase from Acetobacterium woodii directly catalyzing the hydrogenation of CO₂. We also demonstrate a whole-cell system able to produce formate as the sole end product from dihydrogen (H ₂) and CO₂ as well as syngas. This discovery opens biotechnological alternatives for efficient CO₂ hydrogenation either by using the isolated enzyme or by employing whole-cell catalysis.

Full text of DOI:10.1126/science.1244758

Direct and Reversible Hydrogenation of CO2 to Formate by a
Bacterial Carbon Dioxide Reductase
K. Schuchmann and V. Müller
Science 342, 1382 (2013);
DOI: 10.1126/science.1244758
This copy is for your personal, non-commercial use only.
If you wish to distribute this article to others, you can order high-quality copies for your
colleagues, clients, or customers by clicking here.
Permission to republish or repurpose articles or portions of articles can be obtained by
following the guidelines here.
The following resources related to this article are available online at
www.sciencemag.org (this information is current as of December 20, 2013 ):
Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
http://www.sciencemag.org/content/342/6164/1382.full.html
Supporting Online Material can be found at:
http://www.sciencemag.org/content/suppl/2013/12/12/342.6164.1382.DC1.html
A list of selected additional articles on the Science Web sites related to this article can be
found at:
http://www.sciencemag.org/content/342/6164/1382.full.html#related
This article cites 25 articles, 5 of which can be accessed free:
http://www.sciencemag.org/content/342/6164/1382.full.html#ref-list-1
This article has been cited by 1 articles hosted by HighWire Press; see:
http://www.sciencemag.org/content/342/6164/1382.full.html#related-urls
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the
American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright
2
013 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.

REPORTS
whereas typical IMiCa was seen after EMRE re- and MICU2 to the channel activity of MCU (Fig.
expression (Fig. 3F) (6). 4F). The fact that our immunoprecipitated com-
Consistent with EMRE being a core com- plex migrates at the same apparent molecular size
ponent of the uniplex, when EMRE was lost, the as the uniporter from purified mitochondria (Fig.
size of the complex on a native gel was reduced 1A), and recovers the founding members of the
to ~300 kD, similar to that in cells lacking MICU1 complex, suggests that we immunoprecipitated all
5. M. Plovanich et al., PLOS ONE 8, e55785 (2013).
6. D. Chaudhuri, Y. Sancak, V. K. Mootha, D. E. Clapham,
eLife 2, e00704 (2013).
7. J. D. Martell et al., Nat. Biotechnol. 30, 1143–1148
(2012).
8. K. Mallilankaraman et al., Cell 151, 630–644 (2012).
9. G. Csordás et al., Cell Metab. 17, 976–987 (2013).
0. R. Palty et al., J. Biol. Chem. 279, 25234–25240
1
1
(Fig. 4A). The observation that loss of either of its components. It is notable that although other
(
2004).
1. D. Jiang, L. Zhao, D. E. Clapham, Science 326, 144–147
2009).
12. K. Mallilankaraman et al., Nat. Cell Biol. 14, 1336–1343
2012).
3. S. Feng et al., Proc. Natl. Acad. Sci. U.S.A. 110,
1011–11016 (2013).
EMRE or MICU1 reduced the size of the native proteins have been reported to participate in mito-
complex to a similar extent raised the hypothesis chondrial calcium handling—such as LETM1,
that EMRE may mediate interaction between NCLX, UCP2 and 3, MCUR1, and TRPC3—they
MCU and MICU1 and MICU2. To test this, we were not recovered in our proteomic assay, sug-
stably expressed MCU-FLAG, or a control pro- gesting that they play key roles outside of the
tein in HEK-293T, or HeLa wild-type cells, or cells uniplex. A key insight from the current work is
lacking EMRE and performed FLAG immunopre- that EMRE is essential for in vivo uniporter cur-
cipitations. In wild-type cells, MICU1 and MICU2 rent (IMiCa) and that MCU oligomers alone are
immunoprecipitated with MCU-FLAG but not not sufficient for in vivo uniporter activity, in con-
with control SDHB-FLAG (Fig. 4B and fig. S4). trast to what has been suggested using in vitro
However, in cells lacking EMRE, the interaction bilayer studies (4). Why MCU needs association
between MCU-FLAG and MICU1 and MICU2 with EMRE for in vivo calcium conductance re-
was completely lost (Fig. 4B and fig. S4). Sim- quires further investigation. Phylogenetic analy-
ilarly, MCU was not associated with immuno- ses indicate that the uniporter must have been
precipitated MICU1-FLAG in cells lacking EMRE a feature of the earliest mitochondria because
(
(
1
1
1
1
1
4. M. Trenker, R. Malli, I. Fertschai, S. Levak-Frank,
W. F. Graier, Nat. Cell Biol. 9, 445–452 (2007).
5. S. E. Ong et al., Mol. Cell. Proteomics 1, 376–386
(2002).
6. A. Raffaello et al., EMBO J. 32, 2362–2376 (2013).
17. A. Krogh, B. Larsson, G. von Heijne, E. L. Sonnhammer,
J. Mol. Biol. 305, 567–580 (2001).
1
8. M. G. Claros, P. Vincens, Eur. J. Biochem. 241, 779–786
1996).
9. D. J. Pagliarini et al., Cell 134, 112–123 (2008).
(
1
20. A. G. Bick, S. E. Calvo, V. K. Mootha, Science 336, 886
2012).
(
(
Fig. 4C). In the absence of EMRE, MCU still MCU and MICU1 are found within all major
Acknowledgments: We thank Z. Grabarek and D. M. Shechner
for helpful discussions. Y.S. received support from the Helen
oligomerized (Fig. 4, A and B) and interacted with eukaryotic taxa, with lineage-specific losses (20).
MCUb. Moreover, in the absence of EMRE, the EMRE is unique among human uniplex compo- Hay Whitney Foundation. D.C. received support from NIH
interaction between MICU1 and MICU2 was intact nents in that it appears to have emerged more F32HL107021. D.E.C. is an Investigator of the Howard Hughes
Medical Institute. This work was supported by grants to
(
Fig. 4C). These data indicate that loss of EMRE recently and represents a metazoan innovation.
V.K.M from NIH DK080261 and a gift from W. Dan and
Pat Wright.
specifically interrupted the association of MCU Its identification should help us understand how
with MICU1 and MICU2. Furthermore, MCUb, the activity and regulation of this ancient channel
MICU1, and MICU2 appear to be dispensable for evolved.
Supplementary Materials
www.sciencemag.org/content/342/6164/1379/suppl/DC1
Materials and Methods
Figs. S1 to S4
Table S1
MCU-EMRE interaction because in cells lacking
MCUb or MICU1, EMRE was still associated with
immunoprecipitated MCU-FLAG (Fig. 4, D and E).
We propose a model in which EMRE inter-
acts with MICU1 and MICU2 in the IMS and
with MCU oligomers in the inner membrane, thus
linking the calcium-sensing activity of MICU1
References and Notes
1. Y. Kirichok, G. Krapivinsky, D. E. Clapham, Nature 427,
360–364 (2004).
References (21–33)
2. F. Perocchi et al., Nature 467, 291–296 (2010).
3. J. M. Baughman et al., Nature 476, 341–345 (2011).
4. D. De Stefani, A. Raffaello, E. Teardo, I. Szabò, R. Rizzuto,
Nature 476, 336–340 (2011).
10 July 2013; accepted 5 November 2013
Published online 14 November 2013;
10.1126/science.1242993
storage application for a hydrogen-using fuel cell.
An alternative would be the use of a fuel cell
directly using formate (3). Unfortunately, high
temperatures and pressures are often required for
Direct and Reversible Hydrogenation
of CO to Formate by a Bacterial
2
efficient chemical hydrogenation of CO
cently, progress has been made by using an irid-
ium catalyst to reversibly hydrogenate CO at
2
(4). Re-
Carbon Dioxide Reductase
K. Schuchmann and V. Müller*
2
near-ambient conditions, but still with moderate
catalytic rates (5), or by the use of a cobalt-based
catalyst in combination with high concentration
of an expensive base (6).
Storage and transportation of hydrogen is a major obstacle for its use as a fuel. An increasingly
considered alternative for the direct handling of hydrogen is to use carbon dioxide (CO ) as an
Biological systems are often promising alter-
2
intermediate storage material. However, CO
catalysts often require high temperatures, pressures, and/or additives for high catalytic rates.
Here, we present the discovery of a bacterial hydrogen-dependent carbon dioxide reductase from uted reaction in nature, little is known about en-
Acetobacterium woodii directly catalyzing the hydrogenation of CO . We also demonstrate a zymes catalyzing CO reduction activity, and no
enzyme is known to catalyze the direct hydro-
genation of CO alone (7, 8). One isolated for-
mate dehydrogenase (FDH) could be shown to
reduce CO electrochemically, but not with H
uture energy demands require alternatives vealed economically usable possibilities (1). A as the electron donor (9). Acetogenic bacteria are
for the current energy technologies that are promising alternative is the use of CO as storage
based mainly on fossil fuels. Hydrogen is material for hydrogen (2). The near-equilibrium
2
is thermodynamically stable, and developed chemical natives for chemically difficult conversions. Even
though the reduction of CO is a widely distrib-
2
2
2
whole-cell system able to produce formate as the sole end product from dihydrogen (H
as well as syngas. This discovery opens biotechnological alternatives for efficient CO
hydrogenation either by using the isolated enzyme or by employing whole-cell catalysis.
2 2
) and CO
2
2
2
2
2
F
Molecular Microbiology and Bioenergetics, Institute of Molecular
leads to formic acid and Biosciences, Johann Wolfgang Goethe University Frankfurt/Main,
a promising and widely considered option as al- hydrogenation of CO
2
ternative energy feedstock, but research on direct the corresponding base formate, respectively. This Max-von-Laue-Strasse 9, 60438 Frankfurt, Germany
hydrogen storage and transportation has not re- low energy demand is beneficial for a hydrogen *Corresponding author. E-mail: vmueller@bio.uni-frankfurt.de
1382
13 DECEMBER 2013 VOL 342 SCIENCE www.sciencemag.org

REPORTS
promising candidates for such enzymes based on identified four proteins encoded by the gene maximal activity was between 7 and 9 for both
the physiological use of H + CO as the sole cluster Awo_c08190-08260 (fig. S1, B and C). activities (fig. S4). The temperature optimum for
2
2
growth substrate and the production of formate A putative formate dehydrogenase (FdhF1/2) hydrogen oxidation was 30°C, that for formate
–
as intermediate (10, 11). One enzyme was re- and an iron-iron hydrogenase (HydA2) form the oxidation 40°C (fig. S5). HCO3 reductase activ-
ported that can catalyze the hydrogenation of two large subunits accompanied by two small ity with reduced methylviologen as electron do-
CO at the purified state, but it also uses the electron transfer subunits (HycB1/2/3). The ge- nor was catalyzed with 132 U/mg and a K of
2
M
–
3
reduced form of nicotinamide adenine dinucleo- nome encodes a selenium-containing (FdhF2) 37 T 4 mM HCO
(SEM; N = 3 independent
tide phosphate (NADPH) and reduced ferredoxin and a selenium-free (FdhF1) FDH whose genes experiments) (fig. S6).
as electron donors in an electron-bifurcating man- are in tandem on the chromosome. Because cells Searching for an enzyme that directly hydro-
ner (12). were grown in selenium-containing media, most- genates CO , we investigated whether the puri-
2
Here, we report the discovery of a bacterial ly FdhF2 was found. A gene encoding an elec- fied complex can catalyze this reaction. The
hydrogen-dependent carbon dioxide reductase tron transfer protein follows each FDH gene. The enzyme indeed catalyzed the hydrogenation of
(
HDCR) that directly uses dihydrogen for the metal content determined by inductively coupled CO with a rate of 10 mmol formate produced
2
–
1
–1
interconversion of CO2 to formate from the plasma mass spectrometry (ICP-MS) measure- min mg (Fig. 1A). Recently reported chem-
acetogenic bacterium Acetobacterium woodii. ments was 0.6 molybdenum, no tungsten, 0.8 ical catalysts promote the hydrogenation of CO
We targeted this bacterium because it grows selenium, 0.7 zinc and 39 T 6 iron atoms. These at comparable conditions with an initial turnover
with H
ancient pathway for energy conservation. It lives factors (table S1).
under extreme energy limitation, and a direct use By using the artificial electron acceptor methyl- a TOF of 101,600 hours . Thus, the enzyme is
of H for CO reduction would be beneficial for viologen, we were able to measure hydrogenase nearly 1500 times as effective as these chemical
the cell. Moreover, an HDCR was predicted from and FDH activity separately. The isolated enzyme catalysts. The reverse reaction, formate oxida-
the genome sequence (13). catalyzed formate oxidation with an activity of tion to H and CO , was catalyzed with 14 U/mg
The HDCR enzyme was purified from cells ~600 mmol formate min mg and hydrogen (Fig. 1B). The enzyme has two advantages in
2
−1
2
+ CO
2
at ambient conditions using an values are in agreement with the predicted co- frequency (TOF) of ~70 hours (5). The purified
HDCR catalyzes the same reaction at 30°C with
−1
2
2
2
2
–1
–1
–1
–1
2
of A. woodii under anaerobic conditions (14). oxidation with 10,800 mmol H min mg . Spe- industrial applications. First, it does not need an
The purified protein complex is composed of cific activities varied between different enzyme external cofactor as electron carrier. Cofactor-
four different subunits (fig. S1A). It eluted as a preparations due to the high oxygen sensitivity. dependent conversion would raise the cost of
single peak from a gel filtration with an appar- The Michaelis constant (K ) values for formate such a process even if regeneration systems would
M
ent mass around 3500 kD but dissociated during and hydrogen were determined to be 1 T 0.3 mM be used. Second, the reaction is reversible, and
separation by clear native page (fig. S2). The (SEM; N = 2 independent experiments) and thus the directionality is easy to control by sub-
calculated molecular mass of the monomeric 125 T 31 mM (SEM; N = 3 independent exper- strate concentrations.
complex is 169 kD. Peptide mass fingerprinting iments), respectively (fig. S3). The pH range for
Purified HDCR also catalyzed the reduc-
tion of CO with reduced ferredoxin instead of
2
H as electron donor. The rate of ferredoxin-
2
dependent formate production was 0.6 mmol
–
1
–1
formate min mg (Fig. 1C). Ferredoxin was
reduced by CO dehydrogenase (CODH) from
A. woodii, leading to formate production from
CO. The reverse reaction was catalyzed with
–
1
–1
rates of 0.6 mmol ferredoxin min mg and a
for ferredoxin of 13 T 5 mM (SEM; N = 3
K
M
independent experiments) (fig. S7). Industrially
produced hydrogen often contains remaining
CO that is a potent inhibitor of fuel-cell cat-
alysts. Considering A. woodii HDCR as a catalyst
for hydrogen storage, the side activity represents
a possible industrial advantage. Using a combi-
nation of HDCR and CODH would allow the
complete conversion of gas mixtures containing
2 2
H , CO , and CO (Fig. 2).
To get insights into the reaction mechanism
of the HDCR, we searched for compounds that
inhibit partial reactions. CO turned out to be a
potent inhibitor of the hydrogenase activity alone
(fig. S8A). In the presence of 10% CO in the gas
phase, only the FDH module is active. Under
these conditions, formate-dependent ferredoxin
reduction rates increased, showing that the hy-
drogenase moiety is not involved in this electron
transfer (fig. S8B). Ferredoxin reduction with
–1
–1
rates of 0.4 mmol ferredoxin min mg was
also observed, with H as electron donor instead
dinucleotide (NAD)–dependent FDH from Candida boidinii. (B) Hydrogen production from formate (50 mM) of formate when CO was omitted. If CO and
Fig. 1. Reactions catalyzed by the purified HDCR complex. (A) Formate production from H
At the time point indicated, 2 mg HDCR was added. Formate was measured using nicotinamide adenine
2 2
+ CO .
2
2
2
(
10 mg HDCR). All data points are mean T SEM; N = 6 independent experiments. (C) Formate production ferredoxin were present simultaneously, both
with ferredoxin as electron donor (3.5 mg HDCR). (D) Set-up of the measurement from (C). Ferredoxin was were reduced concurrently, although apparent-
reduced by CODH with electrons from CO oxidation. Formate was measured as described for (A).
ly without any coupling because the velocity of
www.sciencemag.org SCIENCE VOL 342 13 DECEMBER 2013
1383

REPORTS
each reaction did not vary in dependence with
the other one.
2
Because the reduction of CO does not re-
quire additional cellular electron carriers, the
conversion is independent of the cell metabo-
lism. This opens the possibility of using the
HDCR in whole-cell catalysis by separating
growth and energy conservation from the hy-
drogenation of CO
cells of A. woodii that produce acetate as the
major end product from H + CO . The energy
2
. To test this, we used whole
2
2
metabolism of A. woodii is strictly sodium ion–
dependent and depends on a sodium ion gradi-
ent across the cytoplasmic membrane (15). We
used this trait to block the energy metabolism
by omitting sodium ions from the buffer or by
adding a sodium ionophore (fig. S9). Under these
conditions, acetate production ceased almost
completely, and formate was produced with an
–
1
–1
initial rate of 2 mmol min mg cell protein
fig. S10). The formate produced was excreted
from the cells. Formate production from H
CO using whole cells was also observed in
other systems, but different enzymes seem to be
responsible there (16–18). For further experi-
Fig. 2. Model of the HDCR from A. woodii. Electrons for CO reduction are either provided by the
hydrogenase subunit HydA2, where hydrogen oxidation takes place, or by reduced ferredoxin. The
latter can be reduced using carbon monoxide and CODH, for example. Electrons are delivered to
2
(
2
+
2
the active site for CO reduction in FdhF2 via the electron-transferring subunits HycB2/3. Iron sulfur
2
clusters are indicated.
ments, we used KHCO
the overall process is almost pH neutral compared
with the production of formic acid from CO
3
as substrate; thereby,
2
.
The genome of A. woodii encodes for a carbo-
anhydrase that allows the rapid interconversion
–
3
–
3
of CO and HCO . Up to 0.45 M HCO , the
2
final formate concentration increased with in-
creasing substrate concentration (Fig. 3). At 1 ×
5
1
0 Pa H
2
, the thermodynamic equilibrium is
–
–
approximately [HCO ] = [HCOO ]. Indeed, up
to 0.45 M HCO the final formate concentration
3
3
–
fits well to the theoretic thermodynamic limit of
the reaction, underlining the independence of the
-
3
carboxylation of CO /HCO from other cellular
2
processes. There is apparently no substantial loss
in unwanted side products or biomass production.
Acetogenic bacteria such as A. woodii are
also able to use syngas, a mixture of H , CO ,
and CO, as substrate for acetogenesis. Syngas
fermentation is an increasingly considered “green”
option for the production of chemicals (19). As
a proof of principle, we tested whether syngas
can also be used as substrate for the production
of formate. Indeed, HDCR in combination with
CODH catalyzed formate production from syngas
with 1.3 U/mg (Fig. 4A). The whole cell system
produced formate from syngas as well, albeit
Fig. 3. Production of formate using whole-cell catalysis. (A) Whole cells of A. woodii were incubated
with H and CO (squares) or H and 0.3 M KHCO in buffer containing 30 mM ETH2120 and 20 mM NaCl. (B)
2 2 2 3
2
2
The final concentration of formate that was produced by whole cells is plotted against the initial concentration
–
of HCO3 used (100% H as gas phase). All data points are mean T SEM; N = 3 independent experiments.
2
with lower activities compared with H
2
+ CO
2
as sub-
2
alone (Fig. 4B). In contrast to H + CO
2
strate, acetate formation could not completely
be inhibited in our test system. Based on the re-
sults with whole cells, a two-step process can
be imagined where the bacteria could be grown
using syngas or a mixture of only H
a substrate, and by disabling the metabolism
the hydrogenation of CO could be induced and
2 2
+ CO as
2
Fig. 4. Formate production from syngas. (A) Purified HDCR (4 mg) together with CODH and ferredoxin
H could be stored as formate (fig. S11). In this catalyzed formate production from syngas. Formate was measured using NAD-dependent FDH from
2
scenario, the greenhouse gas CO
2
is not only C. boidinii. (B) Formate was also produced from syngas by whole cells in the presence of ETH2120 and
used as storage material for H but, additionally, NaCl. At the final time point, 60 mM acetate were produced in addition to formate. All data points are
2
is consumed as growth substrate.
mean T SEM; N = 3 independent experiments.
1384
13 DECEMBER 2013 VOL 342 SCIENCE www.sciencemag.org

REPORTS
We finally want to address the physiological ganisms (22) and syntrophic associations (23).
implications of the properties of HDCR. Based Some gene clusters coding for predicted HDCRs
on the genome sequence, the two FDHs in the vary in the composition compared with the en-
HDCR context seem to be the only enzymes zyme in the present study—for example, the num-
with FDH activity in A. woodii. Thus, the HDCR ber of hydrogenase subunits.
complex is essential for the first step in the re-
15. S. Schmidt, E. Biegel, V. Müller, Biochim. Biophys. Acta
787, 691–696 (2009).
6. S. M. da Silva et al., Microbiology 159, 1760–1769
2013).
1
1
(
17. W. M. Wu, R. F. Hickey, M. K. Jain, J. G. Zeikus,
Arch. Microbiol. 159, 57–65 (1993).
1
8. A. M. Klibanov, B. N. Alberti, S. E. Zale, Biotechnol. Bioeng.
4, 25–36 (1982).
9. P. C. Munasinghe, S. K. Khanal, Bioresour. Technol. 101,
013–5022 (2010).
20. K. Schuchmann, V. Müller, J. Biol. Chem. 287, 31165–31171
2012).
2
ductive acetyl-CoA pathway. In agreement with
this notion are data from proteome analysis,
which revealed no differential expression be-
tween H + CO or fructose as substrate (13).
References and Notes
1
1. L. Schlapbach, A. Züttel, Nature 414, 353–358
5
(
2001).
2
3
4
. S. Enthaler, J. von Langermann, T. Schmidt,
Energy Environ. Sci. 3, 1207–1217 (2010).
. N. M. Aslam, M. S. Masdar, S. K. Kamarudin,
W. R. W. Daud, APCBEE Procedia 3, 33–39 (2012).
. E. Fujita, J. T. Muckerman, Y. Himeda, Biochim. Biophys.
Acta 1827, 1031–1038 (2013).
. J. F. Hull et al., Nat. Chem. 4, 383–388 (2012).
. M. S. Jeletic, M. T. Mock, A. M. Appel, J. C. Linehan, J. Am.
Chem. Soc. 135, 11533–11536 (2013).
7. B. R. Crable, C. M. Plugge, M. J. McInerney, A. J. Stams,
Enzyme Res. 2011, 532536 (2011).
. A. M. Appel et al., Chem. Rev. 113, 6621–6658 (2013).
. T. Reda, C. M. Plugge, N. J. Abram, J. Hirst, Proc. Natl.
Acad. Sci. U.S.A. 105, 10654–10658 (2008).
(
2
2
21. E. G. Matson, X. Zhang, J. R. Leadbetter, Environ.
Acetogenesis from fructose or other substrates
does not yield H directly but reduced ferredoxin.
Microbiol. Rep. 12, 2245–2258 (2010).
2
22. I. A. Pereira et al., Front. Microbiol. 2, 69 (2011).
23. N. Müller, P. Worm, B. Schink, A. J. Stams, C. M. Plugge,
Environ. Microbiol. Rep. 2, 489–499 (2010).
2
Thus, the second H -independent entry path for
electrons is favorable. In addition, CO inhibits
the hydrogenase module, but CO-derived elec-
5
6
Acknowledgments: We are grateful to M. Herzberg and
D. Nies for the ICP-MS measurements and to J. Langer for
protein identification via peptide mass fingerprinting.
This work was supported by grants from the Deutsche
Forschungsgemeinschaft and the German National
Academic Foundation. Part of this work is filed as a patent
trons can still enter the complex for CO
tion via ferredoxin. The ferredoxin entry path also
ensures growth at low H partial pressures. Then
the electron-bifurcating hydrogenase may gener-
ate reduced ferredoxin for CO reduction (20).
2
reduc-
2
8
9
2
(EP-No. 13172411.4). The data reported in this paper
Enzymes similar to the HDCR reported here
seem to play an important role in other bacteria
as well (see table S2 for the occurrence and pre-
vious description of HDCR-like gene clusters
in other prokaryotic genomes). A similar gene
cluster was predicted in the acetogen Treponema
primitia (21). HDCR-type enzymes are not re-
stricted to acetogens, but they, for example, have
also proposed functions in sulfate reducing or-
10. S. L. D. H. L. Drake, C. Matthies, K. Küsel, in
Acetogenesis, H. L. Drake, Ed. (Chapman and Hall,
New York, 1994), pp. 3–60.
are tabulated in the supplementary materials.
Supplementary Materials
www.sciencemag.org/content/342/6164/1382/suppl/DC1
Materials and Methods
Figs. S1 to S11
Tables S1 and S2
1
1. V. Müller, F. Imkamp, A. Rauwolf, K. Küsel, H. L. Drake,
in Strict and Facultative Anaerobes: Medical and
Environmental Aspects, M. M. Nakano, P. Zuber,
Eds. (Horizon Biosciences, Norfolk, 2004), pp. 251–281.
2. S. Wang et al., J. Bacteriol. 195, 4373–4386 (2013).
3. A. Poehlein et al., PLOS ONE 7, e33439 (2012).
4. Materials and methods are available as supplementary
materials on Science online.
1
1
1
References (24–28)
15 August 2013; accepted 28 October 2013
10.1126/science.1244758
the late 1990s and remained in use until 2010. Re-
Genetic and Molecular Basis of Drug sistant parasites were isolated from Brazilian pa-
tients in the 1970s (5, 6) and also selected in the
laboratory from sensitive parasite lines (7). Resist-
ance has a recessive basis and results in a ~500%
reduction in drug sensitivity (8). Genetic com-
plementation experiments demonstrate that the
Resistance and Species-Specific Drug
Action in Schistosome Parasites
same gene determines resistance in both field and
1
,2
3
2
1
Claudia L. L. Valentim, Donato Cioli, Frédéric D. Chevalier, Xiaohang Cao,
laboratory-selected parasites, although whether the
same mutations are responsible is unknown (9).
OXA is species-specific (10, 11), killing S. mansoni
(67 million cases) but not other schistosome species
(S. haematobium, 119 million cases) in Africa or
S. japonicum (1 million cases) in Asia (1).
1
1
3
3
Alexander B. Taylor, Stephen P. Holloway, Livia Pica-Mattoccia, Alessandra Guidi,
Annalisa Basso, Isheng J. Tsai, Matthew Berriman, Claudia Carvalho-Queiroz,
Marcio Almeida, Hector Aguilar, Doug E. Frantz, P. John Hart,
Philip T. LoVerde, *† Timothy J. C. Anderson *†
3
4
4
1
2
5
5
1,6
†
1
2
Oxamniquine resistance evolved in the human blood fluke (Schistosoma mansoni) in Brazil in
the 1970s. We crossed parental parasites differing ~500-fold in drug response, determined
drug sensitivity and marker segregation in clonally derived second-generation progeny, and
identified a single quantitative trait locus (logarithm of odds = 31) on chromosome 6.
A sulfotransferase was identified as the causative gene by using RNA interference knockdown
and biochemical complementation assays, and we subsequently demonstrated independent
origins of loss-of-function mutations in field-derived and laboratory-selected resistant parasites.
These results demonstrate the utility of linkage mapping in a human helminth parasite, while
crystallographic analyses of protein-drug interactions illuminate the mode of drug action and
provide a framework for rational design of oxamniquine derivatives that kill both S. mansoni and
S. haematobium, the two species responsible for >99% of schistosomiasis cases worldwide.
We exploited the S. mansoni genome sequence
(12, 13) and genetic map (14) to identify genome
region(s) that underlie OXA-resistance and to
determine the basis for species-specific drug ac-
tion. The complete life cycle of S. mansoni can be
maintained in the laboratory, and clonal expansion
of larval parasites within the snail host allows
production of thousands of genetically identical
1
Departments of Biochemistry and Pathology, University of Texas
2
Health Science Center, San Antonio, TX 78229, USA. Texas
Biomedical Research Institute, San Antonio, TX 78245, USA.
3
Institute of Cell Biology and Neurobiology, CNR, Rome, Italy.
n the absence of effective vaccines for human cause of river blindness, and Wuchereria bancrofti,
helminth infections, repeated rounds of mass cause of lymphatic filariasis (3, 4). However, Campus, Hinxton, UK. Department of Chemistry, The Univer-
4
Wellcome Trust Sanger Institute, Wellcome Trust Genome
5
6
I
treatment with drug monotherapies are typi- resistance to oxamniquine (OXA) in Schisto- sity of Texas at San Antonio, San Antonio, TX 78249, USA. De-
partment of Veterans Affairs, South Texas Veterans Health Care
cally used for control in most developing coun- soma mansoni, a trematode parasite that infects
tries (1, 2). These programs bring enormous health 67 million people in Africa and South America
benefits but impose strong selection on parasite (1), provides the first and clearest example of nat-
populations, and resistance is suspected for several urally selected drug resistance in a human helminth
System, San Antonio, TX 78229, USA.
These authors contributed equally.
Corresponding author. E-mail: tanderso@txbiomedgenetics.
org (T.J.C.A.); loverde@uthscsa.edu (P.T.L.); pjhart@biochem.
*
†
helminth species, including Onchocerca volvulus, parasite. OXA was the first-line drug in Brazil until uthscsa.edu (P.J.H.)
www.sciencemag.org SCIENCE VOL 342 13 DECEMBER 2013
1385