Current Biology
Magazine
Primer
The nitrogen cycle
Lisa Y. Stein1,* and Martin G. Klotz2,3
Nitrogen is the fourth most abundant
element in cellular biomass, and it
comprises the majority of Earth’s
atmosphere. The interchange between
inert dinitrogen gas (N2) in the extant
atmosphere and ‘reactive nitrogen’
(those nitrogen compounds that
support, or are products of, cellular
metabolism and growth) is entirely
controlled by microbial activities. This
was not the case, however, in the
primordial atmosphere, when abiotic
reactions likely played a significant role
in the inter-transformation of nitrogen
oxides. Although such abiotic reactions
are still important, the extant nitrogen
cycle is driven by reductive fixation of
dinitrogen and an enzyme inventory
that facilitates dinitrogen-producing
reactions. Prior to the advent of the
Haber-Bosch process (the industrial
fixation of N2 into ammonia, NH3) in
1909, nearly all of the reactive nitrogen
in the biosphere was generated and
recycled by microorganisms. Although
the Haber-Bosch process more
than quadrupled the productivity of
agricultural crops, chemical fertilizers
and other anthropogenic sources of
fixed nitrogen now far exceed natural
contributions, leading to unprecedented
environmental degradation.
The significance of nitrogen to
the biosphere and to cellular life is
indisputable; however, our fundamental
knowledge of the microorganisms and
enzymatic processes that transform
nitrogen into its various oxidation states
(Figure 1) remains incomplete. Here, we
outline the major microbial processes of
the nitrogen cycle, the microorganisms
that perform nitrogen transformations,
and the modularity and evolutionary
history of the nitrogen cycle, and
provide a perspective of this cycle, now
and into the future.
Major processes of the nitrogen cycle
The nitrogen cycle has traditionally
been divided into three processes — N2
fixation, nitrification, and denitrification
(Figure 2) — and microbes have
historically been labeled by their
identified participation in one of these
R94 Current Biology 26, R83–R101, February 8, 2016 ©2016 Elsevier Ltd All rights reserved
processes, that is, ‘nitrogen fixers’,
‘nitrifiers’ and ‘denitrifiers’. This
assignment of microbes to specific
eco-physiological cohorts based on
their association with a single process
came under scrutiny when ecologists
found evidence for dissimilatory (that
is, non-assimilatory) reduction of nitrite
to nitric oxides and nitrous oxides in
oxic environments (Figure 2, reaction
6A; called ‘nitrifier denitrification’), as
well as dissimilatory reduction of nitrite
to ammonium (Figure 2, reaction 2;
called ‘respiratory ammonification’) and
dissimilatory oxidation of ammonium in
anoxic environments (Figure 2, reaction
7; called ‘anammox’). Our previous
understanding of the nitrogen cycle
was complicated by the numerous
and diverse scientific approaches and
foci — from the compounds that were
transformed, to the compounds that
were generated, to the environmental
status of reactions that constituted the
processes. The narrow foci in these
approaches — employed for over 100
years in nitrogen-cycle research — have
been overcome in the post-genomic era
due to much-improved instrumentation,
a great wealth of data, and increased
interdisciplinary and international
collaboration. Thus, our understanding
of the nitrogen cycle now consists of
five accepted nitrogen-transformation
flows: ammonification, including
nitrogen fixation, and assimilatory
and dissimilatory reduction of nitrite
(Figure 2, reactions 1 and 2); nitrification
(Figure 2, reactions 3A, 3B and 4);
denitrification, including canonical,
nitrifier-dependent and methane-
oxidation-dependent denitrification
(Figure 2, reactions 6A–D); anammox,
as a form of coupled nitrification–
denitrification (Figure 2, reactions 7A–
C); and nitrite–nitrate interconversion
(Figure 2, reactions 4 and 5). The
general processes of organic matter
mineralization (often mislabeled as
‘ammonification’) and assimilation
(often incorrectly claimed to include
processes that regulate the generation
of ammonium and its uptake) by
cellular life complete the movement of
reactive nitrogen through the biosphere
(Figure 2).
Ammonification
Nitrogen fixation is one of two versions
of ammonification and is accomplished
by bacteria and archaea that encode
nitrogenase enzyme complexes made
Molecule Name Oxidation state
C-NH2 Organic-N Reduced
NH3, NH4+ Ammonia, Ammonium -3
N2H4 Hydrazine -2 More
electrons
NH2OH Hydroxylamine -1
N2 Dinitrogen 0
N2O Nitrous oxide +1
NO Nitric oxide +2
- Fewer
HNO2, NO2 Nitrous acid, Nitrite +3
NO2 Nitrogen dioxide +4
electrons
HNO3, NO3- Nitric acid, Nitrate +5
Oxidized
Current Biology
Figure 1. Nitrogen-cycle intermediates.
Intermediates, representing nine oxidation
states, that donate or accept electrons, there-
by contributing to electron flow and conserva-
tion of energy in participating microbes.
up of the molybdenum–iron protein
dinitrogenase and the vanadium or iron
protein dinitrogenase reductase. Although
there is momentum to genetically
engineer plants to express microbial
nitrogenase complexes or to express the
signaling pathway to attract nodulating
nitrogen-fixing bacteria to colonize
cereal crops, biological nitrogen fixation
remains an activity solely performed by
bacteria and archaea. Nitrogen fixation
is extremely oxygen sensitive, requiring
microorganisms to develop protective
mechanisms, such as spatial decoupling
(compartmentalization or forming
specialized cells), temporal decoupling,
rapid O2 respiration, or maximization of
nitrogenase synthesis and turnover. The
ammonium produced by nitrogen fixation
is either assimilated into biomass or is
further respired by aerobic and anaerobic
ammonia-oxidizing microbes. This
coupling of reaction 1 with either reaction
3 or 7, as shown in Figure 2, occurs
either between nitrogen-cycle-active
microbes in communities or within single
cells, such as in nitrogen fixing, nitrifying
methanotrophic bacteria.
Anaerobic assimilatory (ANRA)
and dissimilatory (DNRA) nitrite
reduction to ammonium (Figure 2,
reaction 2) is the second version of
ammonification and is performed by
both bacteria and fungi. This process
can be linked to nitrate reduction
to nitrite (Figure 2, reaction 5) at the
cellular level or between organisms in
a community. The acronym DNRA was
originally coined to describe an eco-
physiological process in which nitrogen
in the form of nitrate was traceably
removed from soils without either
being lost in the form of nitrogenous
gases (denitrification) or assimilated
Current Biology
Magazine
into microbial biomass. While the
significance of ANRA and DNRA to
global nitrogen cycling facilitated by
fermenting fungi in soils is not well
understood, respiratory ammonification
by bacteria and archaea in diverse
anoxic habitats is well established.
DNRA can generate a cellular proton-
motive force and thus conserve energy
to support cellular growth, although
this feature depends on which enzymes
are coupled. A negative redox potential
(that is, reducing conditions) is the
most important stimulant for DNRA. It
should be noted that the ANRA process
employs identical chemistry to DNRA
but it is facilitated by evolutionarily
unrelated nitrite reductases.
Nitrification
The process of nitrification involves
three cohorts of microorganisms:
cohort I, ammonia oxidizers that
oxidize ammonia to nitrite (nitritation;
Figure 2, reactions 3A and 3B); cohort
II, nitrite oxidizers that oxidize nitrite to
nitrate (nitratation; Figure 2, reaction
4); and cohort III, complete ammonia
oxidizers that oxidize ammonia all
the way to nitrate (comammox;
Figure 2, reactions 3A, 3B and 4).
Although comammox microorganisms
were discovered only in 2015, their
existence was predicted approximately
10 years earlier. Cohorts II and III
include only chemolithotrophic
microbes that can use nitrite and
ammonia, respectively, as sole sources
of energy and reductant for cellular
growth. Whereas nitrite-oxidizing
chemolithotrophs of cohort II are
classified within several classes of
the Proteobacteria and the phylum
Nitrospirae, the comammox bacteria
are, so far, restricted to representatives
in lineage II of the genus Nitrospira.
Nitrifying ammonia-oxidizing microbes
of cohort I include chemolithotrophic
members of Betaproteobacteria and
Gammaproteobacteria, and members
of the Thaumarchaeota. In addition
to these chemolithotrophs, there are
heterotrophic and methanotrophic
microorganisms that proficiently
oxidize ammonium to nitrite, but they
do not gain energy from the process to
support growth. The nitrite and nitrate
produced by aerobic reactions in the
nitrification process can be respired
anaerobically (Figure 2, reactions 2
and 5) or the resulting ammonium can
be assimilated.
Current Biology 26, R83–R101, February 8, 2016 ©2016 Elsevier Ltd All rights reserved
N2 1
6C
7C
N2O
N2H4
Assimilation
6D
6B NH4+/NH3 R~NH2
7B Mineralization
NO 2 3A
NH2OH
7A
6A
NO2- 3B
5 4
NO3 -
Current Biology
Figure 2. Major processes of the nitrogen cycle.
Reactions that comprise the seven major processes of the nitrogen cycle are represented by
the numbered circles. Ammonification may be accomplished either by process 1, reduction of
dinitrogen (also referred to as ‘nitrogen fixation’ or ‘Nif’), or by process 2, dissimilatory nitrite
reduction to ammonium (DNRA). Nitrification is composed of process 3, oxidation of ammonia to
nitrite (also referred to as ‘nitritation’), and process 4, oxidation of nitrite to nitrate (also referred
to as ‘nitratation’). Process 5, reduction of nitrate to nitrite, can be coupled to processes 2, 6 or 7
in a population or a community. Denitrification is shown as process 6, which is also referred to as
‘nitrogen-oxide gasification’. Anammox is shown as process 7, and is also referred to as coupled
‘nitrification–denitrification’.
Denitrification process, a methanotrophic bacterium,
Denitrification describes the process of Candidatus Methylomirabilis oxyfera,
anaerobic respiration of nitrite (NO2-), consumes methane as a source of
nitric oxide (NO), and nitrous oxide energy, reductant and carbon, and
(N2O) to N2 (Figure 2, reactions 6A–C). reduces NO2- to N2 by expressing a
Heterotrophic microbes that can directly nitric oxide dismutase. This enzyme
couple these three reactions with the dismutates two NO molecules to O2 and
reduction of nitrate to nitrite (Figure 2, N2 without producing the intermediate
reaction 5) and perform denitrification N2O (Figure 2, reaction 6D); the O2
from NO3- to N2 are referred to as produced is then used by this bacterium
classical or canonical denitrifiers. It to oxidize methane to methanol while it
is nevertheless recognized that many resides in an anoxic environment, hence
bacteria and archaea express only the name ‘denitrifying intra-oxygenic
partial denitrifying inventories, and methanotroph’.
are missing genes encoding enzymes Anammox
involved in reactions 5, 6B and/or 6C Anammox, or anaerobic ammonium
(Figure 2). Such incomplete pathways oxidation, utilizes the pools of NO2-
can lead to the release of nitrogenous and ammonium (NH4+) to form N2 via
gases such as NO and N2O to the the intermediates NO and hydrazine
environment, or a failure to deplete (N2H4; Figure 2, reactions 7A–C). The
NO3-. This group also includes all recent literature has also suggested
ammonia-oxidizing chemolithotrophic the existence of a modified anammox
bacteria, which reduce NO2- to N2O pathway, in which hydroxylamine
aerobically (Figure 2, reactions 6A and (NH2OH) is used instead of NO for
6B). In addition, several eukaryotes the oxidation of NH4+, but additional
including fungi and foraminifers (a experimentation will be required to
class of protists) can reduce NO2- or validate this pathway. Anammox
NO3- to N2O or N2, respectively. In chemolithotrophy is performed solely
an unusual twist to the denitrifying by the Brocadiaceae bacteria in
R95

the order Planctomycetales within
a specialized organelle called the
anammoxosome. Anammox is
ecologically beneficial for wastewater
treatment as it removes both nitrite
and ammonium simultaneously
without producing N2O, and has been
industrially implemented at full scale.
Anammox is also a major nitrogen-
removal process in the ocean and in
oxygen minimum zones.
From organisms to modularity of the
nitrogen cycle
For hundreds of years, microbiologists
have enriched and isolated
microorganisms from the environment
to connect discrete biological entities to
particular processes. The enzymology
of the above nitrogen-cycle processes
were largely discovered using axenic
cultures of microorganisms, leading
to the impression that these cultured
microbial phylotypes should also be
the main players in charge of those
processes in the environment. For
instance, the first ammonia-oxidizing
bacteria were isolated in the 1890s
and a great proportion of biochemical,
physiological, and genetic knowledge
of ammonia oxidation was derived
from research performed with these
very isolates. For many years, microbial
ecologists relied on gene sequences
and physiological attributes derived
from a small number of isolates to
detect and enumerate ammonia
oxidizers in the environment. However,
in the 2000s, metagenome sequences
showed an irrefutable association
between ammonia monooxygenase
(the first enzyme involved in ammonia
chemolithotrophy) and archaeal 16S
rRNA genes that shifted the focus from
bacterial ammonia-oxidizers to a whole
new research venture based on the
ammonia-oxidizing thaumarchaeota.
Following the discovery of ammonia-
oxidizing thaumarchaeota, the first
strain isolated into pure culture was
Nitrosopumilus maritimus in 2005,
enabling physiological and biochemical
characterization of a completely
different group of ammonia-oxidizing
chemolithotrophs.
Aside from revealing entirely new phyla
of nitrogen-cycling microbes that had
evaded age-old cultivation-dependent
approaches, genome and metagenome
sequencing also revealed that individual
modules, independent of complete
R96 Current Biology 26, R83–R101, February 8, 2016 ©2016 Elsevier Ltd All rights reserved
integrated pathways, are encoded
and expressed by microorganisms to
perform single segments of nitrogen-
cycle metabolism. For instance,
methanotrophic bacteria can express
many genes encoding enzymes with
a known involvement in the processes
of nitrogen fixation, nitrification and
denitrification. Intriguingly, there is
no predictable association between
nitrogen-cycle genes and phylotype
among the methanotrophs. Furthermore,
the extent of nitrogen-cycling inventory
is scattered; some methanotrophs
encode nitrite and nitric oxide reductases
together, some fix nitrogen and encode
a suite of denitrification genes, whereas
others may encode only one reductase
or none. Hence, environmental factors
apparently drive random lateral gene
transfer events to perform tasks to
enhance fitness of a particular species,
thereby leading to niche differentiation.
For instance, the acquisition and
expression of nitrifying modules could
allow certain methanotrophs to survive
high ammonium loads in situations
where NH3 can compete with methane
for access to methane monooxygenase
enzymes. Similarly, expression of
denitrifying modules can allow a
methanotrophic species to grow in nearly
anoxic ecosystems. Nitrogen fixation is
inherently modular, with gene clusters
distributed across numerous bacterial
and archaeal phyla, arguing once again
that habitat conditions select for this trait.
For the majority of history in nitrogen-
cycle research, we have held onto the
concept that organisms of particular
phylotypes are nitrifiers, denitrifiers,
nitrogen fixers, and the like. This
organismal concept is now challenged
by the knowledge that nitrogen-cycle
genes and processes are not necessarily
constrained by phylotype, but rather
endow organisms with the inventory
required to adapt and thrive in ecosystems
where nitrogenous molecules undergo
dynamic changes in concentration and
composition. Envisioning the nitrogen
cycle as a continuum of modular
evolution is consequently the most
parsimonious outlook.
Modular evolution of the nitrogen
cycle
The primordial nitrogen ‘cycle’ was not
closed, but rather was dominated by
coupled abiotic reactions that provided
nitrogen oxides and reduced nitrogen
Current Biology
Magazine
in the form of NH4+, which is required
for nitrogen assimilation into biomass.
The nitrogen oxide reservoir in the
form of NO3- vastly exceeded available
NH4+, which favored the evolution of a
reductase protein inventory involved in
nitrate–nitrite inter-conversion (Figure 2,
reactions 4 and 5) along with numerous
nitrite reductases (Figure 2, reactions
2, and 6A/7A). The suite of reductase
enzymes relied on the bioavailability
of transition metals under anoxic
conditions such as molybdenum (as
found in molybdopterin) and iron (found
in siroheme or heme cytochrome c). The
function and evolution of these protein
complexes and their encoding genes
have been comprehensively reviewed
by Simon and Klotz (2013) with an
emphasis on bioenergetics. With the
exception of reactions labeled 3A, 3B,
4, 7B and 7C in Figure 2, the reactions
that interconvert nitrogen oxides of
different oxidation states (Figure 1) are
reductions, which require the availability
and accessibility of electrons. These
electrons are derived from the quinone
pool, facilitated by quinone-reactive
proteins (QRPs) directly, or via redox
carrier proteins. Therefore, the evolution
of proteins that participate in the
nitrogen cycle is intrinsically connected
to the evolution of QRPs. While QRP
complexes that contain cytochrome b
proteins have been well characterized,
those that include Fe–S clusters
(NapH/NosH family), cytochrome c
proteins (NrfH/NapC/TorC family) or
the enigmatic transmembrane protein
NrfD/PsrC are significantly less well
understood. Interestingly, the evolution
and implementation of these QRPs
in catabolic electron flow are also
modular, interconnected and function
to provide electrons to a variety of
catalytic redox proteins involved in the
sulfur and nitrogen cycles by oxidation
of quinol. Exceptions are the composite
MFc complex that evolved into an
‘alternative complex III’ for branched
electron transfer (quinol oxidase not
serving electrons to enzymes), the
cM552 (CycB) protein, which functions
as a quinone reductase in ammonia-
oxidizing chemolithotrophs, and the
copper-containing analog in ammonia-
oxidizing Thaumarchaeota.
Reactions 3A, 3B, 4, 7B and 7C in
Figure 2 are oxidations, with reactions
3B and 7C being dehydrogenations
performed by homologous multiheme
Current Biology
Magazine
cytochrome c (MMC) proteins.
Interestingly, these dehydrogenases
evolved from MMC reductases that
perform reactions 2 and 6A/7A in
Figure 2. Reactions 6A/7A and leaky
dehydrogenase 3B were likely the main
biotic generators of the NO radical, the
source of nitrosative stress in cells and
otherwise of abiotic origin, which served
as the primordial oxidant before the
advent of molecular oxygen. Abundant
production of NO radicals led to the
emergence of a large complement
of evolutionarily unrelated enzymes
capable of NO detoxification, most of
them being reductases (reaction 6B in
Figure 2), such as the soluble tetraheme
cytochrome c’ and c’-beta proteins,
which were later complemented by
numerous other NO reductases, the
main biotic source of N2O.
We proposed in 2008 that the
key innovation for the closure of the
nitrogen cycle was likely the emergence
of N2H2 hydrolase (reaction 7B in Figure
2) and its subsequent integration
into a redox gradient generated by
reactions 7A and 7C (Figure 2), which
reversed electron flow through the
protein complex and created what is
now known as N2H2 synthase, the key
enzyme in the anammox process. It is
remarkable that this protein complex
is restricted only to the Brocadiaceae
that facilitate the anammox process.
This innovation created a connection
between reduced, oxidized, and
inert nitrogen and thus constituted
the closure and formation of a real
cycle. The onset of nitrogen fixation
is presently still debated, but there is
growing momentum for placing the
evolution of this pathway before the
great oxygenation event, potentially
as a contemporary invention to the
emergence of the anammox pipeline.
Copper was not bioavailable in anoxic
environments, which is the reason why
many enzymes with copper as the
redox-active metal emerged after the
great oxygenation event, including N2O,
NO and NO2- reductases, and functional
ammonia monooxygenase. While there
is an increasingly clear picture on the
evolutionary relationships between
copper-containing membrane-bound
monooxygenases (Cu-MMOs), including
those that oxidize ammonia and short
alkanes such as methane, the origins
of the enzyme subunit proteins remain
elusive. This explosion of nitrogen-
Current Biology 26, R83–R101, February 8, 2016 ©2016 Elsevier Ltd All rights reserved
cycle inventory occurred in parallel
to the explosion of aerobic lifestyles,
which provided more efficient electron
flow into and out of the quinone pool,
while providing additional means
for energy conservation (generation
of proton-motive force). Hence, the
taxonomically unpredictable and
broad distribution of diverse nitrogen-
compound-transformation inventory
emerged gradually over evolutionary
time. This evolution followed diverse
environmental and functional pressures,
resulting in the acquisition, maintenance
or loss of functionally linked inventory,
thereby generating a nitrogen cycle
of modular design, similar to the
modular designs of the cellular central
pathways (glycolysis, TCA, pentose
phosphate pathway) or the Calvin
cycle. It thus comes as no surprise to
those who accept the modular nature
of the nitrogen cycle that anammox
bacteria can oxidize NO2- anaerobically
with an enzyme homologous to the
NO2--oxidizing enzyme of the aerobic
Nitrospira even though the genomic
background (phylotype) of these
two bacterial groups is very distantly
related.
Bioenergetically speaking, the
dissimilatory reactions of the nitrogen
cycle make it the most advanced
biogeochemical nutrient cycle because,
out of nine possible intermediates,
the nitrogen cycle uniquely includes
representative reactions for seven
oxidation states between -3 and +5
(Figure 1). While the sulfur and carbon
cycles also span over nine oxidation
states, they provide only three states
(0, +2, +4) between -2 and +6 and
three states (-2, 0, +2) between -4
and +4, respectively, that are available
for electron-transfer reactions. This
bioenergetics picture is congruent with
our understanding of the timeline for
the emergence of a complete nutrient
cycle, a picture in which the extant
nitrogen cycle represents a more
evolved cycle, cemented by the finding
that significant nitrogen-cycle inventory
evolved from inventory active in the
sulfur and/or carbon cycles.
Modern day nitrogen cycle and
societal challenges
It has become widely recognized,
particularly in analyzing studies of
food production and global climate
change, that the fate of humanity is
intertwined with our ability to control
the nitrogen cycle. On one side,
the Green Revolution of the 20th
century averted a global food crisis,
in part by introducing widespread
synthetic fertilizer usage. To put this
in perspective, the Haber-Bosch
process is responsible for feeding
approximately 48% of the global
human population, equivalent to
supporting four billion births since
1908. On the other side, with our
extremely rapid population growth
and continued increases in fertilizer
usage, we have pushed the nitrogen
cycle beyond sustainability where
now NO3- pollution is responsible
for increasing dead zones in coastal
areas and N2O is considered the
key greenhouse gas to mitigate in
the 21st century and beyond. Control
of the nitrogen cycle is a tricky
business, in part because of our
success with fertilizer-dependent
food production, and also because
we are still discovering how the
microbial processes of the nitrogen-
cycle work. For instance, the relatively
recent discoveries of the anammox
and comammox processes, as
well as the discovery of ammonia-
oxidizing thaumarchaeota (described
above) have revealed organisms
and processes that profoundly
impacted how we view and study
nitrogen transformations and the
environmental factors that control
them. Furthermore, metabolic and
environmental controls on organisms
and processes that release and
accumulate major pollutants, like NO3-
and N2O, remain largely mysterious.
While it is tempting to use genetic
manipulation to engineer nitrogen-
fixing cereal crops or to improve
nitrification inhibitors, technological
fixes will continue to be limited by
our understanding of the underlying
microbiology. It is therefore incumbent
on science and society to pay
attention to the nitrogen cycle and
take our cues from the microbes that
ultimately control it.
ACKNOWLEDGEMENTS
Nitrogen-cycle research in the Stein lab
is supported by the Natural Sciences and
Engineering Research Council of Canada.
Nitrogen-cycle research in the Klotz lab has
been supported during the last two decades
R97

by the NSF, DOE, USDA, the Gordon and
Betty Moore Foundation and institutional
incentive funds from the University of
Louisville and UNC Charlotte.
FURTHER READING
Arp, D.J., and Bottomley, P.J. (2006). Nitrifiers:
More than 100 years from isolation to genome
sequences. Microbe 1, 229–234.
Daims, H., Lebedeva, E.V., Pjevac, P., Han, P.,
Herbold, C., Albertsen, M., Jehmlich, N.,
Palatinszky, M., Vierheilig, J., Bulaev, A., et al.
(2015). Complete nitrification by Nitrospira
bacteria. Nature 528, 555–559.
Erisman, J.W., Sutton, M.A., Galloway, J., Klimont,
Z., and Winiwarter, W. (2008). How a century
of ammonia synthesis changed the world. Nat.
Geosci. 1, 636–639.
Ettwig, K.F., Butler, M.K., Le Paslier, D., Pelletier,
E., Mangenot, S., Kuypers, M.M.M., Schreiber,
F., Dutilh, B.E., Zedelius, J., de Beer, D., et al.
(2010). Nitrite-driven anaerobic methane
oxidation by oxygenic bacteria. Nature 464,
543–548.
Hoffman, B.M., Lukoyanov, D., Yang, Z.-Y., Dean,
D.R., and Seefeldt, L.C. (2014). Mechanism of
nitrogen fixation by nitrogenase: the next stage.
Chem. Rev. 114, 4041–4062.
Kartal, B., Maalcke, W.J., de Almeida, N.M.,
Cirpus, I., Gloerich, J., Geerts, W., den
Camp, H., Harhangi, H.R., Janssen-Megens,
E.M., Francoijs, K.J., et al. (2011). Molecular
mechanism of anaerobic ammonium oxidation.
Nature 479, 127–130.
Klotz, M.G., Schmid, M.C., Strous, M., op den
Camp, H.J., Jetten, M.S., and Hooper, A.B.
(2008). Evolution of an octahaem cytochrome
c protein family that is key to aerobic and
anaerobic ammonia oxidation by bacteria.
Environ. Microbiol. 10, 3150–3163.
Klotz, M.G., and Stein, L.Y. (2008). Nitrifier genomics
and evolution of the nitrogen cycle. FEMS
Microbiol. Lett. 278, 146–456.
Schleper, C., and Nicol, G.W. (2010). Ammonia-
oxidising archaea—physiology, ecology and
evolution. Adv. Microb. Physiol. 57, 1–41.
Simon, J. (2002). Enzymology and bioenergetics
of respiratory nitrite ammonification. FEMS
Microbiol. Rev. 26, 285–309.
Simon, J., and Klotz, M.G. (2013). Diversity and
evolution of bioenergetic systems involved in
microbial nitrogen compound transformations.
Biochim. Biophys. Acta 1827, 114–135.
Stein, L.Y. (2011). Heterotrophic nitrification and
nitrifier denitrification. In Nitrification, B.B. Ward,
D.J. Arp and M.G. Klotz, eds. (Washington, D.C.:
ASM Press), pp. 95–114.
Stein, L.Y., and Klotz, M.G. (2011). Nitrifying and
denitrifying pathways of methanotrophic
bacteria. Biochem. Soc. Tran. 39, 1826–1831.
Ward, B.B. (2011). Nitrification: an introduction and
overview of the state of the field. In Nitrification,
B.B. Ward, D.J. Arp and M.G. Klotz, eds.
(Washington, DC: ASM Press), pp. 3–8.
Zhu-Barker, X., Cavazos, A.R., Ostrom, N.E.,
Horwath, W.R., and Glass, J.B. (2015). The
importance of abiotic reactions for nitrous oxide
production. Biogeochemistry 126, 251–267.
1
Department of Biological Sciences,
University of Alberta, Edmonton, AB, Canada.
2
Department of Biology and School of Earth &
Environmental Sciences, Queens College of
the City University of New York, Flushing,
NY, USA. 3Institute of Marine Microbes &
Ecospheres and State Key Laboratory of
Marine Environmental Science, Xiamen
University, Xiamen, China.
*E-mail: lisa.stein@ualberta.ca
R98 Current Biology 26, R83–R101, February 8, 2016 ©2016 Elsevier Ltd All rights reserved
Correspondence
Morbid attraction
to leopard urine
in Toxoplasma-
infected
chimpanzees
Clémence Poirotte1,*,
Peter M. Kappeler2,
Barthelemy Ngoubangoye3,
Stéphanie Bourgeois4,
Maick Moussodji5,
and Marie J.E. Charpentier1
Parasites are sometimes capable of
inducing phenotypic changes in their
hosts to improve transmission [1].
Toxoplasma gondii, a protozoan that
infects a broad range of warm-blooded
species, is one example that supports
the so-called ‘parasite manipulation
hypothesis’: it induces modifications
in rodents’ olfactory preferences,
converting an innate aversion for cat
odor into attraction and probably
favoring trophic transmission to feline
species, its only definitive hosts [2]. In
humans, T. gondii induces behavioral
modifications such as personality
changes, prolonged reaction times and
decreased long-term concentration
[3]. However, modern humans are not
suitable intermediate hosts because
they are no longer preyed upon by
felines. Consequently, behavioral
modifications in infected people
are generally assumed to be side
effects of toxoplasmosis or residual
manipulation traits that evolved in
appropriate intermediate hosts. An
alternative hypothesis, however, states
that these changes result from parasite
manipulative abilities that evolved
when human ancestors were still under
significant feline predation [3,4]. As
such, T. gondii also alters olfactory
preferences in humans; infected men
rate cat urine, but not tiger urine, as
pleasant while non-infected men do not
[5]. To unravel the origin of Toxoplasma-
induced modifications in humans,
we performed olfactory tests on a
living primate still predated by a feline
species. We found in our closest relative,
the chimpanzee (Pan troglodytes
troglodytes), that Toxoplasma-
infected (TI) animals lost their innate
Current Biology
Magazine
A
Toxoplasma-infected
Non-infected
Chimpanzee's behavioral responses (Least Square Means and SEM)
0.8 p = 0.008
0.7
0.6
p = 0.048
Investigations
0.5
0.4
0.3
0.2
0.1
0
B 0
-1
Approaches
-2
-3
-4
-5
-6 p = 0.003
p = 0.047
Figure 1. Comparison of the behavioral re-
sponses of 33 chimpanzees towards urine
of human, leopard (natural predator) and ti-
ger and lion (non-natural predators) during
behavioral tests based on olfactory cues.
(A) Olfactory investigations performed towards
urine source (for each urine type, n = 9 TI and
n = 24 TN chimpanzees). (B) Approaches dis-
played towards urine source (for each urine
type, n = 40 scans per individual). Least Square
Means (LSM) and SEM are represented. White
bars, TN chimpanzees; black bars, TI chim-
panzees. Only significant differences in LSM
(p < 0.05) are shown for biologically relevant
two-by-two comparisons (intra-treatment
comparisons: same urine type, different para-
site status; and intra-parasite status compari-
sons: same parasite status, different urine type;
see the Supplemental Information for a full
set of biologically relevant comparisons.) For
approach data, high negative values indicate
fewer approaches towards urine source than
low negative values.
aversion towards the urine of leopards
(Panthera pardus), their only natural
predator. By contrast, we observed
no clear difference in the response
of TI and Toxoplasma-non-infected
(TN) animals towards urine collected
from other definitive feline hosts that
chimpanzees do not encounter in
nature. Although the adaptive value of
parasitically induced behavior should be
assessed carefully, we suggest that the
behavioral modification we report could
increase the probability of chimpanzee
predation by leopards for the parasite’s
own benefit. This possible parasite
adaptation would hence suggest that
Toxoplasma-induced modifications in
modern humans are an ancestral legacy
of our evolutionary past.
We performed collective olfactory
tests on 33 chimpanzees (9 TI and
24 TN), living in five captive groups in