Current Biology

Review

Comparative Biology of Oxygen Sensing in Plants

and Animals
Michael J. Holdsworth1 and Daniel J. Gibbs2
1School of Biosciences, University of Nottingham, Loughborough LE12 5RD, UK
2School of Biosciences, University of Birmingham, Edgbaston B15 2TT, UK
Correspondence: Michael.Holdsworth@nottingham.ac.uk (M.J.H.), d.gibbs@bham.ac.uk (D.J.G.)
https://doi.org/10.1016/j.cub.2020.03.021

Aerobic respiration is essential to almost all eukaryotes and sensing oxygen is a key determinant of sur-
vival. Analogous but mechanistically different oxygen-sensing pathways were adopted in plants and meta-
zoan animals, and include ubiquitin-mediated degradation of transcription factors and direct sensing via
non-heme iron(Fe2+)-dependent-dioxygenases. Key roles for oxygen sensing have been identified in
both groups, with downstream signalling focussed on regulating gene transcription and chromatin modi-
fication to control development and stress responses. Components of sensing systems are promising
targets for human therapeutic intervention and developing stress-resilient crops. Here, we review current
knowledge about the origins, commonalities and differences between oxygen sensing in plants and
animals.

Introduction The Ubiquitin Proteasome System as a Hub for Oxygen

Molecular oxygen (O2) is necessary for many core biochemical Sensing across Kingdoms
pathways, most importantly as the final electron acceptor in In metazoans and angiosperms, major mechanisms directing
mitochondrial electron transport, and is therefore essential to changes in gene expression under hypoxia are controlled by hyp-
the vast majority of eukaryotes. Oxygen first appeared in quan- oxia-responsive transcription factors. Their stability is intrinsically
tity on earth as a result of the evolution of oxygenic photosyn- linked to O2 levels, and in oxygenated environments they are poly-
thesis at least 2.3 Ga (billion years ago) (reviewed in [1]) ubiquitylated and rapidly degraded by the 26S proteasome
(Figure 1). Subsequently, as part of the evolution of endosym- (Figure 2). In metazoans, the hypoxia inducible factor (HIF, also
biosis with an ancient cyanobacterial group before 1 Ga, early known as EPAS; see Box 1) heterodimer consists of HIFa and b
eukaryotic algae gained the ability to photosynthesise, leading bHLH-PAS domain subunits (Figure 2). Three HIFa proteins are
to further increases in O2 levels that peaked at over 30% during found in mammals; HIF1a and HIF2a contain amino- and car-
the Carboniferous (360 to 300 Ma). Endosymbiosis with boxy-terminal transactivation domains (NTAD and CTAD), while
purple non-sulphur bacteria (that became mitochondria), which HIF3a lacks the latter [4]. The NTAD of HIFa contains conserved
predated chloroplast endosymbiosis, may have allowed early prolyl residues that are hydroxylated using O2 by prolyl 4-hydrox-
eukaryotes to tolerate O2 and use the energy of mitochondrial lases (PHD, also known as EGLN) [5,6]. This modification then
aerobic respiration to become multicellular. Various hypotheses permits binding of the E3 ubiquitin ligase von Hippel-Lindau pro-
have proposed that increased O2 levels were either highly tein (pVHL) to initiate polyubiquitylation [7,8], leading to HIFa
poisonous and catastrophic for early anaerobic eukaryotes or degradation. Hypoxia limits PHD activity, precluding pVHL bind-
that these organisms were already pre-adapted to deal with ing, thus allowing association with HIFb and re-localisation to
reactive oxygen species that aided evolution of O2 tolerance the nucleus [4] (Figure 2). The CTAD-containing HIFa variants
(discussed in [2]). can also be hydroxylated on asparaginyl residues by Factor Inhib-
Oxygen varies in the environment (for example, declining with iting HIF1a (FIH), which limits HIFa association with transcriptional
increased altitude, or as a result of submergence in water or co-factors [9]. This separate O2-triggered modification also there-
under the soil surface), and also varies internally during develop- fore contributes to inhibition of HIF activity through a parallel hy-
ment or disease [3,4]. It is clear that for such an essential compo- droxylation-dependent but non-proteasomal route.
nent of intracellular biochemistry, sensing and response to In flowering plants, the group VII ETHYLENE RESPONSE
changing O2 levels must be an important feature of multicellular FACTOR transcription factors (ERFVIIs) control anaerobic gene
eukaryotes. In this Review, we focus on biochemical pathways expression under hypoxia [10]. Following co-translational methi-
that evolved in plants and animals to sense and respond to onine excision, in high O2 levels the amino-terminal cysteine of
reduced O2 levels (hypoxia). In both lineages, analogous path- ERFVIIs is converted to Cys-sulfinic acid by PLANT CYSTEINE
ways evolved that target nuclear-located processes for OXIDASEs (PCOs), which leads to amino-terminal (Nt)-arginyla-
response. We highlight the different evolutionary trajectories of tion by ATE [11,12]. Nt-Arg–ERFVIIs are then targeted for protea-
pathways, the importance of dioxygenases as conserved sen- somal degradation by the E3 ligase PROTEOLYSIS 6 (PRT6)
sors of hypoxia, the physiological importance of oxygen-sensing [13,14]. This pathway also requires nitric oxide (NO) [15]. Thus,
and avenues for identification of novel sensors and pathway similarly to HIFa regulation, coupling protein turnover to O2 avail-
components. ability results in ERFVII stabilisation under hypoxia (Figure 2).

R362 Current Biology 30, R362–R369, April 20, 2020 ª 2020 Elsevier Inc.
Current Biology

Review
2.5 2 1.5 1 0.5 Time (Ga) 0 Figure 1. Evolutionary history of core
Proterozoic Phanerozoic components of the HIF and PCO/ADO
oxygen-sensing pathways.
30% (Atmospheric oxygen levels)
20 Ages of key evolutionary events and predicted O2
10 levels at distinct ages of earth history are indicated
0 GOE (billion years ago; Ga). GOE, Great Oxidation
Archaea and bacteria Event, first appearance of significant atmospheric
Eukaryotes O2 levels. Possible times of appearance of oxy-
Viridiplantae gen-sensing pathway components (ovals with
Land plants gene name indicated) are shown based on pres-
Vascular plants ence of similar protein sequences or functional
Angiosperms testing in extant taxonomic groups, and important
Animals functional diversification indicated. Animal-spe-
Metazoans cific components are in greys, plant-specific in
greens.
PCO/ADO pathway components:
PRT6
UBR1 VRN2
PRT1
ATE ? ZPR2

ERFVII
PCO/ decreases in O2 availability would be
ADO
RGS required before its activity is inhibited
Il-32 [27]. PHD/FIH incorporate one oxygen
HIF pathway components: atom into the target HIFa prolyl or
FIH HIF1D asparaginyl residue, while the second
E
HIF1E decarboxylates 2-oxo-glutarate (2-OG)
PHD to produce CO2 and succinate [23,28]
pVHL
(Figure 2). PCOs and ADO also have
Current Biology
high KmO2 values above typical plant
and animal tissue O2 concentrations, but
in contrast to PHDs they are not 2-OG
Recently, a mammalian protein with high similarity to PCO, dependent, they integrate both atoms directly into Nt-Cys to
cysteamine (2-aminoethanethiol) dioxygenase (ADO), was char- generate Cys–sulfinic acid [12,16,29].
acterised and shown to control O2-dependent turnover of non- Metazoans encode multiple PHD isoforms, which are differen-
nuclear REGULATOR OF G PROTEIN SIGNALLING (RGS) 4, 5, tially expressed and have varying subcellular localisations,
and 16 proteins via the mammalian Arg/N-degron pathway although the main mammalian PHD2 variant is cytosolic and
[16]. This highlights an alternative mechanism for O2-sensitive constitutively expressed [4,30]. Flowering plant PCOs have
proteolysis in mammals, equivalent to the predominant system different sensitivities to O2 and pH, and divergent substrate pref-
in plants. erences based on assessment of their activities on peptide se-
There is evidence that alternative pathways can also target quences [29]. Of the five PCOs in Arabidopsis thaliana, PCO4
HIFa and ERFVIIs for degradation, revealing additional proteo- is the most catalytically potent, suggesting that it may be the
lytic mechanisms for fine-tuning their stability [4,17–20]. Further- dominant variant. Apparently without an active oxygen-transport
more, animal PHD and plant PCO enzymes also have non-HIF system, strong gradients of hypoxia exist in plant tissues
and -ERFVII targets, respectively. In Arabidopsis thaliana, the (obvious examples include tubers and seeds) [3,31] and it may
PCO targets LITTLE ZIPPER 2 (ZPR2) and Polycomb repressive be that PCOs with different affinities for O2 operate in different
complex 2 (PRC2) component VERNALIZATION 2 (VRN2) are tissues or at different developmental time points. Interestingly,
subject to ubiquitin-mediated degradation [21,22], whereas hy- a subset of these oxygen-sensing enzymes in animals and plants
droxylation of candidate non-canonical PHD/FIH substrates, are transcriptionally induced by low O2 levels, suggesting that
such as IKKb, p53, and OTUB1, can have different effects on homeostatic mechanisms for dampening the hypoxic response
protein activity and interactions [23]. have evolved in both kingdoms [4,11].
In addition to PHD and PCO/ADO proteins, there are many
The Key Role of Non-Heme Iron (Fe2+)-Dependent other non-heme iron (Fe2+)-dependent dioxygenases in animals
Dioxygenases in Oxygen Sensing and plants [24,32], although several of these, including collagen
The enzymes catalysing both prolyl-/asparaginyl-hydroxylation prolyl hydroxylases and certain JmjC (Jumonji C) domain lysine
(PHD, FIH) and Nt-cysteine oxidation (PCO, ADO) belong to demethylases (KDMs), are unlikely to sense physiological
the non-heme iron (Fe2+)-dependent dioxygenase family, so changes due to their high O2 affinities [23,33]. Nonetheless, it
called because their catalytic sites contain a redox active iron was recently shown that some histone-specific KDMs (KDM5A
directly coordinated to the protein, and incorporate both atoms and 6A) do have KmO2 values in the requisite range for sensing
from O2 into substrates [24]. PHDs function as physiological O2 intracellular O2, and are able to directly modulate the methyl-
sensors due to their high KmO2 values, which for the dominant ation status of chromatin dependent on O2 availability [33,34]
PHD2 isoform (dependent on the length of peptide studied) (Figure 2). Under hypoxic conditions, KDM activity is reduced,
has variably been reported from less than 100 mM to 1700 mM, resulting in enhanced global levels of histone methylation,
much higher than in vivo O2 concentrations [25,26]. In contrast, regulating gene expression and cell fate. The activity of a sepa-
FIH has a higher affinity for O2 than PHDs, indicating that greater rate non-histone KDM (KDM3A), which is involved in the

Current Biology 30, R362–R369, April 20, 2020 R363

 Current Biology

Review

Normoxia

26S proteasome 26S proteasome 26S proteasome

Histone
Ub Ub Ub
demethylation pVHL UBR1 PRT6

ATE ATE
CO2 HO-Pro
OH NO OH NO
succinate O = Cys O = Cys

CO2
succinate
O2 2-OG O2 O2 O2

KDM PHD 2-OG ADO PCO

Hypoxia
[IL-32] [ZPR2]
Pro
KDM Inactive HIFD Cys RGS Cys ERFVII Cys VRN2

PRC2
HIFE

Cardiovasculature
development,
Increased histone HRE HRPE H3K27me3
angiogenesis
methylation

Changes in gene expression Changes in gene expression

Current Biology

Figure 2. A comparison of major oxygen-sensing systems in metazoans and flowering plants.

Mammalian HIFa and plant ERFVII transcription factors are stable under hypoxia where they drive hypoxic gene expression through binding to genes bearing
specific promoter elements (HRE, HRPE). In oxygenated environments, prolyl residues in HIFa are hydroxylated by 2-OG-dependent PHD dioxygenases prior to
ubiquitylation (Ub) by the pVHL E3 ubiquitin ligase, while the amino-terminal Cys of ERFVIIs is converted to Cys-sulfinic acid by 2-OG-independent PCO di-
oxygenases, prior to ATE-mediated arginylation that permits recognition by the PRT6 E3 ubiquitin ligase. ZPR2 stability is also regulated via PCO in plants to
control shoot meristem function. The recently discovered ADO pathway in mammals is equivalent to the PCO pathway in plants and regulates the stability of non-
nuclear RSG and IL-32 substrates that do not directly modulate gene expression. Mammals and angiosperms have contrasting oxygen-regulated mechanisms
controlling histone modifications. In humans, KDM dioxygenases demethylate histones in high O2, but are inhibited under hypoxia; KDMs are also found in plants,
but their oxygen-sensitivity is yet to be established. In plants, stability of the VRN2 subunit of PRC2, a major histone methylating complex, is regulated via PCOs
similarly to ERFVIIs. Acronyms and protein names are defined in the main text and glossary. Hatched blue box highlights the conserved N-degron-based O2
sensing pathways in mammals and plants. Arabidopsis image from Charles Andrès (CC BY-SA 3.0).

R364 Current Biology 30, R362–R369, April 20, 2020

Current Biology

Review

Box 1. Glossary of terms.

that Nt-Cys oxidation required PCO enzyme activity, and genetic
removal of PCO function leads to ERFVII stabilisation and
ACBP acyl-CoA-binding domain-containing protein enhanced hypoxia tolerance. The identification of ADO indicated
ADO cysteamine (2-aminoethanethiol) dioxygenase that oxygen-sensing via this pathway is ancient, predating the
split between animal and plant groups (>1 Ga; Figure 1) [16],
ATE ARGINYL TRANSFERASE
and may indicate that a major mechanism of oxygen-sensing
bHLH basic helix-loop-helix DNA-binding domain
in early eukaryotes was through cysteine dioxygenase control
CKD chronic kidney disease of Nt-Cys oxidation during periods of earth history with compar-
EPAS endothelial PAS domain-containing protein atively low O2 levels. Alternatively, it may suggest that originally
EPO erythropoietin the major function of the pathway was NO sensing, which
ERFVII group VII ETHYLENE RESPONSE FACTOR became coupled to O2 as atmospheric levels rose. PCO-type
FIH factor inhibiting HIF1a Nt-cysteine dioxygenases have not been found in fungi, which
HIF hypoxia-inducible factor
diverged from animals after plants, indicating loss of the capacity
of this group to oxidise Nt-Cys and use this pathway for oxygen-
HRE hypoxia response element
sensing [16]. Genes encoding Met-Cys initiating RGS proteins
HRPE hypoxia responsive promoter element
are only present in vertebrate genomes, and those encoding
IKKb inhibitor of nuclear factor kappa-B kinase INTERLEUKIN-32 (IL-32) are present only in mammalian lineages
subunit beta
(Figure 1).
KDM JmjC (Jumonji C) domain lysine demethylase Although the PCO/ADO branch of the N-degron pathways is
NO nitric oxide ancient in eukaryotes, the HIF pathway is only present in meta-
Nt- amino terminus of the protein zoan animals (choanoflagellates, closest extant relatives to ani-
NTAD/CTAD Amino- and carboxy-terminal transactivation mals, do not contain bHLH-PAS domain proteins [37]) (Figure 1).
domains of HIF A functioning HIF system was identified in the placozoan Tricho-
OTUB1 ovarian tumor domain containing ubiquitin plax adhaerens, representing one of the simplest multicellular
aldehyde binding protein 1 animals [38]. A recent analysis of representatives of basal meta-
PAS Per-Arnt-Sim domain zoa groups porifera (sponges) and ctenophores failed to identify
PCO PLANT (Nt-)CYSTEINE OXIDASE pVHL or PHD-like proteins, or hypoxia-regulated gene expres-
sion [37]. One feature of the evolution of the HIF system appears
PHD(EGLN) proly 4-hydroxlases/Egl nine homolog
to be increased diversification of components in derived evolu-
PHI PHD inhibitor molecule
tionary groups. Whereas T. adhaerens contains single proteins
PRC2 Polycomb repressive complex 2 for each component of the pathway, mammals contain multiple
PRT6 PROTEOLYSIS6 E3 ligase N-recognin variants of HIFa and PHD [38]. The appearance and diversifica-
pVHL von Hippel-Lindau protein E3 ligase tion of a functional HIF pathway, which correlates with large in-
RGS REGULATOR OF G PROTEIN SIGNALLING creases in atmospheric and oceanic O2, may have influenced
SUB1A SUMERGENCE1A the concomitant explosion of animal diversity and size beginning
UBR ubiquitin protein ligase E3 component around the Cambrian period (540 Ma; Figure 1).
N-recognin
VRN2 VERNALIZATION2
Integration of Oxygen Sensing with Downstream
Signalling and Physiology
ZPR2 LITTLE ZIPPER2
Key observations related to major consequences of oxygen
sensing have been the identification of nuclear changes in
response to hypoxia. In both plants and animals, these converge
demethylation of the transcriptional co-activator PGC-1a, also on transcription of hypoxia-related genes and chromatin struc-
connects O2 availability to the regulation of genes linked to mito- ture. In plants, an evolutionarily-conserved core set of hypoxia-
chondrial biogenesis [35], suggesting others await discovery. related genes are activated by ERFVIIs in response to hypoxia-
induced stabilisation, through a conserved hypoxia-responsive
Evolutionary Origins of the Different Oxygen-Sensing promoter element (HRPE) [10]. Similarly, animal hypoxia
Systems response elements (HREs) are bound by HIF factors to enhance
Components of the Arg/N-degron pathway are conserved in eu- low O2 responsiveness [39]. Low O2 levels also influence chro-
karyotes, though distinct evolutionary trajectories are observed. matin structure through the stabilisation of components of chro-
Whereas ATE activity is highly conserved across all major matin-modifying complexes (VRN2 as part of PRC2 [21]), via
groups, E3 ligase functions for recognising distinct destabilising enhanced expression of chromatin modifiers (gene activation
residues (carried out by UBR-type proteins in non-plants) were by HIF [40]), or directly through repression of histone H3 deme-
split early in plant evolution [36] (Figure 1). ERFVIIs are not pre- thylation activity of KDMs [33,34]. In both animal and plant re-
sent in the genomes of the non-vascular land plants Physcomi- sponses, genes encoding biochemical pathways associated
trella patens or Marchantia polymorpha, and VRN2 and ZPR2 with enhanced tolerance of hypoxia are important targets
appeared with angiosperms [21,22]. As the nature of Nt-Cys (including fermentative metabolism, glycolysis and an inhibition
oxidation was for several years obscure, a major advance was of mitochondrial oxidative phosphorylation), and the control of
the identification of the PCOs in A. thaliana [11]. This showed pathways with oxygen-requiring reactions or that occur in

Current Biology 30, R362–R369, April 20, 2020 R365

 Current Biology

Review

hypoxic niches are also important [22,41,42]. The two animal is still unclear exactly where NO acts within the pathway.
cytoplasmic substrates of ADO have been identified Although an in vitro reconstituted mammalian system was
(RGS4,5,16 and IL-32 [16,43]) and give the possibility of more shown to be NO dependent [43], in vitro activity of PCO/ADO
rapid response to declining O2 than transcriptional circuits, since on peptides does not require NO [16,29]. It is possible, therefore,
their immediate stabilisation would trigger a change more that NO influences the activities of enzyme components of the
quickly than responses dependent on increased protein produc- pathway in vivo (ATE, PCO/ADO or UBR1/PRT6), and it was
tion through HIF control of gene expression. Both IL-32 and shown that PRT6 contains a NO-binding domain [53].
RGS4/5 are transcriptional targets of HIF, indicating a possible Factors other than hypoxia can influence oxygen-sensing
interaction between the two sensing systems [16]. Moving for- pathways. A sub-pool of ERFVIIs is stable and sequestered at
ward it will be important to decipher the comparative timescales the plasma membrane through association with ACYL CoA
through which PHD/FIH, KDM and ADO activity leads to cellular BINDING PROTEINs (ACBP) during normoxia [14,54]. Zinc
changes, as this likely contributes to physiologically relevant fine excess in the soil (detrimental to plant growth), inhibits PCO en-
tuning of the overall hypoxia response. zymes, thus causing stabilisation of ERFVIIs [55]. Non-canonical
Analyses of physiological functions reveal the broad reach of mechanisms also control HIF stability; for example, increases in
oxygen-sensing systems, and specific roles are related to the succinate during the progression of certain types of cancer can
different lifestyles of plants and animals. As plants are sessile, allosterically inhibit PHD activity to trigger HIF accumulation un-
a key function of oxygen sensing is related to perception of der normoxia [56,57]. The possible mechanisms influencing O2-
waterlogging and flooding [13,14]. Both stabilised ERFVIIs and responsive factors, and therefore the breadth of possible
VRN2 enhance survival of hypoxia [13,14,21]. It was recently affected physiological processes, will be much wider than those
shown that the plant Cys-initiating substrate ZPR2 is stabilised specifically related to O2 or NO.
by the hypoxic environment of the shoot apical meristem, regu-
lating the production of new leaves [22], and VRN2 also accumu- Pathologies and Interventions of Oxygen Sensing in
lates in hypoxic meristems, where it modulates flowering time Plants and Animals
and root development [44]. In addition, hypoxia-enhanced sta- Oxygen-sensing pathways represent key cellular targets for
bility of ERFVIIs was shown to repress chlorophyll synthesis counteracting diseases and enhancing stress resilience. HIF sig-
(an O2-requiring pathway) in dark-grown seedlings [41], as well nalling controls a range of cellular responses, and also drives
as regulating lateral root development [45]. tumorigenesis and the maintenance of tumour microenviron-
In animals, the HIF pathway plays major roles in O2 homeosta- ment in certain cancers [58]. Thus, interventions that impact
sis, including erythropoiesis (development of red blood cells) and the HIF pathway have the capacity to treat pathologies associ-
angiogenesis (development of new blood vessels) (reviewed in ated with these processes. EPO, a target of HIF, is downregu-
[46]). Similar to ZPR2/VRN2 in plant meristems, HIF1a is stabi- lated in patients with chronic kidney disease (CKD) due to
lised within hypoxic hematopoietic stem cells (that give rise to reduced O2 consumption [59]. Several PHD inhibitor molecules
blood cells) [42]. Stabilised HIF1/2a enhance expression of (PHIs) have been developed that stimulate increased EPO pro-
growth regulators (erythropoietin (EPO) and angiogenic growth duction in CKD patients to counteract renal anaemia [60], acting
factors) and associated components (for example, systems for as 2-OG mimetics or iron chelators to inhibit enzymatic activity
iron uptake and utilisation [46]). An important role of the HIF sys- and increase HIF stability in normoxia [59]. Chemicals that
tem is in adaptation of animals to high altitude, where the partial disrupt other aspects of HIF signalling have also been identified
pressure of O2 is reduced. Genome-wide association studies as potent repressors of cancer progression [61]. For example,
identified allele signatures in human populations associated cancers in patients with VHL disease result from ectopic accu-
with life at high altitudes in the Tibetan Plateau (average altitude mulation of HIF2a [58], and a novel drug that specifically disrupts
4000 m, pO2 13 kPa) for both HIF2a and PHD2. For example, in the HIF2a–HIF2b dimer to downregulate HIF2 signalling was
modern Tibetan populations a variant of EGLN1/PHD2 (Asp4Glu; recently shown to limit tumour progression [62]. The develop-
Cys127Ser) was shown to have a lower KmO2, suggesting that it ment of inhibitory molecules that target discrete HIF or PHD iso-
promotes increased degradation of HIF at high altitude (lower forms, as well as other regulatory points in the HIF signalling
pO2), thus reducing HIF levels to those equivalent to low altitudes pathway, will help to increase therapeutic specificity and efficacy
[47]. Interestingly, one allele of EPAS1/HIF2A enriched in Tibetan of such treatments.
populations appears to have been derived from ancient hominid Genetic manipulation of O2-signalling components in crop
Denisovans [48]. Many studies demonstrate wider roles for the species can increase resistance to waterlogging-induced hyp-
HIF system, indicating that oxygen-sensing by this pathway in- oxia, as shown in barley through genetic reduction in HvPRT6
fluences many aspects of cellular biochemistry, growth and expression/activity [63], while ERFVIIs provide increased toler-
development (discussed in [49]). ance to multiple abiotic stresses [52] and biotic stresses where
Since the PCO/ADO pathway also acts as an NO sensor pathogen-associated hypoxia is an integral factor [64–66]. In
[15,43], the stability of both animal and plant substrates also reg- rice (Oryza sativa), the ERFVII SUB1A-1 is a major regulator of
ulates responses to intracellular NO levels that accompany submergence tolerance that has been bred into high yielding
internal and external stress. For example, destruction of RGS varieties [67]. SUB1A-1 is naturally uncoupled from O2-depen-
proteins to induce cardiomyocyte proliferation can also be dent degradation despite containing Cys2 and downstream
induced by endothelium-derived NO [50]. Stabilisation of Lys residues [13,68], suggesting that the plant oxygen-sensing
ERFVIIs by reduced NO enhances hypoxia tolerance and toler- system has been targeted by natural selection for adaptation
ance to other abiotic stresses (including high salinity) [51,52]. It in wetland environments, and that biotechnological approaches

R366 Current Biology 30, R362–R369, April 20, 2020

Current Biology

Review

could be used to achieve similar outcomes in flooding-suscep- to enhance tolerance to hypoxia for both medical and agricul-
tible crops. tural interventions.

Conclusions and Unresolved Questions ACKNOWLEDGEMENTS

Where do we look for undiscovered oxygen sensors? Based on
structures and domains of already identified proteins, there are This work was supported by a Leverhulme Trust Research Project Grant (RPG-
clear candidates to test as novel components of oxygen-sensing 2017-132) and by the Biotechnology and Biological Sciences Research Coun-
pathways. Plant and animal genomes contain Jumonji C cil [grant numbers BB/M029441/1, BB/R002428/1] to M.J.H., and a European
Research Council Starter Grant [ERC-STG 715441-GasPlaNt] to D.J.G.
domain-containing KDMs in addition to those already shown to
act as oxygen-sensors. Determining those with a physiologically
REFERENCES
relevant (high) KmO2 would be a first step in defining potential
roles as sensors. Although plants do not contain HIFa-like se-
1. Fischer, W.W., Hemp, J., and Johnson, J.E. (2016). Evolution of oxygenic
quences, both plants and animals contain hundreds of proteins photosynthesis. Annu. Rev. Earth Planet. Sci. 44, 647–683.
initiating Met–Cys that could be substrates of PCO/ADO action,
2. Lane, N. (2016). Oxygen: The Molecule that Made the World, Revised 2016
in addition to endopeptidase substrates cleaved to reveal Nt- Edition (Oxford: Oxford University Press).
Cys. Cys2 is evolutionarily constrained in most eukaryote pro-
teomes [69], suggesting that this is an important determinant 3. van Dongen, J.T., and Licausi, F. (2015). Oxygen sensing and signaling.
Annu. Rev. Plant Biol. 66, 345–367.
related to O2/NO sensing. In addition, recently it was hypothes-
ised that mechanisms other than PCO-regulated destabilisation 4. Kaelin, W.G., and Ratcliffe, P.J. (2008). Oxygen sensing by metazoans: the
may act to promote oxygen-sensing in plants, in several cases central role of the HIF hydroxylase pathway. Mol. Cell 30, 393–402.

backed-up by experimental data [70]. 5. Ivan, M., Kondo, K., Yang, H., Kim, W., Valiando, J., Ohh, M., Salic, A.,
Why is N-degron-mediated oxygen sensing not the primary Asara, J.M., Lane, W.S., and Kaelin, W.G., Jr. (2001). HIFalpha targeted
for VHL-mediated destruction by proline hydroxylation: implications for
system in metazoans as it is in angiosperms? The HIF system O2 sensing. Science 292, 464–468.
evolved only in one lineage of animals, whereas the PCO/ADO
pathway evolved early in eukaryotes (Figure 1). Perhaps the un- 6. Jaakkola, P., Mole, D.R., Tian, Y.M., Wilson, M.I., Gielbert, J., Gaskell,
S.J., von Kriegsheim, A., Hebestreit, H.F., Mukherji, M., Schofield, C.J.,
avoidable link of the PCO/ADO pathway to a requirement for NO et al. (2001). Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation
made this pathway unsuitable, or possibly it was not suitable for complex by O2-regulated prolyl hydroxylation. Science 292, 468–472.
large mobile organisms. Lack of transcriptional response to hyp-
7. Iliopoulos, O., Levy, A.P., Jiang, C., Kaelin, W.G., Jr., and Goldberg, M.A.
oxia in the marine sponge Tethya wilhelma indicates that the (1996). Negative regulation of hypoxia-inducible genes by the von Hippel-
ADO branch of the Arg/N-degron pathway does not perform Lindau protein. Proc. Natl. Acad. Sci. USA 93, 10595–10599.
this function in basal animals, though complete anoxia did result 8. Maxwell, P.H., Wiesener, M.S., Chang, G.W., Clifford, S.C., Vaux, E.C.,
in large changes in gene expression [37]. It is unclear what Cockman, M.E., Wykoff, C.C., Pugh, C.W., Maher, E.R., and Ratcliffe,
advantage the coupling of NO- and oxygen-sensing in this P.J. (1999). The tumour suppressor protein VHL targets hypoxia-inducible
factors for oxygen-dependent proteolysis. Nature 399, 271–275.
pathway has; it may be a remnant of evolutionary drivers early
in eukaryote history, where O2 levels were low, which might 9. Lando, D., Peet, D.J., Whelan, D.A., Gorman, J.J., and Whitelaw, M.L.
also suggest early Nt-cysteine dioxygensases had high affinities (2002). Asparagine hydroxylation of the HIF transactivation domain a hyp-
oxic switch. Science 295, 858–861.
for O2, making the pathway primarily important for responding to
changes in levels of intracellular NO. 10. Gasch, P., Fundinger, M., Muller, J.T., Lee, T., Bailey-Serres, J., and Mus-
troph, A. (2016). Redundant ERF-VII transcription factors bind to an
There are several striking commonalities in the major oxygen- evolutionarily conserved cis-motif to regulate hypoxia-responsive gene
sensing systems of angiosperms and metazoans. Both require expression in Arabidopsis. Plant Cell 28, 160–180.
dioxygenases with O2 sensitivity within a physiological range,
11. Weits, D.A., Giuntoli, B., Kosmacz, M., Parlanti, S., Hubberten, H.M., Rie-
both directly target nuclear factors for UPS-mediated destruc- gler, H., Hoefgen, R., Perata, P., van Dongen, J.T., and Licausi, F. (2014).
tion, and both result in large changes in gene expression Plant cysteine oxidases control the oxygen-dependent branch of the
with downstream physiological consequences providing ho- N-end-rule pathway. Nat. Commun. 5, 3425.

meostatic control of O2 response. An important goal of future 12. White, M.D., Klecker, M., Hopkinson, R.J., Weits, D.A., Mueller, C., Nau-
research will be to define the links between O2 affinity of mann, C., O’Neill, R., Wickens, J., Yang, J.Y., Brooks-Bartlett, J.C.,
et al. (2017). Plant cysteine oxidases are dioxygenases that directly enable
pathway dioxygenases and their expression patterns, allowing arginyl transferase-catalysed arginylation of N-end rule targets. Nat. Com-
an understanding of how these enzymes sense all physiologi- mun. 8, 14690.
cally possible internal O2 tensions. The complete gamut of
13. Gibbs, D.J., Lee, S.C., Isa, N.M., Gramuglia, S., Fukao, T., Bassel, G.W.,
influenced processes and interactions is yet to be resolved — Sousa Correia, C., Corbineau, F., Theodoulou, F.L., Bailey-Serres, J.,
at the intracellular level there are clearly similarities of interac- et al. (2011). Homeostatic response to hypoxia is regulated by the N-end
tions between oxygen-sensing pathways and mitochondrial rule pathway in plants. Nature 479, 415–418.
function (key for oxidative phosphorylation), well understood 14. Licausi, F., Kosmacz, M., Weits, D.A., Giuntoli, B., Giorgi, F.M., Voesenek,
for the HIF system, but requiring more understanding for the L.A.C.J., Perata, P., and van Dongen, J.T. (2011). Oxygen sensing in plants
is mediated by an N-end rule pathway for protein destabilization. Nature
PCO/ADO pathway in animals and plants. It is likely that 479, 419–422.
many components of known oxygen-sensing pathways remain
to be discovered, including dioxygenases with novel activities, 15. Gibbs, D.J., Isa, N.M., Movahedi, M., Lozano-Juste, J., Mendiondo, G.M.,
Berckhan, S., Marin-de la Rosa, N., Conde, J.V., Correia, C.S., Pearce,
and PHD/ADO/PCO targets. An important goal of future S.P., et al. (2014). Nitric oxide sensing in plants is mediated by proteolytic
research will be to investigate the use of these components control of group VII ERF transcription factors. Mol. Cell 53, 369–379.

Current Biology 30, R362–R369, April 20, 2020 R367

 Current Biology

Review
16. Masson, N., Keeley, T.P., Giuntoli, B., White, M.D., Puerta, M.L., Perata, 33. Chakraborty, A.A., Laukka, T., Myllykoski, M., Ringel, A.E., Booker, M.A.,
P., Hopkinson, R.J., Flashman, E., Licausi, F., and Ratcliffe, P.J. (2019). Tolstorukov, M.Y., Meng, Y.J., Meier, S.R., Jennings, R.B., Creech, A.L.,
Conserved N-terminal cysteine dioxygenases transduce responses to et al. (2019). Histone demethylase KDM6A directly senses oxygen to con-
hypoxia in animals and plants. Science 365, 65–69. trol chromatin and cell fate. Science 363, 1217–1222.

17. Gibbs, D.J., Conde, J.V., Berckhan, S., Prasad, G., Mendiondo, G.M., and 34. Batie, M., Frost, J., Frost, M., Wilson, J.W., Schofield, P., and Rocha, S.
Holdsworth, M.J. (2015). Group VII ethylene response factors coordinate (2019). Hypoxia induces rapid changes to histone methylation and repro-
oxygen and nitric oxide signal transduction and stress responses in plants. grams chromatin. Science 363, 1222–1226.
Plant Physiol. 169, 23–31.
35. Qian, X., Li, X., Shi, Z., Bai, X., Xia, Y., Zheng, Y., Xu, D., Chen, F., You, Y.,
18. Isaacs, J.S., Jung, Y.J., Mimnaugh, E.G., Martinez, A., Cuttitta, F., and Fang, J., et al. (2019). KDM3A senses oxygen availability to regulate PGC-
Neckers, L.M. (2002). Hsp90 regulates a von Hippel Lindau-independent 1alpha-mediated mitochondrial biogenesis. Mol Cell 76, 885–895 e7.
hypoxia-inducible factor-1 alpha-degradative pathway. J. Biol. Chem.
277, 29936–29944. 36. Till, C.J., Vicente, J., Zhang, H., Oszvald, M., Deery, M.J., Pastor, V., Lilley,
K.S., Ray, R.V., Theodoulou, F.L., and Holdsworth, M.J. (2019). The Arabi-
19. Kong, X., Lin, Z., Liang, D., Fath, D., Sang, N., and Caro, J. (2006). Histone dopsis thaliana N-recognin E3 ligase PROTEOLYSIS1 influences the im-
deacetylase inhibitors induce VHL and ubiquitin-independent proteaso- mune response. Plant Direct 3, e00194.
mal degradation of hypoxia-inducible factor 1alpha. Mol. Cell. Biol. 26,
2019–2028.
37. Mills, D.B., Francis, W.R., Vargas, S., Larsen, M., Elemans, C.P.H., Can-
field, D.E., and Worheide, G. (2018). The last common ancestor of animals
20. Papdi, C., Abraham, E., Joseph, M.P., Popescu, C., Koncz, C., and Szaba-
lacked the HIF pathway and respired in low-oxygen environments. Elife 7,
dos, L. (2008). Functional identification of Arabidopsis stress regulatory
e31176.
genes using the controlled cDNA overexpression system. Plant Physiol.
147, 528–542.
38. Loenarz, C., Coleman, M.L., Boleininger, A., Schierwater, B., Holland,
21. Gibbs, D.J., Tedds, H.M., Labandera, A.M., Bailey, M., White, M.D., Hart- P.H., Ratcliffe, P.J., and Schofield, C.J. (2011). The hypoxia-inducible
man, S., Sprigg, C., Mogg, S.L., Osborne, R., Dambire, C., et al. (2018). transcription factor pathway regulates oxygen sensing in the simplest an-
Oxygen-dependent proteolysis regulates the stability of angiosperm poly- imal, Trichoplax adhaerens. EMBO Rep. 12, 63–70.
comb repressive complex 2 subunit VERNALIZATION 2. Nat. Commun. 9,
5438. 39. Mole, D.R., Blancher, C., Copley, R.R., Pollard, P.J., Gleadle, J.M., Ra-
goussis, J., and Ratcliffe, P.J. (2009). Genome-wide association of
22. Weits, D.A., Kunkowska, A.B., Kamps, N.C.W., Portz, K.M.S., Packbier, hypoxia-inducible factor (HIF)-1 alpha and HIF-2 alpha DNA binding with
N.K., Venza, Z.N., Gaillochet, C., Lohmann, J.U., Pedersen, O., van Don- expression profiling of hypoxia-inducible transcripts. J. Biol. Chem. 284,
gen, J.T., et al. (2019). An apical hypoxic niche sets the pace of shoot mer- 16767–16775.
istem activity. Nature 569, 714–717.
40. Xia, X.B., Lemieux, M.E., Li, W., Carroll, J.S., Brown, M., Liu, X.S., and
23. Strowitzki, M.J., Cummins, E.P., and Taylor, C.T. (2019). Protein hydroxyl- Kung, A.L. (2009). Integrative analysis of HIF binding and transactivation
ation by hypoxia-inducible factor (HIF) hydroxylases: unique or ubiqui- reveals its role in maintaining histone methylation homeostasis. Proc.
tous? Cells 8, E384. Natl. Acad. Sci. USA 106, 4260–4265.

24. White, M.D., and Flashman, E. (2016). Catalytic strategies of the non-heme 41. Abbas, M., Berckhan, S., Rooney, D.J., Gibbs, D.J., Conde, J.V., Correia,
iron dependent oxygenases and their roles in plant biology. Curr. Opin. C.S., Bassel, G.W., Marin-de la Rosa, N., Leon, J., Alabadi, D., et al. (2015).
Chem. Biol. 31, 126–135. Oxygen sensing coordinates photomorphogenesis to facilitate seedling
survival. Curr. Biol. 25, 1483–1488.
25. Ehrismann, D., Flashman, E., Genn, D.N., Mathioudakis, N., Hewitson,
K.S., Ratcliffe, P.J., and Schofield, C.J. (2007). Studies on the activity of 42. Takubo, K., Goda, N., Yamada, W., Iriuchishima, H., Ikeda, E., Kubota, Y.,
the hypoxia-inducible-factor hydroxylases using an oxygen consumption Shima, H., Johnson, R.S., Hirao, A., Suematsu, M., et al. (2010). Regulation
assay. Biochem. J. 401, 227–234. of the HIF-1alpha level is essential for hematopoietic stem cells. Cell Stem
Cell 7, 391–402.
26. Koivunen, P., Hirsila, M., Kivirikko, K.I., and Myllyharju, J. (2006). The
length of peptide substrates has a marked effect on hydroxylation by 43. Hu, R.G., Sheng, J., Qi, X., Xu, Z.M., Takahashi, T.T., and Varshavsky, A.
the hypoxia-inducible factor prolyl 4-hydroxylases. J. Biol. Chem. 281, (2005). The N-end rule pathway as a nitric oxide sensor controlling the
28712–28720. levels of multiple regulators. Nature 437, 981–986.
27. Koivunen, P., Hirsila, M., Gunzler, V., Kivirikko, K.I., and Myllyharju, J. 44. Labandera, A.-M., Tedds, H.M., Bailey, M., Sprigg, C., Etherington, R.D.,
(2004). Catalytic properties of the asparaginyl hydroxylase (FIH) in the ox- Akintewe, O., Kalleechurn, G., Holdsworth, M.J., and Gibbs, D.J. (2020).
ygen sensing pathway are distinct from those of its prolyl 4-hydroxylases. The PRT6 N-degron pathway restricts VERNALIZATION 2 to endogenous
J. Biol. Chem. 279, 9899–9904. hypoxic niches to modulate plant development. New Phytol. https://doi.
org/10.1111/nph.16477.
28. Yeh, T.L., Leissing, T.M., Abboud, M.I., Thinnes, C.C., Atasoylu, O., Holt-
Martyn, J.P., Zhang, D., Tumber, A., Lippl, K., Lohans, C.T., et al. (2017).
45. Shukla, V., Lombardi, L., Lacopino, S., Pencik, A., Novak, O., Perata, P.,
Molecular and cellular mechanisms of HIF prolyl hydroxylase inhibitors
Giuntoli, B., and Licausi, F. (2019). Endogenous hypoxia in lateral root
in clinical trials. Chem. Sci. 8, 7651–7668.
primordia controls root architecture by antagonizing auxin signaling in Ara-
bidopsis. Mol. Plant 12, 538–551.
29. White, M.D., Kamps, J., East, S., Kearney, L.J.T., and Flashman, E. (2018).
The plant cysteine oxidases from Arabidopsis thaliana are kinetically
tailored to act as oxygen sensors. J. Biol. Chem. 293, 11786–11795. 46. Samanta, D., Prabhakar, N.R., and Semenza, G.L. (2017). Systems biology
of oxygen homeostasis. Wiley Interdiscip. Rev. Syst. Biol. Med. 9, e1382.
30. Metzen, E., Berchner-Pfannschmidt, U., Stengel, P., Marxsen, J.H.,
Stolze, I., Klinger, M., Huang, W.Q., Wotzlaw, C., Hellwig-Burgel, T., Jelk- 47. Lorenzo, F.R., Huff, C., Myllymaki, M., Olenchock, B., Swierczek, S., Ta-
mann, W., et al. (2003). Intracellular localisation of human HIF-1 alpha hy- shi, T., Gordeuk, V., Wuren, T., Ri-Li, G., McClain, D.A., et al. (2014). A ge-
droxylases: implications for oxygen sensing. J. Cell Sci. 116, 1319–1326. netic mechanism for Tibetan high-altitude adaptation. Nat. Genet. 46,
951–956.
31. Considine, M.J., Diaz-Vivancos, P., Kerchev, P., Signorelli, S., Agudelo-
Romero, P., Gibbs, D.J., and Foyer, C.H. (2017). Learning to breathe: 48. Huerta-Sánchez, E., Jin, X., Asan, Bianba, Z., Peter, B.M., Vinckenbosch,
developmental phase transitions in oxygen status. Trends Plant Sci. 22, N., Liang, Y., Yi, X., He, M.Z., Somel, M., et al. (2014). Altitude adaptation in
140–153. Tibetans caused by introgression of Denisovan-like DNA. Nature 512,
194–197.
32. McDonough, M.A., Loenarz, C., Chowdhury, R., Clifton, I.J., and Scho-
field, C.J. (2010). Structural studies on human 2-oxoglutarate dependent 49. Pugh, C.W., and Ratcliffe, P.J. (2017). New horizons in hypoxia signaling
oxygenases. Curr. Opin. Struct. Biol. 20, 659–672. pathways. Exp. Cell Res. 356, 116–121.

R368 Current Biology 30, R362–R369, April 20, 2020

Current Biology

Review
50. Jaba, I.M., Zhuang, Z.W., Li, N., Jiang, Y., Martin, K.A., Sinusas, A.J., Pa- 60. Myllyharju, J. (2013). Prolyl 4-hydroxylases, master regulators of the hyp-
pademetris, X., Simons, M., Sessa, W.C., Young, L.H., et al. (2013). NO oxia response. Acta Physiol. (Oxf.) 208, 148–165.
triggers RGS4 degradation to coordinate angiogenesis and cardiomyo-
cyte growth. J. Clin. Invest. 123, 1718–1731. 61. Fallah, J., and Rini, B.I. (2019). HIF inhibitors: status of current clinical
development. Curr. Oncol. Rep. 21, 6.
51. Hartman, S., Liu, Z., van Veen, H., Vicente, J., Reinen, E., Martopawiro, S.,
Zhang, H., van Dongen, N., Bosman, F., Bassel, G.W., et al. (2019). 62. Chen, W., Hill, H., Christie, A., Kim, M.S., Holloman, E., Pavia-Jimenez, A.,
Ethylene-mediated nitric oxide depletion pre-adapts plants to hypoxia Homayoun, F., Ma, Y., Patel, N., Yell, P., et al. (2016). Targeting renal cell
stress. Nat. Commun. 10, 4020. carcinoma with a HIF-2 antagonist. Nature 539, 112–117.

63. Mendiondo, G.M., Gibbs, D.J., Szurman-Zubrzycka, M., Korn, A., Mar-
52. Vicente, J., Mendiondo, G.M., Movahedi, M., Peirats-Llobet, M., Juan,
quez, J., Szarejko, I., Maluszynski, M., King, J., Axcell, B., Smart, K.,
Y.T., Shen, Y.Y., Dambire, C., Smart, K., Rodriguez, P.L., Charng, Y.Y.,
et al. (2016). Enhanced waterlogging tolerance in barley by manipulation
et al. (2017). The Cys-Arg/N-end rule pathway is a general sensor of abiotic
of expression of the N-end rule pathway E3 ligase PROTEOLYSIS6. Plant
stress in flowering plants. Curr. Biol. 27, 3183–3190.
Biotechnol. J. 14, 40–50.
53. Zarban, R., Vogler, M., Wong, A., Eppinger, J., Al-Babili, S., and Gehring, 64. Valeri, M.C., Novi, G., Weits, D.A., Mensuali, A., Perata, P., and Loreti, E.
C. (2019). Discovery of a nitric oxide-responsive protein in Arabidopsis (2020). Botrytis cinerea induces local hypoxia in Arabidopsis leaves. New
thaliana. Molecules 24, 2691. Phytol. https://doi.org/10.1111/nph.16513.

54. Schmidt, R.R., Fulda, M., Paul, M.V., Anders, M., Plum, F., Weits, D.A., , S., Jubault, M., Lariagon, C., Le-
65. Gravot, A., Richard, G., Lime, T., Lemarie
Kosmacz, M., Larson, T.R., Graham, I.A., Beemster, G.T.S., et al. (2018). moine, J., Vicente, J., Robert-Seilaniantz, A., Holdsworth, M.J., et al.
Low-oxygen response is triggered by an ATP-dependent shift in oleoyl- (2016). Hypoxia response in Arabidopsis roots infected by Plasmodio-
CoA in Arabidopsis. Proc. Natl. Acad. Sci. USA 115, E12101–E12110. phora brassicae supports the development of clubroot. BMC Plant Biol.
16, 251.
55. Dalle Carbonare, L., White, M., Shukla, V., Francini, A., Perata, P., Flash-
man, E., Sebastiani, L., and Licausi, F. (2019). Zinc excess induces a hyp- 66. Kerpen, L., Niccolini, L., Licausi, F., van Dongen, J.T., and Weits, D.A.
oxia-like response by inhibiting cysteine oxidases in poplar roots. Plant (2019). Hypoxic conditions in crown galls induce plant anaerobic re-
Physiol. 180, 1614–1628. sponses that support tumor proliferation. Front. Plant Sci. 10, 56.

56. Iommarini, L., Porcelli, A.M., Gasparre, G., and Kurelac, I. (2017). Non-ca- 67. Xu, K., Xu, X., Fukao, T., Canlas, P., Maghirang-Rodriguez, R., Heuer, S.,
nonical mechanisms regulating hypoxia-inducible factor 1 alpha in cancer. Ismail, A.M., Bailey-Serres, J., Ronald, P.C., and Mackill, D.J. (2006).
Front. Oncol. 7, 286. Sub1A is an ethylene-response-factor-like gene that confers submer-
gence tolerance to rice. Nature 442, 705–708.
57. Selak, M.A., Armour, S.M., MacKenzie, E.D., Boulahbel, H., Watson, D.G., 68. Lin, C.C., Chao, Y.T., Chen, W.C., Ho, H.Y., Chou, M.Y., Li, Y.R., Wu, Y.L.,
Mansfield, K.D., Pan, Y., Simon, M.C., Thompson, C.B., and Gottlieb, E. Yang, H.A., Hsieh, H., Lin, C.S., et al. (2019). Regulatory cascade involving
(2005). Succinate links TCA cycle dysfunction to oncogenesis by inhibiting transcriptional and N-end rule pathways in rice under submergence. Proc.
HIF-alpha prolyl hydroxylase. Cancer Cell 7, 77–85. Natl. Acad. Sci. USA 116, 3300–3309.
58. Huang, Y., Lin, D., and Taniguchi, C.M. (2017). Hypoxia inducible factor 69. Gibbs, D.J., Bacardit, J., Bachmair, A., and Holdsworth, M.J. (2014). The
(HIF) in the tumor microenvironment: friend or foe? Sci. China Life Sci. eukaryotic N-end rule pathway: conserved mechanisms and diverse func-
60, 1114–1124. tions. Trends Cell Biol. 24, 603–611.

59. Schodel, J., and Ratcliffe, P.J. (2019). Mechanisms of hypoxia signalling: 70. Holdsworth, M.J. (2017). First hints of new sensors. Nat. Plants 3,
new implications for nephrology. Nat. Rev. Nephrol. 15, 641–659. 767–768.

Current Biology 30, R362–R369, April 20, 2020 R369