Download 60Ma of legume nodulation. What`s new? What`s

Journal of Experimental Botany, Vol. 59, No. 5, pp. 1081–1084, 2008
doi:10.1093/jxb/erm286 Advance Access publication 21 January, 2008
SPECIAL ISSUE REVIEW PAPER
60Ma of legume nodulation. What’s new? What’s changing?
Janet I. Sprent*
Division of Applied and Environmental Biology, University of Dundee, Dundee DD1 4HN, Scotland, UK
Received 5 June 2007; Revised 22 October 2007; Accepted 23 October 2007
Abstract
Current evidence suggests that legumes evolved
about 60 million years ago. Genetic material for nodulation was recruited from existing DNA, often following
gene duplication. The initial process of infection probably did not involve either root hairs or infection
threads. From this initial event, two branched pathways
of nodule developmental processes evolved, one involving and one not involving the development of
infection threads to ‘escort’ bacteria to young nodule
cells. Extant legumes have a wide range of nodule
structures and at least 25% of them do not have infection threads. The latter have uniform infected tissue
whereas those that have infection threads have infected
cells interspersed with uninfected (interstitial) cells.
Each type of nodule may develop indeterminately, with
an apical meristem, or show determinate growth. These
nodule structures are host determined and are largely
congruent with taxonomic position. In addition to
variation on the plant side, the last 10 years have seen
the recognition of many new types of ‘rhizobia’,
bacteria that can induce nodulation and fix nitrogen. It
is not yet possible to fit these into the emerging pattern
of nodule evolution.
Key words: Burkholderia, Cupriavidus, legume, nitrogen
fixation, nodule, rhizobia.
Introduction
Current plant phylogeny places the family Leguminosae
(Fabaceae) in the Eurosid I clade, which also includes all
other higher plants known to form nitrogen-fixing root
nodules (Soltis et al., 2000) as well as many that do not
form these symbioses. Where this predisposition to
nodulate comes from or why it is not always expressed is
not clear nor is it clear why the actinorhizal plants (those
nodulating with the filamentous bacterial genus Frankia)
have nodules with a root-like structure of central vascular
tissue and peripheral infected tissue and legume nodules
have a central infected tissue and a peripheral vascular
system. Pawlowski and Sprent (2007) have reviewed the
differences between legume and actinorhizal nodules and
this topic will not be covered here. Sprent (2007a, b) has
discussed the evolution of nodulation in legumes from
molecular and taxonomic aspects and this paper summarizes and extends these reviews. All generic and specific
names of legumes are those accepted by the International
Legume Database and Information Service (ILDIS).
Life before legumes
Current evidence suggests that legumes evolved about 60
million years ago (Lavin et al., 2005). What could the
older plant groups provide that legumes could capitalize
on to produce nodules? One of the first prerequisites was
the ability of the two partners in the symbiosis to
recognize each other. This is generally agreed to have
developed from the ancient symbiosis between fungi and
land plants, arbuscular mycorrhizas (AM), see, for
example, Szczyglowski and Amyot (2003). In extant
legumes, many of the signals are common between these
symbioses. Another process that may have been ‘hijacked’ is that leading to pollen tube growth. This has
much in common with the growth of infection threads
down root hairs of both legumes and actinorhizal plants.
Recent evidence suggests that gene duplication may have
preceded this modification in function (Rodriguez-Llorente et al., 2004). What is not known is whether or not
those legumes that do not have a hair infection and do not
form infection threads have these duplicated genes.
Another set of genes that has been replicated is that
leading to the formation of haemoglobin. Active legume
nodules are often characterized by their pink colour, due
to the oxygen-carrying pigment haemoglobin. However,
haemoglobin genes are very ancient and plant and animal
* E-mail: [email protected]
ª The Author [2008]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.
For Permissions, please e-mail: [email protected]
1082 Sprent
forms separated long before vascular plants evolved and
extant plants may have one or both of two families of
haemoglobin genes, often referred to as symbiotic and
non-symbiotic forms. Legumes and actinorhizal plants
may have one or more genes from each of these types,
with variations among legumes from different taxonomic
groups. Different forms of haemoglobin may be produced
sequentially and be coupled to the induction of genes
necessary for nitrogen fixation (Downie, 2005; Ott et al.,
2005). There are other examples of gene duplication, for
example, apyrases (Cannon et al., 2003) that will not be
discussed here.
Regulation of differentiation in growing plants appears
to have been highly conserved, possibly dating back over
400Ma (Floyd and Bowman, 2004). Modern molecular
techniques, coupled with more sensitive analytical methods such as GC:MS have led to a resurgence in the study
of the role of plant hormones in nodule processes and
their similarities with, for example, root growth and
nematode gall formation (Mathesius, 2003). For a general
review of this area, see Manoury et al. (2007).
Thus the stage seems to have been well set for the
evolution of nodules, but important questions remain.
Why is nodulation restricted to the Eurosid I clade? What
was the driving force for the ability to form nitrogenfixing symbioses? One suggestion is that it was associated
with a sudden increase in atmospheric carbon dioxide
around the time that legumes evolved (Sprent, 2007b).
How did legumes evolve different nodulation processes?
The next section will consider this question.
How did legumes evolve different nodulation
processes?
Rhizobia and other bacteria can penetrate roots of a variety
of plants. For example, Spencer et al. (1994) found that
rhizobia could enter between cells of potato tissues. Bryan
et al. (1996) suggested that this kind of entry, observed
by them in the non-nodulating caesalpinioid legume
Gleditsia triacanthos was the first stage in the evolution
of nodulation. In their case, invading bacteria appeared
later to be confined within infection threadlike structures,
but these did not penetrate cells. It is important to
distinguish such structures from the transcellular infection
threads that are found in root hair infection and subsequent
nodule development (Brewin, 2004). However, it is
possible to have transcellular infection threads without
having a root hair infection, as seen, for example, in the
mimosoid legume Neptunia plena (James et al., 1992;
Sprent and James, 2007), which is normally infected via
wounds where adventitious roots emerge and in Mimosa
scabrella, where infection may occur between epidermal
cells (Faria et al., 1988). Thus the most likely origin of
infection of legumes by rhizobia is directly through the
epidermis or through breaks, for example, where lateral
roots emerge (often referred to as crack infection). Subsequently, there were two distinct lines of evolution. In
75% of legumes, infection threads are found and these were
a necessary prerequisite for root hair infection. All nodulating mimosoid and caesalpinioid legumes so far examined
and over 50% of papilionoid legumes have transcellular
infection threads in nodules. These infect some, but not all
of the cells in the developing infected region; these infected
cells enlarge greatly, giving the familiar nodule structure as
found in crop species such as peas (Pisum sativum) and
soybean (Glycine max).These two legumes have nodules
with indeterminate and determinate growth, respectively
(Sprent, 2001). Although legumes having crack infection
have been known for many years, most notably in Arachis
hypogaea, the groundnut or peanut, none has been
subjected to the intense molecular study currently centred
on the two model legumes, Lotus japonicus and Medicago
truncatula. This is a pity as they typify a significant section
of papilionoid legumes, the dalbergioid group (Lavin et al.,
2001) and include important tropical forage genera such as
Stylosanthes and prized timber trees such as some species
of Dalbergia and Pterocarpus. Perhaps it is because they
are largely tropical and have not been funded for largescale agricultural production, that they are so poorly
studied. However, it is now becoming clear that many
temperate legumes also lack a hair infection pathway,
bacteria entering between epidermal cells, sometimes after
damage. The most widely studied of these is Lupinus
which is an important grain legume in some areas and
whose nodules frequently encircle the subtending root,
forming a collar-like structure. Collar nodules are also
found in in Lotononis a genus that has good potential as
a forage legume in parts of Australia and South Africa
(Yates et al., 2007). In both crack and epidermal infections,
a few cells in the developing nodule are infected and these
divide repeatedly giving infected tissue lacking uninfected
cells. As with nodules formed after hair infection there are
two further branches of development, determinate (following crack infection, as in Dalbergioid legumes) and
indeterminate (following epidermal infection). If one can
infer infection process from nodule structure, then all those
in which the central tissue is uniformly infected lack a hair
infection process. This would include common temperate
genera in tribe Genisteae, such as Genista, Cytisus, and
Ulex and possibly the related tribe Crotalarieae (Sprent,
2007b). Further, since most of the detailed studies on
recognition processes between host and rhizobia have been
carried out with hair infections, it is pertinent to ask
whether these operate when the infection processes appear
less complex. Evidence now emerging suggests that some
nod genes may not be essential for crack entry (Giraud
et al., 2007).
Endoreduplication in infected cells of nodules has been
a subject of discussion for many years, but recent
evidence suggests that it may be common, at least
60Ma of legume nodulation 1083
in advanced papilionoid legumes. It has been demonstrated in determinate nodules of Lotus japonicus, indeterminate nodules of Medicago truncatula and
M. sativa, and in nodules of Lupinus albus (reviewed in
Maunoury et al., 2007).
Where do the rhizobia fit into all of this?
In the early years of research into legume–rhizobial
interactions, two kinds of nodulating bacteria were
known, slow and fast growing. The former were thought
to be rather promiscuous with the latter showing various
levels of host plant/rhizobial specificity (Nutman, 1987).
These ideas were largely formulated in studies driven
by the need for nitrogen-fixing legumes in countries
such as Australia and the USA whose economies
depended on them. Over the last 30 years, the situation
has changed dramatically. First, the number of genera and
species of ‘rhizobia’ increased greatly and then, in the last
ten years, bacteria outside the family Rhizobiaceae have
been found to nodulate legumes. Other bacteria have been
isolated from legume nodules, but unless they have been
shown to nodulate the host of isolation (or at least some
legumes) their ability to nodulate must be regarded with
suspicion. For example, Chou et al. (2007) isolated
Labrys neptuniae from nodules of the aquatic mimosoid
legume Neptunia oleracea, but were unable to re-infect
the host plant with it.
Current bacterial taxonomy divides a group of gram
negative bacteria, the Proteobacteria into a number of
branches, a, b, c, d, and e (Brenner et al., 2005). The
a-branch has seven orders, one of which is the Rhizobiales,
which in turn has 11 families. Four of these, Bradyrhizobiaceae, Methylobacteriaceae, Phyllobacteriaceae, and
Rhizobiaceae contain genera that can nodulate and fix
nitrogen in association with legumes. The first and last
families correspond to the old categories of slow and fast
growing rhizobia, respectively. In the Methylobacteriaceae,
the genus Methylobacterium is of particular interest. The
species M. nodulans was isolated from some species of
Crotalaria (Sy et al., 2001), but before it was fully
identified was found only to nodulate the species from
which it was isolated, but not some other species of the
genus (Samba et al. 1999). This clear infrageneric
separation of host species has now been echoed by the
report of another species of Methylobacterium, not yet fully
described, which nodulates only some species of the
closely related genus Lotononis (Yates et al., 2007). Such
a clear-cut separation is rare in other nodulating
a-proteobacteria. The b-proteobacterial branch also contains seven orders. One of these, the Burkholderiales is
divided into five families and one of these, the Burkholderiaceae contains two genera, Burkholderia and Cupriavidus
having species known to nodulate legumes. Both genera
have been segregated from the old genus Pseudomonas,
long known to be polyphyletic. Cupriavidus has been
separated from Ralstonia, with a short time spent as
Wautersia and the only species known to nodulate legumes
is Cupriavidus taiwanensis (Chen et al., 2003).
Several species of Burkholderia have been shown to
nodulate legumes, with varying levels of host specificity
(Sprent and James, 2007). Some are capable of fixing
nitrogen ex planta (Elliott et al., 2007a). Although all
the early reports are from mimosoid legumes, there is
now evidence that papilionoid legumes such as Cyclopia
(Elliott et al., 2007b) and Dalbergia (Rasolomampianina
et al., 2005) may also nodulate with Burkholderia spp. It
remains to be seen how common these b-rhizobia are
and whether they are especially important in certain
environments.
What’s next?
Major advances in our understanding of the relations
between legumes and rhizobia are being made using
model legumes, but as Doyle and Luckow (2003) pointed
out, these are only the tip of the iceberg. In the real world,
the legume/rhizobium symbiosis varies greatly, from
being essentially parasitic to highly effective, imposing
various levels of constraints on the partners (Sprent,
2003). These are poorly understood. For many years it
has been the goal of some to have nitrogen-fixing cereals
and other non-legume crops. Others believe that a better
starting point may be to understand why some legumes
cannot nodulate. With modern molecular techniques this
is a realistic goal. During the course of legume evolution,
some species have acquired all the necessary information
to make nodule structures in the absence of rhizobia
(summarized in Sprent and James, 2007). However, such
structures have only been reported for a few papilionoid
species with infection threads and interstitial cells in their
infected tissue, arguably the more complex branch of
nodule evolution.
References
Brenner DJ, Kreig NR, Staley JT, Garrity GM, eds. 2005.
Bergey’s manual of systematic bacteriology, 2nd edn., Vol. 2.
The Proteobacteria, Part C. Springer: New York.
Brewin NJ. 2004. Plant cell wall remodelling in the Rhizobium–
legume symbiosis. CRC Critical Reviews in Plant Science 23,
293–326.
Bryan JA, Berlyn GP, Gordon JC. 1996. Towards a new concept
of the evolution of symbiotic nitrogen fixation in the Leguminosae.
Plant and Soil 186, 151–159.
Cannon SB, McCombie WR, Sato S, et al. 2003. Evolution and
microsynteny of the apyrase gene family in three legume
genomes. Molecular Genetics and Genomics 270, 347–361.
Chen W-M, James EK, Prescott AR, Kierans M, Sprent JI.
2003. Nodulation of Mimosa spp. by the b-proteobacterium
1084 Sprent
Ralstonia taiwanensis. Molecular Plant–Microbe Interactions 16,
1051–1061.
Chou YJ, Elliot GN, James EK, et al. 2007. Labrys neptuniae sp
nov., isolated from root nodules of the aquatic legume Neptunia
oleracea. International Journal of Systematic and Evolutionary
Microbiology 57, 577–581.
Downie JA. 2005. Legume haemoglobins: symbiotic nitrogen
fixation needs bloody nodules. Current Biology 13, 196–198.
Doyle JJ, Luckow MA. 2003. The rest of the iceberg. Legume
diversity in a phylogenetic context. Plant Physiology 131,
900–910.
Elliott GN, Chen W-M, Bontemps C, Chou J-H, Young JPW,
Sprent JI, James EK. 2007b. Nodulation of Cyclopia spp.
(Leguminosae, Papilionoideae) by Burkholderia tuberum. Annals
of Botany 100, 1403–1411.
Elliott GN, Chen W-M, Chou J-H, et al. 2007a. Burkholderia
phymatum is a highly effective nitrogen-fixing symbiont of Mimosa
spp. and fixes nitrogen ex planta. New Phytologist 173, 168–180.
Faria SMde, Hay TG, Sprent JI. 1988. Entry of rhizobia into
roots of Mimosa scabrella Bentham occurs between epidermal
cells. Journal of General Microbiology 134, 2291–2296.
Floyd SK, Bowman JL. 2004. Ancient microRNA target sequences in plants. Nature 428, 485–486.
Giraud E, Moulin L, Vallenet D, et al. 2007. Legume symbioses:
absence of Nod genes in photosynthetic Bradyrhizobia. Science
316, 1307–1312.
James EK, Sprent JI, Sutherland JM, McInroy SG,
Minchin FR. 1992. The structure of nitrogen-fixing root nodules
in the aquatic mimosoid legume Neptunia plena. Annals of
Botany 14, 173–180.
Lavin M, Herendeen PS, Wojciechowski MF. 2005. Evolutionary
rates analysis of Leguminosae implicates a rapid diversification of
lineages during the tertiary. Systematic Biology 54, 574–594.
Lavin M, Pennington RT, Klitgaard BB, Sprent JI, de
Lima HC, Gasson PE. 2001. The Dalbergioid legume (Fabaceae): delimitation of a pantropical monophyletic clade. American Journal of Botany 88, 503–533.
Manoury N, Kondorosi A, Kondorosi E, Mergaert P. 2007. Cell
biology of nodule infection and development. In: Dilworth MJ,
James EK, Sprent JI, Newton WE, eds. Leguminous nitrogenfixing symbioses. The Netherlands: Springer (in press).
Mathesius U. 2003. Conservation and divergence of signalling
pathways between roots and soil microbes: the Rhizobium–
legume symbiosis compared to the development of lateral roots,
mycorrhizal infections, and nematode-induced galls. Plant and
Soil 255, 105–119.
Nutman PS. 1987. Centenary lecture on nitrogen fixation.
Philosophical Transactions of the Royal Society of London Series
B 317, 69–106.
Ott T, van Dongen JT, Günther C, et al. 2005. Symbiotic
leghemoglobins are crucial for nitrogen fixation in legume root
nodules but not for general plant growth and development.
Current Biology 15, 531–535.
Pawlowski K, Sprent JI. 2007. Comparison between actinorhizal
and legume symbioses. In: Pawlowski K, Newton WE, eds.
Actinorhizal symbiosis. The Netherlands: Springer, 261–288.
Rasolomampianina R, Bailly X, Fetiarison R, et al. 2005.
Nitrogen-fixing nodules from rose wood legume trees (Dalbergia
spp.) endemic to Madagascar host seven different genera
belonging to a- and b-Proteobacteria. Molecular Ecology 14,
4135–4136.
Rodriguez-Llorente ID, Pérez-Hormaeche J, El Mounadi K,
Dary M, Caviedes MA, Cosson V, Kondorosi A, Ratet P,
Palomares AJ. 2004. From pollen tubes to infection threads:
recruitment of floral pectic genes for symbiosis. The Plant
Journal 39, 587–598.
Samba RT, de Lajudie P, Gillis M, Neyra M, Barreto MMS,
Dreyfus B. 1999. The diveristy of rhizobia nodulating Crotalaria
spp. from Senegal. Symbiosis 27, 259–268.
Soltis DE, Soltis PS, Chase ME, et al. 2000. Angiosperm
phylogeny inferred from 18S rDNA, rbcL, and atpB sequences.
Botanical Journal of the Linnaean Society 133, 381–461.
Spencer D, James EK, Ellis GJ, Shaw JE, Sprent JI. 1994.
Interaction between rhizobia and potato tissue. Journal of
Experimental Botany 45, 1475–1482.
Sprent JI. 2001. Nodulation in legumes. London, UK: Royal
Botanic Gardens Kew.
Sprent JI. 2003. Mutual sanctions. Nature 422, 672–674.
Sprent JI. 2007a. Evolution and diversity of legumes symbiosis.
In: Dilworth MJ, James EK, Sprent JI, Newton WE, eds.
Leguminous nitrogen-fixing symbioses. The Netherlands: Springer
(in press).
Sprent JI. 2007b. Evolving ideas of legume evolution and
diversity: a taxonomic perspective on the occurrence of nodulation. New Phytologist 174, 11–25.
Sprent JI, James EK. 2007. Legume evolution: where do nodules
and mycorrhizas fit in? Plant Physiology 144, 1–7.
Sy A, Giraud E, Jourand P, et al. 2001. Methylotrophic
Methylobacterium nodulate and fix nitrogen in symbiosis with
legumes. Journal of Bacteriology 183, 214–220.
Szczyglowski K, Amyot L. 2003. Symbiosis, Inventiveness by
recruitment? Plant Physiology 131, 935–940.
Yates RJ, Howieson JG, Reeve WG, Nandasena KG, Law IJ,
Bräu L, Ardley JK, Nistelberger HM, Real D, O’Hara GW.
2007. Lotononis angolensis forms nitrogen-fixing. lupinoid
nodules with phylogenetically unique, fast-growing. pinkpigmented bacteria, which do not nodulate L. bainesii or L. listii.
Soil Biology and Biochemistry 39, 1680–1688.