Journal of Experimental Botany, Vol. 64, No. 13, pp. 4023–4046, 2013
doi:10.1093/jxb/ert306 10.1093/jxb/ert306
Review paper
The evolution of autotrophy in relation to phosphorus
requirement
John A. Raven1,2,*,†
1 2 Division of Plant Sciences, University of Dundee at the James Hutton Institute, Invergowrie, Dundee DD2 5DA, UK†
School of Plant Biology, University of Western Australia, M048, 35 Stirling Highway, Crawley, WA 6009, Australia
* To whom correspondence should be addressed. E-mail: [email protected]
† Permanent address of JAR.
Received 15 April 2013; Revised 6 August 2013; Accepted 20 August 2013
Abstract
The evolution of autotrophy is considered in relation to the availability of phosphorus (P), the ultimate elemental resource
limiting biological productivity through Earth’s history. Work on microbes and plants is emphasized, dealing in turn with
the main uses for P in cells, namely nucleic acids, phospholipids, and water-soluble low molecular mass phosphate esters
plus metabolically active inorganic orthophosphate. There is a greater minimum gene number and minimum DNA content
in autotrophic than in osmochemoorganotrophic archaea and bacteria, as well as a lower rate of biomass increase per
unit P (P-use efficiency) in autotrophs than in osmochemoorganotrophs, in eukaryotes as well as bacteria. This may be
due to the diversion of rRNA from producing proteins common to all organisms to producing highly expressed proteins
specific to autotrophs. The P requirement for phospholipids is decreased in oxygenic photolithotrophs, and some anoxygenic photolithotrophs, by substituting galactolipids and sulpholipids for phospholipids in the photosynthetic, and some
other, membranes. The six different autotrophic inorganic carbon assimilation pathways have varying requirements for
low molecular mass water-soluble phosphate esters. In oxygenic photolithotrophs, there is no clear evidence of a different P requirement for growth in the absence (diffusive CO2 entry) relative to the presence of CO2-concentrating mechanisms (CCMs). P limitation increases the expression of crassulacean acid metabolism (CAM) in facultative CAM plants,
decreases the extent of inorganic carbon accumulation in algae with CCMs, and (usually) their inorganic carbon affinity
and the water-use efficiency of growth of terrestrial plants, and the light-use efficiency of photolithotrophs.
Key words: Anoxygenic photosynthesis, chemolithotrophs, CO2-concentrating mechanisms, oxygenic photosynthesis,
phospholipids, rRNA, Rubisco.
Introduction
Phosphorus (P) is essential for all forms of life, and is widely
perceived by geochemists as the element ultimately limiting global primary productivity (Berner and Berner, 2012;
Williams and Rickaby, 2012). Despite this, P has been considered less than have carbon (C), nitrogen (N), and iron (Fe)
in the evolution of autotrophy in relation to environmental
changes of the last 4 billion years. This could be because one
(or more) of C, N, and Fe is the proximate resource limiting local productivity. In attempting to address this lack of
consideration of P, the paper begins with a consideration
of the origin of life and whether autotrophy is an ancestral
trait. There is then a consideration of the roles of P in organisms, and of elemental stoichiometries in food webs, starting
from autotrophs and proceeding to phago- and osmochemoorganotrophs. P-use efficiency (PUE), measured as the rate
of dry matter (DM) gain per unit P in the organism, or rate
of increase of organism C per unit of organism P, is then
compared for autotrophs and chemoorganotrophs. The
mechanistic reasons for differences in PUE are then related
to P allocation among nucleic acids, phospholipids, and low
© The Author 2013. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved.
For permissions, please email: [email protected]
4024 | Raven
molecular mass water-soluble phosphate esters in chemoorganotrophs and autotrophs, and among variants on photolithotrophy. Finally, there is consideration of the effects of
limitation of growth by P availability on the water-use efficiency of the growth of terrestrial plants, and on acclimatory
variations in the expression of inorganic carbon-concentrating mechanisms in photolithotrophs.
Origin and early evolution of life:
autotrophy and chemoorganotrophy
After a long reign for the hypothesis of a ‘primordial soup’,
involving an abiological origin of organic precursors and chemoorganotrophy as the earliest trophic mechanism of organism
(Haldane, 1929), there are strong arguments for an autotrophic,
and specifically a chemolithotrophic (see Glossary), origin of
life (Lane et al., 2010; Table 1). Accepting this view, the last
universal common ancestor of all extant life on Earth was
a chemolithotrophic autotroph (Lane et al., 2010), so that
these organisms form the baseline for considerations of P
requirements. Derived trophic modes were osmochemoorganotrophic consumers and photolithotrophs such as ferrous
iron-oxidizing anoxygenic photosynthetic bacteria in the ferruginous (dissolved Fe2+ exceeds dissolved S2– in the anoxic
environment) Archean (Table 1) ocean (Crowe et al., 2008).
Chemoorganotrophy, and especially phagotrophy, which
involves the consumption of particulate organic matter (see
Glossary), requires that some of the organic matter ingested is
metabolized in exergonic reactions, which provide the energy
Table 1. Geological time scale, with some key events
Eon
Era
Period
Date (Ma)
Phanerozoic
Phanerozoic
Phanerozoic
Phanerozoic
Coenozoic
Coenozoic
Coenozoic
Mesozoic
Quaternary
Neogene
Palaeogene
Cretaceous
2.6–today
23–2.6
66–23
146–66
Phanerozoic
Mesozoic
Jurassic
200–146
Phanerozoic
Mesozoic
Triassic
251–200
Phanerozoic
Palaeozoic
Permian
299–251
Phanerozoic
Palaeozoic
Carboniferous
359–299
Phanerozoic
Palaeozoic
Devonian
416–359
Phanerozoic
Palaeozoic
Silurian
444–416
Phanerozoic
Phanerozoic
Proterozoic
Palaeozoic
Palaeozoic
Neoproterozoic
Ordovician
Cambrian
Ediacaran
488–444
542–488
635–542
Proterozoic
Neoproterozoic
Cryogenian
850–635
Proterozoic
Proterozoic
Neoproterozoic
Mesoproterozoic
Turonian
Proterozoic
Palaeoproteroic
1000–850
1600–1000
2500–1600
Archean
4000–2500
Hadean
Ends 4000
Comments
Spread of C4 grasslands
First C4 plants
First fossil evidence of angiosperms
First fossil evidence of ectomycorrhizas involved in
organic N, and also P, nutrition
First fossil evidence of angiosperms whose extant
relatives have cluster roots involved in P acquisition
First fossil evidence of (silicified) diatoma
Widespread ginkophytes whose extant example
Ginkgo has arbuscular mycorrhizas
Earliest marine planktonic coccolithophores; some
extant coccolithophores have a high affinity for phosphate
Earliest fossil evidence of cycads whose extant members
are all symbiotic diazotrophs which may increase P requirement
Early members of the clade which gave rise to extant
aquatic and amphibious CAM lycophyte Isoetes
First evidence of roots and of arbuscular mycorrhizas
involved in P acquisition
First fossil evidence of euphyllophyte leaves
First fossil evidence of gymnosperms
First fossil evidence of homoiohydric plants
First fossil evidence of lycophyte leaves
Earliest fossil evidence of terrestrial embryophytes
First fossils of silicified radiolarians
First fossils of marine silicified sponges
Deep ocean oxygenated from Ediacaran till today
Three global glaciations: Kaigas, Sturtian, then Marnonan
Ocean ferruginous with oxygenated surface layer
Ocean sulphidic with oxygenated surface area
First fossil eukaryotic photolithotroph ~1200 Ma
Ocean sulphidic with oxygenated surface layer
Ocean probably ferruginous up to 1900 Ga, then sulphidic,
with oxygenated surface layer after the Global
Oxygenation Event (GOE) ~2350 Ma. Freshwater
cyanobacteria before GOE
Origin of life at or before 3500 Ma
Ferruginous ocean throughout
Based on Raven and Edwards (2001, 2013); Taylor et al. (2009); Planavsky et al (2010); and Raven and Andrews (2010).
The evolution of autotrophy in relation to phosphorus requirement | 4025
for growth and maintenance. For the simplest case in which the
C:N:P stoichiometry of the consumer biomass is the same as
that of the consumed organism (autotrophic for the first trophic
level), the conversion of organic carbon to CO2 (respiration) or
organic matter from which it is more difficult to extract energy
(fermentation) in catabolic energy transduction decreases the
organic C:N and organic C:P ratio ultimately available to the
consumer. Accordingly, the chemoorganotroph would be less
P (or N) limited than the autotroph. However, P limitation of
consumers is found with a high C:P ratio in the food item, for
example growth of the autotroph with low P concentrations or
CO2 enrichment (Vrede et al., 2002). This could be reflected in
the composition of phagotrophs. The argument for osmochemoorganotrophs is more complex, since these organisms acquire
organic C and inorganic or organic N and P by separate transporters in the plasmalemma. These matters are considered by
Hessen et al. (2004) and Vrede et al. (2004).
Phosphorus availability in the biosphere
over the last 3.5 billion years
The P biogeochemical cycle over the last 3.5 billion years
(Table 1) involves physical and chemical weathering of rocks
on land (when continental crust became exposed to the atmosphere), with increasing biological contributions as the land surface was colonized and biological activity intensified (Lenton
and Watson, 2000; Raven and Edwards, 2001, 2013; Saltzman,
2005; Lambers et al., 2008; Raven and Andrews, 2010; Berner
and Berner, 2012; Williams and Rickaby, 2012; Table 1). After
a variable number of cycles through biota on land and in
inland waters exposed to the atmosphere, phosphate entered
the ocean through rivers and, to a smaller extent, as dust.
P is essential for all forms of life on Earth, and an understanding of the factors altering its availability over time is important
in understanding the quantitative role of P in organisms and
in the evolution of life. After the chemolithotrophic last universal common ancestor (Lane et al., 2010), Blake et al. (2010)
argue for a well-developed biogeochemical P cycle in the ferruginous Archean ocean 3.2–3.5 billion years ago (Table 1).
Planavsky et al. (2010) suggest, on the basis of P:Fe ratios
in non-silicoclastic (deposits of broken rocks dominated by
silica) iron deposits, corrected for silicic acid:phosphate competition in the relatively silicic acid-rich ocean, that P was relatively available in the generally ferruginous Archean and early
Palaeoproterozoic ocean. This contrasts with earlier work on
banded iron formations, uncorrected for silicic acid competition, suggesting that P scavenging by ferric oxide meant that
the ocean phosphate concentration in the Archean and early
Proterozoic ferruginous ocean (3.5–1.9 billion years ago, with
a possible sulphidic phase around the Global Oxygenation
Event (GOE) ~2.32 Ga; Bekker et al., 2004; Planavsky et al.,
2010) was only 10–25% of that in present oceans (Bjerrum and
Canfield, 2002) (Table 1). The values of Planavsky et al. (2010)
are preferable, because they are corrected for silicic acid competition in phosphate binding to ferric oxide.
Blank and Sanchez-Baracaldo (2010) showed that cyanobacteria evolved in freshwater, based on the the habitat
requirements of the extant organisms at the base of molecular
reconstructions of cyanobacterial phylogeny. This means that
the GOE ~2.32 billion years ago (Bekker et al., 2004) was permitted by the escape of these first oxygenic photolithotrophs
into the ocean, thus allowing them a global biogeochemical
role (Blank and Sanchez-Baracaldo, 2010) (Table 1). This
scenario means that the origins of oxygenic photolithotrophy
would not have been constrained by oceanic P availability,
but could have been restricted by whatever the (unknown)
local freshwater constraints there were on P supply. Bjerrum
and Canfield (2002; but see Planavsky et al., 2010) suggest
that an increase in ocean sulphate concentration from the
time of the GOE ~2.32 Ga ago led to increased sulphate
reduction and hence increased phosphate availability, since
ferrous sulphide deposits do not sequester phosphate. This
would be to some extent offset by phosphate binding to ferric
oxides deposited in the oxygenated surface ocean, though this
would have been reversed as the ferric iron was reduced in the
deeper sulphidic waters, allowing phosphate to return to the
surface ocean through the thermocline separating cold dark
deep water from the generally illuminated surface waters.
Such a relatively constant surface ocean phosphate concentration agrees with the later analysis by Planavsky et al.
(2010) and suggests little change in phosphate availability in
the surface ocean with the GOE and then until the Snowball
Earth (Cryogenian) episode 730–650 Ma ago when the ocean
was ferruginous, with a possible contemporary model of how
the ferruginous ocean arose and persisted (Mickuchi et al.,
2009) (Table 1). The three widespread glaciations presumably
arose by weathering related to tectonic activity drawing down
greenhouse gases, predominantly CO2, combined with the
lower radiant energy output from the sun earlier in its existence. These widespread glaciations were presumably terminated by the build up of CO2 from vulcanism in the absence
of drawdown by biology and solution in seawater in the very
cold Earth overcoming the high albedo of snow and ice.
Oxygenation of the deep ocean probably did not occur until
the late Proterozoic in the wake of the Marinoan glaciation
at the end of the Snowball Earth (Table 1), with agreement
that there were high phosphate concentrations in the surface
ocean during the Snowball Earth (Shen et al., 2000; Planavsky
et al., 2010; Johnston et al., 2012; Sahoo et al., 2012; see also
Papineau, 2010). After Snowball Earth, the evolution of silicified marine organisms such as benthic sponges from the late
Ediacaran (550 Ma) and planktonic radiolarians from the earliest Cambrian (540 Ma), and then planktonic and benthic diatoms ~120 Ma ago, decreased the silicic acid concentration in
the surface ocean and decreased competition with phosphate
for scavenging by ferric oxide (Planavsky et al., 2010) (Table 1).
This increased scavenging, decreasing the phosphate concentration in the surface ocean (Planavsky et al., 2010). Fluctuation
in redox state phosphate availability in the oceans occurred
through the Palaeozoic, with more stability in the Mesozoic
(Lenton and Watson, 2000; Saltzman, 2005). Cadmium in sediments has been used as a proxy for phosphate in Phanerozoic
seawater (Finkel et al., 2007; Horner et al., 2013) but, in the
absence of proxies for other nutrients. cadmium cannot indicate which is the productivity-limiting resource.
4026 | Raven
An unanswered question is, accepting that P availability
in the ocean declined with the evolution of silicified marine
organisms, was P then the ultimate resource limiting global
productivity? A personal vew is that it probably was, but
much more consideration and evidence is needed.
On land, the increase in area of land vegetated, and in the
depth of rooting (roots evolved in the lower Devonian) with
increasing plant height, as well as arbuscular mycorrhizas
involved in P acquisition, through the Devonian and into the
Carboniferous would have increased biological weathering
(Raven and Edwards, 2001, 2013; Smith and Read, 2008; Raven
and Andrews, 2010). This, increased the quantity of P used by
the terrestrial biosphere and increasing runoff to the oceans until
a new steady state related to the quantity of weatherable surfaceexposed rock (Raven and Edwards, 2001, 2013; Smith and Read,
2008; Raven and Andrews, 2010). Additional characteristics of
vascular land plants which can enhance P acquisition (ectomycorrhizas and taxa which today have representatives with cluster roots) are known from the Cretaceous (Table 1; Raven and
Andrews, 2010). Long periods without tectonic or glacial disturbance can lead to P leaching from terrestrial biomes, so that
P becomes the element limiting terrestrial primary productivity and with a major impact on plant diversity (Lambers et al.,
2008, 2010; Menge et al., 2012; Laliberté et al., 2013).
The roles of phosphorus in cells
In considering the categories of P and its compounds in the
structures and catalysis of cells and organisms, I exclude
P stored as metabolically inactive dissolved inorganic
orthophosphate in vacuoles, or as polyphosphate or as
phytate. The major chemically distinguishable pools of P
in P-replete unicellular and multicellular oxygenic photolithotrophs are, in decreasing order of abundance (i) RNA
(predominantly) and DNA (and a very small quantity of
phosphorylated proteins); (ii) membrane phospholipids; (iii)
low molecular mass water-soluble phosphate esters; and (iv)
metabolically active inorganic orthophosphate (Bieleski,
1968a, 1968b, 1973; Chapin and Bieleski, 1982; Chapin et al.,
1982; Raven, 2012; Veneklaas et al., 2012). These pools are
involved in (i) the storage, replication, and conversion into
catalysts and structural material of genetic material; (ii) the
formation of barriers to free diffusion of most solutes and a
structure into which protein catalysts of transport and other
processes are housed; and (iii) and (iv) intermediate metabolites in energy transduction and in biosynthesis.
Is there a difference in C:N:P ratio and in
P-use efficiency between autotrophs and
chemoorganotrophs?
Background
P-use efficiency (PUE) is defined here as the rate of dry mass
increase per unit P in the organism. Values for PUE can be calculated by multiplying the specific (= relative) growth rate (g
DM increase per organism g DM content d–1) by the organism
DM mol–1 of plant P (Berendse and Aerts, 1987; Raven, 2011,
2012). This latter term, organism DM mol–1 of plant P, is also
used as a measure of PUE (e.g. Veneklaas et al., 2012), and is
appropriate for comparisons among determinate annual species where agriculturally and ecologically important outcome
is biomass, harvestable product, or viable seed produced in the
growing season, but not for work on microorganisms or comparisons among organisms of different sizes (see below). Since
many of the data sets involve specific growth rates of smaller
organisms on a per cell or a per organism carbon basis, it is
also convenient in these cases to work on an organism C basis,
and the atomic C:N:P ratio. In comparing autotrophs and
chemoorganotrophs, it is important to compare organisms of
the same size, since there are differences with organism size in
both specific growth rate and in P per unit C or per unit DM.
Larger organisms have a lower PUE than smaller ones: the
decrease in growth rate with size between cyanobacteria and
flowering plants much more than offsets the slightly lower P
content of biomass in the larger organisms.
The effect of organism size on PUE was shown in the analysis by Raven (2012), who calculated PUE of 250–750 g DM
increase d–1 mol–1 P in the organism for terrestrial leguminous
flowering plants and 1740–10 150 of those units for heterocystous cyanobacteria, with the general trend of increasing
values with N source of N2 lowest, then NO3–, and NH4+/urea
highest. The work of Sanginga et al. (1989) extends the data
set for flowering plants, limited to legumes in Raven (2012),
to the actinorhizal Casuarina equisetifolia, with PUEs of 178
and 156 g DM increase d–1 mol–1 organism P for P-replete
plants with nitrate and dinitrogen as N sources, respectively,
and values of 400 and 210, respectively, with P limitation.
These data extend the range (250–750) for legumes alone to
156–750 g DM d–1 mol–1 plant P for legumes plus Casuarina.
C:N:P and PUE in autotrophic microrganisms
The Redfield Ratio of C 106:N 16:P 1 (by atoms) in marine
phytoplankton is a time-averaged mean value for the world
ocean. This ratio is a very useful baseline (Falkowski, 2000;
Geider and La Roche, 2002) for which a mechanistic basis has
been suggested (Loladze and Elser, 2011). The mean ratio for
29 species of nutrient-replete organisms from a wide phylogenetic range of eukaryotic microalgae and cyanobacteria is C
132:N 18:P 1; that is, with slightly more C and N relative to P
than in the Redfield Ratio (Ho et al., 2003; Quigg et al., 2003,
2011; Falkowski et al., 2004). There is phylogenetic variation
in the ratio, although the number of strains tested in each
clade was necessarily small. Table 2 shows the PUE calculated
from the Redfield Ratio of C:P, and the maximum reported
specific growth for a range of autotrophic microorganism
adjusted to 20 °C assuming a Q10 of 2.
C:N:P and PUE of chemoorganotrophic
microorganisms and metazoan microzooplankton
Fagerbakke et al. (1996) and Cotner et al. (2010) show that
native aquatic and cultured osmochemoorganotrophic bacteria maintain a C:N:P ratio of 50:10:1, although other work
The evolution of autotrophy in relation to phosphorus requirement | 4027
Table 2. Specific growth rates, atomic C:N:P ratios, and PUE values for growth for Archea, Bacteria, and Eukarya of a range of trophic
modes
Organism
Aerobic osmochemoorganotrophic
bacterium Escherichia coli
Aerobic osmochemoorganotrophic
bacterium Escherichia coli1
Aerobic osmochemoorganotrophic
bacteria from a lake2
Aerobic chemolithotrophic bacterium
Alcaligenes eutrophus
Anaerobic chemolithotrophic archean
Methanococcus vannielii
Anoxygenic photolithotrophic
bacterium Chlorobium thiosulphatophilum
Oxygenic photolithotrophic cyanobacterium
Synechococcus leopolensis (as Anacystis nidulans)
Aerobic osmochemoorganotrophic eukaryote
Achlya bisexualis
Aerobic phagochemoorganotrophic eukaryote
Tetrahymena geleii
Oxygenic photolithotroophic eukaryote
Chaetoceros salsugineum4
Oxygenic photolithotrophic eukaryote
Chlorella regularis
Oxygenic photolithotrophic eukaryote
Thalassiosira pseudonana3
Cell
diameter (μm)
Specific growth
rate, mol C
increase mol–1 cell C s–1
C:N:P
PUE, mol C
increase mol–1
cell P s–1
1.5
283 × 10–6
50:10:1
14.2 × 10–3
1.5
146 × 10–6
41:11:1
5.99 × 10–3
138 × 10–6
95:22:1
13.1 × 10–3
1
48 × 10–6
50:10:1
2.40 × 10–3
2
15 × 10–6
50:10:1
0.75 × 10–3
2
27 × 10–6
106:16:1
2.86 × 10–3
1
24 × 10–6
106:16:1
2.54 × 10–3
15
170 × 10–6
106:16:1
18.0 × 10–3
30
88 × 10–6
106:16:1
9.38 × 10–3
4
75 × 10–6
106:16:1
7.95 × 10–3
5
27 × 10–6
106:16:1
2.86 × 10–3
5
11.4 × 10–6
4.72 × 10–6
665:61:1
220:37:1
2.50 × 10–3
3.14 × 10–3
?
Growth rates are from Raven et al. (2013), except where indicated by superscript numbers: 1Makino et al. (2003); 2Makino and Cotner (2004);
Ishimi et al. (2012); 4Perry (1976).
C:N:P for bacteria other than oxygenic photolithotrophic cyanobacteria from Fagerbakke et al. (1996) and Cotner et al. (2010), except where
indicated for cases in which growth rate and C:N:P are available for the same culture in the work of 1Makino et al. (2003) and 2Makino and
Cotner (2004).
C:N:P for oxygenic photolithotrophic cyanobacteria, and for eukaryotes, are assumed to be the Redfield Ratio, except where indicated:
4Perry (1976).
As in Raven et al. (2013), specific growth rates ae normalized to 20 °C assuming a Q10 of 2. Cell diameter is the equivalent spherical diameter,
except for Achlya bisexualis where the hyphal diameter is given.
3
gives a more Redfield-like ratio of 94:17:1 (Keiblinger et al.,
2010), and large variations in C:N:P are seen under extremes
of limitation by the C, the N, or the P substrates for growth
(Chan et al., 2012). For soil microbes as a whole (Archaea,
Bacteria, Eukarya), the C:N:P ratio is 77:7:1, with the usual
large variance (Cleveland and Liptzin, 2007). The bacterial
C:P corresponds to one term in the time-related PUE, namely
organism DM per unit organism P which is about twice as
high for the bacteria as for the algae, so that the PUE defined
as the reciprocal of P content in DM is higher for photolithotrophs than for osmochemoorganotrophs. The maximum
specific growth rate of osmochemoorganotrophic bacteria
is well over twice that of oxygenic cyanobacteria and microalgae, when the rates are normalized to a temperature of
20 °C assuming a Q10 of 2 (Raven et al., 2013). Thus, for the
Archaea and Bacteria, the fastest-growing aerobic osmochemoorganotroph has about six times the specific growth rate
of the fastest-growing aerobic chemolithotroph (Table 2), and
10 times that of the fastest-growing oxygenic photolithotroph
(Table 2), using growth rate data from Raven et al. (2013) and
additional data from Makino et al. (2004) and Makino and
Cotner (2004). Combining these specific growth rates with the
C content per unit organism P suggests that time-based PUE
is five times higher for chemoorganotrophs than for oxygenic
photolithotrophs (Table 2).
For eukaryotic microorganisms, the fastest-growing aerobic osmochemoorganotrophic organism has more than twice
the specific growth rate of the fastest-growing oxygenic photolithotroph; for an aerobic phagochemoorganotroph, the
rate is only slightly higher than for the fastest-growing photolithotroph (Table 2), using growth rate data from Raven
et al. (2013) and additional data from Perry (1976) and Ishimi
et al. (2012). The C:N and C:P of fungal osmochemoorganotrophs and of unicellular phagochemoorganotrophs is similar to that of photolithotrophs (Ho et al.. 2003; Quigg et al.,
2003, 2011; Gruber et al., 2009; Keiblinger et al., 2010). The
PUE of growth of the fastest-growing (Table 1) osmochemoorganotrophs and, by a much smaller margin and probably
insignificantly, phagochemoorganotrophs, is higher than that
of the fastest-growing oxygenic photolithotrophs (Table 2).
There are also C:N:P data for metazoan phagochemoorganotrophs (e.g. Beers, 1966; Gismervik, 1997). Beers (1966)
4028 | Raven
examined the elemental composition of nine taxonomically
defined groups of metazoan zooplankton from the Sargasso
Sea off Bermuda, and found that the range of C:P atomic
ratios was 70:8:1 and 195:47:1 (i.e. ranging from below to
above the Redfield Ratio value of 106:16:1 of their ultimate
food source). It should be noted that the Sargasso Sea is
very oligotrophic which may have altered the C:N:P of phytoplankton and hence of zooplankton. Gismervik (1997)
examined marine planktonic crustaceans from the Oslofjord
and found atomic ratios for C:N:P of between 34:8:1 and
188:40:1. Again, the values range from below to above the
Redfield Ratio, and in this case the environment is less oligotrophic. For planktonic crustaceans in the Baltic Sea, Walve
and Larssson (1999) found C:N:P ratios varying between
84:14:1 and 289:27:1, again spanning the Redfield Ratio for
a more nutrient-rich habitat than the Sargasso Sea. The conclusion is that the C:N:P ratios of marine metazooplankton
can be rather below to rather above the Redfield Ratio for a
range of availabilities of nutrients to phytoplankton in the
environment.
The specific growth rates for metazoan zooplankton,
adjusted to a standard temperature of 20 °C using the appropriate Q10 value (Hansen et al., 1997; see also Hirst et al.,
2003; DeLong et al., 2010), range from 16 × 10–6 s–1 for volumes of 105 μm3 to 3 × 10–6 s–1 for organism volumes of 108
μm3 (metazoan entries in fig. 2B of Hansen et al., 1997).
Assuming a C:P Redfield Ratio of 106:1, this gives PUEs of
1.7 × 10–3 mol C increase mol–1 organisms P s–1 for a volume
of 105 μm3 and 0.32 × 10–3 mol C increase mol–1 organisms
P s–1 for a volume of 108 μm3. These PUEs for phagochemorganotrophs are less than the values for oxygenic photolithotrophs in Table 2. However, the organism volumes for
these metazoans are 105–108 μm3, while the organism (cell)
volumes for the photolithotrophs in Table 2 are <70 μm3,
once more emphasizing the significance of the size of organisms for PUEs. However, for algal cells with a volume of 108
μm3, the growth rate is 4.6 × 10–6 s–1 (Finkel et al., 2010) and,
assuming a Redfield Ratio of C:P, a PUE of 0.49 × 10–3 mol
C increase mol–1 organism P s–1, which is closely similar to
that of the metazoan grazer of the same organism volume,
namely 0.32 × 10–3 mol C increase mol–1 organism P s–1, perhaps reflecting in PUE the discontinuity noted between protists and metazoans in the size scaling of metabolic rates by
DeLong et al. (2010).
Overall, for comparisons within a given range of sizes,
microorganisms show higher PUE for chemoorganotrophs
than for autotrophs (Table 2). It must, however, be emphasized that these calculations, apart from those based on
Perry (1976), Makino et al. (2003), and Makino and Cotner
(2004), involve different organisms for the P content (C:N:P)
data and growth rate data within a trophic mode, and also
assumptions about the C:N:P ratios for methanogens and
for the anoxygenic photolithotroph for which no C:N:P data
could be found (Table 2). Clearly, there is a need for measurements of the specific growth rate and C:P ratio for a range
of organisms of each trophic mode, each under a range of P
availabilities, to test the hypothesized higher PUE for chemoorganotrophs than for autotrophs.
Although there are values for cell size in Table 2, no corrections are made for cell size dependence of specific growth
rate of the organisms. The available evidence suggests that
a simple linear relationship of specific growth rate to log10
cell size with an exponent more negative than –0.2 is not the
rule (Raven, 1994; Bec et al., 2008; Chen and Liu, 2010, 2011;
DeLong et al., 2010; Finkel et al., 2010; Sal and Löpez, 2011;
Kempes et al., 2012: Marañón et al., 2013). A further point is
that the size range in the Bacteria and Archaea considered is
rather small (Table 2); the eukaryote cells are larger and have
a slightly larger size range.
C:N:P and PUE in macrophytes
C:N:P ratios were found from analyses of published values
for green, red, and brown marine macroalgae plus seagrasses
(C 550:N 30:P 1) by Atkinson and Smith (1983), and for seagrasses (C 474:N 24:P 1) by Duarte (1990). In both data sets,
the organisms were collected from the natural environment.
The higher C:N and C:P ratios for the benthic macrophytes
than for microalgae can be related to the consistent presence
of cell walls in the macrophytes and, perhaps, to nutrient limitation in the natural environment. In no case do there seem to
be have been measurements of C:N:P in the non-photosynthetic roots (seagrasses) or rhizoids (a few macroalgae such
as Caulerpa: Raven, 1984b).
For terrestrial vascular plants, the possibility of separating
the C:N:P ratio in photosynthetic from that in non-photosynthetic parts has been realized. This allows the estimation of
the PUEs of the two parts. Assuming isometric growth (i.e. a
constant ratio of contributions to total biomass of the photosynthetic and non-photosynthetic parts during growth; Raven,
2001), the ratio of the PUEs equals the C:P ratios. More data
are available for shoots and leaves than for below-ground parts:
it is more difficult to obtain complete below-ground structures
for plants growing in soil than is the case for leaves, and ease of
harvesting complete root systems from hydroponically growth
plants must be set against the artificial nature of hydroponic
growth for the great majority of terrestrial vascular plants.
Broadley et al. (2004) examined the elemental contents of
the shoots of a range of flowering plants and found significant
phylogenetic variation, and a mean ratio of C 144:N 17:P 1
(by atoms). While the N:P ratio is closely similar to the values
for microalgae (Falkowski, 2000; Geider and LaRoche, 2002;
Falkowski and Raven, 2007; Quigg et al., 2011), the C:N and
C:P ratio is significantly higher than in the microalgae, presumably related to the presence of a cell wall of high C:N and
C:P in all flowering plants, but not in many of the microalgae.
Watanabe et al. (2007) found significant phylogenetic variation in the elemental composition of the leaves of flowering
plants. These important data sets do not have comparative
values for below-ground parts. Wang et al. (2010) present a
compilation of meta-analyses of data on C:N:P ratios for
leaves (Wright et al., 2004) and fine (<5 mm diameter) roots
(Jackson et al., 1997; cf. Gordon and Jackson, 2000). The
compilations used by Wang et al. (2010) involve 11 biomes
for leaves, while for roots eight of the 11 biomes were lumped
together. With this proviso, the mean leaf elemental ratio
The evolution of autotrophy in relation to phosphorus requirement | 4029
is C 924:N 33.2:P 1, and the mean root elemental ratio is C
2229:N 33.2:P 1. The leaf ratios of C:P and N:P are higher
than those found by Broadley et al. (2004) and Watanabe
et al. (2007), presumably as a result of analysis of more
sclerophyllous leaves than for solely mesophytic herbaceous
leaves, and/or due to nutrient (N, P) limitation. Interpretation
of the leaf and root C:N:P ratios requires that the root data
contain meristematic tissues, while the leaf data are almost
entirely for mature leaves. It would be expected that the C:N
and C:P ratios would be lower in meristematic tissue where
protein synthesis, and the necessary RNAs, predominate over
the production of mechanical tissue. This means that the C:N
and C:P ratios for leaves would be lower if expanding leaves
were considered. Interestingly, the C:N:P of total fine roots
(live plus dead) is closely similar to that of live fine roots
(Jackson et al., 1997), suggesting negligible movement of N
and P back to the main roots from senescent fine roots (see
Robinson, 1990; Veneklaas et al., 2012) relative to the movement of C, despite the significantly greater fraction of C than
of N and P in the unrecoverable cell wall material.
Other, more limited data sets include that of Zheng et al.
(2012) on grasslands in Inner Mongolia. Leaves and roots
samples from three communities, using only the ungrazed
control treatment, show lower C:N:P ratios in the roots than
in the leaves, in agreement with the outcome of the analysis
by Wang et al. (2010). However, some data obtained using
hydroponically grown plants give more mixed results (Allen
and Raven, 1987; Allen et al., 1988).
An important paper (Yu et al., 2012) reports specific
growth rates and C:N:P ratios for above- and below-ground
structures of three species from the Inner Mongolian grasslands; these data, and the derived PUE values, are presented
in Table 3. Despite their contrasting phylogenies and ecophysiologies (a C3 and a C4 grass and a C3 chenopod), the C:P
ratio and the PUE values are much higher for below-ground
than above-ground parts for all three species (Table 3). This
is also in agreement with the higher PUE in chemoorganotrophic than photolithotrophic structures.
Conclusion on differences in C:N:P and PUE among
trophic modes
There is some evidence of a lower C:P in chemorganotrophs
than in otherwise comparable photolithotrophs, and rather
stronger evidence for a higher PUE in osmochemorganotrophs than in photolithotrophs when organisms with the
highest PUE in each trophic mode are compared.
Are there differences in P pools between
autotrophs and chemoorganotrophs?
Is there a systematic difference in nucleic acids, phospholipids, and water-soluble low molecular mass phosphate esters
plus metabolically active inorganic phosphate between autotrophs and chemoorganotrophs (Table 4)?
Nucleic acids: DNA
Genome sizes can be compared in terms of the number of base
pairs and the number of genes. On both of these grounds, the
smallest known genome of a chemoorganotrophic bacterium
is smaller than the smallest known genome for an oxygenic
photolithotrophic bacterium (Raven et al., 2013). However,
there is a very large range of sizes of genomes for chemoorganotrophs and for photolithotrophs among Bacteria and
among Eukarya, and no general conclusions can be drawn
as to the genome sizes of chemoorganotrophs, chemolithotrophs, and photolithotrophs (Raven et al., 2013; see also
Hodgson et al., 2010). As is discussed in more detail under
differences in P pools among autotrophs, DNA can be <10%
of the total nucleic acids.
Nucleic acids: RNA
Appealing to optimal allocation of P, Raven (2012) rationalized the lower PUE of some cyanobacteria and leguminous
plants when growing on dinitrogen than when growing on
Table 3. Specific growth rate, mol C:mol N:mol P, and PUE for above- and below-ground parts of an N- and of a P-limited C3 grass, C4
grass, and C3 chenopod
Plant and part
Specific growth rate,
mol C increase mol–1 cell C s–1
Mol C:mol
N:mol P
PUE, mol C increase
mol–1 P s–1
Leymus chinensis above-ground
(perennial C3, Poaceae)
Leymus chinensis below-ground
(perennial C3, Poaceae)
Cleistogenes sqarrosas above-ground
(perennial C4, Poaceae)
Cleistogenes sqarrosa below-ground
(perennial C4, Poaceae)
Chenopodium glaucum above-ground
(annual C3, Chenopodiaceae)
Chenopodium glaucum below-ground
(annual C3 Chenopodiaceae)
0.66 × 10–6 (N lim)
0.60 × 10–6 (P lim)
0.86 × 10–6 (N lim)
0.79 × 10–6 (P lim)
0.61 × 10–3 (N lim)
0.64 × 10–3 (P lim)
0.64 × 10–6 (N lim)
0.58 × 10–6 (P lim)
0.91 × 10 –6 (N lim)
0.81 × 10–6 (P lim)
0.91 × 10–6 (N lim)
0.81 × 10–6 (P lim)
258:22:1
258:19:1
1123:24:1
698:16:1
185:17:1
152:9:1
1230:23:1
994:16:1
199:8.9:1
185:13:1
1230:22:1
1123:17:1
0.17 × 10–3
0.16 × 10–3
0.97 × 10–3
0.55 × 10–3
0.11 × 10–3
0.097 × 10–3
0.79 × 10–3
0.58 × 10–3
0.21 × 10–3
0.15 × 10–3
1.12 × 10–3
0.91 × 10–3
Data are from fig. 5 of Yu et al. (2012).
4030 | Raven
Table 4. Outline of possible interactions of P with the autotrophic CO2 assimilation pathway
Role of P
Relationship to C assimilation pathway
References
RNA
Organisms with C assimilation pathways with higher
rates of CO2 assimilation per unit protein committed
to the pathway have less protein to allocate to other
growth-related processes. Under P limitation, with
protein synthesis limited by rRNA availability, the PUE
of growth (rate of biomass gain per unit P in biomass)
would decrease when the subsistence quota for P is
approached. Savings of up to 10% are possible for
CCM relative to CO2 diffusion, or for antennae based
on chlorophyll/carotenoid pigment–protein complexes
relative to phycobiliproteins
Replacement of phospholipid by galactolipid and/
or S-lipid might alter permeability to CO2 with
implications for diffusive CO2 entry, in parallel with any
aquaporin-like CO2 channels, in C3 plants and in CO2
entry for cyanobacterial CCMs, and for leakage from
CCMs. Not quantifiable, as there are no comparative
date on the CO2 permeability of galactolipid and/or
S-lipid membranes and for phospholipid membrane.
Possible saving on P in organism with C assimilation
pathways involving fewer P-esters could have a higher
PUE of growth. There is little scope for economizing
on P in P-deficient organisms by decreasing P-ester
concentrations, since this would apparently require
higher enzyme content, with a corresponding increase
in the requirement for RNA for protein synthesis
Possible decreased P requirement for growth if
decreased P availability causes expression of a
more P-efficient pathway, or variant on a
pathway; not readily quantified
Raven (1984a, b, 2011); Flynn et al. (2010);
Bar-Even et al. (2010, 2012); Raven et al. (2012);
Veneklaas et al. (2012)
Phospholipids
P-esters (low molecular
mass water-soluble P esters)
Regulation of expression of
C assimilation pathway
combined N as related to the diversion of rRNA (and hence
P) from other essential functions into synthesis of nitrogenase, including replacement of nitrogenase damaged by oxygen, and to the synthesis of catalysts and structures related
to the protection of nitrogenase from oxygen. This effect
would be particularly evident when the growth rate was limited by P supply; this is what was found for the effect of N
sources in the cases examined by Raven (2012), and see also
Sanginga et al. (1989; discussed above). Raven et al. (2013)
extend this to the difference between autotrophs and chemoorganotrophs, arguing that more unique highly expressed
proteins are involved in autotrophy than in chemoorganotrophy (see Ellis, 1979; Raven, 1984a, b, 1991, 2013a; Geider
and La Roche, 2002; Losh et al., 2013). This would mean
a lower PUE for autotrophs than for chemoorganotrophs,
as was shown above for bacterial and archaeal examples.
However, rRNA measurements are needed to test the correlation further, and then to attempt to falsify the hypothesis.
It is clear that optimal allocation of P in rRNA, at least as
indicated by the growth rate hypothesis, is more readily seen
in chemoorganotrophs than in photolithotrophs (Sterner
and Elser, 2002; Vrede et al., 2002, 2004; Karpinets et al.,
2004; Matzek and Vitousek, 2009; Nicklisch and Steinberg,
2009; Flynn et al., 2010; Reef et al., 2010, 2012; Loladze and
Elser, 2011). rRNA requirements will be returned to below
Raven et al. (2012); Veneklaas et al. (2012
in supplementary information)
Veneklaas et al. (2012, in supplementary
information)
Raven et al. (2012) (only relates to the
phenomenon, not the specific
mechanism mentioned)
in the context of the possibility of variations with pathways
among autotrophs.
Phospholipids
Phosphorylated lipids are very widespread among the
polar lipids of biological bilayer membranes, regardless of
whether the lipids have an ether (Archaea) or ester (Bacteria,
Eukarya) bond between the fatty alcohol (Archaea) or fatty
acid (Bacteria, Eukarya) and glycerol. However, there are a
number of instances of dominance of glycolipids, for example in methanogenic Archaea (Koga et al. 1993), and of
glycolipids, namely the galactolipids monogalactosyldiacylglycerol (MGDG) and digalactosyldiacylglycerol (DGDG)
and the sulpholipid sulphoquinovosyldiacylglycerol in photosynthetic membranes of some anoxygenic photosynthetic
bacteria, and almost all oxygenic photolithotrophs (Raven,
1989; Imhoff and Bias-Imhoff, 1995; Selstam and Campbell,
1996; Rexroth et al., 2011; Yuzawa et al., 2012; Mizoguchi
et al., 2013).
Under P deficiency, an anoxygenic photosynthetic bacterium (Benning et al., 1993, 1995), freshwater cyanobacteria
and eukaryotic algae (Khozin-Goldberg and Cohen, 2006;
Bellinger and van Mooy, 2012), some marine cyanobacterial
and eukaryotic phytoplankton (Van Mooy et al., 2006, 2009;
The evolution of autotrophy in relation to phosphorus requirement | 4031
Martin et al., 2011; Popendorf, 2011a, b; Dyhrman et al.,
2012), and flowering plants (Lambers et al., 2012; Venekelaas
et al., 2012; Okazaki et al., 2013) replace phospholipids in
non-photosynthetic membranes with the glycolipids common
in photosynthetic membranes, namely sulpho- and galactolipids (Table 4). The functional results of such replacements, in,
for example, the permeability of the lipid bilayer membrane,
and the function of integral membrane proteins, are not yet
clear (supplementary information in Veneklaas et al., 2012).
Such substitution of phospholipids is not limited to
autotrophs, Minnikin et al. (1974) found that P limitation of the chemoorganotroph Pseudomonas diminuta led
to the replacement of acidic phospholipid by acidic glycolipids, while the chemoorganotroph Bacillus acidocaldarius has constitutive replacement of phospholipid by
sulpholipid (Langworthy et al., 1976), as does Bacillus coahuilensis from a very P-limited habitat (Souza et al., 2008).
Sulpholipid has also been reported from the chemoorganotrophic bacterium Magicaulis spp. (Abraham et al., 1999).
Furthermore, Popendorf et al. (2011a) showed a significant
content of MGDG in chemoorganotrophic marine bacteria
from the low-P Sargasso Sea, and suggested that the occurrence of MGDG in chemoorganotrophs is a function of P
deficiency. Löpez-Lara et al. (2005) and Geske et al. (2012)
found similar results for the Proteobacteria Sinorhizobium
meliloti and Agrobacterium meliloti, respectively, while sulpholipid occurs constutively in some members of the proteobacterial Rhizobiaceae (Cedergreen and Hollingsworth,
1994). The occurrence of substitutes for phospholipids in the
Rhizobiacese is, as pointed out by Benning (1998), perhaps
not surprising since some members of the Rhizobiaceae are
photosynthetic. It should be emphasized that the distinction
between chemoorganotrophs and photolithotrophs is not
absolute, with many photosynthetically competent organisms functioning as mixotrophs, for example the phagomixotrophs in aquatic habitats (Flynn et al., 2013) and a range of
mixotrophs on land (Schmidt et al., 2013). In the context of
P, plastidic as well as aplastidic small protists in the North
Atlantic tropical gyre obtain much of their P by phagotrophy,
rather than uptake of inorganic phosphate across the plasmalemma (Hartmann et al., 2011).
It would be particularly interesting in the context of economizing on P in autotrophs to know if there is any replacement of phospholipid by glycolipids (including sulpholipids)
in mycorrhizal fungi, either constitutively or in response to
P deficiency. This applies especially to arbuscular mycorrhizas which have a particular role in P acquisition (Smith and
Read, 2008). However, there seem to be no relevant data for
the extraradicular hyphae of mycorrhizas, although there
are data for the glycolipid and phospholipid content of vesicles of an arbiscular mycorrhizal fungus (Jabaji-Hare et al.,
1984). In soils of very low P availability, the relative utility
of the alternatives of arbuscular mycorrhizas and of cluster
roots (Lambers et al., 2008, 2010) might be influenced by
differences in the capacity to recover P to the longer lived
parts of the plant from senescent cluster roots or arbuscular mycorrhizas. Plasmalemma lipids would be particularly
difficult to resorb since resorption requires maintenance of
plasmalemma functions. An analogous, but better documented, situation is found with N acquisition by ectomycorrhizas (Smith and Read, 2008): here the non-recoverable
N-containing component is chitin in the fungal cell walls with
a C:N ratio of eight, while the equivalent structural element
in the plant structures, cellulose, contains no N.
Low molecular mass water-soluble phosphate esters
There are at least five autotrophic CO2 fixation pathways
found in Archaea and Bacteria, in addition to the Benson–
Calvin cycle (the C-reduction cycle) found in many bacterial autotrophs as well as in all photosynthetic eukaryotes
(Raven, 2009; Bar-Even et al., 2010, 2012; Raven et al., 2012).
These pathways have a variable number, sometimes zero, of
low molecular mass water-soluble phosphate esters, which
are unique to that pathway. Do these additional phosphate
esters add significantly to the total content of these low
molecular mass water-soluble phosphate esters per unit biomass? Data only seem to be available for organisms with the
Benson–Calvin cycle. Bieleski (1968a, b; see also Chapin and
Bieleski, 1982; Chapin et al., 1982) examined these phosphate
esters in the duckweed Spirodela: his results showed that any
contributions made by the specific phosphate esters ribulose1,5-bisphosphate and sedoheptulose-1,7-bisphosphate, which
are not involved in the oxidative pentose phosphate cycle or
glycolysis, was very small, since they are included in the ‘other
compounds’ not listed by name which amount to 7.5% of the
total low molecular mass water-soluble phosphate esters.
Despite this small contribution of ribulose-1,5-bisphosphate
to the total pool of phosphate esters, it is worth pointing out
the large range of K0.5 values of Rubisco (ribulose bisphosphate carboxylase-oxygenase) for ribulose-1,5-bisphosphate
of a wide range of green algae and aquatic and terrestrial
embryophytes: the range is from 10 mmol m–3 for Clematis sp.
to 136 mmol m–3 for Stipa mollis (Yeoh et al., 1981). This range
of values relates to phylogeny rather than to the environment
or the photosynthetic pathway. Regardless of the cause of the
genetically determined variation, it is clear that some plants
can potentially carry out photosynthesis at the same rate as
others but with less than a tenth of the concentration of ribulose-1,5-bisphosphate, regardless of whether the photosynthetic apparatus usually functions with Rubisco-saturating
concentrations of ribulose-1,5-bisphosphate or whether the
concentration of this substrate is typically near the concentration corresponding to K0.5. This argument requires that
there is no trade-off between affinity and the substrate-saturated rate for Rubisco for ribulose-1,5-bisphosphate of the
kind found for CO2 (Tcherkez et al., 2006) (Table 4).
For observed K0.5 values of phosphate esters for their
enzymes, there is a potential trade-off in (optimal) P allocation
between low molecular mass water-soluble phosphate esters
and rRNA (Veneklaas et al., 2012). For those low molecular mass water-soluble phosphate esters that are present at
concentrations which are subsaturating for the enzyme(s)
for which they are substrate(s), maintenance of the metabolic flux through the pathway(s) involving the enzymes(s) if
the concentration of the low molecular mass water-soluble
4032 | Raven
phosphate ester substrate(s) is decreased would require a
compensatory increase in the content of the enzyme(s) and
hence (with optimal allocation) of the rRNA needed to synthesize the increased content of enzymes (Veneklaas et al.,
2012). There is unlikely to be a decreased P requirement for
metabolism from such a decrease in the concentration of low
molecular mass water-soluble phosphate esters. If, however,
the low molecular mass water-soluble phosphate esters are
well above the saturating concentrations for the enzymes for
which they are substrates, then the concentration of those
particular low molecular mass water-soluble phosphate esters
could be decreased without decreasing the flux through the
pathway(s) or needing synthesis of additional enzyme protein
and hence of the additional rRNA. Of course, it is possible
that a certain concentration of low molecular mass water-soluble phosphate esters in compartments such as the cytosol,
plastid stroma, and mitochondrial matrix might be needed
for some reason apart from their role as enzyme substrates or
activity modifiers, for example as acid–base buffers (Raven,
2013b) (Table 4).
Conclusions on differences in P pools between
autotrophs and chemoorganotrophs
Osmochemorganotrophic bacteria have a smaller minimum
genome size than do autotrophic bacteria, but with a large
range of genome sizes in both trophic groups. RNA content
is probably more closely related to growth rate than in photolithotrophs. Replacement of phospholipids by glycolipids
is more common in photolithotrophs than in chemoorganotrophs. There is no evidence of differences in allocation to
low molecular mass water-soluble phosphate esters in chemoorganotrophs relative to autotrophs.
Are there differences in phosphorus pools
among autotrophs?
Is there a systematic difference in nucleic acid, phospholipids,
and water-soluble low molecular mass phosphate esters plus
metabolically active inorganic phosphate among autotrophs
depending on the autotrophic pathway used?
Nucleic acids: DNA
The smallest known genome of oxygenic photolithotrophs is for a strain of the very small (0.5 μm equivalent
spherical diameter) marine planktonic cyanobacterium
Prochlorococcus marina which expresses a CO2-concentrating
mechanism (CCM) and which occurs in very low-P habitats
(Raven et al., 2013). For photolithotrophic eukaryotes, the
smallest gene number (but not the smallest genone) is for the
acidophilic red microalga Cyanidioschyzon merolae; rather
smaller genomes, and slightly greater gene numbers are found
in Ostreococcus spp., some strains of which grow in low-P
environments (Raven et al., 2013).
Greilhuber et al. (2006) suggest that the smallest known
genome size among flowering plants is that of Genlisea margaretae (Lentibulariaceae), a C3 carnivorous plant from
low-P environments, with 63 Mbp; although Soltis et al.
(2003) report the lower value of 49 Mbp for Cardamine amara
(Brassicaceae), Johnston et al. (2005) cite a value of 221 Mbp
for this species. For another family tolerant of low P availability, the Proteaceae, genome sizes for six species range
from 807 Mbp to 946 Mbp (Soltis et al., 2003; Morgan and
Westoby, 2005), which is smaller than the largest genome
of the carnivorous family Lentibulariaceae (1620 Mbp).
Relating genome size to tolerance of low P availability is very
much a ‘fits where it touches’ exercise.
For multicellular plants such as those considered in the
previous paragraph, the positive correlation of genome size,
and hence nucleus size, with cell size (see Soltis et al., 2003;
Connolly et al., 2008) means that, for a plant of a given size,
more P per nucleus does not necessarily mean more P in
DNA per plant since there are fewer, larger cells. Zhu et al.
(2005) showed that there was an increased copy number of
rRNA genes with increasing cell size in phytoplankton, since
larger cells need more copies of the genes for highly expressed
proteins to supply protein in the larger growing cells which
have a very small specific growth rate decrement with increasing cell size (see fig. 3 of Finkel et al., 2010). It would be of
interest to see if there is a correlation of rRNA copy number in the genome with cell size. There are necessarily relatively few highly expressed proteins, since few proteins from
a proteome of thousands of different proteins can be highly
expressed, defining highly expressed proteins (or functionally
related groups of a few proteins) as those contributing >1%
of all proteins (Raven et al., 2013). This argument also applies
to coenocytic organisms, such as the glomeromycote arbuscular mycorrhizal fungi, some of which have very small (for
eukaryotes) genomes—as little as 15 Mbp (Hijri and Sanders,
2004a, b). This cannot necessarily be used to indicate that
arbuscular mycorrhizas represent a means of P acquisition
with minimal P use in DNA, since the small genome size may
well be offset by a greater number of nuclei per unit of fungal
biomass.
Soltis et al. (2003) point out that basal angiosperms, and
many of the basal members of major flowering plant clades,
have small genomes with a general increase through polyploidy and, especially, proliferation of retrotransposons,
although there are examples of decreasing genome size in
several clades. As well as these phylogenetic signatures in
genome size, there are ecophysiological considerations other
than P limitation related to small genome sizes; an example is
the smaller size of stomatal guard cells with smaller genomes,
permitting more rapid opening and closing even if the change
from Fick’s law diffusion to Knudsen diffusion (where the
width of stomatal pores is less than the mean free path of gas
molecules) requires smaller guard cells than are likely from
the smallest known plant genomes (Hodgson et al., 2010).
The conclusion on gene number and genome size is that there
is little evidence that economizing on P is a significant selective pressure in minimizing genome size, perhaps because of
the generally small fraction (≤5%) of non-storage P which
is in DNA (see Raven et al., 2013, the section ‘Differences
among P pools in autotrophs and chemoorganotrophs?’, and
the following paragraph).
The evolution of autotrophy in relation to phosphorus requirement | 4033
As a link to the discusson below on RNA, it should be
pointed out that RNA is typically thought to account for
most of the total nucleic acid pool. Bieleski (1973) gives a typical RNA:DNA ratio of 10 in young flowering plant leaves.
However, there are reports of much lower RNA:DNA ratios.
Lourenço et al. (1998) found a RNA:DNA ratio of ~1 in the
small marine cyanobacterium Synechococcus, and Bertilsson
et al. (2003) showed that P-limited cultures of the very small
marine cyanobacterium Prochlorococcus had over half of
the cellular P in DNA, so that the RNA:DNA ratio must be
significantly less than 1 to accommodate the very small cell
P allocation to phospholipid (van Mooy et al., 2006, 2009)
and the unknown allocation to low molecular mass watersoluble phosphate esters and metabolically active inorganic
phosphate. This large fraction of the nucleic acid pool which
is DNA in these very small cells might relate to the fraction
of the cell taken up by the admittedly very small genomes
(Raven, 1994; Raven et al., 2013), although such effects would
probably only be significant in smaller cells..
Reef et al. (2010) found a RNA:DNA ratio of 0.8–17 in
the vascular cambium, and 5.3–14.3 for leaves, of mangroves.
Reef et al. (2012) found RNA:DNA ratios of 1.5–8 for macroalgae as they were collected from a coral reef (i.e. an oligotrophic habitat), while the range became 6–8 for these algae
when nutrient enriched. Nicklisch and Steinberg (2009) measured RNA:DNA ratios in eukaryotic microalgae, but only
give relative units so true RNA:DNA ratios cannot be determined. In all three of these latter data sets, the RNA:DNA
ratio increases with growth rate, but there are interspecific differences in the RNA:DNA ratio as a function of growth rate.
Nucleic acids: RNA
The possibility here is that there is a smaller allocation of P
to rRNA relating to autotrophy if there is a smaller requirement for protein synthesis in some variants of autotrophy,
especially with highly expressed proteins such as Rubisco and
the family of apoproteins of light-harvesting pigments.
For the apoproteins of light-harvesting pigments, there is
a larger mass of protein per mole of chromophore in phycobiliproteins than when the chromophores are chlorophylls
and carotenoids (Raven, 1984a, b). This greater requirement
for apoprotein on a mole of chromophore basis means that,
other things being equal, phycobilin-containing algae (cyanobacteria, glacocystophytes, rhodophytes, and cryptophytes)
would be expected to have a greater allocation of RNA to
phycobiliprotein synthesis than is the case for apoproteins of
alage with chlorophyll and carotenoid chromophores. There
seem to be no data with which this prediction can be tested.
For Rubisco, an obvious case is that of oxygenic photosynthetic organisms using a CCM to supply CO2 to Rubisco at a
higher steady-state concentration than would be the case with
diffusive supply of CO2 from an air equilibrium bulk aqueous phase for aquatic organisms, or the bulk atmosphere for
terrestrial organisms. This permits, as shown below, a lower
content of Rubisco per unit total protein or per unit biomass
and, when the additional protein related to the CCM is offset against the decreased protein resulting from a much lower
expression of the photorespiratory C oxidation cycle, there is
a lower overall requirement for protein related to autotrophic
inorganic C assimilation.
The enzyme Rubisco is the core autotrophic carboxylase
in all oxygenic photosynthetic organisms, and is involved in
>99.5% of the inorganic C assimilated in primary producers (chemolithotrophs as well as photolithotrophs) (Johnston
et al., 2009; Raven, 2009) with a global annual flux of at least
10 Pmol CO2 handled by Rubisco (Field et al., 1998; Raven,
2009, 2013a). This flux requires a large quantity of Rubisco
as a result of the enzyme’s low substrate-saturated specific
catalytic rate on a protein mass basis, a CO2 affinity which
means that the carboxylation activity is not saturated with
CO2 at the present atmospheric concentration, and expression of oxygenase as well as carboxylase activities (Tcherkez
et al., 2006; Savir et al., 2010). There are major constraints
on how these properties can vary independently among the
range of Rubiscos (Tcherkez et al., 2006; Savir et al., 2010;
Young et al., 2012).
A major difference in the quantity of Rubisco in photosynthetic organisms results from a reliance on diffusive entry
of CO2 from the medium to Rubisco rather than the occurrence of one of a variety of CCMs. In general terms, the former mechanism (‘C3 physiology’) involves a large quantity of
Rubisco with a low substrate-saturated specific reaction rate
and a high affinity for CO2 and a high CO2:O2 selectivity, with
a total Rubisco CO2-saturated catalytic activity in the organism several times the rate of photosynthesis in the present
atmosphere which is set by the light-saturated catalytic capacity of the thylakoid reactions (Tcherkez et al., 2006; Raven
et al., 2012). The latter mechanism (CCM) involves a smaller
amount of Rubisco with a higher substrate-saturated specific
reaction rate, a lower affinity for CO2, and a lower CO2:O2
selectivity, and a total Rubisco CO2-saturated catalytic activity in the organism similar to that needed for the light-saturated rate of photosynthesis in the present atmosphere and
the light-saturated catalytic capacity of the thylakoid reactions (Tcherkez et al., 2006; Raven et al., 2012). Table 5, which
is an extended version of table 1 in Raven (2013a), gives the
Rubisco content for a range of oxygenic photosynthetic
organisms on the basis of Rubisco protein as a fraction of
total protein (for cyanobacteria and algae) and Rubisco N as
a fraction of total leaf N (for vascular land plants).
How can the leaf Rubisco N as a fraction of the total leaf
N values for the vascular plants in Table 4 best be compared
with whole-organism (whole-cell) Rubisco protein as a fraction of total protein values for cyanobacteria and algae?
Rubisco N on a leaf blade N basis is less than Rubisco N as a
fraction of leaf protein N, to an extent indicated in Spinacia
oleracea by the finding that, by the approximation that the
‘thylakoid’ plus ‘soluble’ N fractions as a minimum estimate
of protein, 0.6–0.8 of leaf N is protein, with the range relating to variation in N and light availability. Working in the
opposite direction, protein occurs in the rest of the plant
where the fraction of Rubisco protein, contributed by green
reproductive structures, aerial stems, and petioles, is much
less than in leaf blades. Equating, as a first approximation,
total organic N as an (over) estimate of protein, organic N in
4034 | Raven
Table 5. Rubisco protein as a fraction of total protein and Rubisco N as a fraction of total N for cyanobacteria, algae, and plants with
different inorganic C acquisition mechanisms
Organisms, carbon
assimilation pathway
Rubisco N as
fraction of total N
C3 physiology flowering plants
0.095–0.28 (leaves)
C3 physiology flowering plant
Chenopodium album
Five species of C3 physiology
flowering plants
C3 physiology conifer
Pinus pinaster
0.10–0.26 (leaves)
0.158–0.259 (leaves, low CO2)
0.120–0.217 (leaves, high CO2)
0.087–0.309 (leaves)
C4-NAD-ME dicotyledon
Amaranthus retroflexus
0.040–0.080
Seven C4-NAD-ME
monocotyledons
0.040–0.129; 0.042–0.084
Seven C4-NADP-ME
monocotyledons
0.043–0.088; 0.044–0.074
Cyanobacteria and microalgae,
probably all with CCMs
Microalgae: five species of marine
diatoms, two species of marine
prymnesiophytes, one species of
freshwater green, all with CCMs
Rubisco protein as
fraction of total protein
0.024–0.120 (0.16, 0.23)
0.02–0.06; <0.025
the rest of the plant is 0.22–0.40 and 0.34–0.53 of the total in
the plant in Ricinus communis (Allen and Raven, 1988) and
Phaseolus vulgaris (Allen et al. 1989); in each case, the range is
a function of the N source. The indications from this limited
comparison are that leaf blade organic N can be 0.47–0.78 of
the total plant organic N, while leaf blade protein is 0.6–0.8
of the total leaf N, so the expression of Rubisco N as a fraction of leaf blade N is likely to overestimate Rubisco N as
a fraction of total plant protein. However, a more detailed
analysis is needed before it can be concluded that the Rubisco
N:leaf N ratios needs downward correction for comparison
with the Rubisco:total organism protein ratios in Table 5.
A further complication is that there are processes in vascular plants which have no clear analogue in cyanobacteria and
microalgae: an obvious example is long-distance transport,
which adds to the protein budget of vascular plants.
While Table 4 has no data for plants using the crassulacean acid metabolism (CAM) variant of CCM, there are
estimates of Rubisco protein as a fraction of soluble protein (not total protein) in the facultative CAM bromeliad
Guzmania monostachia, with Rubisco as 0.35 of soluble
protein in plants expressing C3 and 0.43 in plants expressing CAM (Maxwell, 2002), although Rubisco protein on a
Comments
References
Lowest value is for a
shade-adapted plant
Range is for low to high
N supply to plants
Low CO2 300 ppm
High CO2 900 ppm
Lower values for older
needles low in the canopy,
higher values for younger
needles high in the canopy
First value is for low N
supply, second value is
for high N supply
First range is for low N
supply, second range is
for high N supply
First range is for low N
supply, second range is
for high N supply
Range is for 15 values, many
for high and low CO2 cultures
of the same organism, with
two higher values Some
values involve an assumed
value for chlorophyll a per
total pigment
Range is for nutrient-replete
laboratory cultures; values for
field cultures and nutrientlimited laboratory cultures are
all below 0.025
Evans (1989)
Sage et al. (1987)
Sage et al. (1989)
Warren and Adams (2001)
Sage et al. (1987)
Ghannoum et al. (2005)
Ghannoum et al. (2005)
Raven (1991)
Losh et al. (2013)
leaf area basis is lower in the CAM than in the C3 physiology
phenotype. Niewiasomska et al. (2011) examined the CAM
and C3 physiology phenotypes of the facultative CAM dicotyledon Mesembryanthemum crystallinum and found that the
CAM phenotype had a smaller fraction of soluble protein
from leaves contributed by Rubisco than did the C3 phenotype. Further work is clearly needed to provide data on CAM
plants which are comparable with the values in Table 4 for
other CCM organisms.
The smaller quantity of Rubisco in photolithotrophs using
CCMs is often reflected in a greater N-use efficiency (NUE;
defined, similarly to PUE defined above, as the rate of biomass gain per unit N in the organism) in C4 plants (Long,
1999; Sage and Zhu, 2011), and predicted (with support from
the very limited relevant data available) for cyanobacteria
and algae with CCMs (Raven, 1991; Raven et al., 2012). This
greater NUE means that the decreased content of Rubisco,
and of the machinery used in the photorespiratory C oxidation cycle(s) in organisms with CCMs than in C3 terrestrial
plants, is not quantitatively offset by the need for additional
N in the catalysts and structures required by CCMs. On the
basis of optimal allocation of P (as used above), it would
be predicted that less rRNA is needed in the production of
The evolution of autotrophy in relation to phosphorus requirement | 4035
proteins related to inorganic C acquisition and assimilation in
CCM organisms than in those with C3 physiology, so leaving
more for the synthesis and other proteins and hence a higher
overall PUE. This is considered below under ‘Differences in
phosphorus-use efficiency as a function of autotrophic pathway’, with the conclusion that there is no clear evidence for
such a higher PUE in C4 plants.
More work is needed, with due allowance made for phylogenetic bias among the organisms examined, since the possibility of a high PUE for photosynthetic organisms with
CCMs could have global biogeochemical implications, as will
now be outlined.
Consideration of the global quantitative significance of
Rubisco requires that the lower fraction of Rubisco in algae
and vascular plants with CCMs is considered in relation to
the contribution of these organisms to global primary productivity. Essentially all of the net marine planktonic primary
productivity of at least 47.5 P g C year–1 (Field et al., 1998;
Raven, 2009) involves organisms with CCMs, and CCMs
also function in many of the contributors to coastal benthic net primary productivity (~1 P g C year–1: Field et al.,
1998). CCMs are as common in polar as in warmer waters
(Mitchell and Beardall, 1996; Beardall and Roberts, 1999;
Raven et al., 2002a, b; Cassar et al., 2004; Tortell et al., 2006;
see also Henley et al., 2012), although the arguments as to
the smaller, or absent, competitive advantage of CCMs at
low temperatures used for terrestrial C4 plants (Long, 1999;
Sage and Zhu, 2011) also apply to marine primary producers (Raven et al., 2002a, b). On land, Field et al. (1998) suggest a net global primary productivity of 56 P g C year–1.
Global values for natural abundance stable C isotope fractionation suggest that 21% of terrestrial primary productivity
is due to C4 plants (Lloyd and Farquhar, 1994). Ehleringer
et al. (1997) based their conclusion that C4 plants contribute 18% to global terrestrial primary productivity on estimates of productivity per unit area of the various habitats,
the fraction of the productivity attributable to the C3 and C4
photosynthesis, and the total area of the habitats. The agreement between the two outcomes is extremely satisfactory (if
not surprising), granted the different assumptions involved.
There are non-C4 terrestrial plants with CCMs, namely CAM
vascular plants and terrestrial free-living and lichenized algae
and hornworts, which would have made a contribution to the
low C isotope discrimination (C4 plants) category. However,
in view of the possible variance of the estimates by Lloyd and
Farquhar (1994) and by Ehleringer et al, (1997), the 3% difference between the estimates should not be attributed to the
terrestrial photosynthetic organisms, other than C4 plants,
with CCMs.
The analysis in the preceding paragraph suggest that at
least half of global primary productivity involves CCMs, and
the organisms with CCMs have less Rubisco per unit biomass,
and a smaller fraction of total protein allocated to Rubisco,
than in C3 terrestrial plants. The values in Table 4 suggest
that the organisms with CCMs generally have less than half
of the Rubisco content of terrestrial C3 plants, so that the
global quantity of Rubisco is, then, less than three-quarters,
and possibly less than five-eighths, of what would been the
case if all primary producers using Rubisco had the quantity
of Rubisco found in C3 terrestrial plants. Again appealing
to optimal allocation of P, this lower global Rubisco content should decrease the global use of P in rRNA. However,
the limited data on PUE (as a surrogate for the allocation to
RNA, the major single use of metabolically active P in photosynthetic organisms of organisms with and without CCMs)
do not, as discussed above, show uniform differences between
the two metabolic types.
Could there be any alternative autotrophic CO2 assimilation pathways which function in oxygenic photosynthetic
organisms in the present atmosphere with a lower requirement for protein in the CO2 assimilation pathway, and hence
for P in rRNA? Of the five well-characterized alternatives
to Rubisco and the Benson–Calvin cycle (Raven, 2009; BarEven et al., 2010, 2012; Raven et al., 2012), some are ruled
out by their O2 sensitivity or their low affinity for inorganic C
(Raven, 2009; Bar-Even et al., 2010, 2012; Raven et al., 2012).
All of the alternative pathways have a lower running cost (mol
ATP and mol NADPH per mol CO2 assimilated) than the
Rubisco-based system when account is taken of the energetic
costs of the photorespiratory C oxidation pathway (PCOC)
related to diffusive CO2 entry, or of a CCM (Raven, 2009;
Bar-Even et al., 2010, 2012, Raven et al., 2012). An important
aspect of the energetic running costs is that organisms with
CCMs do not have complete suppression of the PCOC, for
example Lacuesta et al. (1997) for a C4 flowering plant and
Roberts et al. (2007) showing significant quantitative differences in the flux through glycolate and glycine in two species
of diatom from the genus Thalassiosira with indistinguishable
inorganic C concentration dependence of autotrophic inorganic C assimilation. Bar-Even et al. (2012) point out that
lower energy inputs mean that the overall reaction sequence
is closer to thermodynamic equilibrium, with a greater extent
of back-reactions and hence a greater need for enzyme protein than for a reaction sequence further from thermodynamic equilibrium. Bar-Even et al. (2012) also point out that
the enzymes with the lowest specific reaction sequence in the
pathway are typically carboxylases, so it is not just Rubiscobased autotrophic CO2 assimilation pathways which have a
high protein allocation to carboxylases. The enzyme phosphenolpyruvate carboxylase (PEPc), the widespread anaplerotic carboxylase, which is also, through gene duplication and
different regulatory properties of the resulting protein, the
initial carboxylase of C4 photosynthesis, has a high substratesaturated specific reaction rate; however, the content of PEPc
in leaves of the C4 plant Amaranthus retroflexus is half that
of Rubisco, although the PEPc value is not corrected for the
anaplerotic isoform of the enzyme (Sage et al., 1987).
In terms of the combination of CO2 affinity, O2 insensitivity, and energetic running costs, the most plausible alternative
to the Rubisco-based system is the 3-hydroxypropionate bicycle (Bar-Even et al., 2010, 2012). However, even this pathway would only represent a decrease in protein requirement
for a given rate of CO2 assimilation to about two-thirds of
that needed for the Rubisco-based system with the PCOC
(Bar-Even et al., 2010, 2012), and a rather smaller decrease in
protein requirement relative to a Rubisco-based system with a
4036 | Raven
CCM (Raven, 1991, 2013a; Long, 1999; Sage and Zhu, 2011).
Such decreases in protein requirement may mean, with optimal allocation of P, a lower rRNA content needed to support
protein synthesis, with a predicted increase in PUE. However,
we have seen above that the available data on PUE in organisms with and without CCMs did not support this optimalitybased prediction.
A further aspect of possible variations in RNA, and hence P,
costs among autotrophs is the rate of protein turnover: faster
protein turnover means a greater requirement for P in RNA.
Bulk protein turnover rates in oxygenic photolithotrophs
were measured, and reviewed, by Quigg and Beardall (2003).
Pre-proteomic analyses of turnover of individual proteins
focused on the D1 protein (psbA gene product) and associated
proteins in the reaction centre of photosystem II (Quigg and
Beardall, 2004; Raven, 2011, 2012) and to Rubisco in C3 and
C4 plants (Ferreira and Davies, 1987a, b; Esquivel et al., 1998;
Irving and Robinson, 2006a, b). Proteomics has allowed the
simultaneous measurements of the turnover of ≥10 proteins
in prasinophycean (Ostreocococcus: Martin et al., 2012) and
chlorophycean (Chlamydomonas: Mastrobuoni et al., 2012)
green microalgae. Minimizing the RNA, and hence P, cost
of protein turnover requires as constant as possible an environment, for example the absence of high irradiance episodes
that could cause photodamage to photosystem II (e.g. the
strains of Prochlorococcus that live at the deep chlorophyll
maximum of low-latitude low-P regions of the ocean; Raven
et al., 2013).
Clearly, more work is needed to test the relationship
between growth rate, protein turnover rate, and rRNA content. However, it is important to recall that the relationship
of rRNA content and net protein synthesis rate in algae and
cyanobacteria does not, in many cases, agree with the growth
rate hypothesis and optimal allocation of P (Nicklisch and
Steinberg, 2009; Flynn et al., 2010), in contrast to many chemoorganotrophic microorganisms (Karpinets et al. 2006).
Low molecular mass water-soluble phosphate esters
We have seen in the comparison of autotrophs with chemoorganotrophs that the six most common autotrophic CO2
fixation pathways found in Archaea, Bacteria, and Eukarya
(Raven, 2009; Bar-Even et al., 2010, 2012; Raven et al., 2012)
have a variable number, sometimes zero, of low molecular
mass water-soluble phosphate esters that are unique to that
pathway. The discussion above suggests that this variation is
likely to have a very small impact on the total low molecular
mass water-soluble phosphate ester component of the total P
requirement of autotrophic organisms. C4 and CAM plants
have the involvement of the additional C3 carboxylic acid
ester phosphoenolpyruvate in the autotrophic inorganic C
assimilation mechanism compared with the photosynthetic
carbon reduction cycle with diffusive CO2 entry to Rubisco
or a CCM based on active transport across membranes of
an inorganic C species or of protons. However, phophoenolpyruvate is also produced universally in oxygenic photolithotrophs in glycolysis or by phosphate,pyruvate dikinase,
and consumed in the production of pyruvate or in anaplerotic
aspects of anabolism, so it is ubiquitous in the cells of oxygenic photolithotrophs.
It is therefore unlikely that the difference among CCMs,
and between CCMs and CO2 diffusion, will be reflected in a
significant influence on the pool size of low molecular mass
water-soluble phosphate esters.
Differences in phosphorus-use efficiency
as a function of autotrophic pathway
There are very few data sets obtained with the objective of
investigating PUE of the metabolism or growth of autotrophs
as a function of the autotrophic pathway used, and they all
involve the comparison of C3 and C4 flowering plants. Halsted
and Lynch (1996) examined 12 species, four C4 monocots (all
grasses, of which only three species were used in comparisons
because the other showed extreme P limitation responses),
two C3 monocots (both grasses), a C3–C4 intermediate monocotyledon (a grass), two C4 dicotyledons, two C3 dicotyledons, and a C3–C4 intermediate dicotyledon. The C3 species
had a larger dry weight than the C4 species when grown for
7.5 weeks under P-replete conditions, but there was no significant difference in DM accumulation under P-limiting conditions. No whole-plant P concentrations are given, so PUE as
the rate of DM gain per unit plant P could not be calculated.
However, the growth rate of C4 plants was less limited by P
deficiency than was the case for C3 plants. Photosynthetic
PUE (PPUE), with the units of μmol CO2 fixed s–1 g–1 (or
mol–1) leaf P, showed no significant difference between C3 and
C4 plants. The main conclusion from this work was that photosynthesis of the grasses is more P efficient than that of the
dicotyledons.
Other data on C3 and C4 grass responses to P limitation are
provided by Ghannoum and Conroy (2007) and Ghannoun
et al. (2008). Ghannoum and Conroy (2007) examined a C3
grass and two C4 grasses. While P deficiency decreased the
plant dry mass accumulation rate to a similar extent in all
three species, PPUE was higher in the two C4 grasses than
in the C3 grass; the reverse was true of the rate of DM accumulation per unit plant P. Ghannoum et al. (2008) examined
four tropical grasses, one of which was C3 and the others were
C4, with one each from the NAD-ME (NAD-malic enzyme),
NADP-ME (NADP-malic enzyme), and PEPCK (phosphoenolpyruvate carboxykinase) subtypes. The C4 grasses had
higher PPUE, expressed on the basis of leaf P, as was found
by Ghannoum and Conroy (2007), but contrasting with the
findings of Halsted and Lynch (1996).
Mantiana et al. (2008) measured PPUEs for C4 plants
from an African savannah/wetland mosaic; values were significantly higher than the mean for C3 plants (Wright et al.,
2004), but the authors caution that a larger C4 data set is
needed to examine further the effects of P on C4 relative to
C3 photosynthesis. While the work of Yu et al. (2012) only
deals with one C3 and one C4 perennial grass species (and an
annual C3 chenopod), the PUE for the C4 species is less than
that of the C3 species for growth under both P-limited and
N-limited conditions.
The evolution of autotrophy in relation to phosphorus requirement | 4037
Attempts were made to estimate the PUE of growth of
natural populations of three freshwater isoetid life form vascular plants, Isoetes lacuustris and Litorella uniflora, which
take up almost all of their CO2 through their roots and use
CAM, and Lobelia dortmanna, which also takes up almost
all of its CO2 through the roots but uses C3 photosynthesis
(Richardson et al., 1964; Christensen and Sand-Jensen, 1998;
Christensen et al., 1998). However, the wide variations in P
concentration and problems with estimating growth rates
under natural conditions meant that no significant differences
could be found within the high variance data sets. Comparing
the [14C]inorganic C assimilation rates per unit plant P in the
dark plus light for the natural populations of the C3 L. dortmanna and the CAM I. lacustris does not yield significant differences, again within the rather large variances (Richardson
et al., 1984).
There is clearly a need for more work, using both total
plant P and leaf P, as well as measurements of the four main
P pools, for a more detailed analysis of the determinants of
variations of PPUE and the PUE of growth in terms of the
relationship of photosynthetic and growth rates to nucleic
acid P, phospholipid P, low molecular mass water-soluble
phosphate esters and metabolically active inorganic phosphate, and stored (metabolically inactive) phosphate. The
interpretations could be related to the possible differences in
uses of the various P pools in C3 and C4 plants, as discussed
above. In particular, we note the absence of clear evidence
of an increased P paralleling the higher NUE in many photolithotrophs with CCMs, which would be predicted from
the decreased requirement for rRNA if less protein has to be
made.
Water-use efficiency of vascular plants
with different photosynthetic pathways
under P limitation
Water and P can co-limit plant growth, so an important
aspect of P deficiency in terrestrial plants is its relationship to
water loss per unit DM gain. The typical response of nutrient
deficiency in terrestrial vascular plants is a decrease in wateruse efficiency (WUE), measured as g DM gain g–1 water lost
in transpiration (Raven et al., 2004). While almost all of the
data relate to N deficiency, there are some data indicating a
decrease in WUE in P-limited C3 plants (Raven et al., 2004).
There are some additional data on P deficiency effects on
WUE. Conroy et al. (1988) examined the effect of P limitation of growth rate on WUE efficiency of the C3 tree Pinus
radiata. Limiting P supply decreased WUE at all three levels of water supply and at both CO2 levels (330 and 660 mm3
DM–3). Brück et al. (2000) examined the C4 Pennisetum
glaucum (pearl millet), and found a decreased WUE with P
limitation of growth for well-watered, but not for droughted,
plants. Singh et al. (2008) found that P limitation decreased
the WUE of the C3 herb Trifolium repens. Cernusak et al.
(2011) examined 19 C3 woody tropical tree seedlings and a
liana, and found a negative correlation between shoot P:C
ratio and WUE; that is, the reverse of the ‘typical’ response
of a decreased WUE with decreasing plant P concentration.
Cernusak et al. (2011) attributed their findings to a P movement to the root surface driven by transpirational water flow.
A tentative conclusion is that the conflicting results arise from
situations in which processes internal to the plant dominate
and WUE decreases with plant P concentration, and situations where mass flow of P with the transpiration stream
external to the plant dominate, when WUE increases with
decreasing plant P concentration.
Phosphorus effects on the expression
of CCMs
There is some information on the effects of P availability on
the expression of CAM in facultative CAM plants, which
can express either C3 or CAM photosynthesis, as well as on
algae in relation to the expression of CCMs. C4 photosynthesis in land plants is constitutive, so there are no parallel
data sets on the effects of P availability on the extent of C4
expression.
Starting with algae, Kozłowska-Szerenos et al. (2004)
examined the effects of P deficiency on the rate of photosynthetic oxygen production under a range of light and CO2
levels in the green (trebouxiophycean) freshwater microalga
Chlorella vulgaris. The cells were grown in originally airequilibrated media at pH 6.8 in vessels, which were subsequently sealed with bacteriological cotton wool, and with no
subsequent attempts to re-equilibrate the cultures with the
atmosphere, so the culture medium was at or below air equilibrium CO2 during cell growth. While P deficiency decreased
the initial slope (α) of the photosynthetic rate on a chlorophyll basis versus an irradiance relationship, the light-saturated rate of photosynthesis was higher in the low-P cells.
The affinity of whole-cell photosynthesis for inorganic carbon was higher in the low-P than in the control cells, with
K0.5 values significantly decreased by ~30% in the low-P cells
(Table 6).
Beardall et al. (2005) grew the green (trebouxiophycean)
freshwater microalga Chlorella emersonii at a range of P concentrations in air-equilibrated culture media. In contrast to
the results of Kozłowska-Szerenos et al. (2004) with C. vulgaris, the K0.5 increased (affinity decreased) with increasing P
limitation, although this was only significant at the lowest P
level. Beardall et al. (2005) also measured the accumulation
ratio (internal:external) for CO2 achieved by the CCM, and
found that this was more than halved at the lowest relative
to the highest P treatment. A less effective CCM in the low-P
cells is also indicated by the increased inhibition of photosynthesis by oxygen (air equilibrium versus a quarter of air
equilibrium) for low-P than for higher P cells, and also the
significantly more negative δ13C values of cell organic matter
for the cells grown under lower relative to higher P supply. On
all four of the grounds tested, the CCM was less effective in
low-P cultures (Table 6).
Xu and Gao (2009) examined the red (florideophycean)
marine macroalga Gracilaria lemanaeformis. The thalli
were grown in air-equilibrated seawater at two P levels. The
4038 | Raven
Table 6. Effects of phosphorus limitation on the expression of CCMs in a cyanobacterium, an algae, and facultative CAM flowering
plants
Organism
CCF
K0.5 CO2
O2 inhibition
δ13C
CAM ΔpH
References
Cylindrospermospsis raciborskii
(Cyanobacterium)
Chlorella vulgaris
(Trebouxiophyceae)
Chlorella emersonii
(Trebouxiophyceae)
Chlamydomonas acidophila
(Chlorophyceae)
Gracilaria lemanaeiformis
(Florididiophyceae)
Mesembryanthemum crystallinum
(Magnoliophytopsida)
Clusia minor (Magnoliophytopsida)
NM
Decrease
NM
NM
NA
Wu et al. (2012)
NM
Decrease
NM
NM
NA
Decrease
Increase
Increase
NA
Decrease
Increase
NM
Decrease
(more –ve)
NM
Kozłowska-Szerenos
et al. (2004)
Beardall et al. (2005)
NA
NM
Increase
NM
NM
NA
NM
NM
NM
NM
Increase
Spijkerman (2011);
Spijkerman et al. (2011)
Xu and Gao (2009);
Xu et al. (2010)
Paul and Cockburn (1990)
NM
NM
NM
NM
Increase
Maiquetía et al. (2009)
NM, not measured; NA, not applicable.
experimental treatment also had UV-A and UV-B treatments,
but for the controls [photosynthetically active radiation
(PAR) alone] the K0.5 values for total inorganic carbon were
about doubled in the low-P treatments. Xu et al. (2010) also
used G. lemanaeiformis, this time grown in air levels of CO2
(370 μl l–1) or high CO2 (720 μl l–1) combined with natural
seawater (0.5 mmol m–3) and high (30 mmol m–3) inorganic
phosphate concentrations. In the low P treatments, the K0.5
for total inorganic C was higher at the lower CO2 treatment,
with no effects of CO2 alone on the inorganic C K0.5 value.
Specific growth rate measurements show a significantly lower
value for the low-CO2 low-P treatment than for the other
three combinations, among which there were no significant
differences (Table 6).
Spijkerman (2011) and Spijkerman et al. (2011) used
the acidophilic freshwater (chlorophycean) green alga
Chlamydomonas acidophila grown in a factorial experimental design with varied supplies of CO2 and P; the work
also used Fe treatments which are not considered here. The
inorganic C-concentration factor (inside:outside) or CCF
increased with the P content of the cells, in both the lowand high-CO2 treatment. The highest CCFs were found in
low-CO2-grown cells with high P supply, with the lowest
CCFs for CO2- and P-co-limited cells. The CCF values did
not correlate well with the affinity for external inorganic C
in photosynthesis since CO2-limited cells always had a low
K0.5 for CO2. Spijkerman (2011) suggests that the CCF correlates with inorganic C porter density in the plasmalemma,
since a high CCF parallels a high potential for photosynthesis, and does not parallel the affinity of the porters for
inorganic C. Spijkerman et al. (2011) showed that there is
kinetically independent co-limitation for CO2 and for P in
C. acidophila (Table 6).
The final relevant data set on algae and cyanobacteria is that of Wu et al. (2012), who investigated inorganic C–P interactions in the freshwater cyanobacterium
Cylindrospermopsis raciborski, again using a factorial experimental design. Low C and low P gave the lower specific
growth rate, and a significantly higher rate for high C and
low P, and even (significantly) higher rates for the two highP treatments. In this case, the K0.5 for inorganic C only varies
with P supply, with a significantly higher K0.5 at the higher P
concentrations, with no effect of inorganic C supply within
a P treatment. These results resemble those of KozłowskaSzerenos et al. (2004), namely a high affinity for inorganic
C at the lower P concentration, but not the other data sets
(Table 6).
The conclusion is that, in the two cases where it was
measured, the CCF decreased with P limitation, and in the
majority of cases examined the affinity for external inorganic
carbon decreased with P limitation (Table 6). A rather different approach was taken by Garcia et al. (2013) who measured
cellular growth rates at high light and low light combined
with high (2100 AD?) CO2 and low (last glacial maximum)
CO2 as a function of P supply. The maximum growth rate of
P-replete increases in the order low light–low CO2, low light–
high CO2, high light–low CO2, high light–high CO2 (Garcia
et al., 2013). These authors also measured the CO2 and N2
fixation rates per unit cell P under the four sets of light and
CO2 conditions, and found that both rates were highest for
P-replete cells in high light–high CO2, with all other treatments giving lower values. Garcia et al. (2013) suggest that
high light–high CO2 permits a greater allocation of energy to
processes other than the CCM.
The other potential role of P in acclimation of inorganic
C acquisition is in facultative CAM plants, which can switch
from C3 to CAM. Paul and Cockburn (1990) examined the
interaction of phosphate and nitrate with salinity in the
expression of CAM in Mesembryanthemum crystallinum. P
deficiency led to some CAM expression in the absence of
salinity treatment, and also increased CAM expression after
induction by salinity treatment (Paul and Cockburn, 1990)
(Table 5). Maiquetía et al. (2009) showed that P deficiency
promoted the induction of CAM by drought in seedlings of
Clusia minor; addition of mycorrhizal fungi, or of phosphate,
did not promote the induction of CAM by drought (Table 5).
The evolution of autotrophy in relation to phosphorus requirement | 4039
A final point concerns the effects of P deficiency on the
constitutively expressed C4 CCM in terrestrial vascular plants
(Jacob and Lawlor, 1991, 1992). P deficiency decreases the
maximum (light- and CO2-saturated) rate of photosynthesis,
and the carboxylation efficiency (increment of photosynthetic rate per unit increase in external CO2 concentration)
resulting from effects on metabolism in photosynthetic cells
rather than effects on diffusion through stomata (Jacob and
Lawlor, 1991, 1992, 1993a, b). The decreased photosynthetic
capacity is not a function of limitation by electron transport, but rather of ATP availability in regenerating ribulose1,5-bisphosphate (Jacob and Lawlor, 1993b). This decreased
affinity for inorganic C under P deficiency resembles many of
the effects noted for cyanobacteria and algae. However, similar effects are found for C3 terrestrial vascular plants, albeit
based on a lower P-replete CO2 affinity than is the case for C4
plants (Jacob and Lawlor, 1991, 1992).
The discussion above shows that P deficiency decreases the
CCF in the two algae with CCMs that have been tested, and
that most data sets show decreased affinity for external inorganic C in photosynthesis in cyanobacteria and algae. P deficiency increases the expression of CAM in facultative CAM
plants, either alone or in conjunction with decreased water
availability. The CO2 affinity of both C3 and C4 terrestrial
flowering plants is decreased by P deficiency.
Conclusions on differences on phosphorus allocation
among photolithotrophs
Most evidence shows no greater PUE in C4 than in C3 plants.
The effect of P availability on the expression of CCMs is variable; CAM expression is increased by P deficiency, while in most
cases algal CCMs show lower expression under P deficiency.
Interactions of phosphorus and
photosynthetically active radiation in
organisms with CCMs
Hessen et al. (2002) examined the growth of cultures of the
freshwater green alga Selenastrum capricornutum under a
range of P concentrations and of PAR levels. The cell C:P
ratio decreased with increasing external phosphate and/or
decreasing PAR, so that the PUE is lower for growth at low
PAR. Hill et al. (2009) examined growth of diatom-dominated communities of stream algae in a range of PAR and
P treatments. The effects of external P were decreased at low
PAR, and PAR effects were decreased at low external P, but
there was only weak evidence of PAR–P interactions. As
with the work on Selenastrum, PUE is lower at low growth
PAR, although there is the complication of changes in species composition of the communities. A decreased PUE
with decreasing growth rate as a function of availability
of a non-P resource is to be expected, since there is a baseline content of P-containing components (e.g. membrane
lipids) for survival. The absence of clear evidence for kinetic
interaction of P and PAR recalls the absence of such interactions between P and CO2 in Chlamydomonas acidophila
(Spijkerman et al., 2011). P-deficient cells have a lower PAR
use efficiency, as indicated in the work discussed in this paragraph, in Kozłowska-Szerenos et al. (2004), and in the work
of Jacob and Lawlor (1991) on flowering plants with and
without CCMs.
Conclusions
The interactions in evolution between autotrophy and P
availability are complex, but some conclusions can be drawn.
The PUE of autotrophs (specifically photolithotrophs)
is lower than that for chemoorganotrophs in comparisons
of the fastest-growing Bacteria and Eukarya of the two
trophic modes that are otherwise as closely alike as possible. However, there is a very wide range of PUEs in each
trophic mode and so a large range of overlap. More experiments are needed to measure specific growth rates and C:P
ratios in parallel for a range of organisms of each trophic
mode, each under a range of P supplies, to test further the
hypothesis of a higher PUE in chemoorganotrophs than in
autotrophs. Among rapidly growing organisms, the higher
PUE in chemoorganotrophs is unlikely to be a result of a
smaller P content in DNA because of the small fraction of
P in DNA, and the finding that the smaller genome size of
autotrophic than of chemoorganotrophic bacteria is not
necessarily the case in eukaryotes. Chemoorganotrophs generally have closer to optimal allocation of P to rRNA than
do photolithotrophs, especially under P-limited conditions,
but there is a greater occurrence of non-phospholipid polar
lipids in membranes in oxygenic photolithotrophs than in
many chemoorganotrophs.
Among photolithotrophs, there seem to be no data available
for comparison of PUE of growth between anoxygenic and
oxygenic bacterial photolithotrophs, although the resourcesaturated maximum specific growth rate of the anoxygenic
photolithotroph Chlorobium sp. exceeds that of oxygenic photolithotrophic cyanobacteria. There is little possibility that
the different numbers of low molecular mass water-soluble
phosphate esters specific to the six well-described pathways
of autotrophic inorganic carbon assimilation alter the PUE
of autotrophic inorganic carbon assimilation or of growth.
The data on PUE of growth and of photosynthesis for C3
and C4 flowering plants give very limited support for a greater
PUE of C4 than C3 plants, though the major difference noted
was the greater PUE of grasses than dicotyledons independent of C fixation pathway, correlated with the site of P storage. The WUE of growth of terrestrial photolithotrophs is
generally decreased by P limitation. For photolithotrophs
with facultative expression of CCMs, P limitation increases
CAM expression in combination with other predisposing factors such as salinity or drought. For algae and cyanobacteria,
the most complete data sets show a decrease in the CCF and
the affinity for extracellular inorganic C when P is limiting,
but some data on other organisms show a higher inorganic
C affinity under P limitation. PUE is decreased in organisms
with decreased supply of (non-P) resources, such as CO2 and
PAR.
4040 | Raven
Glossary
Anoxygenic photolithotroph. A photolithotroph which
uses an electron donor other than H2O (e.g. H2, H2S, Fe2+)
as electron donor and generates, respectively, no oxidant,
S or SO42–, and Fe3+. Only obligate in non-cyanobacterial
Bacteria. This paper shows that anoxygenic photolithotrophs have lower phosphorus-use efficiencies than do
aerobic osmochemoorganotrophs.
Autotroph. An organism using exergonic inorganic chemical reactions, or photons, as the energy source for growth,
and inorganic chemicals taken up on a molecule by molecule
basis across the plasmalemma to supply nutrient elements
Chemolithotroph. An organism using the catalysis of
exergonic inorganic chemical reactions as the energy
source for growth, and inorganic chemicals taken up on
a molecule by molecule basis across the plasmalemma
to supply nutrient elements. Only found in Archaea and
Bacteria. This paper shows that anoxygenic chemolithotrophs have lower phosphorus-use efficiencies than do
aerobic osmochemoorganotrophs.
Chemoorganotroph. An organism using the catabolism of
organic compounds as their energy source for growth, and
organic compounds as the source of carbon and, in many
cases, the source of nitrogen. Heterotroph is a synonym of
chemoorganotroph.
Mixotroph. An organism combining autotrophy and
chemoorganotrophy; usually applied to organisms which
combines phototrophy with chemoorganotrophy.
Osmochemoorganotroph. A chemoorganotrophic organism taking up organic and inorganic nutrients on a
molecule by molecule basis across the plasmlamma
as the means of acquiring carbon and other elements.
Saprochemoorganotroph is a synonym of osmochemoorganotroph. This paper shows that osmochemoorganotrophs have lower phosphorus-use efficiencies than do
autotrophs.
Oxygenic photolithotroph. A photolithotrophic which uses
H2O as electron donor and generates O2. This paper shows
that oxygenic photolithotrophs have lower phosphorususe efficiencies than do aerobic osmochemoorganotrophs.
Phagochemoorganotroph. A chemoorganotrophic organism taking up particles of live or dead organic matter as
the means of acquiring carbon and other elements; limited to Eukarya. This paper shows that phagochemoorganotrophs have lower phosphorus-use efficiencies than
do osmochemoorganotrophs and (marginally) oxygenic
photolithotrophs.
Photolithotroph. An organism using photons as the
energy source for growth and inorganic chemicals taken
up on a molecule by molecule basis across the plasmalemma to supply nutrient elements.
Comments from Professor Hans Lambers on an earlier version of this manuscript have been invaluable. The comments
of two anonymous reviewers, and of the editor (Dr Colin
Osborne), have greatly improved the manuscript. The
University of Dundee is a Scottish Registered Charity, No.
SC15096.
References
Abraham W-R, Strompi C, Bennasar A, Vanncaneyt M,
Snauwaert C, Swings J, Smit J, Moore ERB. 2002. Phylogeny of
Maricaulis Abraham et al., 1999 and proposal of Maricaulis virginiensis
sp.nov., M. parjimensis sp. nov., M. washingtonesns sp.nov and
M. salignorans sp. nov. International Journal of Systematics and
Evolutionary Microbiology 52, 2191–2201.
Allen S, Raven JA. 1987. Intracellular pH regulation in Ricinus
communis grown with ammonium or nitrate as N source: the role of
long distance transport. Journal of Experimental Botany 38, 580–596.
Allen S, Raven JA, Sprent JI. 1988. The role of long-distance
transport in intracellular pH regulation in Phaseolus vulgaris grown
with ammonium or nitrate as nitrogen source, or nodulated. Journal of
Experimental Botany 39, 513–528.
Atkinson MR, Smith SV. 1983. C:N:P ratio of benthic marine plants.
Limnology ad Oceanography 28, 568–584.
Bar-Even A, Noor E, Lewis NE, Milo R. 2010. Design and analysis
of synthetic carbon fixation pathways. Proceedings of the National
Academy of Sciences, USA 107, 8889–8894.
Bar-Even A, Noor E, Milo R. 2012. A survey of carbon fixation
pathways through a quantitative lens. Journal of Experimental Botany
63, 2325–2342.
Beardall J, Roberts S. 1999. Inorganic carbon acquisition by two
Antarctic macroalgae, Porphyra endiviifolia (Rhodophyta, Bangiales)
and Palmaria decipiens (Rhodophyta, Palmariales). Polar Biology 21,
310–315.
Beardall J, Roberts S, Raven JA. 2005. Regulation of inorganic
carbon assimilation by phosphorus limitation in the green alga
Chlorella emersonii. Canadian Journl of Botany 83, 859–864.
Bec B, Collos Y, Vaquer A, Mouillot D, Souche P. 2008.
Growth rate peaks at intermediate size in marine photosynthetic
picoeukaryotes. Limnology and Oceanography 53, 863–867
Beers JR. 1966. Studies on the chemical composition of the major
zooplankton groups in the Sargasso Sea off Bermuda. Limnology and
Oceanography 11, 520–328.
Bekker A, Holland HD, Wang PL, Rumble D III., Stein H J,
Hannah JL, Coetzee LL, Beukes NJ. 2004. Dating the rise of
atmospheric oxygen. Nature 427, 117–120.
Bellinger BJ, van Mooy BAS. 2012. Nonphosphorus lipids in
periphyton reflect available nutrients in the Florida Everglades, USA.
Journal of Phycology 49, 303–311.
Acknowledgements
Benning C. 1998. Biosynthesis and function of the sulfolipid
sulfoquinovosyldiacylglycerol. Annual Review of Plant Physiology and
Plant Molecular Biology 49, 55–75.
Discussions with Professors Kingsley Dixon, R. John Ellis,
Hans Lambers and George Lorimer have been very helpful.
Benning C, Beatty IT, Prince RC, Sommerville CR. 1993.
The sulfolipid sulfoquinovosyldiacylglycerol is not required for
The evolution of autotrophy in relation to phosphorus requirement | 4041
photosynthetic electron transport in Rhodobacter sphaeroides but
enhances growth under phosphate limitation. Proceedings of the
National Academy of Sciences, USA 90, 1561–1565.
Benning C, Huang Z-H, Gage DA. 1995. Accumulation of a novel
glycolipid and betaine lipid in cells of Rhodobacter sphaeroides under
phosphate limitation. Archives of Biochemistry and Biophysics 317,
103–111.
Berendse F, Aerts R. 1987. Nitrogen use efficiency. A biological
meaningful definition? Functional Ecology 1, 293–296.
Berner EK, Berner RA. 2012. Global environment: water, air and
geochemical cycles, 2nd edn. Princeton, NJ: Princeton University
Press.
Bertilsson S, Berglund O. Karl DM, Chishold SW. 2003. Elemental
composition of marine Prochlorococcus and Synechococcus:
implications for the ecological stoichiometry of the sea. Limnology and
Oceanography 48, 1721–1731.
Bieleski RL. 1968a. Levels of phosphate esters in Spirodela. Plant
Physiology 43, 1297–1308.
Chapin FS, Bieleski RL. 1982. Mild phosphorus stress in barley and
related low-phosphorus-adapted barleygrass phosphorus fractions
and phosphorus absorption and growth. Physiologia Plantarum 54,
309–317.
Chapin FS III, Follett JH, O’Connor KF. 1982. Growth, phosphate
absorption and phosphorus chemical forms in two Echinochloa
species. Journal of Ecology 70, 305–321.
Chen BZ, Liu HB. 2010. Relationship between phytoplankton growth
and cell size in surface oceans: interactive effect of temperature,
nutrients and grazing. Limnology and Oceanography 55, 965–972.
Chen BZ, Liu HB. 2011. Comment: unimodal relationships between
phytoplankton mass-specific growth rate, and size: a reply to the
comment of Sal and Polez-Umitia. Limnology and Oceanography 56,
1956–1958.
Christensen KK, Sand-Jensen K. 1998. Precipitated iron and
manganese plaques restrict root uptake of phosphorus in Lobelia
dortmanna. Canadian Journal of Botany 76, 2158–2163.
Bieleski RL. 1968b. Effect of phosphorus deficiency on levels
of phosphorus compounds in Spirodela. Plant Physiology 43,
1309–1316.
Christensen KK, Jensen HS, Anderson FØ,Wigand C, Holmer M.
1998. Interference between root plaque formation and phosphorus
availability for isoetids in sediments of oligotrophic lakes. Hydrobiologia
43, 107–128.
Bieleski RL. 1973. Phosphate pools, phosphate transport and
phosphate availability. Annual Review of Plant Physiology 24,
225–252.
Cleveland CC, Liptzin D. 2007. C:N:P ratio in soils: is there a
‘Redfield Ratio’ for the microbial biomass? Biogeochemistry 85,
235–252.
Bjerrum CJ, Canfield DE. 2002. Ocean productivity before about
1.9 Gyr limited by phosphorus adsorption onto iron oxides. Nature
417, 159–162.
Connolly JA, Oliver MJ, Beaulieu JM, Knight CA, Tomanek L,
Moline MA. 2008. Correlated evolution of genome size and cell volume
in diatoms (Bacillariophyceae). Journal of Phycology 44, 124–131.
Blake RE, Chang SJ, Lepland A. 2010. Phosphate oxygen isotope
evidence for a temperate and biologically active Archean ocean.
Nature 464, 1029–1033.
Conroy JP, Küppers M, Küppers B, Virgaona J, Barlow EWR.
1988. The influence of CO2 enrichment, phosphorus deficiency and
water stress on the growth, conductance and water use of Pinus
radiata D. Don. Plant, Cell and Environment 11, 91–98.
Blank CE, Sanchez-Baracaldo P. 2010. Timing of morphological
and ecological innovations in the cyanobacteria—a key to
understanding the rise of atmospheric oxygen. Geobiology 8, 1–23.
Broadley MR, Bowen HC, Cotterill HL, Hammond JP, Meacham
MC, Mead A, White PJ. 2004. Phylogenetic variation in the shoot
mineral concentration of angiosperms. Journal of Experimental Botany
55, 321–336.
Brück H, Payne WA, Sattelmacher B. 2000. Effects of phosphorus
and water supply on yield, transpirational water-loss efficiency, and
carbon isotope discrimination of pearl millet. Crop Science 40,
120–125.
Cassar N, Laws EA, Bidigare RP, Popp BN. 2004. Bicarbonate
uptake by Southern Ocean phytoplankton. Global Biogeochemical
Cycles 18, GB2003.
Cedergreen RA, Hollingsworth RI. 1994. Occurrence of
sulfoqunivosyl diacylglycerol in some members of the family
Rhizobiaceae. Journal of Lipid Research 35, 1452–1461.
Cernusak LA, Winter K, Turner BL. 2011. Transpiration modulates
phosphorus acquisition in tropical tree seedlings. Tree Physiology 31,
878–885.
Chan L-K, Newton RJ, Sharma S, Smith CB. Rayapati P,
Limando AJ, Meiel C, Moran MA. 2012. Transcriptional changes
underlying elemental stoichiometry shifts in a marine heterotrophic
bacterium. Frontiers in Microbiology 3, 159.
Cotner JE, Hall EK, Scott IT, Heldal M. 2010. Freshwater bacteria
are stoichiometrically flexible with a nutrient composition similar to
seston. Frontiers in Microbiology 1, 32.
Crowe SA, Jones C, Katser S, et al. 2008. Photoferrotrophs thrive
in an Archean ocean analogue. Proceedings of the National Academy
of Sciences, USA 105, 15938–15943.
DeLong JP, Pike JD, Moses MER, Sibly RM, Brown JH. 2010.
Shifts in metabolic scaling across major transitions of life. Proceedings
of the National Academy of Sciences, USA 107, 12941–12945
Duarte CM. 1990. Seagrass nutrient content. Marine Ecology
Progress Series 67, 201–207.
Dyhrman ST, Jenkins BG, Rynearson TA, et al. 2012.
The transcriptome and proteome of the diatom Thalassiosira
pseudonana reveal a diverse phosphorus stress response. PLoS
One 7, e33768.
Ehleringer JR, Cerling TE, Helliker BR. 1997. C4 photosynthesis,
atmospheric CO2 and climate. Oecologia 112, 285–299.
Ellis RJ. 1979. Most abundant protein in the world. Trends in
Biochemical Sciences 4, 241–244.
Esquivel MG, Ferreira RB, Teixeira AR. 1998. Protein degradation
of C3 and C4 plants with particular reference to ribulose bisphosphate
carboxylase and glycolate oxidase. Journal of Experimental Botany
49, 807–816,
4042 | Raven
Evans JR. 1989. Photosynthesis and nitrogen relationships in leaves
of C3 plants. Oecologia 78, 9–19.
nitrogen-use efficiency in NADP-malic enzyme relative to NAD-malic
enzyme C4 grasses. Plant Physiology 137, 638–650.
Fagerbakke KM, Heldal M, Norland S. 1996. Content of carbon,
nitrogen, oxgyen, sulphur and phosphorus in native aquatic and
cultured bacteria. Aquatic Microbial Ecology 10, 15–27.
Ghannoum O, Paul MJ, Ward JI, Beale MH, Corol D, Conroy
JP. 2008. The sensitivity of photosynthesis to phosphorus deficiency
differs between C3 and C4 tropical grasses. Functional Plant Biology
35, 213–221.
Falkowski PG. 2000. Rationalising elemental ratios in unicellular
algae. Journal of Phycology 36, 3–6.
Falkowski PG, Katz MT, Knoll AH, Quigg A, Raven JA,
Schofield O, Taylor FJR. 2004. The evolution of modern eukaryotic
phytoplankton. Science 395, 354–360.
Falkowski PG, Raven JA. 2007. Aquatic photosynthesis , 2nd edn.
Princeton, NJ: Princeton University Press.
Ferreira RB, Davies DD. 1987a. Protein degradation in Lemna with
particular reference to ribulose bisphosphate carboxylase I. The effect
of light and dark. Plant Physiology 83, 869–877.
Ferreira RB, Davies DD. 1987b. Protein degradation in Lemna with
particular reference to ribulose bisphosphate carboxylase II. The effect
of nutrient starvation. Plant Physiology 83, 878–883.
Field CH, Behrenfeld MJ, Randerson JT, Falkowski P. 1998.
Primary production of the biosphere: integrating terrestrial and oceanic
components. Science 281, 237–240.
Finkel ZV, Beardall J. Flynn KJ, Quigg A, Raven JA, Rees TAV.
2010. Phytoplankton in a changing world: cell size and elemental
stoichiometry. Journal of Plankton Research 32, 118–137.
Finkel ZV, Quigg AS, Chiampi RK, Schofield OE, Falkowski PG.
2007. Phylogenetic diversity in cadmium:phosphorus ratio regulation
by marine phytoplankton. Limnology and Oceanography 52,
1131–1138.
Flynn KJ, Raven JA, Rees TAV, Finkel Z, Quigg A, Beardall J.
2010. Is the growth rate hypothesis applicable to microalgae? Journal
of Phycology 46, 1–12.
Flynn KJ, Stoecker DK, Mitra A, Raven JA, Glibert PM,
Hansen PJ, Granéli E, Burkholder JM. 2013. A case of mistaken
identification: the importance of mixotrophs and the clarification of
plankton functional-classification. Journal of Plankton Research 35,
3–11.
Garcia NS, Fu F-X, Hutchins DA. 2013. Colimitation of the
unicellular photosynthetic diazotroph Crocosphaera watsonii by
phosphorus, light, and carbon dioxide. Limnology and Oceanography
58, 1501–1512.
Geider RJ, La Roche J. 2002. Redfield revisited: variability of C:N:P
in marine microalgae and its biochemical basis. European Journal of
Phycology 37, 1–17.
Geske T, von Dorp K, Dörman P. Hölzi G. 2012. Accumulation
of glycolipids and other non-phosphorus lipids in Agrobacterium
tumefaciens grown under phosphate deprivation. Glycobiology 23,
69–80.
Ghannoum O, Conroy JP. 2007. Phosphorus deficiency inhibits
growth in parallel with photosynthesis in a C3 (Panicum laxum) but not
two C4 (P. coloratum and Cenchrus ciliaris) grasses. Functional Plant
Biology 34, 72–81.
Ghannoum O, Evans JR, Chow WS, Andrews TJ, Conroy JP,
von Caemmerer S. 2005. Faster Rubisco is the key to superior
Gismervik I. 1997. Stoichiometry of some marine planktonic
crustaceans. Journal of Plankton Research 19, 279–285.
Gordon WS, Jackson, RB. 2000. Nutrient concentration in fine
roots. Ecology 81, 275–280.
Greilhuber J, Borsch T, Müller K, Warberg A. 2006. Smallest
angiosperm genolems found in Lentibulariaceae, with chromosomes
of bacterial size. Plant Biology (Stuttgart) 8, 770–777.
Gruber DF, Tuorto S, Taghon GL. 2009. Growth phase and
elemental stoichiometry of bacterial prey influences ciliate grazing
selectivity. Journal of Eukaryotic Microbiology 156, 466–471.
Haldane JBS. 1929. The origin of life. Rationalist Annual 3, 3–10.
Halsted M, Lynch J. 1996. Phosphorus responses of C3 and C4
grasses. Journal of Experimental Botany 47, 497–505.
Hansen PJ, Bjørnsen PK, Hansen BE. 1997. Zooplankton
grazing and growth scaling 2–2000 μm size range. Limnology and
Oceanography 42, 687–701.
Hartman M, Grob C, Scanlon DJ, Martin AP, Burkhill PH,
Zubkov MV. 2011. Comparison of phosphate uptake rates by
the smallest plastidic and aplastidic protists in the North Atlantic
subtropical gyre. FEMS Microbiology Ecology 78, 327–335.
Henley SF, Annett AL, Ganeshram RS, Carson DS, Weston K,
Crosta X, Tait A, Dougans J, Fallick AE, Clarke A. 2012. Factors
influencing the stable carbon isotope composition of suspended and
sinking organic matter in the coastal Antarctic sea ice environment.
Biogeosciences 9, 1137–1157.
Hessen DO, Ferøvjo PJ, Andersen T. 2002. Light, nutrients and
P:C ratio in algae: grazer performance related to food quality and
quantity. Ecology 83, 1886–1898.
Hessen DO, Ågren GI, Anderson TR, Elser JJ, de Rutter PC.
2004. Carbon sequestration in ecosystems: the role of stoichiometry.
Ecology 85, 1179–1192.
Hijri M, Sanders IR. 2004a. The arbuscular mycorrhizal fungus
Glomus intraradices is haploid and has a small genome at the lower
limit of eukaryotes. Fungal Genetics and Biology 41, 253–261.
Hijri M, Sanders IR. 2004b. Low gene copy number shows that
arbuscular mycorrhizal fungi inherit genetically different fungi. Nature
433, 160–163.
Hill WR, Fanta SE, Roberts BJ. 2009. Quantifying phosphorus
and light effects on stream algae. Limnology and Oceanography 54,
368–380.
Hirst AG, Roff JC, Lampitt RS. 2003. A synthesis of growth rates
of marine epipelagic invertebrate zooplankton. Advances in Marine
Biology 44, 1–142.
Ho T-Y, Quigg A, Finkel ZV, Milligan AJ, Wyman K, Falkowski PG,
Morel FMM. 2003. The elemental stoichiometry of some marine
phytoplankton. Journal of Phycology 39, 1124–1159.
The evolution of autotrophy in relation to phosphorus requirement | 4043
Hodgson JG, Sharafi M, Jlili A, et al. 2010. Stomatal vs. genome
size in angiosperms: the somatic tail wagging the genome dog?
Annals of Botany 105, 573–584.
Keiblinger KM, Hall EK, Wanek W, et al. 2010. The effect of
resource quantity and resource stoichiometry on microbial carbon-use
efficiency. FEMS Microbiology Ecology 73, 430–440.
Horner TJ, Lee, RBY, Henderson GM, Rickaby REM. 2013.
Nonspecific uptake and homeostasis drive the oceanic cadmium
cycles. Proceedings of the National Academy of Sciences, USA 110,
2500–2505.
Kempes CP, Dutkiewicz S, Follows MJ. 2012. Growth, metabolic
partitioning, and the size of microorganisms. Proceedings of the
National. Academy of Sciences, USA 109, 495–500.
Imhoff JF, Bias-Imhoff U. 1995. Lipids, quinones and fatty acids
of anoxygenic phototrophic bacteria. In: Blankenship RF, Madigan
MT, Bauer CE, eds. Anoxygenic photosynthetic bacteria . Dordrecht:
Kluwer Academic Publishers, 179–205.
Irving LJ, Robinson D. 2006a. A dynamic model of Rubisco turnover
in cereal leaves. New Phytologist 169, 493–504.
Irving LJ, Robinson D. 2006b. On modelling Rubisco turnover:
dynamics and applicability. New Phytologist 170, 104–106.
Ishimi K, Kawamura T, Yamamoto A, Tada K, Harrison PJ.
2012. Extremely high growth rate of the small diatom Chaetoceros
salsugineum isolated from an estuary in the eastern Seto Island sea,
Tapan. Journal of Phycology 48, 1284–1288.
Jabaji-Hare S, Deschere A, Kendrick B. 1984. Lipid content and
composition of a vesicular-arbuscular mycorrhizal fungus. Mycologia
76, 1024–1030.
Jackson, WS, Mooney HA, Schulze E-D. 1997. A global budget for
fine root biomass, surface area, and nutrient contents. Proceedings of
the National Academy of Sciences, USA 94, 7262–7366.
Jacob J, Lawlor DW. 1991. Stomatal and mesophyll limitation of
photosynthesis in phosphate deficient sunflower, maize and wheat.
Journal of Experimental Botany 42, 1003–1011.
Jacob J, Lawlor DW. 1992. Dependence of photosynthesis in
sunflower and maize leaves on phosphate supply, ribulose-1,5bisphosphate carboxylase/oxygenase and ribulose-1.5-bisphosphate
pool size. Plant Physiology 98, 801–807.
Jacob J, Lawlor DW. 1993a. Extreme phosphorus deficiency
decreases the in vivo CO2/O2 specificity factor of ribulose
1,5-bisphosphate carboxylase-oxygenase in intact leaves of
sunflower. Journal of Experimental Botany 44, 1635–1641.
Jacob J, Lawlor DW. 1993b. In vivo photosynthetic electron
transport does not limit photosynthetic capacity in phosphate-limited
sunflower and maize leaves. Plant, Cell and Environment 16, 785–795.
Johnston DT, Poulton SW, Goldberg T, Sergeev VN,
Podkovyrov V, Vorob’eva NG, Bekker A, Knoll AH. 2012. Late
Ediacaran redox stability and metazoan evolution. Earth and Planetary
Science Letters 336, 25–35.
Khozin-Goldberg I, Cohen Z. 2006. The effect of phosphate
starvation on the lipid and fatty acid composition of the fresh water
eustigmatophyte Monodus subterraneus. Phytochemistry 67,
696–701.
Koga Y, Nishihara M, Morii H, Akagawa-Matsushita M. 1993.
Ether polar lipids of methanogenic bacteria: structures, comparative
aspects, and biosynthesis. Microbiological Reviews 57, 164–182.
Kozlowska-Szerenos B, Zieliński P, Maleszewski S. 2004.
Involvement of glycolate metabolism in acclimation of Chlorella
vulgaris cultures to low phosphate supply. Plant Physiology and
Biochemistry 38, 727–734.
Lacuesta M, Dever LV, Muñoz-Rueda A, Lea PJ. 1997. A study
of photorespiratory ammonia production in the C4 plant Amaranthus
edulis, using mutants with altered photosynthetic capacities.
Physiologia Plantarum 99, 447–455.
Laliberté E, Grace JB, Huston MA, Lambers H, Teste FP,
Turner BL, Wardle DA. 2013. How does pedogenesis drive plant
diversity? Trends in Ecology and Evolution 28, 331–340.
Lambers H, Brundrett MC, Raven JA, Hopper SD. 2010. Plant
mineral nutrition in ancient landscapes: high plant species diversity
on infertile soils is linked to functional diversity for nutritional strategy
(Marschner Review). Plant and Soil 334, 11–31.
Lambers H, Cawthray GR, Giavalisco P, Kuo J, Laliberté E,
Pearse SJ. Scheible W-R, Stitt M, Teste F. Turner BL. 2012.
Proteaceae from severely phosphorus-impoverished soils extensively
replace phospholipids with galactolipids and sulfolipids during leaf
development to achieve high photosynthetic phosphorus-useefficiency. New Phytologist 196, 1098–1108.
Lambers H, Raven JA, Shaver G, Smith SE. 2008. Plant nutrientacquisition strategies change with soil age. Trends in Ecology and
Evolution 23, 95–103.
Lane N, Allen JF, Martin W. 2010. How did LUCA make a living?
Chemiosmosis in the origin of life. Bioessays 32, 271–280.
Langworthy TA, Mayberry WR, Smith PF. 1976. A sulfolipid and
novel glucosamidyl glycolipid in the extreme thermoacidophile Bacillus
acidocaldarius. Biochimica et Biophysica Acta 431, 550–569.
Lenton TM, Watson AJ. 2000. Redfield revisited. 1. Regulation of
nitrate, phosphate and oxygen in the ocean. Global Biogeochemical
Cycles 14, 225–245.
Johnston DT, Wolfe-Simon F, Pearson A, Knoll AH. 2009.
Anoxygenic photosynthesis modulated Proterozoic oxygen and
sustained Earth’s middle age. Proceedings of the National Academy of
Sciences, USA 106, 16925–16929.
Lloyd J, Farquhar GD. 1994. 13C discrimination during CO2
assimilation by the terrestrial biosphere. Oecologia 112, 285–299.
Johnston JS, Pepper AE, Hall AE, Chen ZJ, Hodnett G,
Drabek J, Lopez R, Price J. 2005. Estimation of genome size in
Brassicaceae. Annals of Botany 95, 229–235.
Loladze I, Elser JJ. 2011. The origins of the Redfield nitrogen-tophosphorus ratio are in a homeostatic protein to RNA ratio. Ecology
Letters 14, 244–250.
Karpinets TV, Greenwood DJ, Sand CE, Ammons JT. 2006.
RNA:protein ratio of the unicellular organisms as a characteristic of
phosphorous (sic) and nitrogen stoichiometry of cellular reqirements
for ribosomes for protein synthesis. BMC Biology 4, 20.
Long SP. 1999. Environmental responses. In: Sage RF, Monson RK,
eds. C4 plant biology . San Diego: Academic Press, 215–249.
Löpez-Lara IM, Gao JL, Soto MJ, Solarez-Perez A.
Weissenmajer B, Thomas-Shlenkamp C, Verroios GP, Oates J,
4044 | Raven
Geiger C. 2005. Phosphorus-free membrane lipids of Sinorhizobium
meliloti are not required for the symbiosis with alfalfa but contribute to
increased cell yields under phosphorus-limited conditions of growth.
Molecilar Plant-Microbe Interactions 18, 973–982.
Losh JL, Young JN, Morel FMM. 2013. Rubisco is a small fraction
of total protein in marine phytoplankton. New Phytologist 198, 52–58.
Lourenço SO, Barbarion U, Lanfer Mercquez UM, Aidar E.
1998. Distribution of intracellular nitrogen in marine microalgae: basis
for calculation of nitrogen to protein conversion factors. Journal of
Phycology 36, 798–811.
Maiquetia M, Cáceres A, Herrera A. 2009. Mycorrhization and
phosphorus nutrition affect water relations and CAM induction by
drought in seedlings of Clusia minor. Annals of Botany 103, 525–532.
Makino W, Cotner JB. 2004. Elemental stoichiometry of a
heterotrophic bacterial community in a freshwater lake: implications
for growth- and resource-dependent variations. Aquatic Microbial
Ecology 34, 33–41.
Makino W, Cotner JB, Sterner RW, Elser JJ. 2003. Are bacteria
more like plants or animals? Growth rate and resource depence of
bacterial C:N:P stoichiometry. Functional Ecology 17, 121–130.
Mantiana KB, Arneth A, Veenendaal EM, Wohland P, Wolski P,
Kolle O, Wagner M, Lloyd J. 2008. Photosynthetic properties of
C4 plants growing in an African savanna/wetland mosaic. Journal of
Experimental Botany 59, 3941–3952.
Marañón E, Cermeño P, López-Sandoval DC, Rodríguez-Ramos
T, Sobrino C, Huete-Ortega M, Blanco JM, Rodríguez J.
2013. Unimodal size scaling of phytoplankton growth and the size
dependence of nutrient uptake and use. Ecology Letters 16, 371–379.
Martin P, van Mooy BAS, Heithoff A, Dyrhman ST. 2011.
Phosphorus supply drives rapid turnover of phospholipids in the
diatom Thalassiosira pseudonana. ISME Journal 5, 1057–1060.
Martin SF, Munagapati VS, SalvoChimside E, Kerr LE,
Le Bihan T. 2012. Proteome turnover in the green alga Ostreococcus
tauri by time course 15N labelling mass spectrometry. Journal of
Proteome Research 11, 476–486.
Mastrobuoni G, Ingandg S, Pietzke M, Aβmus HF, Wenzel M,
Schule WX, Kempa S. 2012. Proteome dynamics and early salt
stress response of the photosynthetic organism Chlamydomonas
reinhardtii. BMC Genomics 13, 215.
Matzek Y, Vitousek PM. 2009. N:P stoichiometry and protein:RNA
ratios in vascular plants: an evaluation of the growth-rate hypothesis.
Ecology Letters 12, 765–761.
Maxwell K. 2002. Resistance is useful; diurnal patterns of
photosynthesis in C3 and crassulacean acid metabolism epiphytic
bromeliads. Functional Plant Biology 29, 679–687.
Menge DNL, Hedin LO, Pascale SW. 2012. Nitrogen and
phosphorus limitation over long term ecosystems development in
terrestrial ecosystems. PLoS One 7, e42045.
Mikucki JA, Pearson A, Johnston DT. Turchyn AV, Farquahr J,
Schrag DP, Anbar AD, Priscu JC, Lee PA. 2009. A contemporaray
microbially maintained subglacial ferrous ‘Ocean’. Science 324, 397–400.
Minnikin DE, Abdolrahimzadeh H, Baddiley J. 1974. Replacement
of acidic phospholipids by acidic glycolipids in Pseudomonas dininuta.
Nature 249, 268.
Mitchell C, Beardall J. 1996. Inorganic carbon uptake by an
Antarctic sea-ice diatom, Nitzschia frigida. Polar Biology 21, 310–315.
Mizoguchi T, Harada J, Yoshitomi T, Tamiaka H. 2013. A variety of
glycolipids in green photosynthetic bacteria. Photosynthesis Research
114, 179–188.
Morgan HD, Westoby M. 2005. The relationship between nuclear
DNA content and leaf strategy in seed plants. Annals of Botany 96,
1321–1330.
Nicklisch A, Steinberg CEW. 2009. RNA–protein and RNA–DNA
ratios by flow cytometry and their relationship to growth limitation of
selected planktonic algae in culture. European Journal of Phycology
44, 297–308.
Niewiadomska E, Bilger W, Gruca M, Mulisch M, Miszalski Z,
Krupinska K. 2011. CAM-related changes in chloroplastic metabolism
of Mesembryanthemum crystallinum L. Planta 233, 275–285.
Okazaki Y, Otsuki T, Kobayashi M, Sawai S, Kamide Y,
Kusano M, Aoki T, Hirai MY, Saito K. 2013. A new class of of plant
lipid is essential for protection against phosphorus depletion. Nature
Communications 4, 1510.
Papineau D. 2010. Global biogeochemical changes at both ends of
the Proterozoic: insights from phosphorites. Geobiology 10, 166–181.
Paul MJ, Cockburn W. 1990. The stimulation of CAM activity in
Mesembryanthemum crystallinum in nitrate- and phosphate-deficient
conditions. New Phytologist 114, 391–398.
Perry MJ. 1976. Phosphorus utilization by an oceanic diatom in
phosphate limited chemostat culture and in the oligotrophic waters
of the central North Pacific. Limnology and Oceanography 21,
88–107.
Planavsky NJ, Rouxel OJ, Bekker A, Lalonde SJ, Konhauser KO,
Reinhard CT, Lyons TW. 2010. The evolution of the marine
phosphate reservoir. Nature 467, 1088–1090.
Popendorf KJ, Lomas MW, Van Mooy BAS. 2011a. Microbial
sources of intact polar diacylglycerolipids in the Western North Atlantic
Ocean. Organic Geochemistry 42, 803–811.
Popendorf KJ, Tanaka T, Poujo-Pay M, Lagaria A, Courties C,
Conan P, Oriol L, Sofen LE, Moutin T, Van Mooy RAS. 2011b.
Gradients of intact polar diacylglycerolipids across the Mediterranean
Sea are related to phosphate availability. Biogeosciences 8,
3733–3745.
Quigg A, Beardall J. 2003. Protein turnover in relation to
maintenance metabolism at low photon flux in two marine microalgae.
Plant, Cell and Environment 26, 693–703.
Quigg A, Finkel ZV, Irwin AJ, Rosenthal Y, Ho T-Y, Reinfelder JR,
Schofield O, Morel FMM, Falkowski PG. 2003. The evolutionary
inheritance of elemental stoichiometry in marine phytoplankton. Nature
425, 291–294.
Quigg A, Irwin AJ, Finkel ZV. 2011. Evolutionary inheritance of
elementary stoichiometry in phytoplankton. Proceedings of the Royal
Society B: Biological Sciences 278, 526–534.
Raven JA. 1984a. A cost–benefit analysis of photon absorption by
photosynthetic unicells. New Phytologist 98, 593–625.
Raven JA. 1984b. Energetics and transport in aquatic plants . New
York: A.R. Liss.
The evolution of autotrophy in relation to phosphorus requirement | 4045
Raven JA. 1989. Function of lipids: lipids of the photosynthetic
apparatus. In: Ratledge C, Wilkinson SG, eds. Microbial lipids , Vol. 2.
London: Academic Press, 439–486..
Raven JA. 1991. Physiology of inorganic C acquisition and
implications for resource use efficiency by marine phytoplankton—
relation to increased CO2 and temperature. Plant, Cell and
Environment 14, 79–794.
Raven JA. 1994. Why are there no picoplankton O2 evolvers with
volumes less than 10–19 m3? Journal of Plankton Research 16,
565–580.
Raven JA. 2001. An aquatic perspective on the concepts of Ingestad
relating plant nutrition to plant growth. Physiologia Plantarum 113,
301–307
Raven JA. 2009. Contributions of anoxygenic and oxygenic
phototrophy and chemolithotrophy to carbon and oxygen fluxes in
aquatic environments. Aquatic Microbiology Ecology 56, 177–192.
Raven JA. 2011. The cost of photoinhibition. Physiologia Plantarum
142, 87–104.
Raven JA. 2012. Protein turnover and plant RNA and phosphorus
requirements in relation to nitrogen fixation. Plant Science 188–189,
25–35.
Raven JA. 2013a. Rubisco: still the most abundant protein on Earth?
New Phytologist 198, 1–3.
Raven JA. 2013b. Half a century of pursuing the pervasive proton.
Progress in Botany 74, 3–34.
Raven JA, Andrews M. 2010. Evolution of tree nutrition. Tree
Physiology 30, 1050–1071.
Raven JA, Beardall J, Giordano M, Maberly SC. 2012. Algal
evolution in relation to atmospheric CO2: carboxylases, carbon
concentrating mechanisms and carbon oxidation cycles. Philosophical
Transactions of the Royal Society B: Biological Sciences 367,
493–507
Raven JA, Beardall J, Larkum AWD, Sanchez-Baracaldo P.
2013. Interactions of photosynthesis with genome size and function.
Philosophical Transactions of the Royal Society B: Biological Sciences
368, 2012064.
Raven JA, Edwards D. 2001. Roots: evolutionary origin and
biogeochemical significance. Journal of Experimental Botany 52,
281–401.
Raven JA, Edwards D. 2013. Photosynthesis in early land plants:
adapting to the terrestrial environments. In: Hanson DT, Rice SK, eds.
Photosynthesis of bryophytes and early land plants. Berlin: Springer.
Raven JA, Handley LL, Wollenweber B. 2004. Plant nutrition and
water use efficiency. In: Bacon M, ed. Water use efficiency in plant
biology . Oxford: Blackwell Science, 162–188.
Raven JA, Johnston AM, Kübler JE, et al. 2002a. Seaweeds in
cold seas: evolution and carbon acquisition. Annals of Botany 90,
525–536.
Raven JA., Johnston AM, Kübler JE, et al. 2002b. Mechanistic
interpretation of carbon isotope discrimination by marine macroalgae
and seagrasses. Functional Plant Biology 29, 355–378.
Reef R, Ball MC, Feller IC, Lovelock CE. 2010. Relationship
among RNA:DNA ratio, growth and elemental stoichometry of
mangrove trees. Functional Ecology 24, 1064–1072.
Reef R, Pandolfi JM, Lovelock CE. 2012. The effect of nutrient
enrichment on the growth, nucleic acid concentrations, and elemental
stoichiometry of coral reef macroalgae. Ecology and Evolution 2,
1985–1995.
Rexroth S, Mullineaux CW, Ellinger D, Sendiko E, Rögner M,
Koenig F. 2011. The plasma membrane of the cyanobacterium
Gloeobacter violaceous contains segregated bioenergetics domains.
The Plant Cell 23, 2379–2390.
Richardson K, Griffiths H, Reed ML, Raven JA, Griffiths NM.
1984. Inorganic carbon assimilation in the isoetids, Isoetes lacustris
L. and Lobelia dortmanna L. Oecologia 61, 115–121.
Roberts K, Granum E, Leegood RC, Raven JA. 2007. C3 and C4
pathways of photosynthetic carbon assimilation in marine diatoms are
under genetic, not environmental, control. Plant Physiology 145, 230–235.
Robinson D. 1990. Phosphorus availability and cortical senescence
in cereal roots. Journal of Theoretical Biology 145, 257–265.
Sage RF, Pearcy RW, Seeman JR. 1987. The nitrogen use
efficiency of C3 and C4. III. Leaf nitrogen effects on the activity of
carboxylating enzymes in Chenopodium album (L.) and Amaranthus
retrofexus (L.). Plant Physiology 85, 355–359.
Sage RF, Sharkey TD, Seeman JR. 1989. Acclimation of
photosynthesis to elevated CO2 in five C3 physiology plants. Plant
Physiology 89, 590–596.
Sage RF, Zhu X-G. 2011. Exploiting the engine of C4 photosynthesis.
Journal of Experimental Botany 62, 2989–3000.
Sahoo SK, Planavsky NJ, Kendall B, Wang X, Shi X, Scott C,
Anbar A, Lyons TW, Joang G. 2012. Ocean oxygenation in the
wake of the Marinoan glaciation. Nature 489, 546–549.
Sal S, López-Umitia A. 2011. Temperature, nutrient and size-scaling
of phytoplankton growth in the sea. Limnology and.Oceanography 56,
1952–1955.
Saltzman MR. 2005. Phosphate, nitrogen, and the redox evolution of
the Paleozoic ocean. Geology 33, 573–576.
Sanginga N, Danso SKA, Bowen GD. 1989. Nodulation and growth
responses of Allocasuarina and Carsuarina species to phosphorus
fertilization. Plant and Soil 118, 125–132.
Savir Y, Noor E, Milo R, Tlusty T. 2010. Cross-species analysis
traces adaptation of Rubisco toward optimality in a low-dimensional
landscape. Proceedings of the National Academy of Sciences, USA
107, 3475–3480.
Schmidt S, Raven JA, Paungfoo-Lonhienne S. 2013. The
mixotrophic nature of photosynthetic plants. Functional Plant Biology
(in press).
Selstam E, Campbell D. 1996. Membrane lipid composition of the
unusual cyanobacterium Gloeobacter violaceaous sp. PCC7421,
which lacks sulfoquinovosyl diacylglycerol. Archives of Microbiology
166, 125–135.
Shen Y, Schidlowski M, Chu X. 2000. Biogeochemical approach to
understanding phosphorogenic events in the terminal Proterozoic to
Cambrian. Palaeogeography, Palaeoclimatology, Palaeoecology 158,
99–108.
Singh DK, Sale PG, Pallaghy CK, Mckenzie BM. 2008.
Phosphorus concentrations in the leaves of defoliated white clover
affects abscisic acid formation and transpiration in drying soil. New
Phytologist 146, 249–259.
4046 | Raven
Smith SE, Read D. 2008. Mycorrhizal symbiosis . Amsterdam: Elsevier.
Soltis DE,Soltis PS, Bennett MD, Leitch IJ. 2003. Evolution of
genome size in the angiosperms. American Journal of Botany 80,
1596–1603.
Souza V, Egiuartae LE, Siefert J, Elser JJ. 2008. Microbial
endemism: does phosphorus limitation enhance speciation? Nature
Reviews Microbiology 6, 559–564.
Spijkerman E. 2011. The expression of carbon concentrating
mechanism in Chlamydomonas acidophila under variable phosphorus,
iron and CO2 concentrations. Photosynthesis Research 109, 179–189.
Spijkerman E, de Castro F, Gaedke U. 2011. Independent
colimitation for carbon dioxide and inorganic phosphorus. PLoS One
6, e28219.
Sterner RW, Elser JJ. 2002. Elemental stoichiometry: the biology of
elements from molecules to the biosphere. Princetion, NJ: Princeton
University Press.
Tcherkez GGB, Farquhar GD, Andrews TA. 2006. Despite slow
catalysis and confused substrate specificity, all ribulose bisphosphate
carboxylases may be nearly perfectly optimised. Proceedings of the
National Academy of Sciences, USA 103, 7246–7251.
Tortell PD, Payne C, Gueguen G, Strzepek RF, Boyd PW, Rost B.
2006. Inorganic carbon uptake by Southern Ocean phytoplankton.
Limnology and Oceanography 53, 1266–1278.
Van Mooy BAS, Fedricks HF, Pedler BE, et al. 2009.
Phytoplankton in the ocean substitute phospholipids in response to
phosphorus scarcity. Nature 458, 69–72.
Van Mooy BAS, Rocap G, Fedricks HF, Evans CT, Devol AH.
2006. Sulfolipids dramatically decrease phosphorus demand by
picocyanobacteria in oligotrophic environments. Proceedings of the
National Academy of Sciences, USA 103, 8607–8612.
Warren CR, Adams MA. 2001. Distribution of N, Rubisco and
photosynthesis in Pinus pinaster and acclimation to light. Plant, Cell
and Environment 51, 1147–1157.
Watanabe T, Broadley MR, Jansen S, White PJ, Takada J,
Satake K, Takamatsu T, Yuah SJ, Osaki N. 2007. Evolutionary
control of leaf elemental composition in plants. New Phytologist 174,
516–523.
Williams RJP, Rickaby R. 2012. Evolutions destiny. Co-evolution of
the environment and life. London: Royal Society of Chemistry.
Wright IJ, Reich PB, Westoby M, et al. 2004. The leaf economics
spectrum worldwide. Nature 428, 821–827.
Wu Z, Zend B, Li R, Song L. 2012. Combined effects of carbon
and phosphorus on the invasive cyanobacterium Cylindrospermopsis
raciborski. Phycologia 51, 144–150.
Xu Z, Gao K. 2009. Impacts of UV radiation on growth and
photosynthetic carbon assimilation in Gracilaria lemaneiformis
(Rhodophyta) under phosphorus limited and replete conditions.
Functional Plant Biology 36, 1057–1064.
Xu Z, Zou D, Gao K. 2010. Effects of elevated CO2 and phosphorus
supply on growth, photosynthesis and nutrient uptake in the marine
macroalga Gracilaria lemaneiformis (Rhodophyta). Botanica Marina
53, 123–129.
Yeoh H-H, Badger MR, Watson L. 1981. Variations in kinetic
properties of ribulose-1.5-bisphosphate carboxylases among plants.
Plant Physiology 67, 1151–1155.
Young JN, Rickaby REM, Kaparalov B, Filatove D. 2012. Adaptive
signals of algal Rubisco reveals a history of ancient atmospheric CO2.
Philosophical Transactions of the Royal Society B: Biological Sciences
367, 483–482.
Veneklaas EJ, Lambers H, Bragg J, et al. 2012. Opportunities for
improving phosphorus-use efficiency in crop plants. New Phytologist
195, 306–320.
Yu G, Wu H, He H, Lü X, Elser JJ, Wu J, Han X. 2012. Testing the
growth rate hypothesis in vascular plants with above- and belowground biomass. PLoS One 7, e32162
Vrede T, Dobberfuhl DR, Kooijman DAlM, Elser JJ. 2004.
Fundamental connections among organism C:N:P stoichiometry,
molecular composition, and growth. Ecology 85, 1217–1229.
Yuzawa Y, Nixhihara H, Haraguchi T, Masuda S, Shimojima M,
Simoyama A, Yuasa H, Okada N, Ohta H. 2012. Phylogeny of
galactolipid synthase homologs together with their enzymatic analyses
revealed a possible origin and divergence time for photosynthetic
membrane biogenesis. DNA Research 19, 91–102.
Vrede T, Persson J, Aronsen G. 2002. The effect of food quality
(P:C ratio) on RNA:DNA ratio and somatic growth rate of Daphnia.
Limnology and Oceanography 47, 487–494.
Walve J, Larsson U. 1999. Carbon, nitrogen and phosphorus
stoichiometry of crustacean zooplankton in the Baltic Sea: implications
for nutrient recycling. Journal of Plankton Research 21, 2309–2321.
Zheng S, Ren H, Li W, Lan Z. 2012. Scale-dependent effects of
grazing in plant C:N:P stoichiometry and linkages to ecosystem
functioning in the Inner Mongolian grassland. PLoS One 7,
e51570.
Wang YP, Law RH, Pak B. 2010. A global model of carbon, nitrogen
and phosporus cycles for the terrestrial biosphere. Biogeoscience 7,
2261–2287.
Zhu F, Massara P, Not F, Marie D, Vaulot D. 2005. Mapping of
picoeukatyotes in marine ecosystems with quantitative PCR of the
18S rRNA gene. FEMS Microbiology Ecology 52, 79–92.