Journal of Experimental Botany, Vol. 59, No. 7, pp. 1525–1541, 2008
doi:10.1093/jxb/erm297 Advance Access publication 2 February, 2008
SPECIAL ISSUE REVIEW PAPER
Multiple Rubisco forms in proteobacteria: their functional
significance in relation to CO2 acquisition by the CBB cycle
Murray Ronald Badger1,2,* and Emily Jane Bek
1
Molecular Plant Physiology Group, Research School of Biological Sciences, The Australian National University,
Canberra, ACT, Australia
2
ARC Centre of Excellence in Plant Energy Biology, Research School of Biological Sciences, The Australian
National University, Canberra, ACT, Australia
Received 8 August 2007; Revised 19 September 2007; Accepted 22 October 2007
Abstract
Rubisco is the predominant enzymatic mechanism in
the biosphere by which autotrophic bacteria, algae,
and terrestrial plants fix CO2 into organic biomass via
the Calvin–Benson–Basham reductive pentose phosphate pathway. Rubisco is not a perfect catalyst,
suffering from low turnover rates, a low affinity for its
CO2 substrate, and a competitive inhibition by O2 as an
alternative substrate. As a consequence of changing
environmental conditions over the past 3.5 billion
years, with decreasing CO2 and increasing O2 in the
atmosphere, Rubisco has evolved into multiple enzymatic forms with a range of kinetic properties, as well
as co-evolving with CO2-concentrating mechanisms to
cope with the different environmental contexts in
which it must operate. The most dramatic evidence of
this is the occurrence of multiple forms of Rubisco
within autotrophic proteobacteria, where Forms II, IC,
IBc, IAc, and IAq can be found either singly or in
multiple combinations within a particular bacterial
genome. Over the past few years there has been
increasing availability of genomic sequence data for
bacteria and this has allowed us to gain more extensive insights into the functional significance of this
diversification. This paper is focused on summarizing
what is known about the diversity of Rubisco forms,
their kinetic properties, development of bacterial CO2concentrating mechanisms, and correlations with metabolic flexibility and inorganic carbon environments in
which proteobacteria perform various types of obligate
and facultative chemo- and photoautotrophic CO2
fixation.
Key words: Autotrophic bacteria, chemotrophic bacteria, CO2concentrating mechanism, photosynthesis, proteobacteria,
Rubisco.
Introduction
Rubisco is the predominant enzymatic mechanism in the
biosphere by which autotrophic bacteria, algae, and
terrestrial plants fix CO2 into organic biomass (Tabita,
1999; Atomi, 2002). Its biochemical role is chiefly as
the central part of the Calvin–Benson–Basham (CBB)
reductive pentose phosphate pathway where it combines
CO2 with ribulose 1,5-bisphosphate to form two molecules of 3-phosphoglycerate. Hence, there has been
tremendous research interest in its biochemistry, catalytic
function, protein structure, and evolution, and a number of
reviews have appeared over recent years that have covered
various aspects of this research (Cleland et al., 1998;
Spreitzer and Salvucci, 2002).
The imperfection of Rubisco
It is well recognized that Rubisco is not a perfect catalyst
in terms of its enzymatic properties and the ecological and
cellular environments in which it is situated (Tcherkez
et al., 2006). These imperfections are to be found in three
major areas. Firstly, its catalytic turnover rate is slow in
relation to the rates of CO2 fixation which occur in many
autotrophic cells and this requires an investment of
a significant amount of protein in this single catalytic step.
This is especially so in higher plants, and it is telling that
it is regularly cited as the most abundant protein on earth.
A second impediment is that it has a low affinity for CO2
* To whom correspondence should be addressed. E-mail: [email protected]
ª The Author [2008]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.
For Permissions, please e-mail: [email protected]
1526 Badger and Bek
as a substrate, although this has varied with evolution,
which will be elaborated on below, and there has been
a continual evolutionary struggle to improve Rubisco
performance, by both increases in catalytic affinity for
CO2 and addition of active CO2-acquisition mechanisms
which help it to perform (Badger et al., 1998). Finally, it
has suffered from the fact that molecular O2 is an
alternative substrate for the reaction (Rubisco oxygenase)
and produces P-glycolate as a product. As a result, O2
inhibits the catalytic affinity for CO2 and cells must deal
with the toxic P-glycolate thus produced (see Cleland
et al., 1998; Spreitzer and Salvucci, 2002).
The major evolutionary forces which have acted on the
evolution of Rubisco must be seen in terms of its
imperfections listed above. Over the past 3.5 billion years,
the earth’s biosphere has seen a dramatic change in the
concentration of its substrates, with oxygenic photosynthesis causing O2 to rise from an anaerobic starting point
to reach as high as 35% O2 in the late Paleozoic (Berner,
2001; Berner et al., 2003), and is currently 21%. There
have also been significant concurrent declines in CO2,
from an initial concentration in the percentage range, to
fall below 0.02% 280 Mya (Mora et al., 1999), and is
currently 0.037% and rising (Berner and Kothavala,
2001). In addition, nitrogen as a limiting nutrient resource
has varied tremendously in terrestrial and aquatic environments. Evolution has thus had to address how Rubisco is
optimized with regard to CO2 limitation, inhibition by O2,
and by the scarcity of nitrogen. Evolutionary changes in
the enzyme which reflect responses to these pressures are
evident in a significant decline in Km(CO2), increases in
the specificity factor (which is a measure of the selectivity
for CO2 versus O2), and significant variance in Vmax for
the carboxylase reaction (Tcherkez et al., 2006). From an
organismal perspective, there has also been the evolution
of a diverse array of active CO2-concentrating mechanisms (CCMs) which elevate CO2 at the active site of
Rubisco thus largely minimizing its imperfections (Badger
and Price, 1992; Badger et al., 1998).
The evolutionary diversification of Rubisco types and
co-evolution between Rubisco properties and cellular
CCMs is most evident in CO2-fixing bacteria. Over the
past few years there has been increasing availability of
genomic sequence data for bacteria and this has allowed
us to gain more extensive insights into this co-evolution.
This review is focused on summarizing what is known
about the diversity of Rubisco forms, their kinetic
properties, development of CCMs, and correlations with
the environment. In particular, the co-existence of multiple Rubisco types in many proteobacteria is explored.
Multiple Rubisco forms
Table 1 summarizes the recognized presence, structure,
and function of four related Rubisco enzyme forms in
archaea, bacteria, and eukaryotes. Of these, only Forms I,
II, and III have been shown to have RuBP-dependent
CO2-fixing ability (for review, see Ashida et al., 2005).
Form IV enzymes have been termed ‘Rubisco-like’
proteins and have been shown to be functional in the
methionine salvage pathway in many bacteria (Ashida
et al., 2003). There are suggestions that Form IV Rubisco
may have been the evolutionary progenitor of CO2-fixing
Rubisco as it is known today (Ashida et al., 2005). Form
III Rubiscos have to date only been found in archaea, and
their CO2-fixing role has been somewhat uncertain due to
the lack of an identifiable phosphoribulokinase in archaea
(Watson et al., 1999; Finn and Tabita, 2004), although
some archaea have recently been identified with potential
phosphoribulokinase genes (Mueller-Cajar and Badger,
2007). Most recently, it has been shown to participate in
a pathway for AMP metabolism, with RuBP being
supplied from AMP and phosphate (Sato et al., 2007).
Rubisco Forms I and II are the enzymes which are directly
involved in carbon metabolism associated with the CBB
cycle and have a well-recognized autotophic CO2-fixing
function which supports growth. These enzymes are found
in a wide range of chemo-, organo-, and phototrophs,
Table 1. Rubisco enzyme forms and their phylogenetic distribution
Information presented in this table was obtained from various review sources including Tabita (1999, 2004) and Ashida et al. (2005). The nature of
Form IAc, IAq, and IBc Rubiscos is described in the text. Table S1 available at JXB online presents a detailed view of this phylogenetic distribution
showing the presence of Forms I, II, III, and IV Rubisco genes in archaea, bacteria, algae, and plants.
Rubisco form
Green
Red
Form
Form
Form
Form
IA
IB
IC
ID
Form II
Form III
Form IV: ‘Rubisco-like‘
Macromolecular
structure
Phylogenetic occurrence
Enzymatic function
L8S8
L8S8
L8S8
L8S8
L2
L10
L2?
Cyanobacteria (Form IAc), proteobacteria (Forms IAcs and IAq)
Cyanobacteria (Form IBc), chlorophyte (green) algae, higher plants (Form IB)
Proteobacteria
Heterokont and haptophyte algae (non-green)
Proteobacteria, archaea, and dinoflagellate algae
Archaea
Bacteria, archaea, including both photosynthetic and non-photosynthetic
bacteria
CBB cycle
CBB cycle
CBB cycle
CBB cycle
CBB cycle
RuPP pathway
Methionine
salvage pathway
Multiple Rubisco forms in proteobacteria 1527
occurring in bacteria, algae, and plants. Form I enzymes
show considerable diversification and have been divided
into four distinct clades. Forms IA and B belong to
a ‘Green’ grouping and are found in proteobacteria,
cyanobacteria, green algae, and higher plants. Forms IC
and D are members of a ‘Red’ group, occurring in
proteobacteria and non-green algae.
Table 2 shows a more complete view of the existence of
different Form I and II Rubisco in proteobacteria and
cyanobacteria. The data in the table are based on
identification of Rubisco genes in publicly available
genome sequence information, together with other information gathered from papers and protein database
sequences. Table S1 in Supplementary data available at
JXB online extends this table to include a summary of the
presence of Forms I, II, II, and IV Rubisco genes in
archaea, bacteria, and eukaryotes. The nomenclature used
for Rubisco forms in Table 2 is somewhat different from
previous terminology, particularly in relation to Form I
enzymes. As will be discussed below, Form IA enzymes
are divided into two distinct types, IAc and IAq, based on
two distinct types of small subunits and gene arrangements (see Figs 1, 3). Form IB enzymes have been
subclassified into IB and IBc to indicate the Form IBc in
cyanobacteria which is associated with carboxysomes
(Badger et al., 2002). References to Form IC and D, II,
III, and IV enzymes are in accordance with others (Tabita,
2004; Ashida et al., 2005).
From Table 2 and Table S1 in Supplementary data
available at JXB online, it is evident that all obligate
photoautotrophic organisms employing the CBB cycle
possess a single form of Rubisco which has evolved to
serve the primary role of CO2 fixation during autotrophic
capture of CO2 by the CBB cycle from the external
environment. This includes cyanobacteria with either
Form IAc or IBc, chlorophyte algae with Form IB,
chromophyte and rhodophyte algae (including haptophytes and heterkonts) with Form ID, dinoflagellates with
Form II, and higher plants with Form IB (see Table S1 in
Supplementary data available at JXB online). The general
conclusion from this might be that when you have one
primary carbon source for CO2 fixation (CO2 or HCO–3)
then specialization in Rubisco becomes a primary driving
force, particularly when this is associated with a chloroplast structure as it is in algae and higher plants.
For archaea, the predominant specialization in a Form
III enzyme is most likely because of its proposed special
role in AMP metabolism rather than autotrophic CO2
fixation (Sato et al., 2007). However, the apparent
presence of a Rubisco which may be more related to
Form II enzymes in Methanococcoides burtonii (see Fig.
S1 in Supplementary data available at JXB online) raises
an interesting question about the role of this enzyme in
this species. However, this species appears to have the
genes associated with AMP metabolism (Sato et al., 2007)
and it would seem most likely that it was also associated
with this pathway.
The proteobacteria in general stand apart from these
other organisms in having genes for up to three different
forms of Rubisco including Form IAc, Form IAq, one or
more Form IC, and Form II (Table 2). The absence of
Form IB Rubisco from proteobacteria is interesting, given
that cyanobacteria have evolved to contain either Form
IAc or Form IBc. This situation may indicate that Form IB
evolved more recently than the other Form I Rubiscos,
being derived from Form IA. This has been previously
speculated on and is linked to questions about the
evolution of two types of carboxysomes in proteobacteria
and cyanobacteria (Badger et al., 2002; Badger and Price,
2003).
The reasons for multiple Rubiscos in proteobacteria
must presumably lie in the flexibility of the CO2-fixing
lifestyles associated with their generally plastic metabolism which includes CO2 derived from organic and
inorganic substrates and environments where the levels of
CO2 and O2 can fluctuate from high to low levels, and
they have opted for a strategy where different forms of
Rubisco are used under different environmental situations.
The following part of this review will focus on how the
presence and function of multiple forms of Form I and II
Rubisco in proteobacteria can be interpreted from knowledge of the properties of the different forms of Rubisco
and the metabolic behaviour and ecological environments
of particular organisms.
A distinction between two different Form IA
Rubiscos
Previous studies and reviews have classed all Form IA
Rubiscos under one grouping. However, with multiple
bacterial genome sequences becoming available, examination of Form IA Rubisco gene structure arrangements
within the genomes clearly indicates that there are two
different types of Form IA structures. This is illustrated in
Fig. 1 where Rubisco gene region structures from a
number of bacteria are shown. Gene region structures for
two Form IA enzymes from Thiomicrospira crunogenea
are shown. One of the Form IA L and S gene pairs are
associated with the genes coding for cbbQ1 and cbbO1.
This Rubisco has been termed Form IAq. The other Form
IA L and S gene pair is associated with a standard set of
a-carboxysome genes (Cannon et al., 2001, 2003) and is
presumably associated with the formation of a carboxysome structure. This has been termed Form IAc. A third
Rubisco Form II is found near to the Form IAq gene set,
associated with a second cbbQ/O gene pair. Such an
arrangement of genes has been previously seen in Hydrogenovibrio marinus (Hayashi et al., 1999; Yoshizawa
et al., 2004), where the same three sets of Rubisco genes
are found.
1528 Badger and Bek
Table 2. The presence of various Form I and II Rubiscos in proteobacteria and cyanobacteria together with information about their
metabolism and the ecological habitats with which they are associated
The documentation of the presence of these Rubisco forms was performed largely using the JGI Integrated Microbial Genomes browser (http://
img.jgi.doe.gov/cgi-bin/pub/main.cgi) together with other information extracted from papers where the presence of multiple or single Rubisco forms
is described. The sequence nomenclature differs somewhat from previous studies in the subdivision of Forms 1A and 1B. Form IAq and Form IAc
are distinguished in Fig. 2A and B by the amino acid sequences of the small subunits and their respective genome associations with cbbQ (for 1Aq)
or a-carboxysome (Form 1Ac) genes. Form 1Bc genes are associated with b-carboxysome genes. Information on metabolism and ecological habitats
was compiled from the DOE microbial genome project descriptions (http://genome.jgi-psf.org/mic_home.html), NCBI genome project database
(http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db¼genomeprj), and Prescott et al. (2005). The potential presence of carboxysomes is indicated in
the ‘Carb’ column (Y/N).
Group
Rubisco form
Carb Species
IBc IAc IAq IC
Alpha
Oxygen
Metabolism
Rhodospirillum
rubrum ATCC 11170
Acidophilum
cryptum JF-5
Facultative
anaerobe
Facultative
anaerobe
Facultative
photomixotroph
Facultative
chemolithomixotroph
Bradyrhizobium
japonicum USDA 110
Bradyrhizobium
sp. BTA1
Nitrobacter
hamburgensis
Nitrobacter
winogradskyi Nb-255
Rhodopseudomonas
palustris CGA009
Rhodopseudomonas
palustris BisA53
Rhodopseudomonas
palustris BisB18
Rhodopseudomonas
palustris BisB5
Rhodopseudomonas
palustris HaA2
Sinorhizobium
meliloti 1021
Xanthobacter
autotrophicus Py2
Paracoccus
denitrificans 1222
Rhodobacter
sphaeroides 2.4.1
Rhodobacter
sphaeroides
ATCC17025
Dechloromonas
aromatica RCB
Aerobic
Facultative
anaerobe
Rhodoferax
ferrireducens
Facultative
anaerobe
N
Hydrogenophaga
pseudoflava
Facultative
anaerobe
Burkholderia
xenovorans LB400
Ralstonia
eutropha H16
Rubrivivax
gelatinosus PM1
Aerobic
Alpha
IC
II N
Alpha
IC
N
Alpha
IAc
2*IC
Y
Alpha
IAc
IC
Y
Alpha
IAc
IC
Y
Alpha
IC
II N
Alpha
IC
II N
Alpha
IC
II N
Alpha
IAq IC
II N
Alpha
IC
II N
Alpha
IC
N
Alpha
IC
N
Alpha
IC
N
Alpha
IC
II N
Alpha
IAq IC
II N
Beta
II N
Beta
II N
IAq
Beta
IC
N
Beta
2*IC
N
Beta
2*IC
N
Beta
Beta
IAq
IAc
Taxonomy
II
II N
Beta
Ecological habitat
N
II Y
Cupriavidus
metallidurans CH34
Thiomonas
intermedia K12
Aerobic
Aerobic
Facultative
anaerobe
Facultative
anaerobe
Facultative
anaerobe
Facultative
anaerobe
Facultative
anaerobe
Facultative
anaerobe
Aerobic
Aerobic
Facultative
anaerobe
Facultative
anaerobe
Facultative
anaerobe
Aerobic
Facultative
anaerobe
Facultative
anaerobe
Aerobic
Multiple; freshwater Rhodospirillales,
Rhodospirillaceae
Mine site, low pH, Rhodospirillales,
heavy metalRhodospirillaceae
contaminated
habitats
Organotroph
Root/soil
Rhizobiales,
Bradyrhizobiales
Facultative photoStem/root/soil
Rhizobiales,
organomixotroph
Bradyrhizobiales
Facultative
Soil/water
Rhizobiales,
chemolithomixotroph
Bradyrhizobiales
Facultative
Soil/water
Rhizobiales,
chemolithomixotroph
Bradyrhizobiales
Photoorganomixotroph Temperate
Rhizobiales,
soil/water
Bradyrhizobiales
Facultative photoSoil/water
Rhizobiales,
organomixotroph
Bradyrhizobiales
Facultative
Soil/water
Rhizobiales,
photoorganomixotroph
Bradyrhizobiales
Facultative photoSoil/water
Rhizobiales,
organomixotroph
Bradyrhizobiales
Facultative photoSoil/water
Rhizobiales,
organomixotroph
Bradyrhizobiales
Organotroph
Soil/root nodules
Rhizobiales;
Rhizobiaceae
Organotroph
Soil/sludge
Rhizobiales,
Xanthobacteriaceae
Photoorganotroph
Soil and sewage
Rhodobacterales;
sludge
Rhodobacteraceae
Facultative
Multiple
Rhodobacterales;
photomixotroph
Rhodobacteraceae
Facultative
Multiple
Rhodobacterales;
photomixotroph
Rhodobacteraceae
Organomixotroph—
oxidizes benzene
anaerobically
Organomixotroph—
reduces Fe(III)
Multiple: anoxic
water sediments
Crater Lake,
Oregon, depth
18 feet—
anoxic sediments
Facultative
Soils, hydrogen
chemolithomixootroph oxidizing, organic
substrates
Facultative
PCB-containing
chemoorganomixotroph landfill/soils
Chemoorganotroph
PCB-containing
soils
Photomixotroph
Freshwater, sewage,
and activated
sludge
Obligate
Decantation tank
chemolithotroph
of zinc factory
Obligate
Soil and aquatic,
chemolithoautotroph
both marine and
freshwater
Rhodocyclales;
Rhodocyclaceae
Burkholderiales;
Comamonadaceae
Burkholderiales;
Comamonadaceae
Burkholderiales;
Burkholderiaceae
Burkholderiales;
Burkholderiaceae
Burkholderiales
Burkholderiales;
Burkholderiaceae
Burkholderiales
Multiple Rubisco forms in proteobacteria 1529
Table 2. (Continued)
Group
Rubisco form
Carb Species
IBc IAc IAq IC
Beta
IAq
Beta
Beta
Beta
IAq
N
N
IAc
Y
Gamma
IC
Metabolism
Ecological habitat
Taxonomy
Facultative
anaerobe
Obligate
chemolithoautotroph
Hydrogenophilales;
Hydrogenophilaceae
Aerobic
Obligate
chemolithotroph
Soil and aquatic,
both marine and
freshwater
Soil, ammonia
oxidizer
Nitrosomonadales;
Nitrosomonadaceae
Aerobic
Obligate
chemolithotroph
Soil, sewage,
freshwater, marine
Nitrosomonadales;
Nitrosomonadaceae
Aerobic
Obligate
chemolithoautotroph
Aerobic–
anaerobic
Obligate
chemolithoautotroph
Commonly found
Nitrosomonadales;
in strongly eutrophic Nitrosomonadaceae
environments
Marine
Chromatiales;
Chromatiaceae
Aerobic
Obligate
photolithoautotroph
II
II N
IC
Oxygen
N
Thiobacillus
denitrificans
sp. ATCC 25259
Nitrosospira
multiformis
ATCC25196
Nitrosomonas
europaea
ATCC 19718
Nitrosomonas
eutropha C91
Nitrosococcus
oceani
ATCC 19707
Allochromatium
vinosum
Gamma
IAc IAq
?
Gamma
IAq
N
Halorhodospira
halophila SL1
Aerobic
Obligate
chemolithoautotroph
Gamma
IAq
N
Alkalilimnicola
ehrlichei MLHE-1
Aerobic–
anaerobic
Facultative
chemolithoautotroph
Nitrococcus
mobilis NB-231
Halothiobacillus
neapolitanus
Methylococcus
capsulatus Bath
Aerobic
Obligate
chemolithoautotroph
Obligate
chemolithoautotroph
Organotroph
Gamma
IAc
Y
Gamma
IAc
II Y
Gamma
IAq
N
Gamma
IAc IAq
II Y
Gamma
IAc IAq
II Y
Gamma
IAc IAq
Y
a-Cyano
IAc
b-Cyano IBc
Aerobic
Aerobic
Hydrogenovibrio
marinus
Thiomicrospira
crunogena XCL-2
Acidithiobacillus
ferrooxidans
Aerobic–
anaerobic
Aerobic–
anaerobic
Aerobic
Obligate
chemolithoautotroph
Obligate
chemolithoautotroph
Obligate
chemolithoautotroph
Y
a-Cyanobacteria:
e.g Prochlorococcus sp.
Aerobic
Obligate
photolithoautotroph
Y
b-Cyanobacteria:
e.g. Synechococcus
PCC7942
Aerobic–
Obligate
microaerobic photolithoautotroph
The amino acid alignments for all bacterial Form IA and
Form IB small subunits are shown in Fig. 2. A major
difference between the proteins coded for by the Form IA
genes is to be found in the small subunits of the L8S8
enzymes. Figure S2 in Supplementary data available at
JXB online also shows an alignment and phylogenetic tree
for the corresponding large subunits, to demonstrate the
much smaller differences in this protein subunit. All the
Form IAq small subunit sequences are characterized by
a six-amino-acid insertion at the N-terminal end of the
protein. This holds for all the genes which are present in
the database and where known gene arrangements are
available. Interestingly, the alignment in Fig. 2 showing
an equivalent insertion is also missing from Form IBc
small subunits in cyanobacteria. Where pairs of Form IAc
and IAq Rubiscos are present in the same organism
Freshwater lakes
and intertidal
sandflats
Alkaline,
hypersaline
aquatic
Alkaline,
hypersaline, Mono
Lake, California
Marine, surface
waters
Saline aquatic/soil
Soils, able to use
methane and
sugars for carbon
Marine
Marine,
hydrothermal vents
Soil and aquatic,
both marine and
freshwater
Generally open
ocean marine
environments
Freshwater,
marine, lichens,
crusts, mats
Chromatiales;
Chromatiaceae
Chromatiales;
Ectothiorhodospiraceae
Chromatiales;
Ectothiorhodospiraceae
Chromatiales;
Ectothiorhodospiraceae
Chromatiales;
Halothiobacillaceae
Methylococcales;
Methylococcaceae
Thiotrichales;
Piscirickettsiaceae
Thiotrichales;
Piscirickettsiaceae
Acidithiobacillales;
Acidithiobacillaceae
Cyanobacteria
Cyanobacteria
(Hyrogenovibrio marinus, Thiomicrospira crunogenea,
Chromatium vinosum, and Acidithiobacillus ferrooxidans), the six-amino-acid insertion is always present in
the Form IAq small subunit. There are two main groupings of Form IA small subunit sequences, one containing
the a-cyanobacteria, and the other the chemoautotrophic
proteobacteria. However, the chemoautotrophs Hydrogenovibrio marinus, Thiomicrospira crunogenea, and
Chromatium vinosum group phylogenetically with the
cyanobacteria as far as the Form IAc sequences are
concerned. Figure S3 in Supplementary data available at
JXB online is included for completeness, showing amino
acid sequence alignments of bacterial Form I small
subunits together with a phylogenetic tree. Again the
distinction between Form IAc and Form IAq is readily
apparent.
1530 Badger and Bek
Fig. 1. The structural gene arrangement for various forms of
Rubisco found in bacteria. Forms IAq, IAc, and Form II L and S
genes are shown for Thiomicrospira crunogenea. A Form IC L and S
gene structure from Bradyrhizobium japonicum is shown as well as
Form IBc from Synechococcus PCC7942. The contextual gene arrangements found with these Rubisco small and large subunit genes were
obtained using the JGI Integrated Microbial Genomes browser (http://
img.jgi.doe.gov/cgi-bin/pub/main.cgi) and are discussed further in
the text.
The major difference between the two Form IA Rubisco
small subunit types is obviously that the Form IAc is
associated with carboxysomes while the Form IAq
probably is not. Does this mean that this insertion is in
some way associated with the ability or inability to form
carboxysomes? From the crystal structure of the Form IAc
Rubisco from Halothiobacillus neapolitanus (1SVD; CA
Kerfeld et al., unpublished data, 2005) and the crystal
structure of the cyanobacteria Form IBc enzyme (Newman
et al., 1993; Newman and Gutteridge, 1994) it can be
deduced that this region of the small subunit sequence
occurs at a position which would be on the top exterior
face of each small subunit, with potential interactions
between adjoining small subunits and perhaps external
proteins. Given that there are predicted interactions
between the Rubisco holoenzyme and b-carboxysome
shell proteins (Ludwig et al., 2000; Long et al., 2005) an
insertion at this position could disrupt these interactions if
they were occurring with the small subunits. Interestingly,
in the Form IBc enzyme, this insertion point would be
immediately adjacent to the conserved KERRY amino
acid domain region of the small subunit, which has been
shown to be present in the three or four small subunit
repeat regions present in the b-carboxysomal shell protein
CcmM (Ludwig et al., 2000). This region has been
implicated in interactions between CcmM carboxysome
protein and the large subunit of Rubisco, but the reality is
not clear. However, the large carboxysomal shell proteins
of a-carboxysomes, CsoS2 and CsoS3, show no analogous Form IA small subunit repeats (data not shown) so
the analogy between these types of interactions is not
clear at present.
The presence of two Form IA enzymes in Chromatium
vinosum (Allochromatium vinosum) (Table 2) was noted
some 20 years ago (Viale et al., 1990; Kobayashi et al.,
1991), but its significance was previously unrecognized.
Unfortunately, there is no genome sequence for this
organism nor extension of sequence around the two gene
sets. However, the analysis of small subunits here would
clearly predict that Chromatium may have a set of
a-carboxysome genes and even functional carboxysomes.
No carboxysomes have been reported for Chromatium;
however, sulphur granules similar in structure to carboxysomes with a surrounding protein membrane have been
described (Jordan and Chollet, 1985). The reality of
carboxysomes in this species remains to be established.
Another interesting observation concerns the Rubisco
gene arrangement and type in Thiobacillus denitrificans
(Fig. 3). This species has a Form IAq Rubisco and a set of
a-carboxysome genes that are somewhat removed from
the Form IAq gene set and closer to the Form II gene.
The above analysis would suggest that Thiobacillus
denitrificans should not possess functional carboxysomes.
The fact that carboxysomes have not been observed in this
species (Cannon et al., 2003) has been noted and is
consistent with these predictions. Perhaps Thiobacillus
denitrificans originally contained two different Form IA
Rubisco gene sets as Acidithiobacillus ferrooxidans does,
but they have been lost as this species has modified its
ecological niche range (see discussion below).
Previous assessment of Rubisco forms in bacteria has
aggregated all Form IA Rubiscos together without making
any distinction between those which are part of carboxysome structures and those that are not. Unfortunately,
this distinction is not easily made on large subunit
sequences which have generally been the basis of enzyme
form classification. However, there is a very important
relevance to recognizing the present or absence of
carboxysomes with respect to recognizing if certain
bacteria are undertaking an active CCM to enhance the
effectiveness of CO2 capture. CCMs have been recognized in gamma- proteobacteria with Form IAc Rubiscos
and carboxysomes (Baker et al., 1998; Dobrinski et al.,
2005) and it can be reasonably assumed that they will
occur regularly when a Form IAc Rubisco is found in the
genome.
Multiple Rubisco forms in proteobacteria 1531
Fig. 2. (A) An alignment of the amino acid sequences of proteobacteria and cyanobacteria Form IA and B small subunits. The sequences were
aligned with ClustalX 1.8.1. The amino acid sequences used for this alignment were obtained from the JGI Integrated Microbial Genomes browser
(http://img.jgi.doe.gov/cgi-bin/pub/main.cgi) and the NCBI protein database (http://www.ncbi.nlm.nih.gov/entrez/). The full species names are given
in the accompanying phylogram tree (B) and the amino acid sequences are listed in Table S2 available at JXB online. Form IAq Rubisco sequences
are boxed. (B) An NJ unrooted tree of Rubisco Form I A and B subunits aligned in (A). The guide tree was produced using ClustalX 1.8.1 and drawn
using TreeView 1.5.1.
Gene arrangements associated with Form I and
II Rubiscos
Previous reviews have spent some time focusing on the
flanking genes which are associated with different
Rubisco genes in bacteria (Kusian and Bowien, 1997;
Shively et al., 1998; Tabita, 1999; Dubbs and Tabita,
2004). The possible variation is illustrated by the arrangements shown in Figs 1 and 3, and a summary of these
gene arrangement patterns is given in Table 3.
Form II Rubiscos are found in two arrangements. One
arrangement is in association with the cbb genes which
code for enzymatic components of the CBB cycle (Cbb
metabolic). In the second arrangement, the Form II
Rubisco large subunit gene is associated with the
downstream presence of the cbbQ and cbbO genes.
Previous analyses of these genes and the function of their
protein products in Hydrogenovibrio marinus (Hayashi
et al., 1999) have indicated that they may be involved in
some form of post-translational regulation of both Form II
and Form IAq enzymes (see below) in these species, with
the CbbQ protein possessing two nucleotide-binding
motifs showing interactions with ATP and Mg2+ (Hayashi
and Igarashi, 2002).
These LII-cbb(metabolic) and LII-cbbQ-CbbO gene
arrangements are well correlated with the metabolic
functioning of the organisms in which they occur (Kusian
and Bowien, 1997; Hayashi et al., 1999; Tabita, 1999;
Scott et al., 2006). Facultative autotrophs which can
switch between CBB cycle CO2 fixation and external
organic substrates as sources of carbon have the
L-Cbb(metabolic) arrangement, showing clustering with
Cbb metabolic genes. This presumably allows the expression of the CBB cycle to be tightly regulated by the
availability of organic and inorganic substrates. On the
other hand, obligate autotrophs must continually express
the CBB cycle and close regulation with Rubsico is not
necessary. In these cases the Cbb metabolic genes are
located elsewhere in the genome.
It appears as though all Rubisco Form IC LS genes are
found in association with cbbX located downstream. This
gene codes for a protein of unknown function but is
thought to be involved with some form of posttranslational activation of the Rubisco (Li and Tabita,
1997). This is generally, but not always, correlated with
the presence of a number of cbb metabolic genes in the
same cluster and the fact that Form IC enzymes appear to
be associated with facultative autotrophs (mixotrophs) (see
Table 2). These species include the beta-proteobacteria
Rubrivivax gelatinosus, Cupriavidus necator, and
Burkholderia xenovorans, and the alpha-proteobacteria
Bradyrhizobium japonicum, Bradyrhizobium BTA1,
Rhodopseudomonas palustris (various strains), Nitrobacter
hamburgensis, Nitrobacter winogradsky, Xanthobacter
1532 Badger and Bek
Fig. 2. (Continued )
flavus, Paracoccus denitrificans, Rhodobacter sphaeroides, Rhodobacter capsulatus, and Sinorhizobium meliloti (deduced from gene arrangements accessed at JGI
Integrated Microbial Genome Browser, http://img.jgi.doe.
gov/cgi-bin/pub/main.cgi). Two exceptions to this correlation appear to be the beta-proteobacterium Nitrosospira
multiformis and the gamma-proteobacterium Nitrosococcus
oceanii, which are both obligate chemolithoautotrophs and
have a single Form IC Rubisco gene set.
The distinguishing feature of the Form IAq gene arrangement is the universal presence of two genes downstream of the LS gene pair (Kusian and Bowien, 1997;
Multiple Rubisco forms in proteobacteria 1533
Thiobacillus denitrificans ATCC25259
cbbO1, cbbQ1
cbbR1
rbcS, rbcL
Form IAq
cbbR3
cbbR2
cbbO2, cbbQ2 rbcL
Form II
csoS3, csoS2
α-carboxysome cluster
Fig. 3. The structural gene arrangement for Rubisco Form I and II genes in Thiobacillus denitrificans ATCC25259. The contextual gene
arrangements found with these Rubisco small and large subunit genes were obtained using the JGI Integrated Microbial Genomes browser (http://
img.jgi.doe.gov/cgi-bin/pub/main.cgi) and are discussed further in the text.
Table 3. A summary of gene arrangements associated with Form I and II Rubiscos in bacteria and their correlation with autotrophic
CO2 fixation and metabolic type
The conclusions arrived at in this table are discussed in more detail in the text.
Rubisco
form
Gene arrangement
Metabolic types
Other comments
Form II
Form IC
L-cbb(metabolic)
L-cbbQ-cbbO
LS-cbbX
Generally together with cluster of other cbb metabolic genes
cbb metabolic genes located elsewhere in genome
Generally together with cluster of other cbb metabolic genes
Form IAq
Form IAc
LS-cbbQ-cbbO
LS-a-carboxysome genes
Form IBc
LS 6 b-carboxysome
genes
Facultative autotrophs–mixotrophs
Obligate chemolithotrophs
Generally facultative
autotrophs/mixotrophs
Obligate and facultative chemolithotrophs
Obligate and facultative photo
and chemolithotrophs
Obligate photolithotrophs
Shively et al., 1998; Tabita, 1999). These two genes code
for CbbQ and CbbO and the distribution of the cbb metabolic genes elsewhere in the genome. Form IAq Rubisco
appears predominantly in obligate chemolithotrophs
such as the beta-proteobacteria Thiobacillus denitrificans,
Cupravidus metallidurans, and Nitrosomonas europaea,
and the gamma-proteobacteria Hyrogenovibrio marinus,
Thiomicrospira crunogenea, and Acidithiobacillus
ferrooxidans. However, it is also found in the alphaproteobacterium Rhodopseudomonas BisB5, the betaproteobacterium Hydrogenophaga pseudoflava, and the
gamma-proteobacterium Methylococcus capsulatus, which
are facultative in their use of organic and inorganic
electron donors for CO2 fixation.
Form IAc Rubiscos are found in both obligate and
facultative autototrophs (photo- and chemolithotrophs)
and are always associated together with a cluster of
a-carboxysome genes. The presence of Rubisco in acarboxysomes is presumably part of the CCM syndrome
found in these organisms (Badger and Price, 2003;
Dobrinski et al., 2005). The presence of Form IAc
Rubisco and a-carboxysomes in both a-cyanobacteria and
proteobacteria has been attributed to lateral gene transfer
events (Badger et al., 2002). The cbb metabolic genes are
found elsewhere in the genome.
Form IBc Rubisco is found only in b-cyanobacteria and
the LS genes may be associated with the b-carboxysome
gene cluster. In those species where they are not adjacent,
cbb metabolic genes located elsewhere in genome
cbb metabolic genes located elsewhere in genome
b-Carboxysome genes may be elsewhere in genome, cbb
metabolic genes elsewhere in genome
a b-carboxysome gene cluster is located elsewhere in the
genome. b-Cyanobacteria are the only bacteria with the
Form IB lineage of Rubisco.
Kinetic properties of Form I and II Rubiscos
Over the past 3.5 billion years, the earth’s atmosphere has
evolved so that oxygen has been elevated from 0% to
21% and CO2 has declined from >1% to very limiting
levels in the past 10 000 years (<0.025%) (Berner, 2001;
Berner and Kothavala, 2001; Berner et al., 2003). It is
clear that these atmospheric and environmental changes
have caused the evolution of the kinetic properties of
Rubisco both between and within form types (Badger and
Andrews, 1987; Badger et al., 1998; Tabita, 1999). This
evolution seems to be have been driven by the need to
optimize kinetic properties to match the nature of CO2
substrate availability and the O2 levels of the catalytic
environments in which they operate. However, the
evolution of Rubisco has also been intimately linked with
the development of CCMs which serve to capture decreasing levels of external inorganic carbon and elevate
CO2 around the catalytic site of the enzyme (Badger et al.,
1998).
Table 4 shows a summary of some of the relevant
kinetic properties of the various Form I and II Rubisco
proteins identified in this review and the potential
1534 Badger and Bek
presence of CCMs. Within each of the forms there is often
a considerable spread of kinetic values for Kc. This spread
may be associated with the exact nature of the CO2
environment in which Rubisco operates. If the external
CO2 is generally high during the CO2-fixation period in
which the Rubisco is active or there is intervention of
a CCM, then it is likely that higher Kc values will have
evolved. This is illustrated for the spread of values in nongreen algae, green algae, and higher plants. In all cases the
higher Kc values are found in species with active CCMs
while the low values are associated with what appears to
be passive acquisition of CO2 by photosynthesis (Badger
et al., 1998).
Form II Rubiscos are characterized by a low Srel value
(indicating a low discrimination against O2 as an alternative substrate), a poor affinity for CO2, and a relatively
high kcat. Based on these properties it has been generally
concluded that Form II enzymes are adapted to functioning in low-O2 and high-CO2 environments. There is no
clear evidence in proteobacteria that Form II Rubiscos
operate in conjunction with an active CCM, but they are
found in gamma-proteobacteria species, such as Thiomicrospira crunogenea and Hydrogenovibrio marinus
(Table 2), with multiple Rubisco forms where CCM
mechanisms clearly operate at some time in the life cycle
under some environmental conditions (Dobrinski et al.,
2005) (see discussion below).
Form IC enzymes show improved Srel values and this is
generally assumed to be the first measure of the evolution
of enzymes that are better adapted to increased levels of
O2 (Tabita, 1999). They have medium to low affinities for
CO2, indicating an adaptation to environments with
medium to high CO2 but with O2 present. Even though
they are closely related to Form ID enzymes associated
with red and non-green algae, they do not show the high
Srel and low Kc values found with some of these enzymes.
Form IAq enzymes show kinetic properties which
overlap with those found for Form IC Rubiscos. Unlike
their closely related Form IAc counterparts, they are not
associated with carboxysomes and are most probobly
adapted to environments with medium to high CO2 but
with O2 present. They are present together with Form IAc
enzymes in some gamma-proteobacteria, including Hydrogenovibrio marinus, Thiomicrospira crunogenea, and
Acidithiobacillus ferrooxidans (Table 2), where clearly
there is the potential for inorganic carbon accumulation at
some times (Cannon et al., 2001; Dobrinski et al., 2005).
Form IAc Rubiscos are somewhat poorly studied with
regard to their kinetic properties. They appear to have Srel
values that are less than the values for the Form IBc
Rubiscos associated with b-carboxysomes, and probably
medium to low affinity for CO2, although this is based on
a very limited sample (see Table S3 in Supplementary
data available at JXB online). Studies with the acyanobacteria Synechococcus and Prochlorococcus
marine species and the gamma-proteobacterium Thiomicrospira crunogenea (Hassidim et al., 1997; Dobrinski
et al., 2005) clearly show they are part of a CO2concentrating mechanism and are able to function at low
external CO2 levels as a consequence of this.
Finally, the Form IBc Rubiscos from b-cyanobacteria
are well studied, and have been shown to have medium
Srel values and the lowest affinities for CO2. They function
as part of a well-characterized b-carboxysome-associated
CCM, enabling efficient photosynthesis at very low levels
of external CO2.
The carbon fixation strategies and growth
habitats of proteobacteria
Within the bacteria, there are a range of metabolic
approaches to utilizing the CBB cycle and Rubisco for
CO2 fixation resulting in increased biomass. Two major
types of autotrophic behaviour can be discerned among
the various chemo- and phototrophs, namely obligate and
facultative. Obligate autotrophs are generally solely dependent on CO2 and the CBB cycle as the sole source of
Table 4. Summary of the kinetic properties of Form I and II Rubiscos found in bacteria and eukaryotes
The data are summarized from the more extensive table with references in Table S3 in Supplementary data available at JXB online. The potential
presence of CO2 concentrating mechanisms (CCMs) and carboxysomes being associated with Rubisco forms is also shown.
Rubisco
form
Species occurrence
Srel (Vc3Ko)/
(Kc3Vo)
kcat(C)
(s1 site1)
II
IC
ID
IAq
IAc
Proteobacteria
Proteobacteria
Non-green algae
Proteobacteria
Cyanobacteria,
proteobacteria
Cyanobacteria
9–15
44–75
129–238
30–53
33–37
5.7
2–3.2
1.3–1.6
3.7
6.5
38–56
54–83
78–90
IBc
IB
Green algae
Plants
Kc (lM)
CCM
Functional niche
80–247
22–100
6–59
29–138
72
–/+
–?
–/+
–?
++
2.6–11.4
130–180
++
2.9–4.2
12–38
10–32
–/+
–/+
Medium to high CO2, when O2 is low
Medium to high CO2 levels with O2 present
Low CO2, high O2, with or without a CCM
Medium to low CO2 with O2 present
Low CO2, low–high O2, with CCM/carboxysome
syndrome
Low CO2, low–high O2, with CCM/carboxysome
syndrome
Low CO2, high O2, with or without a CCM
Multiple Rubisco forms in proteobacteria 1535
net carbon for growth, and these largely include the
chemo- and photolithoautotrophs. Facultative autotrophs
show metabolic flexibility allowing these organisms to
grow on organic substrates as alternative carbon and
energy sources (Kusian and Bowien, 1997; Shively et al.,
1998). The organotrophs use only organic substrates but
use the CBB cycle for CO2 fixation and synthesis
processes.
Obligate chemo- (e.g. many beta- and gammaproteobacteria) and photoautotrophs in Table 2 can be
seen as specialists in their growth strategies, and their
metabolism is optimized for oxidation of a small number
of reduced donors (Shively et al., 1998). In chemoautotrophs, the electron donors include H2, reduced
nitrogen (e.g. NH+4 , NO–2), sulphur (e.g. S2O2–3, H2S),
metals (e.g. Fe2+, Mn2+), and carbon compounds (e.g. CO,
CH4, CH3OH). These reduced donors are produced from
various sources, including anthropogenic, biological, and
geological processes, which include industry and agriculture, anaerobic metabolism in sediments and animal guts,
and volcanic activity (Shively et al., 1998, 2001). Photoautotrophs can use either anoxygenic (PSI driven) photosynthesis with electrons derived from reduced inorganic or
organic substrates; or oxygenic (PSI plus PS2) photosynthesis with electrons derived from H2O. Anoxygenic
photosynthesis is found within the proteobacteria, while
oxygenic photosynthesis is confined to cyanobacteria.
By contrast, facultative autotrophic bacteria display
a great metabolic diversity which allows them to grow on
a wide range of substrates. A clear distinction with
obligate autotrophs is that catabolic (including the CBB
pathway) and anabolic pathways of carbon metabolism
are inducible (Kusian and Bowien, 1997; Shively et al.,
1998; Dubbs and Tabita, 2004). Many of these bacteria
have the ability to grow mixotrophically, enabling
situations where autotrophic and heterotrophic growth
metabolisms can be simultaneously engaged.
Obligate and facultative autotrophs occupy considerably
different ecological niches as a result of their different
approaches to acquiring reduced substrate electron donors.
Obligate photo- and chemoautotrophs grow where there is
a continuous or fluctuating supply of reduced inorganic
compounds and a low abundance and turnover of organic
donors. Conversely, facultative chemoautotrophs grow
best when both inorganic and organic reduced compounds
are present.
One important consideration with chemoautotrophs is
that the reduced energy substrates, such as H2 and reduced
sulphur compounds, are produced anaerobically, in sediments or hydrothermal vents, for example, and these
subsequently accumulate and diffuse to aerobic zones.
Many chemoautotrophic bacteria require O2 as an electron
acceptor and are therefore limited to the oxic layer where
they compete with chemical oxidation of the reduced
inorganic electron donors. This results in these bacteria
positioning themselves at the oxic/anoxic interface of
sediments, stratified lakes, or hydrothermal vents. This
interface is usually highly dynamic and would be
expected to present a variable environment where fluctuations in reduced compounds, CO2, and O2 would be
expected (Shively et al., 1998; Overmann, 2001). An
added limitation for photoautotrophs is that they must also
gain access to adequate light and this further determines
their potential distribution.
Correlations between Rubisco complement and
ecological niche
The above considerations result in autotrophic proteobacteria growing in a range of environments which differ
widely in the spatial and temporal variation in CO2 and
O2 which influence the effectiveness of Rubisco in fixing
CO2. Table 2 shows combined information about Rubisco
complement, metabolic type, and the nature of the
environments in which included bacteria generally occur.
Using the general conclusions about functional niches
which may be occupied by the different Form I and II
Rubiscos (Table 4), the following conclusions and speculations between Rubisco complement and ecological
adaptation can be drawn for bacterial species shown in
Table 2.
Alpha-proteobacteria
The Rubisco-containing alpha-proteobacteria encompass
a number of species in the Rhodospirilalles, Rhizobiales,
and Rhodobacteriales. They are an extremely flexible
metabolic group including members that can perform
anoxygenic photosynthesis (the non-sulphur purple bacteria), aerobic or anaerobic respiration, and fermentation.
During phototrophy and chemo-autotrophy they can use
a number of electron donors including H2 and organic
carbon compounds, while acquiring new carbon through
the CBB cycle. However, the CBB cycle performs two
distinct functions in these organisms, acting to supply new
carbon during autotrophic growth and as an electron sink,
which participates in maintaining redox balance during
heterotrophic growth on organic substrates (Dubbs and
Tabita, 2004). This is particularly true for phototrophs
where excess electrons are produced as a result of
oxidation of reduced organic substrates.
The alpha-proteobacteria inhabit environments which
vary considerably in their O2 and CO2 levels. Inhabiting
soil, sludge, water, and plant roots and stems, some
environments can be anaerobic and high in CO2 (sludge,
water-filled soils), high in CO2 but aerobic (aerated soils,
plant root nodules), or low in CO2 and aerobic (plant stem
nodules). It is not surprising then that, overall, this
group of bacteria contains the four main proteobacterial
1536 Badger and Bek
Rubiscos, Forms II, IC, IAq, and IAc (Table 2), and the
complement in each species can be interpreted in terms of
an adaptation to the particular characteristics of the
environment they inhabit.
The presence of Form II and Form IC enzymes has been
extensively studied in a number of the non-sulphur purple
photoautotrophic bacteria shown in Table 2, including
Rhodospirillim rubrum, Rhodopseudomonas palustris,
Rhodobacter sphaeroides, and Rhodobacter capsulatus
(Dubbs and Tabita, 2004). The response of the expression
of both Rubisco forms has been well studied, being
related to the metabolic state and redox state of the cells
and the CO2 and O2 levels (see Kusian and Bowien, 1997;
Dubbs and Tabita, 2004). The expression of Form II
Rubisco varies somewhat between species, and can be
highly expressed under photoheterotrophic, photoautotrophic, and chemoautotrophic growth conditions (Dubbs
and Tabita, 2004). However, in general it does appear that
the engagement of Form II is favoured under conditions of
high CO2 (>1.5%) and low O2. Form IC Rubisco has been
shown to be induced during photo-autotrophic growth at
medium levels of CO2 (<1.5% CO2; Dubbs and Tabita,
2004). Based on the kinetic properties of the Rubisco
forms (Table 4), it would be reasonable to expect that
Form II Rubiscos would be favoured at high CO2 and low
O2, while Form IC would be utilized when environments
became more aerobic with somewhat reduced levels of
CO2.
While most of the non-purple sulphur phototrophic
bacteria have Form IC and Form II Rubiscos, there is
variation. Rhododspirillum rubrum has only one Form II
gene and presumably is restricted to engaging the CBB
cycle under conditions of high CO2 and low O2. By contrast, Rhodopseudomonas palustris BisB5 and Rhodobacter
sphaeroides ATCC 17025 have Form II, Form IC, and
Form IAq Rubisco gene sets. A clear distinction between
the kinetic advantages of Form IC versus Form IAq
enzymes is not immediately obvious (see Table 4). One
could speculate that Form IAq may be more associated
with growth under photolithoautotrophic conditions as
these genes are not associated with other cbb metabolic
genes and may be engaged when inorganic substrates are
oxidized. Under these conditions the CO2 may be reduced
(0.1–1%) and there may be more O2.
The photoorganotroph Bradyrhizobium sp. BTA1 is an
interesting variant in that it has a Form IAc Rubisco
suggesting a CCM and operation of Rubisco under low
CO2 conditions. Bradyrhizobium BTA1 is a stem-nodulating
bacterium of legumes (Giraud et al., 2000) and as such
may perform anoxygenic photosynthesis in a relatively
aerobic and low CO2 (<0.1%) environment, being
surrounded by the green photosynthetic tissues of the
plant stem and having immediate access to the gaseous
atmosphere surrounding the stem. Thus Form IAc
Rubisco may be important in the light during photo-
organotrophic growth, while Form IC Rubisco may be
used in the dark when CO2 levels are presumably much
higher.
The response of the CBB cycle and Rubisco expression
is likely to be similar between the photo- and chemoautotrophs in the alpha-proteobacteria, except that the
need for the CBB cycle to dissipate excess electrons will
be less critical (Kusian and Bowien, 1997; Dubbs and
Tabita, 2004). The facultative chemotroph Acidophilum
cryptum has Form II and Form IC Rubiscos, suggesting
similar roles to the phototrophs. By contrast, the organotrophs Bradyrhizobium japonicum and Xanthobacter
autotrophicus only have Form IC, suggesting engagement
of Rubisco under conditions of medium CO2 and suboxic
O2 as may be found in soils and root nodules. However, it
is significant that two of the facultative chemolithomixotrophs, Nitrobacter hamburgensis and Nitrobacter
winogradski, have Form IAc Rubisco and a-carboxysome
genes as well as Form IC Rubisco. This suggests the
potential engagement of CCMs under low CO2 conditions. They grow by oxidizing both NO2 and, to a lesser
extent, organic substrates, and inhabit soil and water
environments (Starkenburg et al., 2006). This would
suggest that Form IAc Rubisco and a CCM may be
engaged under nitrite growth, particularly in water where
CO2 may be reduced, while Form IC may be more
important during growth on organic substrates.
Beta-proteobacteria
The beta-proteobacteria encompass largely obligate and
facultative chemotrophs but they also include some
photomixotrophs. Members of the Burkholderales and
Nitrosomonadales are the predominant representatives of
this group. They are generally less flexible in their
metabolic strategies than many of the alpha-proteobacteria
and this is related to the greater number of obligate
chemolithotrophs represented in this group. The four
proteobacterial Rubisco forms (II, IC, IAq, and IAc) are
also found in various members of the group.
The regulation of Rubisco expression has been studied
extensively in the chemo-organotroph Ralstonia eutropha
(Bowien and Kusian, 2002). It has only Form IC Rubisco
which is induced under conditions of autotrophic growth
on organic substrates and is repressed during heterotrophic
growth. Burkholderia xenovorans appears similar in
oxidizing organic substrates and having Form IC Rubisco.
Engagement of Rubisco under medium CO2 with some O2
present is suggested. The photomixotroph Rubrivivax
gelatinosus has two Form IC Rubiscos and homologies
with the non-purple sulphur alpha-proteobacteria. However, the absence of a Form II Rubisco suggests CBB
cycle activity under moderate CO2 levels with suboxic
levels of O2 rather than anaerobic conditions.
Multiple Rubisco forms in proteobacteria 1537
Several of the obligate chemolithotrophs have a single
Form IAq Rubisco. This includes Cupriavidus metallidurans, Thiobacillus denitrificans, and Nitrosomonas
europaea. This is consistent with the obligate and more
specialized metabolic nature of these bacteria. However,
as discussed above, the advantage of a Form IAq Rubisco
is not immediately obvious but is correlated with chemolithotrophic-driven CO2 fixation at moderate CO2 and O2
levels. Thiobacillus denitrificans is interesting (see Fig. 3)
in that it has a Form IAq Rubisco and a-carboxysome
genes nearby. This may have been a remnant of a previous
metabolic state which was adapted to lower CO2 environments or a transition to a future state. Hydrogenophaga
pseudoflava, a CO and H2 oxidizer, also has a single IAq
Rubisco, and would suggest that autotrophic metabolism
using the CBB cycle is primarily associated with chemolithotrophic metabolism oxidizing H2. Nitrospira multiformis is an exception being an obligate chemolithotroph,
deriving its energy from ammonia oxidation, but having
a single Form IC Rubisco. This is also true of other
Nitrospira species (Utaker et al., 2002). Whether this is
associated with advantages of this particular Rubisco,
such as higher CO2 environments, or a result of stochastic
phylogenetic events remains to be determined.
Form IAc Rubisco and a-carboxysome genes are present
in the obligate chemolithotrophs Nitrosomonas eutropha
and Thiomonas intermedia K12. Nitrosomonas eutropha,
like Nitrosomonas europaea is an ammonia-oxidizing
chemotroph, and is found in eutrophic environments (Chain
et al., 2003). The distinction between it and N. europaea
must be found in the nature of the levels of CO2 commonly
found during CO2 fixation. Presumably N. eutropha
occupies aerobic but eutrophic environments where other
CO2-fixing activity must deplete the inorganic carbon and
thus a high affinity CO2-fixation system is needed.
Thiomonas intermedia is a sulphur-oxidizing bacterium
and contains both a Form IAc and a Form II Rubisco. This
could be explained by variation in its environment between
anaerobic soils with high CO2 and low O2 and more
aerobic aquatic environments with low CO2.
Gamma-proteobacteria
The gamma-proteobacteria encompass a range of obligate
photo- and chemolithotrophs, and some organotrophs
which encompass families that include Chromatiales,
Thiotrichales, and Acidothibacillales. Form IAq and Form
IAc Rubiscos are dominant in this group, but there are
some occurrences of Form IC and Form II enzymes. The
photosynthetic (anoxygenic photosynthesis) purple sulphur bacteria are contained within the Chromatiales, but
of these only Allochromatium vinosum is represented in
Table 2. Species in the subfamily Chromatiaceae generally inhabit freshwater lakes and intertidal sandflats, while
species in the Ectothiorhodospiraceae are associated with
hypersaline waters. Many of the species oxidize reduced
sulphur with and without the aid of anoxygenic photosynthesis.
Evidence for the potential participation of Form IAc
Rubisco and carboxysomes in a high-affinity CCM in
chemolithotrophs comes most strongly for some species
in this group. Carboxysomes were originally discovered in
the Thiobacilli, most of which are included in this group
(Cannon et al., 2001, 2003). Insertional mutagenesis of
carboxysome genes in Halothiobacillus neapolitanus
resulted in a high CO2 requirement for growth (Baker
et al., 1998); active accumulation of inorganic carbon has
been demonstrated in Thiomicrospira crunogenea, and the
affinity for inorganic carbon is increased when the cells
are grown at limiting CO2 (Dobrinski et al., 2005), and
Form IAc Rubisco is induced when Hydrogenovibrio
marinus is grown at low CO2 (Yoshizawa et al., 2004).
This evidence strongly supports the notion that Form IAc
Rubisco and carboxysomes are a part of an active
inorganic carbon acquisition mechanism in many chemoautotrophs.
Single genes for Form IAq Rubisco are found in the two
members of the Ectothiorhodospiraceae, Halorhodospira
halophila and Alkalilimnicola ehrlichei, which inhabit
alkaline hypersaline lakes. These species probably occur
at the chemocline within these environments, oscillating
between anaerobic and microaerobic conditions, and
experiencing variable CO2 levels dependent on positioning within the chemocline interface. Even though these
lakes are alkaline, and bicarbonate and carbonate rather
than CO2 may dominate the inorganic carbon pool, the
CO2 may be sufficiently high and growth relatively slow
to allow fixation by the Form IAq Rubisco but not high
enough to favour the use of a Form II enzyme. Nitrococcus
mobilis in this group is a nitrite-oxidizing bacterium and
contains a single Form IAc Rubisco and a-carboxysome
genes. This species inhabits marine surface-water environments and as such experiences a more aerobic and reduced CO2 environment. The presence of carboxysomes
and a CCM is consistent with this.
Allochromatium vinosum has been shown to have
a dominant Form IAq Rubisco (Viale et al., 1989;
Kobayashi et al., 1991). However, it is also apparent that
it contains a second Form IAc Rubisco which is probably
associated with the potential to make carboxysomes. The
observation that the Form IAc genes seem weakly
expressed may indicate that this is a gene that is not used
for significant growth; however, it may also indicate that it
is only expressed at low CO2 levels which have not yet
been tested. However, the fact that it occurs indicates that
other species in this group may have flexibility to grow in
medium and low CO2 conditions.
The presence of Form IAc and Form II Rubiscos in the
halotolerant sulphur-oxidizing species Halothiobacillus
1538 Badger and Bek
neopolitanus would indicate a constant potential for this
species to inhabit higher O2 and lower CO2 environments
than for the Form IAq species, as well as photosynthesis
under anaerobic and high CO2 conditions. By contrast,
the acid-tolerant and iron-oxidizer Acidithiobacillus
ferrooxidans possesses both a Form IAc and a Form IAq
Rubisco. This would suggest a potential adaptation for
variable CO2 and O2 levels, giving it the ability to fix
carbon both at low CO2 and ambient O2 as well as
medium CO2 and lower O2 levels, but an absence from
anaerobic environments.
Finally, within the Thiotrichales, Hydrogenovibrio
marinus and Thiomicrospira crunogenea both have
a complement of Form II, Form IAq, and Form IAc
Rubiscos. As mentioned above, both these species have
provided evidence for the dynamic nature of the expression and function of these genes in relation to the environment. In Hydrogenovibrio, although no cell physiology
has been conducted, Form II Rubisco was expressed at
very high CO2 (15%), Form IAq was expressed at 2%
CO2, and Form IAc became dominant when grown at
0.15% CO2. Somewhat unexpectedly, all three forms were
expressed at 0.03% CO2 (Yoshizawa et al., 2004). In
Thiomicrospira, Rubisco expression has not been examined, but the cell physiology responded to growth at
limiting CO2 by increasing the affinity of CO2 fixation for
external inorganic carbon and actively accumulating
inorganic carbon (Dobrinski et al., 2005). This was
interpreted as clear evidence for a CCM associated with
carboxysome function. In the case of Thiomicrospira,
a CCM may serve to increase the supply of CO2 in
a hydrothermal-vent habitat. It was speculated that
temperature differences between the bottom water (2 C)
and hydrothermal efflux (35 C) may create turbulent
eddies in the water column which could expose cells to
oscillations in CO2 (ranging from 2 lM to 20 lM) and O2
levels. In both these species, the Form II Rubiscos are
presumably expressed in the more anaerobic environments
at high CO2.
The Rubisco genes have been surveyed within a range of
obligate sulphur-oxidizing chemolithotrophs closely related to Thiomicrospira crunogenea and Hydrogenovibrio
marinus (Tourova et al., 2006). There is a high occurrence
of Form IAc Rubiscos in these species, indicating a general
use of CCMs and carboxysomes to acquire CO2 in this
larger group. One interesting fact is that Thiomicrospira
crunogenea is known as the fastest-growing mesophyllic
chemolithoautotroph (Jannasch et al., 1985), and indeed
high growth rates are common for many of these sulphuroxidizing bacteria (Tourova et al., 2006). The employment
of CCMs is often most critical in fast-growing species
where uptake of external inorganic carbon can exceed
the supply rate to the surface of the cell, resulting in
Ci-limiting conditions and the need to use both CO2 and
HCO–3 as carbon sources (Badger, 1987).
Rubisco genes present in different environments
There have been a number of studies surveying the
presence of the different Rubisco form genes in bacterial
assemblages from different environments, which have
made some inferences about their phylogenetic, metabolic,
and ecological significance. Sampling of marine phytoplankton communities has shown the presence of Form IA
and IB Rubiscos associated with a- and b-cyanobacteria,
together with Form ID Rubiscos associated with nongreen algae such as diatoms and coccolithophorids
(Pichard et al., 1997a, b; Paul et al., 1999). Analysis of
deep-sea environments, including hydrothermal vents, has
identified the presence of Form III Rubisco genes from
archaea and Rubisco Forms II, IAc and IAq, IB, IC, and
ID (Elsaied and Naganuma, 2001; Elsaied et al., 2007).
The Form II and Form IA Rubiscos appeared most highly
related to members of the Thiobacilli, Thiotrichales,
where two or three Rubisco forms may be present in the
same genome (see above). Rubisco genes have also been
found in bacteria from an anoxic hypersaline basin (van
der Wielen, 2006). Here there was evidence for Form
I and II Rubiscos most closely related to obligate chemolithoautotroph Thiobacillus species and Hydrogenovibrio
marinus. Sampling from a groundwater aquifer, which is
anaerobic to microaerobic, has found the presence of
Form II and Form IA Rubiscos again most closely related
to obligate chemolithoautotrophs such a Halothiobacillus,
Acidothiobacillus, and Thiobacillus species (Alfreider
et al., 2003).
Terrestrial sites such as soils (Selesi et al., 2005; Tolli
and King, 2005) and volcanic deposits (Nanba et al.,
2004) have also been sampled. These reveal the presence
of facultative and obligate chemolithotrophs. Agricultural
and pine forest soils showed a narrow diversity of Form
IA Rubiscos related to Nitrobacter species, and a high
diversity of Form IC Rubiscos related to a range of alphaand beta-proteobacteria (Selesi et al., 2005; Tolli and
King, 2005). Volcanic deposits showed Form IC Rubisco
related to a number of facultative chemolithotrophs from
the alpha-proteobacteria. However, microbial mats overlying the volcanic material were dominated by Form IA
Rubiscos related to Thiobacilli species (Nanba et al.,
2004).
Conclusions
It is apparent that genes for all five forms of Form I and
Form II Rubisco are widely distributed among proteobacteria and cyanobacteria. This is most obvious in proteobacteria where up to three different forms can exist within
the one bacterial genome. There are two ways to view the
presence of different and multiple Rubiscos within an
organism. First, it can be viewed as a simple result of
phylogeny and stochastic inheritance. However, it can
Multiple Rubisco forms in proteobacteria 1539
also be viewed as a specific result of the adaptive
evolution to particular ecological environments that each
bacterium has undergone. This review has attempted to
examine the extent to which the complement of Rubisco
genes within a single bacterial species may fit with the
environmental conditions and lifestyle it is adapted to.
Bacteria with single Rubisco genes can be seen as
specialists, with CO2 fixation by the CBB cycle occurring
under a fairly constant and predictable set of environmental conditions relating to CO2 and O2 levels. Multiple
Rubisco gene sets can be found in bacteria experiencing
a more flexible lifestyle with respect to autotrophic and
heterotrophic metabolism, as well as large fluctuations
in the CO2 and O2 within the environment. It is interesting
to note that this indicates a contrasting difference in evolutionary strategies employed by obligate photoautotrophs
such as cyanobacteria, algae, and plants which possess
a single Rubisco form, and proteobacteria. In obligate
photoautotrophs, adaptation of Rubisco is based on the
evolution of the kinetic properties of a single Rubisco and
the flexible implementation of some form of CCM. The
notion that bacteria may employ a number of different
Rubisco forms, together with the implementation of a
CCM to achieve evolutionary adaptation and acclimation
is quite different. However, in reaching any clear conclusions about the role of different Rubisco forms in most
species, it is evident there is a real lack of good data
which describe the ecophysiology of proteobacteria in
relation to the CO2 and O2 levels in their environments,
and more studies of this nature relating gene expression of Rubisco form to environment and physiology
are needed.
Supplementary data
Figure S1. Alignment and phylogentic tree of Bacterial
Rubisco Form I and Form II Large Subunit amino acid
sequences. Alignments and phylogenetic tree were produced as described in Figure 2.
Figure S2. Alignment and phylogenetic tree of Bacterial Rubisco Form IA and IB Large Subunit amino acid
sequences. Alignments and phylogenetic tree were produced as described in Figure 2.
Figure S3. Alignment and phylogenetic tree of Bacterial Rubisco Form I Small Subunit amino acid sequences.
Alignments and phylogenetic tree were produced as
described in Figure 2.
Table S1. The presence of various Form I, II, III and IV
Rubiscos in Bacteria, Eukaryotes and archaea.
Table S2. The species names and amino acid sequences
for Rubisco large and small subunits used for alignments
and phylogenetic trees used in the paper.
Table S3. The kinetic properties of Form I and II
Rubiscos found in Bacteria and Eukaryotes.
References
Alfreider A, Vogt C, Hoffmann D, Babel W. 2003. Diversity of
ribulose-1,5-bisphosphate carboxylase/oxygenase large-subunit
genes from groundwater and aquifer microorganisms. Microbial
Ecology 45, 317–328.
Ashida H, Danchin A, Yokota A. 2005. Was photosynthetic
RuBisCO recruited by acquisitive evolution from RuBisCO-like
proteins involved in sulfur metabolism? Research in Microbiology 156, 611–618.
Ashida H, Saito Y, Kojima C, Kobayashi K, Ogasawara N,
Yokota A. 2003. A functional link between RuBisCO-like
protein of Bacillus and photosynthetic RuBisCO. Science 302,
286–290.
Atomi H. 2002. Microbial enzymes involved in carbon dioxide
fixation. Journal of Bioscience and Bioengineering 94, 497–505.
Badger MR. 1987. The CO2-concentrating mechanism in aquatic
phototrophs. In: Hatch MD, Boardman NK, eds. The biochemistry of plants, Vol. 10. San Diego, CA: Academic Press, 219–274.
Badger MR, Andrews TJ. 1987. Co-evolution of Rubisco and
CO2 concentrating mechanisms. In: Biggins J, ed. Progress in
photosynthesis research, Vol. III. Dordrecht: Martinus Nijhoff
Publishers, 601–609.
Badger MR, Andrews TJ, Whitney SM, Ludwig M,
Yellowlees DC, Leggat W, Price GD. 1998. The diversity and
co-evolution of Rubisco, plastids, pyrenoids and chloroplastbased CCMs in the algae. Canadian Journal of Botany 76, 1052–
1071.
Badger MR, Hanson DT, Price GD. 2002. Evolution and diversity
of CO2 concentrating mechanisms in cyanobacteria. Functional
Plant Biology 29, 161–173.
Badger MR, Price GD. 1992. The CO2 concentrating mechanism
in cyanobacteria and microalgae. Physiologia Plantarum 84,
606–615.
Badger MR, Price GD. 2003. CO2 concentrating mechanisms in
cyanobacteria: molecular components, their diversity and evolution. Journal of Experimental Botany 54, 609–622.
Baker SH, Jin S, Aldrich HC, Howard GT, Shively JM. 1998.
Insertion mutation of the Form I cbbL gene encododing ribulose
bisphosphate carboxylase/oxygenase (Rubisco) in Thiobacillus
neapolitantus results in expression of Form II rubisco, loss of
carboxysomes, and increased CO2 requirement for growth.
Journal of Bacteriology 180, 4133–4139.
Berner RA. 2001. Modeling atmospheric O2 over Phanerozoic
time. Geochimica et Cosmochimica Acta 65, 685–694.
Berner RA, Beerling DJ, Dudley R, Robinson JM,
Wildman RA. 2003. Phanerozoic atmospheric oxygen. Annual
Review of Earth and Planetary Sciences 31, 105–134.
Berner RA, Kothavala Z. 2001. GEOCARB III: a revised model
of atmospheric CO2 over phanerozoic time. American Journal of
Science 301, 182–204.
Bowien B, Kusian B. 2002. Genetics and control of CO2
assimilation in the chemoautotroph Ralstonia eutropha. Archives
of Microbiology 178, 85–93.
Cannon GC, Bradburne CE, Aldrich HC, Baker SH,
Heinhorst S, Shively JM. 2001. Microcompartments in prokaryotes: carboxysomes and related polyhedra. Applied and Environmental Microbiology 67, 5351–5361.
Cannon GC, Baker SH, Soyer F, Johnson DR, Bradburne CE,
Mehlman JL, Davies PS, Jiang QL, Heinhorst S, Shively JM.
2003. Organization of carboxysome genes in the thiobacilli.
Current Microbiology 46, 115–119.
Chain P, Lamerdin J, Larimer F, et al. 2003. Complete genome
sequence of the ammonia-oxidizing bacterium and obligate
chemolithoautotroph Nitrosomonas europaea. Journal of Bacteriology 185, 2759–2773.
1540 Badger and Bek
Cleland WW, Andrews TJ, Gutteridge S, Hartman FC,
Lorimer GH. 1998. Mechanism of Rubisco – the carbamate as
general base. Chemical Reviews 98, 549–561.
Dobrinski KP, Longo DL, Scott KM. 2005. The carbonconcentrating mechanism of the hydrothermal vent chemolithoautotroph Thiomicrospira crunogena. Journal of Bacteriology 187,
5761–5766.
Dubbs JM, Tabita FR. 2004. Regulators of nonsulfur purple
phototrophic bacteria and the interactive control of CO2 assimilation, nitrogen fixation, hydrogen metabolism and energy generation. FEMS Microbiological Reviews 28, 353–376.
Elsaied HE, Kimura H, Naganuma T. 2007. Composition of
archaeal, bacterial eukaryal RuBisCO genotypes in three Western
Pacific arc hydrothermal vent systems. Extremophiles 11, 191–202.
Elsaied H, Naganuma T. 2001. Phylogenetic diversity of ribulose1,5-bisphosphate carboxylase/oxygenase large-subunit genes from
deep-sea microorganisms. Applied Environmental Microbiology
67, 1751–1765.
Finn MW, Tabita FR. 2004. Modified pathway to synthesize
ribulose 1,5-bisphosphate in methanogenic archaea. Journal
of Bacteriology 186, 6360–6366.
Giraud E, Hannibal L, Fardoux J, Vermeglio A, Dreyfus B.
2000. Effect of Bradyrhizobium photosynthesis on stem nodulation of Aeschynomene sensitiva. Proceedings of the National
Academy of Sciences, USA 97, 14795–14800.
Hassidim M, Keren N, Ohad I, Reinhold L, Kaplan A. 1997.
Acclimation of Synechococcus strain WH7803 to ambient CO2
concentration and to elevated light intensity. Journal of Phycology 33, 811–817.
Hayashi NR, Arai H, Kodama T, Igarashi Y. 1999. The cbbQ
genes, located downstream of the form I and form II RubisCO
genes, affect the activity of both RubisCOs. Biochemical and
Biophysical Research Communications 265, 177–183.
Hayashi NR, Igarashi Y. 2002. ATP binding and hydrolysis and
autophosphorylation of CbbQ encoded by the gene located
downstream of RubisCO genes. Biochemical and Biophysical
Research Communications 290, 1434–1440.
Jannasch HW, Wirsen CO, Nelson DC, Robertson LA. 1985.
Thiomicrospira crunogenea sp. nov., a colorless sulfur-oxidising
bacterium from a deep-sea hydrothermal vent. International
Journal of Systematic Bacteriology 35, 422–424.
Jordan DB, Chollet R. 1985. Subunit dissociation and reconstitution of ribulose-1,5-bisphosphate carboxylase from chromatiumvinosum. Archives of Biochemistry and Biophysics 236, 487–496.
Kobayashi H, Viale AM, Takabe T, Akazawa T, Wada K,
Shinozaki K, Kobayashi K, Sugiura M. 1991. Sequence and
expression of genes encoding the large and small subunits of
ribulose 1,5-bisphosphate carboxylase/oxygenase from Chromatium vinosum. Gene 97, 55–62.
Kusian B, Bowien B. 1997. Organisation and regulation of CBB
CO2 assimilation genes in autotrophic bacteria. FEMS Microbiology Reviews 21, 135–155.
Li LA, Tabita FR. 1997. Maximum activity of recombinant
ribulose 1,5-bisphosphate carboxylase/oxygenase of Anabaena
sp. strain ca requires the product of the rbcx gene. Journal of
Bacteriology 179, 3793–3796.
Long BM, Price GD, Badger MR. 2005. Proteomic assessment of
an established technique for carboxysome enrichment from
Synechococcus PCC7942. Canadian Journal of Botany 83,
746–757.
Ludwig M, Sültemeyer D, Price GD. 2000. Isolation of
ccmKLMN genes from the marine cyanobacterium, Synechococcus sp PCC7002 (Cyanobacteria), and evidence that CcmM is
essential for carboxysome assembly. Journal of Phycology 36,
1109–1118.
Mora CI, Driese SG, Colarusso LA. 1999. Middle to Late
Paleozoic atmospheric CO2 levels from soil carbonate and
organic matter. Science 271, 1105–1107.
Mueller-Cajar O, Badger MR. 2007. New roads lead to Rubisco
in archaebacteria. Bioessays 29, 722–724.
Nanba K, King GM, Dunfield K. 2004. Analysis of facultative
lithotroph distribution and diversity on volcanic deposits by
use of the large subunit of ribulose 1,5-bisphosphate carboxylase/oxygenase. Applied Environmental Microbiology 70,
2245–2253.
Newman J, Branden CI, Jones TA. 1993. Structure determination
and refinement of ribulose 1,5-bisphosphate carboxylase/oxygenase from Synechococcus PCC6301. Acta Crystallographica
D Biological Crystallography 49, 548–560.
Newman J, Gutteridge S. 1994. Structure of an effector-induced
inactivated state of ribulose 1,5-bisphosphate carboxylase/oxygenase: the binary complex between enzyme and xylulose 1,5bisphosphate. Structure 2, 495–502.
Overmann J. 2001. Diversity and ecology of phototrophic sulfur
bacteria. Microbiology Today 28, 116–118.
Paul JH, Pichard SL, Kang JB, Watson GMF, Tabita FR. 1999.
Evidence for a clade-specific temporal and spatial separation in
ribulose bisphosphate carboxylase gene expression in phytoplankton populations off Cape Hatteras and Bermuda. Limnology and
Oceanography 44, 12–23.
Pichard SL, Campbell L, Carder K, Kang JB, Patch J,
Tabita FR, Paul JH. 1997a. Analysis of ribulose bisphosphate
carboxylase gene expression in natural phytoplankton communities by group-specific gene probing. Marine Ecology – Progress
Series 149, 239–253.
Pichard SL, Campbell L, Paul JH. 1997b. Diversity of the
ribulose bisphosphate carboxylase/oxygenase form I gene (rbcL)
in natural phytoplankton communities. Applied Environmental
Microbiology 63, 3600–3606.
Prescott LM, Harley JP, Klein DA. 2005. Microbiology. Boston,
MA: McGraw Hill.
Sato T, Atomi H, Imanaka T. 2007. Archaeal type III RuBisCOs
function in a pathway for AMP metabolism. Science 315, 1003–
1006.
Scott KM, Sievert SM, Abril FN, et al. 2006. The genome of
deep-sea vent chemolithoautotroph Thiomicrospira crunogena
XCL-2. PLoS Biology 4, e383.
Selesi D, Schmid M, Hartmann A. 2005. Diversity of green-like
and red-like ribulose-1,5-bisphosphate carboxylase/oxygenase
large-subunit genes (cbbL) in differently managed agricultural
soils. Applied Environmental Microbiology 71, 175–184.
Shively JM, English RS, Baker SH, Cannon GC. 2001. Carbon
cycling: the prokaryotic contribution. Current Opinion in
Microbiology 4, 301–306.
Shively JM, Vankeulen G, Meijer WG. 1998. Something from
almost nothing – carbon dioxide fixation in chemoautotrophs.
Annual Review of Microbiology 52, 191–230.
Spreitzer RJ, Salvucci ME. 2002. Rubisco: structure, regulatory
interactions, and possibilities for a better enzyme. Annual Review
of Plant Biology 53, 449–475.
Starkenburg SR, Chain PS, Sayavedra-Soto LA, et al. 2006.
Genome sequence of the chemolithoautotrophic nitrite-oxidizing
bacterium Nitrobacter winogradskyi Nb-255. Applied Environmental Microbiology 72, 2050–2063.
Tabita FR. 1999. Microbial ribulose 1,5-bisphosphate carboxylase/
oxygenase: a different perspective. Photosynthesis Research 60,
1–28.
Tabita FR. 2004. Research on carbon dioxide fixation in
photosynthetic microorganisms (1971–present). Photosynthesis
Research 80, 315–332.
Multiple Rubisco forms in proteobacteria 1541
Tcherkez GG, Farquhar GD, Andrews TJ. 2006. Despite slow
catalysis and confused substrate specificity, all ribulose bisphosphate carboxylases may be nearly perfectly optimized. Proceedings of the National Academy of Sciences, USA 103, 7246–7251.
Tolli J, King GM. 2005. Diversity and structure of bacterial
chemolithotrophic communities in pine forest and agroecosystem
soils. Applied Environmental Microbiology 71, 8411–8418.
Tourova TP, Spiridonova EM, Berg IA, Kuznetsov BB,
Sorokin DY. 2006. Occurrence, phylogeny and evolution of
ribulose-1,5-bisphosphate carboxylase/oxygenase genes in obligately chemolithoautotrophic sulfur-oxidizing bacteria of the
genera Thiomicrospira and Thioalkalimicrobium. Microbiology
152, 2159–2169.
Utaker JB, Andersen K, Aakra A, Moen B, Nes IF. 2002.
Phylogeny and functional expression of ribulose 1,5-bisphosphate
carboxylase/oxygenase from the autotrophic ammonia-oxidizing
bacterium Nitrosospira sp. isolate 40KI. Journal of Bacteriology
184, 468–478.
van der Wielen PW. 2006. Diversity of ribulose-1,5-bisphosphate
carboxylase/oxygenase large-subunit genes in the MgCl2-
dominated deep hypersaline anoxic basin discovery. FEMS
Microbiology Letters 259, 326–331.
Viale AM, Kobayashi H, Akazawa T. 1989. Expressed genes for
plant-type ribulose 1,5-bisphosphate carboxylase/oxygenase in
the photosynthetic bacterium Chromatium vinosum, which possesses two complete sets of the genes. Journal of Bacteriology
171, 2391–2400.
Viale AM, Kobayashi H, Akazawa T. 1990. Distinct properties of
Escherichia coli products of plant-type ribulose-1,5-bisphosphate
carboxylase/oxygenase directed by two sets of genes from the
photosynthetic bacterium Chromatium vinosum. Journal of
Biological Chemistry 265, 18386–18392.
Watson GMF, Yu JP, Tabita FR. 1999. Unusual ribulose 1,5bisphosphate carboxylase/oxygenase of anoxic archaea. Journal
of Bacteriology 181, 1569–1575.
Yoshizawa Y, Toyoda K, Arai H, Ishii M, Igarashi Y. 2004.
CO2-responsive expression and gene organization of three
ribulose-1,5-bisphosphate carboxylase/oxygenase enzymes and
carboxysomes in Hydrogenovibrio marinus strain MH-110.
Journal of Bacteriology 186, 5685–5691.