physiology, genetics, and biochemistry of carbon metabolism in the alphaproteobacterium ...
TRANSCRIPT
Physiology, Genetics and Biochemistry of Carbon Metabolism in the α-Proteobacterium 1 Sinorhizobium meliloti 2 3 4 5 6 7 8 9 10
By: Barney A. Geddes and Ivan J. Oresnik* 11
12 13 14 15 16 17 18 19 20 21 22 23 24 25 *Corresponding author 26
Ivan Oresnik 27 Dept of Microbiology, 28 University of Manitoba 29 Winnipeg, MB 30 R3T 2N2 31
32
Page 1 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
33 Abstract 34
A large proportion of genes within a genome encode proteins that play a role in metabolism. The 35
α-proteobacteria are a ubiquitous group of bacteria that play a major role in a number of 36
environments. Carbon metabolism has been studied in Rhizobium for well over 50 years at a 37
biochemical as well as at a genetic level. Here we review the pre and post genomic metabolic 38
literature of the α-proteobacterium Sinorhizobium meliloti. The review provides an overview of 39
carbon metabolism that is useful to readers interested in this organism as well as to those 40
working in other organisms that do not follow other model system paradigms. 41
Keywords: Rhizobium, alpha-proteobacteria, carbon metabolism, model system 42
Introduction 43
The ability to utilize a broad range of carbon sources is important for a versatile organism 44
with several growth habits. Rhizobia engage in a dimorphic lifecycle wherein they must thrive in 45
the complex environment of the soil, as well as intracellularly during nitrogen fixing symbiosis. 46
It is therefore not surprising that the genome of S. meliloti contains a large proportion of genes 47
dedicated to carbon catabolism and substrate transport (Galibert et al. 2001). Correspondingly, S. 48
meliloti is capable of the catabolism of, and growth on many carbon sources including a wide 49
range of monosaccharides, polysaccharides, organic acids and some aromatic compounds 50
(Jordan 1984; Stowers 1985). 51
This review of the literature primarily focuses on what is currently known about S. 52
meliloti metabolism in relation to its biology, with particular attention paid to central carbon 53
metabolism and other intermediary pathways that are present. Since the metabolism of S. meliloti 54
is different from the established microbial models of metabolism that have been described (ie, E. 55
coli; B. subitlis) our objectives are two-fold; first to establish S. meliloti as a model organism for 56
Page 2 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
studying carbon catabolism in the α-proteobacteria and other related bacteria, and second, we 57
review the understanding of the influence of carbon metabolism on the biology of S. meliloti 58
during life in the rhizosphere and during the invasion of its host plants. In an effort to focus the 59
topic, we will not consider the area of bacteroid metabolism in great detail since this area has 60
been the focus of many excellent reviews (Lodwig and Poole 2003; Udvardi and Poole 2013). 61
We focus here on central metabolism as well as peripheral sugar and sugar alcohol catabolism. A 62
recent review extensively summarizes amino acid catabolism in rhizobia (Dunn 2014). 63
Metabolically, rhizobia have been sub-divided into fast-growing and slow-growing 64
groups based on their growth rates. Particular metabolic schemes that correlate with either group 65
have been identified. The slow-growing rhizobia include rhizobia isolated from legumes of 66
tropical origin, such as Bradyrhizobium spp. The fast-growing rhizobia include rhizobia isolated 67
from legumes of temperate origin such as Sinorhizobium spp. and Rhizobium spp. (Stowers 68
1985). In general the pathways described in this review as defined in S. meliloti may be 69
representative of other members of the fast-growing clade. Similarly, findings in other members 70
of the fast-growing rhizobia will be used to augment our understanding of S. meliloti catabolism. 71
72
Sinorhizobium meliloti Central Carbon Metabolism 73
Glycolysis in S. meliloti proceeds through a cyclic Entner-Doudoroff (ED) pathway that 74
utilizes the upper Embden-Meyerhoff-Parnas (EMP) pathway in a gluconeogenic manner. S. 75
meliloti also contains a complete pentose phosphate (PP) pathway and tricarboxylic acid (TCA) 76
cycle. S. meliloti Rm1021 is one of the few rhizobia to contain an operational Calvin-Benson-77
Bassham (CBB) cycle that allows formate dependent autotrophic growth. The genetic and 78
Page 3 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
biochemical evidence for the functions of these pathways is described below. A summary of the 79
predicted and verified genes that encode the enzymes in these pathways is presented in Table 1. 80
Entner-Doudoroff pathway 81
It is well established that Rhizobia possess an ED pathway for hexose catabolism 82
(Stowers 1985). The ED pathway is a widely distributed, alternative pathway of glycolysis to 83
the EMP pathway used in E. coli. ED catabolism proceeds via oxidation of glucose-6-phosphate 84
(G6P) to 6-phosphogluconolactone using G6P dehydrogenase. 6-phosphogluconolactone is 85
resolved to 6-phosphogluconate (6PG) by 6-phosphogluconolactonase. 6PG is dehydrated using 86
6PG dehydratase to 2-keto-3-deoxy-6-phosphogluconate (KDPG), which is split by KDPG 87
aldolase into glyceraldehyde-3-phosphate (G3P) and pyruvate that enter the lower EMP pathway 88
(Figure 1) (Conway 1992). 6PG dehydratase and KDPG aldolase are the two key enzymes of the 89
ED pathway that direct carbon from hexoses into assimilation via pyruvate, rather than through 90
the pentose phosphate pathway. 91
Early studies demonstrated the presence of enzymes for the ED pathway in a broad range 92
of rhizobial species. ED enzymes were found at particularly highly levels in fast-growing 93
rhizobia (Martinez de Drets et al. 1974). Enzymes of the ED pathway are highly induced in S. 94
meliloti during growth with glucose compared to succinate (Finan et al. 1988; Irigoyen et al. 95
1990). Consistent with its role as the main pathway of hexose assimilation, the ED genes have 96
been shown to be among the most highly expressed genes during growth with glucose in S. 97
meliloti (Barnett et al. 2004). 98
Mutants have been isolated in S. meliloti that lacked both NAD+ and NADP
+ linked G6P 99
dehydrogenase activities (Cerveñansky and Arias 1984). These mutants were unable to grow 100
using glucose, fructose, ribose, xylose, mannitol, or sorbitol as sole carbon sources but were able 101
Page 4 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
to grow using gluconeogenic carbon sources such as succinate (Cerveñansky and Arias 1984). 102
Notably, they were also able to grow using galactose or L-arabinose as a sole carbon source, 103
supporting alternate modes of assimilation for these sugars. The inability to grow using most 104
hexoses and pentoses supports the role of the ED pathway as the primary mode of glycolysis. 105
Some mutants that lacked G6P dehydrogenase activity also lacked 6PG dehydratase /KDPG 106
aldolase activity, suggesting that the dedicated ED enzymes may be encoded in the same genetic 107
locus as G6P dehydrogenase (Cerveñansky and Arias 1984). Such a locus was identified 108
adjacent to the genes for α-glucoside utilization (Willis and Walker 1999). The locus contains 109
edd, predicted to encode 6PG dehydratase, pgl, predicted to encode 6-phosphogluconolactonase 110
and zwf, predicted to encode G6P dehydrogenase (Table 1). A strain carrying a zwf mutation was 111
later shown to lack G6P dehydrogenase activity, reinforcing this region as an ED locus in S. 112
meliloti (Barra et al. 2003). A gene encoding KDPG aldolase has not been experimentally 113
identified, but two putative KDPG aldolase genes are encoded on the chromosome (eda1, eda2) 114
(Table 1) (Galibert et al. 2001). 115
Several lines of evidence point to the functioning of a cyclic ED pathway in S. meliloti, 116
where G3P is cycled through the upper EMP pathway into hexoses that can be used for 117
biosynthesis or recycled through the ED pathway. The cyclic ED pathway is in contrast to the 118
linear ED pathway of E. coli (mainly for the catabolism of gluconate), where the end product 119
G3P is catabolized to pyruvate through the lower EMP pathway (Conway 1992). Experiments 120
by two different groups using chromatography-mass spectrometry and nuclear magnetic 121
resonance with labeled carbon compounds clearly demonstrated that some of the G3P produced 122
by ED is converted back into high-molecular weight compounds through the upper EMP 123
pathway (Fuhrer et al. 2005; Portais et al. 1999). These studies also confirmed that glucose was 124
Page 5 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
degraded primarily through the ED pathway, and that glycolysis through the EMP pathway was 125
absent (Fuhrer et al. 2005; Portais et al. 1999). A survey of carbon flux in a number of 126
microorganisms identified a cyclic ED pathway in S. meliloti, as well as pseudomonads. Further, 127
the ED pathway was the predominant route of glucose catabolism in most microorganisms that 128
possessed both ED and EMP pathways (Fuhrer et al. 2005). Early studies also identified a cyclic 129
ED pathway in slow-growing rhizobia (Stowers 1985). 130
Embden-Meyerhoff-Parnas pathway 131
Evidence is consistent with the EMP pathway being used in a gluconeogenic role rather 132
than a glycolytic role in S. meliloti (Figure 2). The dedicated glycolytic enzyme of the EMP 133
pathway, phosphofructokinase, has not been detected in cell-free extracts of S. meliloti (Arias et 134
al. 1979; Irigoyen et al. 1990). Consistent with the absence of detectable phosphofructokinase 135
activity, the genome is not predicted to contain an ATP-dependent pfk gene (Capela et al. 2001). 136
The remaining enzymes of EMP pathway are predicted to be encoded by the genome (Table 1) 137
(Galibert et al. 2001). The activities of many of these enzymes have been detected, including 138
phosphoglucose isomerase, fructose-bisphosphate aldolase, triose-phosphate isomerase, G3P 139
dehydrogenase, 3-phosphoglycerate kinase, phosphoglycerate mutase and enolase (Arias et al. 140
1979; Finan et al. 1988; Irigoyen et al. 1990). In contrast to the enzymes of the ED pathway, 141
which showed substantial induction during growth with glucose compared to succinate, the EMP 142
pathway enzymes showed similar levels of expression during growth with either carbon source 143
(Arias et al. 1979; Finan et al. 1988; Irigoyen et al. 1990). 144
Mutants of genes in the upper and lower EMP pathway have been isolated. Mutants that 145
were deficient in phosphoglucose isomerase were unable to grow using carbon sources that enter 146
central metabolism either directly through fructose-6-phosphate (F6P), including fructose, 147
Page 6 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
mannose, sorbitol and mannitol, or indirectly through the PP pathway, including ribose and 148
xylose (Arias et al. 1979). Growth phenotypes on these carbon sources are consistent with their 149
catabolism through the cyclic ED pathway following conversion of F6P to G6P by 150
phosphoglucose isomerase. 151
Several mutants have been isolated in genes encoding the enzymes of the lower EMP 152
pathway. Carbon utilization phenotypes in these mutants demonstrate the importance of the 153
lower EMP pathway in the gluconeogenesis of carbon sources that enter central metabolism 154
through the TCA cycle. Mutants of enolase, G3P dehydrogenase and 3-phosphoglycerate kinase 155
were isolated based on a genetic screen for the inability to utilize succinate as a sole carbon 156
source. These mutants were unable to grow using TCA cycle intermediates or pyruvate as sole 157
carbon sources (Finan et al. 1988), but were capable of growth using glucose as a sole carbon 158
source; consistent with glycolysis occurring through the ED pathway. However, growth using 159
glucose as a sole carbon source was slower in these mutants than in the wild-type, suggesting 160
that although the EMP pathway is not required for glycolysis, it may be advantageous to 161
catabolize some of the G3P synthesized by the ED pathway through the lower EMP pathway. 162
Based on the absence of 3-phosphoglycerate kinase activity in some G3P dehydrogenase mutants 163
it was suggested that gap and pgk may be located in the same genetic locus (Finan et al. 1988). 164
The genome sequence of S. meliloti Rm1021 supports this hypothesis (Table 1) (Galibert et al. 165
2001). 166
Unlike E. coli and B. subtilis, S. meliloti and many fast-growing rhizobia contain two 167
triose-phosphate isomerase genes. Mutants in both of the triose-phosphate isomerase genes (tpiA 168
and tpiB) have been isolated in S. meliloti (Poysti and Oresnik 2007). Both genes were shown to 169
encode functional triose-phosphate isomerases that were required for the catabolism of carbon 170
Page 7 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
sources that enter central metabolism through dihydroxyacetone-3-phosphate (DHAP). TpiA 171
was required for growth using glycerol and rhamnose as sole carbon sources, whereas TpiB was 172
specifically required for erythritol catabolism. A double mutant was unable to grow using 173
gluconeogenic carbon sources such as succinate as sole carbon sources, demonstrating that they 174
are required for gluconeogenesis (Poysti and Oresnik 2007). Labeled carbon experiments also 175
demonstrated the cycling of G3P synthesized during the catabolism of glucose by the cyclic ED 176
pathway through triose-phosphate isomerase and fructose-bisphosphate aldolase (Fuhrer et al. 177
2005; Portais et al. 1999). 178
Taken together, the EMP pathway performs several roles in central metabolism. It is 179
required for gluconeogenesis of carbon sources that enter central metabolism through the TCA 180
cycle. The EMP pathway also participates in the cycling of G3P synthesized by the ED pathway 181
during growth on hexoses as well as the conversion of pentoses and hexoses that enter central 182
metabolism through F6P into G6P for glycolysis by the ED pathway. 183
Pentose phosphate pathway 184
The pentose phosphate (PP) pathway is comprised of oxidative and non-oxidative 185
branches. The first two steps of the oxidative branch are shared with the ED pathway. G6P 186
dehydrogenase and 6-phosphogluconolactonase synthesize 6PG from G6P. The dedicated 187
enzyme of the PP pathway is 6PG dehydrogenase. 6PG dehydrogenase catalyzes the irreversible 188
decarboxylation of 6PG to ribulose-5-phosphate (R5P) (Figure 3). The presence or absence of 189
6PG dehydrogenase activity has been used to classify rhizobia as fast- or slow-growers. Fast-190
growing rhizobia, including S. meliloti, contain a complete PP pathway as indicated by the 191
presence of 6PG dehydrogenase, while most slow-growing rhizobia do not (Martínez de Drets 192
and Arias 1972). 193
Page 8 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
The non-oxidative branch of the PP pathway is a series of reversible reactions that 194
interchange R5P with F6P and G3P (Figure 3). The activities of transketolase and transaldolase 195
have been shown to be present in cell-free extracts of S. meliloti (Cerveñansky and Arias 1984). 196
A proteomic study found a probable transketolase and ribose-5-phosphate epimerase to be 197
expressed under free living conditions as well as in the bacteroid (Djordjevic 2004). 198
The active cycling of the PP pathway has been demonstrated by labeling experiments 199
(Fuhrer et al. 2005; Gosselin et al. 2001; Portais et al. 1999). Complicated labeling patterns were 200
observed that were consistent with the reversible action of the non-oxidative branch of the PP 201
pathway using 13
C-NMR with glucose or fructose (Gosselin et al. 2001; Portais et al. 1999). 202
Flux through the PP pathway to supply PP intermediates for biosynthesis was also observed 203
using gas chromatography with radiolabelled carbon (Fuhrer et al. 2005). 204
Overall, the study of the PP pathway is limited in S. meliloti. The genes encoding the PP 205
pathway have not been characterized, and the regulation of the PP pathway is unclear. Possible 206
candidates for the genes encoding all components of the PP pathway can be identified in the 207
genome sequence and are presented in Table 1. 208
Calvin-Benson-Bassham cycle 209
The ability to fix carbon dioxide autotrophically is generally associated with the presence 210
of ribulose-1,5-bisphosphate carboxylase activity. Across the Rhizobiaceae only Bradyrhizobium 211
and Sinorhizobium have been shown to have this activity (Hanus et al. 1979; Manian and O'Gara 212
1982a; Pickering and Oresnik 2008). Whereas B. japonicum is capable of chemolithoautotrophic 213
growth (Hanus et al. 1979), autotrophic growth by S. meliloti has only been demonstrated in the 214
presence of another form of reduced carbon (Manian and O'Gara 1982a, b; Pickering and 215
Oresnik 2008). The genes cbbFPTALSX and ppe predicted to encode determinants of the CBB 216
Page 9 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
pathway are found on pSymB along with a presumed LysR type positive regulator, cbbR, that is 217
divergently transcribed. Together these genes could encode all the enzymes necessary for a 218
traditional CBB pathway except for those that are shared with lower portion of the EMP pathway 219
or PP pathway. In support of this hypothesis, growth in a medium containing formate and 220
bicarbonate was shown to be dependent upon the cbb operon (Pickering and Oresnik 2008). 221
Formate mediated autotrophic growth was also dependent on one of the two triose-phosphate 222
isomerase genes (tpiA or tpiB) found in the S. meliloti genome, suggesting that fixed carbon is 223
removed from the cycle by the conversion of GAP to DHAP and condensing this to form 224
fructose-1,6-bisphosphate (Pickering and Oresnik 2008). 225
The annotation of the Rm1021 genome suggests that three formate dehydrogenases are 226
present (fdoGHI, fdsABCDG and SMa0487) (Barnett et al. 2001). It was demonstrated that the 227
fdsABCDG locus was absolutely necessary for formate-dependent autotrophic growth, whereas a 228
mutation that affected the pSymA encoded fdoGHI locus reduced the amount of carbon fixed to 229
about 40% of wild-type levels (Pickering and Oresnik 2008). The third annotated formate 230
dehydrogenase encoding gene, SMa0487, did not appear to have formate dehydrogenase activity, 231
and did not affect autotrophic growth (Pickering and Oresnik 2008). 232
Tricarboxylic acid cycle 233
S. meliloti is known to contain a complete TCA cycle (Figure 4) (Dunn 1998; Stowers 234
1985). All the enzymes of the TCA cycle have been shown to be expressed under free-living 235
growth conditions, and in symbiotic association (Djordjevic 2004). Mutants that were deficient 236
in α-ketoglutarate dehydrogenase (Duncan and Fraenkel 1979) and succinate dehydrogenase 237
(Gardiol et al. 1982) activities have been previously isolated. More recently the genes encoding 238
many of the TCA cycle components in S. meliloti have been identified (Table 1). Citrate 239
Page 10 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
synthase, aconitase and isocitrate dehydrogenase are encoded by gltA, acnA and icd respectively 240
(Koziol et al. 2009; McDermott and Kahn 1992; Mortimer et al. 1999). Knockouts of each of 241
these genes have been constructed and shown to lack the associated enzyme activity, suggesting 242
that they uniquely encode each of these enzymes in S. meliloti (Koziol et al. 2009; McDermott 243
and Kahn 1992; Mortimer et al. 1999). These three genes appear scattered throughout the 244
chromosome as ORFan genes in S. meliloti (Galibert et al. 2001). 245
Random mutagenesis identified a locus that contains many of the remaining TCA cycle 246
genes (Dymov et al. 2004). In this study a Tn5-tac1 mutant was isolated in mdh that was shown 247
to encode malate dehydrogenase. The genes encoding succinyl-CoA synthetase (sucCD) and the 248
E1/E2 components of α-ketoglutarate dehydrogenase (sucAB) complex were shown to be encoded 249
immediately downstream of mdh (Dymov et al. 2004). Although not cotranscribed, the genes 250
predicted to encode succinic dehydrogenase complex (sdhABCD) and the E3 component of α-251
ketoglutarate dehydrogenase, dihydrolipoyl dehydrogenase (lpdA2), are also in close proximity. 252
The genes encoding the pyruvate dehydrogenase complex (pdhABC) are cotranscribed elsewhere 253
on the chromosome (Cabanes et al. 2000). Although dihydrolipoyl dehydrogenase can be shared 254
between the pyruvate dehydrogenase and α-ketoglutarate dehydrogenase complexes (Mattevi et 255
al. 1992), a second dihydrolipoyl dehydrogenase (lpdA1) is encoded near pdhABC and a 256
mutation that was polar on its transcription abolished pyruvate dehydrogenase activity (Soto et 257
al. 2001). 258
Gluconeogenesis 259
Growth on TCA cycle intermediates, or carbon sources that enter central metabolism via 260
the TCA cycle, is carried out by gluconeogenesis. Gluconeogenesis requires the EMP pathway 261
pathway (as has been described above) as well as a dedicated gluconeogenic enzyme to 262
Page 11 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
synthesize intermediates in the lower EMP pathway from TCA cycle intermediates. One such 263
enzyme that has been identified in S. meliloti and other rhizobia is phosphoenolpyruvate (PEP) 264
carboxykinase (Dunn 1998). PEP carboxykinase catalyzes the conversion of oxaloacetate to PEP 265
(Figure 4). PEP carboxykinase activity has been demonstrated in S. meliloti, and mutants that 266
were deficient in PEP carboxykinase activity grew very slowly using gluconeogenic carbon 267
sources (Finan et al. 1988). PEP carboxykinase was shown to be encoded by pckA in S. meliloti 268
(Table 1) (Østeräs et al. 1995). 269
An alternative gluconeogenic route is through the cooperative action of pyruvate 270
orthophosphate dikinase and malic enzyme (Figure 4). Pyruvate orthophosphate dikinase 271
catalyzes the ATP-dependent conversion of pyruvate to PEP. Increased pyruvate orthophosphate 272
dikinase activity was associated with suppressors to pckA mutants for growth on succinate 273
(Østeräs et al. 1997). Pyruvate orthophosphate dikinase is encoded by pod (Table 1). Although a 274
pod mutant alone showed no significant phenotype, a double mutant of pckA and pod was 275
completely unable to utilize gluconeogenic carbon sources (Østeräs et al. 1997). Malic enzyme 276
catalyzes the decarboxylation of malate to form pyruvate. S. meliloti contains both an NAD+ 277
dependent and an NADP+ dependent malic enzyme (Driscoll and Finan 1993, 1996, 1997). 278
These are encoded by dme and tme, respectively (Table 1) (Driscoll and Finan 1993, 1996). The 279
suppression of a pckA mutant by pyruvate orthophosphate dikinase was dependent on the 280
presence of dme (Østeräs et al. 1997). However, dme and tme mutants alone showed no 281
significant carbon utilization phenotype suggesting that PEP carboxykinase is the predominant 282
gluconeogenic route in free-living S. meliloti. Interestingly, evidence is consistent with the use 283
of malic enzyme rather than PEP carboxykinase for gluconeogenesis in the bacteroid (Driscoll 284
and Finan 1993, 1997; Finan et al. 1988). 285
Page 12 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Anaplerotic pathways 286
Glycolytic growth requires an anaplerotic enzyme to resupply the TCA cycle with 287
intermediates that are removed during amino acid biosynthesis or gluconeogenesis. Pyruvate 288
carboxylase fulfills this role in S. meliloti (Dunn et al. 2001). Pyruvate carboxylase catalyzes the 289
carboxylation of pyruvate to form oxaloacetate (Figure 4). Pyruvate carboxylase is encoded by 290
pyc in S. meliloti. Consistent with its role as an anaplerotic enzyme, pyc mutants were unable to 291
grow using glucose or pyruvate as sole carbon sources, but were capable of growth using 292
succinate as a sole carbon source (Dunn et al. 2001). 293
Growth with acetate as a carbon source also requires an anaplerotic pathway in order to 294
bypass the decarboxylating steps of the TCA cycle to allow a net accumulation of carbon. In S. 295
meliloti this is accomplished by the glyoxylate shunt (Figure 4). The glyoxylate shunt involves 296
the cleavage of isocitrate to form succinate and glyoxylate by isocitrate lyase, followed by the 297
synthesis of malate from glycoxylate and acetyl-CoA by malate synthase (Dunn 1998). Both of 298
these enzymes have been detected and are expressed during growth with acetate (Duncan and 299
Fraenkel 1979). These enzymes are encoded by aceA and glcB, respectively, and mutations in 300
these genes abolished isocitrate lyase and malate synthase activities in S. meliloti (Ramírez-301
Trujillo et al. 2007). 302
Peripheral Carbon Metabolism in S. meliloti 303
The ability to utilize a broad range of carbon sources is important for survival in a diverse 304
array of growth conditions. Rhizobia must be able to survive in the complex environment of the 305
soil, intracellularly during infection of the plant and throughout the symbiosis. It is therefore 306
unsurprising that a large portion of its genome is dedicated to carbon catabolism and transport 307
(Galibert et al. 2001; Mauchline et al. 2006). This section discusses peripheral pathways of 308
Page 13 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
metabolism of some of these sugars and sugar alcohols, as defined in S. meliloti. The predicted 309
entry-points of these sugars and sugar alcohols into the pathways of central metabolism 310
described in the previous section are presented Figure 5. 311
Disaccharide catabolism and transport 312
An early feature that differentiated fast-growing from slow-growing rhizobia was the 313
capacity to utilize disaccharides. Studies showed that fast-growing rhizobia including 314
Sinorhizobium and Rhizobium spp. were capable of utilizing lactose, sucrose, maltose, trehalose 315
and cellobiose as sole carbon sources, whereas slow-growers such as Bradyrhizobia were not 316
(Glenn and Dilworth 1981; Jordan 1984; Martínez de Drets and Arias 1972; Martinez de Drets et 317
al. 1974; Stowers 1985). More recently targeted and genomic approaches have identified many 318
of the loci involved in β- and α -galactoside and glucoside catabolism in S. meliloti. 319
β-galactosides (lactose and lactulose) 320
Two different β-galactosidase activities have been demonstrated in S. meliloti, however 321
only one of the activities was inducible by lactose. The inducible β-galactosidase was shown to 322
be required for growth using lactose as a sole carbon source (Niel et al. 1977). Lactose uptake 323
has been shown to be inducible in S. meliloti and other fast-growing rhizobia (Ucker and Signer 324
1978). Uptake of lactose was not inhibited by sucrose, maltose, trehalose or cellobiose 325
suggesting that it uses a unique transporter (Glenn and Dilworth 1981). The genes for lactose 326
catabolism are carried on the megaplasmid pSymB and are adjacent to the locus for α-327
galactoside utilization (Charles and Finan 1991; Charles et al. 1990). The lactose locus contains 328
the β-galactosidase lacZ, genes that presumably encode a lactose ABC transporter, lacEFGK, 329
and a LacI type negative regulator, lacR (Jelesko and Leigh 1994). Lactose regulation proceeds 330
Page 14 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
differently than E. coli in S. meliloti as β-galactosidase activity is not inducible by IPTG (Ucker 331
and Signer 1978). 332
A locus that is necessary for growth using lactulose, a β-galactoside composed of 333
galactose and fructose, as a sole carbon source has recently been identified on pSymB. This 334
locus contains a putative ABC transporter encoded by SMb20929-20931 (Finan personal 335
communication). Induction of the putative sugar binding protein SMb20931 by several α- and β-336
galactosides including lactulose has previously been demonstrated (Mauchline et al. 2006). To 337
our knowledge it is currently unclear how lactulose catabolism proceeds in S. meliloti. 338
339
α-galactosides (melibiose and raffinose) 340
The locus required for the utilization of α-galactosides including melibiose and raffinose 341
is located on pSymB adjacent to the β-galactoside locus (Charles and Finan 1991; Charles et al. 342
1990; Gage and Long 1998). The locus contains an ABC transporter (agpABCD) and two 343
putative α-galactosidases, melA and agaL2. The ABC transporter was required for growth on 344
melibiose and raffinose, but not galactose or glucose. Further, a mutant of the gene encoding the 345
periplasmic binding protein agpA was shown to be unable to transport raffinose. Transcription 346
of agpA was down regulated by SyrA, and upregulated by galactose and α -galactosides (Gage 347
and Long 1998). Induction of genes in the locus by galactose and α -galactosides is mediated by 348
an AraC-type transcriptional regulator encoded by the upstream gene agpT. Mutants of agpT 349
were unable to grow using melibiose or raffinose as sole carbon sources (Bringhurst and Gage 350
2000). 351
352
α- and β-glucosides (sucrose, maltose, trehalose and cellobiose) 353
Page 15 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Early studies of disaccharide uptake and hydrolysis provided evidence that the uptake of 354
sucrose and maltose was facilitated by an active process, and that sucrose invertase activity was 355
inducible in S. meliloti (Glenn and Dilworth 1981). The α-glucosides sucrose, maltose and 356
trehalose were able to compete for active uptake with sucrose and maltose, whereas lactose or 357
the β-glucoside cellobiose were not (Glenn and Dilworth 1981). This early work suggested that 358
sucrose, maltose and trehalose share a common uptake system, whereas cellobiose may be 359
transported by a separate system. A locus that encoded α-glucoside transport and catabolism was 360
isolated on an S. meliloti cosmid that conferred the ability to utilize sucrose, maltose and 361
trehalose to Ralstonia eutrophia (Willis and Walker 1998). The locus consisted of a set of genes 362
that map to the chromosome of S. meliloti annotated as aglEFGAK, where aglEFGK and aglA 363
are predicted to encode an ABC transport system and α-glucosidase respectively. However, S. 364
meliloti strains carrying mutations in the agl region were not impaired in α-glucoside utilization, 365
even in backgrounds that also carried mutations in the agp α-galactoside region (Willis and 366
Walker 1999). 367
Evidence for redundancy in α-glucoside transport systems has more recently been 368
presented; a locus was isolated by Tn5 mutagenesis based on an impaired ability to utilize 369
trehalose as a sole carbon source. The locus containing the Tn5 mapped to pSymB and contained 370
four putative ABC transport genes thuEFGK (Jensen et al. 2002). Uptake experiments in a thuE 371
mutant background showed reduced uptake of radiolabelled sucrose, maltose and trehalose. 372
Furthermore, a double mutant of thuE, aglE was abolished for sucrose, trehalose and maltose 373
utilization showing that aglEFGK and thuEFGK together encode the two major α-glucoside 374
transport systems in S. meliloti (Jensen et al. 2002). 375
Page 16 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Further work characterizing trehalose catabolism showed that mutants of two putative 376
catabolic genes in the locus, thuAB were strongly impaired in their ability to utilize trehalose as a 377
sole carbon source (Jensen et al. 2005). These genes had little similarity to genes encoding 378
known enzymes of trehalose catabolism and were later shown to encode enzymes for a novel 379
pathway of trehalose catabolism through a 3-ketotrehalose intermediate (Ampomah et al. 2013; 380
Avetisyan et al. 2013). The transport genes thuEFGK and catabolic genes thuAB were further 381
shown to be involved in the transport and catabolism of maltitol and the sucrose isomers 382
leucrose, palatinose and trehalulose (Ampomah et al. 2013). 383
Hexoses 384
The catabolism of most hexoses requires a functional cyclic ED pathway. This is 385
illustrated by mutations in G6P dehydrogenase that are unable to grow using fructose, mannose 386
or glucose as sole carbon sources. Galactose catabolism does not follow this scheme, nor does it 387
enter glycolysis through the Leloir pathway as in E. coli. Rather it is catabolized through an 388
analogous pathway to the ED pathway: the De Ley-Doudoroff pathway (Arias and Cerveñansky 389
1986). 390
Glucose 391
The most direct point of entry for glucose into the ED pathway is phosphorylation by 392
hexose kinase (Figure 5). However, glucose can also be oxidized by a gluconate bypass which 393
consists of a periplasmic pyrroloquinoline quinone (PQQ)-dependent glucose dehydrogenase and 394
gluconate kinase (Gosselin et al. 2001; Portais et al. 1999). Although this pathway exists, 395
glucose dehydrogenase mutants did not have a slower growth rate than wild type when grown 396
with glucose as a sole carbon source (Gosselin et al. 2001). Growth experiments in chemostat 397
cultures containing glucose and succinate indicated that glucose is preferentially oxidized while 398
Page 17 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
gluconate accumulates in the culture (Bernardelli et al. 2001). A genetic locus that encodes an 399
ATP-independent periplasmic transporter that is required for gluconate utilization has been 400
identified on pSymA (gntABC) (Steele et al. 2009). 401
402
Galactose 403
Based on the observation that a G6P dehydrogenase mutant was unable to grow with 404
most hexoses, but was able to grow with galactose as a sole carbon source, it was hypothesized 405
that the De Ley-Doudoroff pathway was used in S. meliloti (Cerveñansky and Arias 1984). The 406
De Ley-Doudoroff pathway of galactose catabolism was first discovered in Pseudomonas 407
saccharophila (De Ley and Doudoroff 1957; Lessie and Phibbs 1984). Briefly, galactose is 408
oxidized to galactono-γ-lactone, and resolved to galactonate by a lactonase. Galactonate is 409
dehydrated to 2-keto-3-deoxygalactonate. 2-keto-3-deoxygalactonate is phosphorylated forming 410
2-keto-3-deoxy-6-phosphogalactonate (KDPGal), which is split by an aldolase reaction into G3P 411
and pyruvate (Figure 5). Cell free extracts of S. meliloti L5-30 were assayed for the presence of 412
four out of five of the enzymes required for the De Ley-Doudoroff pathway (Arias and 413
Cerveñansky 1986). Activities of each of these enzymes were demonstrated, consistent with De 414
Ley-Doudoroff pathway operation. A chemically induced mutant unable to grow using galactose 415
as a sole carbon source was found to be missing KDPGal aldolase activity (Arias and 416
Cerveñansky 1986). 417
More recently the genetic determinants involved have been identified in S. meliloti 418
(Geddes and Oresnik 2012b). The catabolic genes are found in an operon of 5 genes whose 419
annotation correspond to the complete pathway (Geddes and Oresnik 2012b). However, mutant 420
analysis demonstrated that only 2 of the genes, dgoK and dgoA, encoding the 2-keto-3-421
Page 18 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
deoxygalactonate kinase and KDPGal aldolase were absolutely necessary for growth using 422
galactose as a sole carbon source (Geddes and Oresnik 2012b). The dehydrogenase (galD) was 423
shown to have enzyme activity with galactose as a substrate, but a second inducible galactose 424
dehydrogenase that was also capable of using L-arabinose as a substrate was also found (Geddes 425
and Oresnik 2012b). Recent data has shown that this galactose/L-arabinose dehydrogenase is 426
encoded by SMc00588 (Geddes et al., in review). 427
Transport competition experiments showed that although a mutant of the arabinose 428
transporter araABC only showed a modest reduction in growth rate using galactose as a sole 429
carbon source, both galactose and glucose competed with arabinose transport (Geddes and 430
Oresnik 2012b). It was hypothesized that AraABC serves as a multiple monosaccharide 431
transporter as homologues do in Brucella (Alvarez-Martinez et al. 2001) and Agrobacterium (He 432
et al. 2009), 433
434
Fructose and Mannose 435
Fructose enters central metabolism through phosphoglucose isomerase (pgi) (Arias et al. 436
1979), following phosphorylation by fructose kinase (Figure 5). Fructose kinase activity has been 437
demonstrated in S. meliloti and mutants deficient in fructose kinase activity have been isolated 438
and shown to be unable to grow using fructose as a sole carbon source (Gardiol et al. 1980). The 439
pathway for D-mannose catabolism has been characterized in S. meliloti L5-30. Both mannose 440
kinase and mannose-phosphate isomerase activities were detected and mutants that lacked these 441
activities were isolated (Arias and Cerveñansky 1982). These activities are consistent with the 442
conversion of mannose to mannose-6-phosphate, and subsequently fructose-6-phosphate where it 443
is able to enter central metabolism (Figure 5). 444
Page 19 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
The transport of both mannose and fructose has been shown to be an active process 445
(Arias and Cerveñansky 1982; Gardiol et al. 1980). A genetic locus encoding a high affinity 446
fructose ABC transporter has been identified in S. meliloti following a Tn5 mutagenesis by 447
screening for mutants with the inability to utilize fructose as a sole carbon source (Lambert et al. 448
2001). The ABC transporter is encoded on the chromosome by frcBCA. Both ribose and 449
mannose competed for binding to the fructose sugar binding protein FrcB, however mutants in 450
the transporter were still capable of growth using these sugars as sole carbon sources (Lambert et 451
al. 2001). The locus also contains a putative fructose kinase frcK, however FrcK is not 452
homologous to the demonstrated fructose kinase of R. leguminosarum (Fennington and Hughes 453
1996). The gene encoding pgi may also be encoded nearby (Galibert et al. 2001). However, the 454
functions of pgi, or the putative fructose catabolism genes have not been experimentally verified. 455
456
Pentoses: 457
The inability of strains carrying mutations in G6P dehydrogenase and phosphoglucose 458
isomerase mutants to grow using ribose and xylose is consistent with catabolism through the 459
cyclic ED pathway via the pentose phosphate pathway. L-arabinose catabolism has been 460
characterized because its entry into central metabolism in S. meliloti is through the TCA cycle 461
rather than the PP pathway, or glycolysis as in E. coli (Duncan and Fraenkel 1979). 462
463
Arabinose 464
In the slow-growing rhizobia such as B. japonicum, L-arabinose is catabolized through 465
non-phosphorylated intermediates leading to the production of pyruvate and glycolaldehyde 466
(Pedrosa and Zancan 1974). In this pathway, L-arabinose catabolism begins in analogy to 467
Page 20 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
galactose with oxidation to arabinolactone, followed by hydration forming L-arabonate and 468
dehydration to 2-keto-3-deoxy-L-arabonate which is split by an aldolase reaction to pyruvate and 469
glycolaldehyde (Pedrosa and Zancan 1974). Glycolaldehyde is subsequently catabolized through 470
an oxalate degradation pathway (Koch et al. 2014). In contrast, S. meliloti L5-30 has been shown 471
to grow gluconeogenically on L-arabinose, where it enters central metabolism through α-472
ketoglutarate (Duncan and Fraenkel 1979). Here, the first steps of arabinose degradation are 473
conserved leading to the production of 2-keto-3-deoxy-L-arabonate. At this point, α-474
ketoglutarate semialdehyde dehydrogenase oxidizes 2-keto-3-deoxy-L-arabonate to α-475
ketoglutarate through which it enters central metabolism (Figure 5) (Duncan and Fraenkel 1979). 476
Surveys of several other fast-growing rhizobia showed that they all contained α-ketoglutarate 477
semialdehyde dehydrogenase rather than 2-keto-3-deoxy-L-arabonate aldolase, suggesting that 478
gluconeogenic growth on L-arabinose is a common theme among fast-growing rhizobia (Duncan 479
and Fraenkel 1979). 480
A genetic locus has been identified on pSymB that is required for L-arabinose catabolism 481
in S. meliloti Rm1021. The locus contains genes for both arabinose catabolism and transport 482
(Poysti et al. 2007). Genes predicted to encode a hydratase, dehydratase and α-ketoglutarate 483
semialdehyde dehydrogenase are all contained within the locus but a gene encoding L-arabinose 484
dehydrogenase was notably absent (Poysti et al. 2007). The ABC transporter encoded by the 485
locus AraABC was shown to be required for intracellular accumulation of radiolabelled L-486
arabinose (Poysti et al. 2007). 487
488
Ribose and Xylose 489
Page 21 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Ribose kinase activity has been demonstrated in cell-free extracts of S. meliloti L5-30, 490
and a mutant was isolated that lacked ribose kinase activity (Duncan 1981). However, the 491
location of the gene encoding ribose kinase in S. meliloti has not been described. In many 492
organisms the catabolism of xylose proceeds by isomerization to xylulose followed by a 493
phosphorylation to yield xylulose-5-phosphate. Xylose isomerase activity has been 494
demonstrated in S. meliloti and a mutant was isolated that lacked this enzyme activity and was 495
unable to grow using xylose as a sole carbon source (Duncan 1981). A strain carrying a mutation 496
in a gene annotated as encoding xylulose kinase, xylB, has also been isolated in S. meliloti 497
Rm1021 and shown to be unable to utilize xylose as a sole carbon source (Geddes and Oresnik 498
2012a). A putative xylulose kinase (xylA) is also encoded immediately upstream of xylB on the 499
chromosome (Galibert et al. 2001). 500
These data are consistent with the hypothesis that the primary route of ribose and xylose 501
catabolism is through the PP pathway, however there is evidence that there may be a secondary 502
“metabolic bypass” for ribose and xylose catabolism (Duncan 1981). This was suggested based 503
on the ability of ribose kinase and xylose isomerase mutants to continuously incorporate 14
C 504
labeled ribose and xylose at rates consistent with the wild-type over a 22 hour period (Duncan 505
1981). More recently, we have shown that a mutant wherein metabolism is blocked in both 506
glycolysis (pyc) and gluconeogenesis (tpiA, tpiB) was capable of growth using either ribose or 507
xylose as sole carbon sources (Geddes and Oresnik 2012a). Other carbon sources that enter 508
central metabolism through common intermediates, such as xylulose in the case of D-arabitol, 509
were not capable of growing around these lesions (Geddes and Oresnik 2012a). Taken together 510
these physiological and genetic data support the idea of a metabolic bypass for the catabolism of 511
ribose and xylose. 512
Page 22 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
513
Methylpentoses: 514
The biochemical pathways for the methylpentoses fucose and rhamnose have been 515
determined in E. coli and Salmonella (Lin 1996). Although these sugars are very similar in 516
structure, their catabolism is carried out by two parallel pathways in these organisms. Essentially 517
the sugars are isomerized to their keto derivatives, phosphorylated, and broken into two 3 carbon 518
compounds by sugar-specific aldolases yielding DHAP and lactaldehyde, the lactaldehyde is 519
converted to pyruvate. 520
Rhamnose 521
Rhamnose transport and catabolism has been extensively characterized in R. 522
leguminosarum bv. trifolii (Oresnik et al. 1998). The R. leguminosarum locus consists of 9 523
genes that are arranged in two divergent operons. The genes at this locus consisted of a negative 524
regulator (rhaR), the components of an ABC transporter (rhaSTPQ), a mutarotase (rhaU), as 525
well as an isomerase (rhaI), kinase (rhaK) and dehydrogenase/aldolase (rhaD) (Richardson et al. 526
2004). It was shown that in R. leguminosarum, transport of rhamnose is dependent on the ABC 527
transporter (Richardson et al. 2004). Proper functioning of the transporter is dependent on the 528
sugar kinase RhaK (Richardson and Oresnik 2007). Utilizing a linker scanning mutagenesis the 529
ability of RhaK to affect transport was shown to be distinct from its ability to act as a kinase 530
(Rivers and Oresnik 2013). Genetic and biochemical evidence support the hypothesis that RhaK 531
could directly phosphorylate rhamnose (Richardson et al. 2004; Richardson and Oresnik 2007). 532
The complement of enzymes in the locus is consistent with the capacity to catabolize rhamnose 533
in a similar manner to E. coli, however the direct phosphorylation of rhamnose by RhaK is not. 534
Page 23 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Further work should be carried out to resolve these differences and determine whether the 535
pathway of rhamnose catabolism in rhizobia may be unique. 536
The evidence for rhamnose catabolism is more disparate in S. meliloti. The genes for 537
rhamnose catabolism were first suggested based on similarity when the R. leguminosarum locus 538
was sequenced (Richardson et al. 2004). The locus was subsequently shown to be necessary for 539
rhamnose utilization and induced in response to rhamnose as a proof of principle on a genome-540
wide attempt to provide functionality to ABC transporters (Mauchline et al. 2006). A triose-541
phosphate isomerase mutant (tpiA) was impaired in its ability to grow using rhamnose as a sole 542
carbon source, suggesting that it is catabolized to DHAP and lactaldehyde as in E. coli (Figure 5) 543
(Poysti and Oresnik 2007). 544
Fucose 545
The locus used for fucose transport and utilization is found on pSymB (SMb21103-546
SMb21113). The evidence for this is as follows: SMb21103-SMb21106 encode a putative ABC 547
transporter that has been shown to be induced by D and L-fucose (Mauchline et al. 2006); a 548
mutant of the putative short-chain dehydrogenase SMb21111 has been shown to be unable to use 549
either D- or L-fucose as sole carbon sources (Jacob et al. 2008); deletions spanning these regions 550
were shown to result in an inability to utilize D-fucose as a sole carbon source (Milunovic et al. 551
2014) and SMb21108 encodes a protein that is highly similar to RhaU which has been 552
crystallized and shown to be a rhamnose mutarotase (Richardson et al. 2008). 553
The locus contains a putative racemase (SMb21107 /manR), a mutarotase (SMb21108), 554
two dehydrogenases (SMb21109 and SMb21111), two putative dehydratases (SMb21110 and 555
SMb21113), and a gene that encodes an enzyme capable of cleaving a carbon-carbon bond 556
(SMb21112 /hpaG). Based on these annotations, it is possible that fucose is broken down 557
Page 24 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
through non-phosphorylated intermediates since a carbohydrate kinase is not present. Precedent 558
for fucose breakdown using a non-phosphorylated pathway has previously been shown in 559
pseudomonads where fucose is broken down to pyruvate and lactaldehyde (Dahms and Anderson 560
1972a, b, c, d). The ability of tpiA and tpiB mutants to grow using fucose as a sole carbon source 561
is consistent with such an entry-point, rather than DHAP as in E. coli (Poysti and Oresnik 2007). 562
Sugar alcohol transport and catabolism. 563
Rhizobia have been historically known for their ability to grow efficiently using sugar 564
alcohols. Traditional media used for isolating rhizobia from nature contains high concentrations 565
of mannitol to help enrich for rhizobia (Vincent 1970). Sugar alcohol catabolism in S. meliloti 566
has been studied for many years and the catabolism of several different sugar alcohols have been 567
characterized genetically and biochemically. 568
Inositol 569
The cyclic six carbon polyol inositol has been suggested to be prevalent in soils and an important 570
substrate for S. meliloti based on the detection of inositol dehydrogenase activity in S. meliloti 571
and S. fredii strains grown in solutions extracted from a wide range of soils (Wood and Stanway 572
2001). The metabolism of myo-inositol and its isomers scyllo- and chiro-inositol has been 573
shown to be encoded by three different loci in S. meliloti; idhA, encoded on pSymB, and the 574
chromosomally encoded iolA and SMc01163-iolRCDEB loci (Kohler et al. 2010). The 575
metabolism of inositol in S. meliloti is thought to proceed through a similar pathway to that 576
characterized in Bacillus subtilis (as first identified in Klebsiella aerogenes) wherein the end-577
products of its catabolism are DHAP and acetyl-CoA (Figure 5) (Kohler et al. 2010; Yoshida et 578
al. 2008). Activities of the enzymes encoding the first two steps of the pathway (myo- 579
dehydrogenase and 2-keto myo-inositol dehydratase) have been demonstrated in Rhizobium 580
Page 25 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
leguminosarum bv. viciae (Poole et al. 1994). The inositol dehydrogenase that carries out the 581
oxidation of myo- and chiro-inositol in S. meliloti is encoded by idhA (Galbraith et al. 1998; 582
Kohler et al. 2010). The remaining enzymes catalyzing inositol catabolism breakdown in S. 583
meliloti are thought to be encoded by iolE, iolD, iolB, iolC and iolA (Kohler et al. 2010). 584
SMc01163 (iolY) is required for the catabolism of scyllo-inositol but not myo- or chiro-inositol, 585
and is predicted to encode a scyllo-inositol dehydrogenase that can oxidize scyllo-inositol to 2-586
keto-myo-inositol which then proceeds through a common pathway with myo- and chiro- inositol 587
(Kohler et al. 2010). Regulation of inositol loci is carried out by IolR which negatively regulates 588
the transcription of the idhA and SMc01163-iolREDBCA loci but not iolA (Kohler et al. 2011). 589
Evidence for induction of inositol genes by an intermediate of myo-inositol catabolism has been 590
presented in R. leguminosarum bv. viciae (Fry et al. 2001). A myo-inositol ABC transporter has 591
been identified in R. leguminosarum, bv. viciae and designated IntABC (Fry et al. 2001). In S. 592
meliloti, two ATP-dependent transport systems were shown to be inducible by myo-inositol. One 593
of these systems, encoded by ibpA-iatA-iatP (SMb20712-SMb20714) shares high identity with 594
IntABC of R. leguminosarum (Mauchline et al. 2006). A S. meliloti strain carrying a mutation in 595
these genes was unable to grow using inositol as a sole carbon source (Boutte et al. 2008). 596
597
Sorbitol, Mannitol and D-Arabitol 598
Mannitol is widely considered to be a preferred carbon source for fast-growing rhizobia 599
(Jordan 1984; Vincent 1970). An early study using extracts of mannitol-grown S. meliloti 600
identified two different inducible polyol dehydrogenase activities. One of these acted specifically 601
on sorbitol, while the other acted on mannitol and D-arabitol (Martinez de Drets and Arias 602
1970). A mutation of a putative sorbitol dehydrogenase on the chromosome, SMc01500 (smoS), 603
Page 26 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
resulted in the inability to utilize D-sorbitol, D-mannitol or D-arabitol (Jacob et al. 2008). 604
Additionally, a putative mannitol dehydrogenase, mtlK is encoded immediately downstream of 605
smoC (Galibert et al. 2001). Genes encoding an ABC transporter (smoEFGK) immediately 606
upstream of smoS and mtlK have been shown to be induced by sorbitol and mannitol (Mauchline 607
et al. 2006). Therefore, it seems likely that this locus may encode both the transport and 608
catabolism of these polyols. Sorbitol and mannitol are oxidized to fructose in S. meliloti and D-609
arabitol is oxidized to D-xylulose (Martinez de Drets and Arias 1970). A mutant deficient in 610
fructose kinase activity in S. meliloti L5-30 was unable to grow using mannitol or sorbitol as sole 611
carbon sources, consistent with these polyols entering central metabolism through fructose 612
(Figure 5) (Gardiol et al. 1980). D-xylulose derived from D-arabitol oxidation presumably 613
enters central metabolism through the PP pathway as in xylose catabolism. Consistent with this, 614
a mutant of xylB was unable to grow using D-arabitol as a sole carbon source (Geddes and 615
Oresnik 2012a). 616
Dulcitol 617
A region of pSymB has been shown to be required for the utilization of dulcitol as a sole 618
carbon source (Charles and Finan 1991; Milunovic et al. 2014). The most likely candidate genes 619
for dulcitol catabolism in this region are SMb21375-SMb21377, encoding a putative ABC 620
transporter. Transcriptional fusions to SMb21377 showed significant induction by dulcitol 621
(Mauchline et al. 2006). Although dulcitol dehydrogenase is present in R. leguminosarum bv. 622
trifolii (Primrose and Ronson 1980), an NAD+ or NADP
+ dependent dulcitol dehydrogenase was 623
not detected in the extracts of dulcitol grown S. meliloti cells (Martinez de Drets and Arias 624
1970). Therefore, dulcitol may be catabolized by a unique route in S. meliloti. 625
Erythritol, Adonitol and L-arabitol 626
Page 27 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
The catabolism of the four carbon polyol erythritol has been characterized in Brucella 627
abortus and R. leguminosarum bv. viciae (Sangari et al. 2000; Sperry and Robertson 1975a, b; 628
Yost et al. 2006). The biochemical pathway of erythritol catabolism was elucidated in B. 629
abortus. Erythritol catabolism proceeds through phosphorylation to erythritol-1-phosphate, 630
followed by oxidation to erythrulose-1-phosphate. Further oxidations generating 3-keto-631
erythrose-4-phosphate and 3-keto-erythronate-4-phosphate, followed by decarboxylation yield 632
DHAP where it enters central metabolism (Sperry and Robertson 1975a, b). A genetic locus was 633
identified that encoded erythritol kinase (eryA), erythritol-1-phosphate dehydrogenase (eryB) and 634
erythrulose-1-phosphate dehydrogenase (eryC) organized as an inducible operon with a putative 635
negative regulator (eryD) (Sangari et al. 2000). A homologous operon was later identified in R. 636
leguminosarum and shown to be necessary for erythritol utilization (Yost et al. 2006). The R. 637
leguminosarum erythritol locus was further shown to contain a divergently transcribed ABC-638
transporter (eryEFG) that was necessary for growth using erythritol as a sole carbon source (Yost 639
et al. 2006). 640
The erythritol locus of S. meliloti was identified during the characterization of the triose-641
phosphate isomerase genes tpiA and tpiB. A putative locus of erythritol catabolism that 642
contained orthologs to eryA, eryB, eryC and eryD was identified adjacent to tpiB in the genome 643
(Poysti and Oresnik 2007). It was found that although S. meliloti contains homologs to all of 644
eryABCD, the content and arrangement of its genetic locus of erythritol catabolism was 645
dramatically different (Geddes et al. 2013). Despite these differences, the genes eryA, eryB and 646
eryC were shown to be necessary for erythritol catabolism in S. meliloti, suggesting that 647
erythritol is catabolized through a similar pathway to that elucidated in Brucella (Geddes et al. 648
2010). An ABC transporter encoded in the locus that is not homologous to the erythritol 649
Page 28 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
transporter of R. leguminosarum and Brucella was shown to be necessary for the transport of 650
erythritol. 651
In addition to erythritol catabolism, the locus was shown to encode all of the components 652
necessary for the transport and catabolism of adonitol and L-arabitol (Geddes and Oresnik 653
2012a). Adonitol and L-arabitol were shown to compete for transport with erythritol confirming 654
that the transporter encoded by the locus transports all three polyols and adonitol and L-arabitol 655
were shown to be phosphorylated by the erythritol kinase EryA as the first step in their 656
catabolism (Geddes and Oresnik 2012a). Adonitol and L-arabitol catabolism was shown to be 657
further facilitated by a number of other genes interspersed through the locus: rbtABC; and 658
rbtABC, lalA respectively (Geddes and Oresnik 2012a). Based on annotation, and genetic data a 659
putative pathway of adonitol and L-arabitol catabolism was proposed wherin adonitol and L-660
arabitol enter central metabolism through the pentose phosphate intermediate D-ribulose-5-P, 661
however more biochemical evidence is necessary to confirm this pathway (Geddes and Oresnik 662
2012a). 663
664
Glycerol 665
Glycerol is the most widely used carbon source by rhizobia (Stowers 1985). Glycerol 666
catabolism has been studied in B. japonicum and R. leguminosarum bv. viciae, where it was 667
found that glycerol is catabolized by glycerol kinase and α-glycerolphosphate dehydrogenase to 668
DHAP (Arias and Martinez de Drets 1976). As well, in R. leguminosarum bv. viciae, an operon 669
containing an ABC transporter for glycerol (glpSTPQUV), the glycerol kinase glpK, and the 670
glycerol-3-phosphate dehydrogenase, glpD have been identified and were all shown to be 671
Page 29 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
necessary for the ability to utilize glycerol as a sole carbon source (Ding et al. 2012). This 672
operon is conserved across many of the α-proteobacteria. 673
In S. meliloti glycerol catabolism has not been well studied. S. meliloti contains an 674
orthologous locus to the R. leguminosarum locus, however the kinase glpK is not present (Ding 675
et al. 2012). The gene encoding glycerol kinase is divergently transcribed from bdhA and is 676
found on pSymB (Aneja and Charles 1999). The induction of these predicted glycerol 677
catabolism genes has been demonstrated during growth of S. meliloti using glycerol as a sole 678
carbon source (White et al. 2012). Taken together this suggests that the catabolism of glycerol 679
occurs in an identical manner to that of R. leguminosarum where it enters central metabolism 680
through DHAP which is subsequently converted to GAP (Figure 5). Consistent with this, S. 681
meliloti strains carrying tpiA mutations are unable to grow using glycerol as a sole carbon source 682
(Poysti and Oresnik 2007). Presumably some carbon from glycerol is then cycled through the 683
ED pathway, while the rest is catabolized to pyruvate. This is supported by the observation that 684
strains carrying mutations in either eno and gap are able to grow slowly with glycerol as a sole 685
carbon source; indicating that the lower half of EMP pathway is not required but is beneficial for 686
growth (Finan et al. 1988). 687
Regulation of Carbon Metabolism 688
Succinate and other C4 dicarboxylic acids may be the preferred carbon sources for S. 689
meliloti and R. leguminosarum. The regulation of the genes involved in the catabolism and 690
transport of disaccharides, hexoses and sugar alcohols supports this, where many are subject to 691
succinate-mediated catabolite repression (SMCR) (de Vries et al. 1982; Gage and Long 1998; 692
Poole et al. 1994; Ucker and Signer 1978). In R. leguminosarum glucose mediated catabolite 693
repression-like phenomena have also been reported during growth with sugar alcohols (Poole et 694
Page 30 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
al. 1994; Ronson and Primrose 1979). Glucose mediated catabolite repression of the α -695
galactoside locus has been demonstrated in S. meliloti (Gage and Long 1998). 696
SMCR is manifested as diauxic growth in S. meliloti during growth with lactose and is 697
independent of cyclic nucleotides (Ucker and Signer 1978). SMCR has been proposed to 698
function by an inducer exclusion mechanism where the presence of succinate in the cell 699
transduces a signal that prevents the accumulation of inducing molecules that result in the 700
expression of sugar transport and uptake systems (Bringhurst and Gage 2002). 701
Although S. meliloti does not contain a complete PTS system for sugar transport, it does 702
contain several components of a nitrogen-type PTS system (PTSNtr
) including EINtr
, Hpr and 703
EIIANtr
that is thought to control SMCR (Pinedo et al. 2008; Pinedo and Gage 2009). Evidence 704
is consistent with a model wherin EI phosphorylates HIS-22 of Hpr-yielding Hpr-P. Hpr-P 705
subsequently phosphorylates EIIA-like components (Goodwin and Gage 2014; Untiet et al. 706
2013). Phosphorylation and dephosphorylation of SER-53 of Hpr by the kinase/phosphatase 707
HprK has been shown to influence SMCR (Pinedo and Gage 2009). EINtr
activity is modulated 708
by interaction with glutamine (Goodwin and Gage 2014). SMCR is also influenced by a two-709
component system encoded by SMa0113/SMa0114 (Garcia et al. 2010). Notably, components 710
of the PTSNTR
system in R. leguminosarum have been suggested to globally regulate ATP 711
transporters (Prell et al. 2012). 712
Another emerging layer of global regulation affecting carbon metabolism is through the 713
action of small regulatory RNAs. Strains carrying mutations in the RNA chaperone Hfq have 714
been shown to display the differential accumulation of a large number of mRNA transcripts (Gao 715
et al. 2010; Torres-Quesada et al. 2010). These include several transcripts associated with the 716
transport and utilization of carbon compounds as well as central metabolism. For example, the 717
Page 31 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
reduced expression of genes associated with alpha-glucoside and glycerol transport and the 718
increased expression of pckA were reported by two independent studies (Gao et al. 2010; Torres-719
Quesada et al. 2010). It is expected that future studies will more clearly dissect the role of 720
specific small RNAs in the regulation transport and metabolism (Torres-Quesada et al. 2013; 721
Torres-Quesada et al. 2014). 722
723
Biological relevance of carbon metabolism 724
Metabolism is fundamental to the physiology of any organism. S. meliloti can be found 725
as a saprophytic organism, colonizing a root surface, or in an association with the host plant. The 726
ability to catabolize organic compounds is important at all stages of its lifecycle including free-727
living in the rhizosphere, during the infection process of its host and in the intracellular nitrogen-728
fixing bacteroid form. 729
Free-living catabolism (competition for nodule occupancy) 730
Prior to invasion of the plant, S. meliloti must grow and thrive saprophytically in the 731
rhizosphere. The effectiveness of rhizobial inocula in agriculture is limited by the inability of 732
commercial inocula to compete with indigenous rhizobia for nodule occupancy of legume crops 733
(Triplett and Sadowsky 1992). The ability to thrive in the rhizosphere and on the root surface 734
will be influenced by the ability to extract energy from carbon compounds that are available. 735
Most rhizobia, including S. meliloti, contain large multi-partite genomes. The large 736
genomic size and complexity is reflective of a large metabolic capacity that could be 737
advantageous towards adaptation to the diverse environment of the soil. Consistent with this 738
idea, the metabolic potential of Rhizobium strains has been correlated with increased 739
competitiveness (Wiebo et al. 2007). The S. meliloti genome contains a large number of ABC 740
Page 32 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
transporters that are inducible by organic compounds (Galibert et al. 2001; Mauchline et al. 741
2006). Of the two megaplasmids contained by S. meliloti, pSymA has been described as a 742
symbiotic plasmid, encoding the genes required for nod factor biosynthesis and nitrogen fixation 743
(Barnett et al. 2001). The second megaplasmid, pSymB encodes genes for exopolysaccharide 744
biosynthesis as well as a large number of solute uptake systems and carbon metabolism genes 745
(Finan et al. 2001). It it has been suggested that pSymB may play a primary role in the ability to 746
thrive in the diverse environment of the soil during saprophytic growth (Finan et al. 2001). 747
Evidence for plasmids providing a competitive advantage in this way has been 748
demonstrated in rhizobia. In R. leguminosarum bv. viciae and Rhizobium etli CFN42, plasmid 749
curing experiments showed that strains lacking certain plasmids were less competitive for nodule 750
occupancy than wild-type strains (Brom et al. 1992; Hynes and McGregor 1990; Hynes and O' 751
Connell 1990). This idea has been reinforced by an experiment analyzing the transcriptional 752
response of R. leguminosarum bv. viciae grown in the rhizosphere of its host legume (pea), a 753
non-host legume (alfalfa) and a non-legume (sugar beet) (Ramachandran et al. 2011). Here, one 754
of the six plasmids contained by R. leguminosarum, pRL8, was shown to be a pea rhizosphere 755
specific plasmid. A large proportion of genes up-regulated specifically in the pea rhizosphere are 756
encoded on this plasmid, and mutants of many of these genes showed reduced competitiveness 757
for pea rhizosphere colonization (Ramachandran et al. 2011). Overall, carbon metabolism in the 758
rhizosphere in R. leguminosarum was shown to be dominated by organic acids with a bias 759
towards those that consist of one or two carbon atoms (Ramachandran et al. 2011). 760
There has been no directed study of the carbon compounds that make up the rhizosphere 761
of the legume hosts of S. meliloti. The R. leguminosarum transcriptional responses that were 762
specific to the alfalfa rhizosphere suggest that the C4 dicarboxylic acid malonate and the 763
Page 33 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
polysaccharide arabinogalactan may be prevalent (Ramachandran et al. 2011). A gfp fusion to 764
the melA promoter of S. meliloti, which is induced by galactose and galactosides, was used as a 765
biosensor to show that galactosides are released from alfalfa seeds during germination, and from 766
roots of alfalfa seedlings in sterilized and unsterilized soil (Bringhurst et al. 2001). The 767
arabinose locus has also been shown to be induced by alfalfa seed exudate (Poysti et al. 2007). 768
One of the two putative ribulose-5-phosphate epimerase genes, ppe, and the sugar binding 769
protein frcB of the fructose transporter are induced by alfalfa root exudate (Zhang and Cheng, 770
2006). Transcription of lacZ reporter gene fusions to the trehalose catabolic gene thuB and the 771
PQQ-dependent glucose dehydrogenase gcd was observed during growth on the alfalfa root 772
surface in addition to other stages of infection (Bernardelli et al. 2008; Jensen et al. 2005). 773
An alternate approach to identifying carbon sources that may be important for the ability 774
to thrive in the rhizosphere has been the isolation of genetic determinants of competition (Triplett 775
and Sadowsky 1992). Several different carbon utilization mutants have been isolated that have 776
been shown to be impaired in competitiveness for nodule occupancy in S. meliloti and R. 777
leguminosarum. The ability to utilize rhamnose is plasmid encoded in R. leguminosarum and 778
mutants that are unable to utilize rhamnose show a severe competitive defect (Oresnik et al. 779
1998). Mutants that were unable to catabolize inositol showed reduced competitiveness for 780
nodule occupancy in both S. meliloti Rm1021 and R. leguminosarum as well as S. fredii 781
USDA191 (Fry et al. 2001; Jiang et al. 2001; Kohler et al. 2010). 782
Some strains of S. meliloti and R. leguminosarum are capable of both the synthesis and 783
catabolism of inositol derivatives called rhizopines (L-3-O-methyl-scyllo-inosamine and scyllo-784
inosamine). Rhizopine is synthesized in the nodules formed by these species, and exuded into the 785
rhizosphere where it can be catabolized by other saprophytically growing rhizobia (few other 786
Page 34 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
bacteria possess the ability to catabolize rhizopine) (Murphy et al. 1995). This has led to the 787
“rhizopine concept” which suggests an altruistic synthesis of rhizopine in the nodules by rhizobia 788
for catabolism by their siblings in the rhizosphere or during invasion, providing them with a 789
competitive advantage (Murphy et al. 1995). Consistent with this concept, mutants in rhizopine 790
catabolism genes had a reduced ability to compete for nodule occupancy (Gordon et al. 1996). 791
Mutants that were unable to catabolize erythritol, glycerol and homoserine all reduced 792
competitiveness for nodule occupancy in R. leguminosarum bv. viciae (Ding et al. 2012; 793
Vanderlinde et al. 2013; Yost et al. 2006). The ability to catabolize the amino acid proline was 794
shown to contribute to competitiveness in S. meliloti under drought conditions (Jiménez-Zurdo et 795
al. 1995; van Dillewijn et al. 2001). Mutants of the PQQ-dependent glucose dehydrogenase 796
showed a reduced competitiveness for nodule occupancy in S. meliloti (Bernardelli et al. 2008). 797
However, given that glucose catabolism would still be intact in this mutant it is possible that the 798
competitive defect may be a result of other effects. For example, gluconate production by a 799
periplasmic glucose dehydrogenase is involved in inorganic phosphate solubilization as well as 800
the regulation of biocontrol traits in Pseudomonas spp. (de Werra et al. 2009). 801
In most cases there is not significant evidence showing that competition is occurring 802
specifically in the rhizosphere. An alternative possibility is that the carbon compounds may be 803
utilized during portions of the infection process further downstream. It has been hypothesized 804
that this may be the case for both rhamnose and inositol in R. leguminosarum- based on the lack 805
of significant differences in competitiveness observed at the level of root colonization and 806
rhizosphere growth, respectively (Fry et al. 2001; Oresnik et al. 1998). 807
Carbon catabolism and invasion 808
Page 35 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Successful penetration of the infection thread by S. meliloti requires both continued 809
biosynthesis of Nod Factor as well as biosynthesis of symbiotic exopolysaccharide (Jones et al. 810
2007). The requirements for active biosynthesis of such macromolecules as well as for active 811
proliferation in both the curled colonized root hair and the infection thread suggests the rhizobia 812
must be in metabolically active to allow successful invasion. Despite the importance of this 813
portion of the life cycle of S. meliloti, the carbon sources that are utilized to fuel invasion have 814
not been elucidated. 815
It has been postulated that growth in the infection thread may be fueled by the storage 816
polymer polyhydroxybutyrate (PHB) in S. meliloti (Charles et al. 1997). PHB accumulates as 817
granules during normal growth in S. meliloti and other indeterminate nodule forming rhizobia 818
that disappear during bacteroid differentiation (Lodwig et al. 2005). It has therefore also been 819
hypothesized that PHB may be an energy store to drive bacteroid differentiation (Prell and Poole 820
2006). Consistent with the hypothesis that PHB may help fuel growth in the infection thread, 821
mutants that were unable to synthesize or catabolize PHB were less competitive than the wild-822
type for nodule occupancy (Aneja et al. 2005; Willis and Walker 1998). However, mutants that 823
were unable to synthesize PHB were at a more severe competitive disadvantage than mutants 824
that were unable to catabolize PHB (Aneja et al. 2005). Therefore, it was suggested that PHB 825
synthesis may play a role in removing inhibitory metabolic intermediates during invasion (Aneja 826
et al. 2005). The study of the role of PHB in S. meliloti is complicated because rhizobia are also 827
capable of accumulating the storage polymer glycogen in addition to PHB (Lodwig et al. 2005). 828
To resolve how these storage polymers function in the interactions of S. meliloti with Medicago 829
spp. a study was conducted wherein mutants in PHB synthesis (phbC) were constructed in 830
concert with mutants in the two putative glycogen synthases in S. meliloti (glgA1, glgA2) (Wang 831
Page 36 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
et al. 2007). Glycogen was not detectable in glgA1 mutants, demonstrating that glgA1 encodes 832
the glycogen synthase in S. meliloti (Wang et al. 2007). Glycogen levels were elevated in a 833
phbC mutant, whereas PHB accumulation was reduced in the glgA1 mutant (Wang et al. 2007). 834
Mutants in glycogen biosynthesis appeared to be unaffected in invasion of Medicago spp., but 835
did show reduced nitrogen fixation phenotypes. Consistent with a reduced ability to compete for 836
nodule occupancy, nodule formation delays were observed in phbC mutants, however EPS 837
synthesis was significantly reduced in phbC mutants. Therefore, it is possible that PHB plays a 838
role in fueling EPS biosynthesis during invasion (Wang et al. 2007). 839
Although the mechanism is unclear, the synthesis of exopolysaccharide is critical for 840
invasion by S. meliloti (Jones et al. 2007). EPS synthesis has also been shown to affect 841
competition for nodule occupancy in other rhizobia (Triplett and Sadowsky 1992). The common 842
precursor for the biosynthesis of the sugar subunits for EPSI, EPSII, and cyclic β-glucan is 843
glucose-6-P. Isotopic labeling experiments have demonstrated that during growth on hexoses, 844
much of the carbon used for the biosynthesis of these polysaccharides is degraded through the 845
ED and PP pathways and reformed before being used for biosynthesis (Gosselin et al. 2001; 846
Portais et al. 1999). These results demonstrate a link between central metabolism and 847
biosynthesis of these important symbiotic molecules. Consistent with the role of central 848
metabolism in exopolysaccharide biosynthesis, strains carrying phosphoglucose isomerase 849
mutants were 80% reduced in exopolysaccharide production relative to wild-type (Arias et al. 850
1979). 851
Some studies have suggested a link between the regulation of carbon metabolism and 852
exopolysaccharide biosynthesis. The α-galactoside locus was isolated on the basis of identifying 853
genes down regulated by SyrA which is also known to effect exopolysaccharide regulation 854
Page 37 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
(Gage and Long 1998). Mutations in components of the PTS-like SMCR system (hpr and eIIA) 855
also showed dramatic effects on exopolysaccharide production (Pinedo et al. 2008). Notably, 856
these components are encoded in a locus with exoS and chvI; ExoS and ChvI comprise a two-857
component system that regulates exopolysaccharide biosynthesis in S. meliloti (Cheng and 858
Walker 1998). In addition to dramatic effects on exopolysaccharide production, null mutants of 859
exoS and chvI showed an inability to grow using over 21 different carbon sources (Bélanger et al. 860
2009). During a global study of ChvI binding sites, a ChvI binding site was identified upstream 861
of the putative galactose dehydrogenase gal/SMc00588 (Bélanger and Charles 2013). Despite 862
these links it is unclear how carbon metabolism, or its regulation might influence 863
exopolysaccharide biosynthesis during symbiosis. It is of note that it has been recently shown 864
that mutants unable to catabolize galactose are more competitive for nodule occupancy (Geddes 865
and Oresnik 2012c). 866
Osmoprotection of S. meliloti can be accomplished by the accumulation of trehalose in 867
the cytosol during growth in hyperosmotic conditions and the accumulation of cyclic β-glucan in 868
the periplasm during growth in hypoosmotic conditions (Breedveld et al. 1993; Dylan et al. 869
1990a). Symbiotic phenotypes have been associated with the inability to synthesize either cyclic 870
β-glucan, or trehalose. Mutants in cyclic β-glucan biosynthesis were impaired in their ability to 871
nodulate the host plant; however, suppressor mutants that regained the ability to effectively 872
nodulate alfalfa did not regain the ability to synthesize cyclic β-glucan and remained sensitive to 873
hypoosmotic growth conditions (Dylan et al. 1990b). 874
Mutants of three trehalose biosynthetic loci showed reduced osmotic tolerance and a 875
triple mutant of all three loci showed a significantly reduced ability to compete for nodule 876
occupancy (Domínguez-Ferreras et al. 2009). Therefore, tolerance to hyperosmotic conditions 877
Page 38 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
may be more important than tolerance to hypoosmotic conditions during invasion. An intriguing 878
relationship between trehalose biosynthesis and catabolism during invasion has been 879
demonstrated in S. meliloti where mutants of the trehalose catabolism genes thuA and thuB 880
showed an increased ability to compete for nodule occupancy despite a reduced ability to 881
colonize alfalfa roots (Jensen et al. 2005). This increased competitive advantage occurs at the 882
level of root hair infection (Jensen et al. 2005). It was suggested that the increased ability to 883
compete for nodule occupancy was the result of enhanced tolerance to osmotic stress in the 884
infection thread due to the incidental accumulation of trehalose in the catabolic mutants (Jensen 885
et al. 2005). Trehalose biosynthesis is also tied to central metabolism as glucose-6-P 886
dehydrogenase was shown to be required for trehalose to be an efficient osmoprotectant (Barra et 887
al. 2003). 888
889
Bacteroid metabolism 890
Evidence is overwhelmingly in favor of C4 dicarboxylic acids as the carbon source in the 891
bacteroid. The dicarboxylic acids are derived from sucrose photosynthate in plant cells, and 892
transported through the symbiotic membrane to the bacteroid. There is evidence that the 893
symbiotic membrane of legumes contains a dicarboxylic acid transporter, however a genetic 894
locus encoding the transporter has not yet been identified (Udvardi and Poole 2013). Bacterial 895
genetics has defined the components for dicarboxylic acid transport and catabolism in S. meliloti. 896
Mutants that are unable to transport or catabolize dicarboxylic acids display phenotypes 897
consistent with normal infection and bacteroid development, but produce bacteroids that are 898
devoid of nitrogen fixation (Fix -) (Lodwig and Poole 2003). In contrast, mutants of hexose or 899
polyol catabolism, as well as the key central catabolism enzymes G6P dehydrogenase and 900
Page 39 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
pyruvate carboxylase show no significant effect on nitrogen fixation in bacteroids (Cerveñansky 901
and Arias 1984; Dunn et al. 2001; Stowers 1985). 902
Dicarboxylic acid transporters (Dct) were first identified in R. leguminosarum where they 903
were shown to be necessary for growth on succinate, fumarate and malate as well as being 904
necessary for nitrogen fixation in nodules (Finan et al. 1983; Ronson et al. 1981). In S. meliloti, 905
dct mutants were subsequently isolated and also shown to be necessary for effective nitrogen 906
fixation (Bolton et al. 1986; Engelke et al. 1987; Finan et al. 1988; Jiang et al. 1989; Watson et 907
al. 1988; Yarosh et al. 1989). The C4 dicarboxylic acid transporter DctA, is a member of the 908
major facilitator superfamily of transporters. 909
C4 dicarboxylic acids are catabolized through the TCA cycle. Consistent with this, the 910
TCA cycle has been shown to be indispensible for nitrogen fixation in many rhizobia. Mutants 911
of several TCA cycle enzymes in S. meliloti have been shown to display a Fix - phenotype. 912
These include succinate dehydrogenase (Gardiol et al. 1982) citrate synthase (Koziol et al. 2009), 913
isocitrate dehydrogenase (McDermott and Kahn 1992), and malate dehydrogenase (Dymov et al. 914
2004). Function of the TCA cycle during growth on dicarboxylates requires the active synthesis 915
of acetyl-CoA from pyruvate. In S. meliloti bacteroids this role is predominantly performed by 916
NAD+ dependent malic enzyme (dme) rather than PEP carboxykinase (pck) or NADP
+ dependent 917
malic enzyme (tme). Mutants of dme display a Fix – phenotype (Driscoll and Finan 1993). 918
Mutants of tme show no symbiotic phenotype, and despite encoding enzymes with the same 919
biochemical activity that could functionally replace each other in vitro, overexpressed tme was 920
unable to complement dme mutants for functional symbiosis in planta (Driscoll and Finan 1996; 921
Mitsch et al. 2007). Mutants of pckA did show a reduced nitrogen fixation phenotype (Finan et 922
al. 1991). However, significant PEP carboxykinase activity has not been detected in bacteroids, 923
Page 40 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
whereas dme has been shown to be constitutively expressed (Driscoll and Finan 1996; Finan et 924
al. 1991). 925
In addition to the synthesis of pyruvate for acetyl-CoA production, evidence is consistent 926
with the requirement of gluconeogenesis to synthesize essential sugar intermediates for 927
biosynthesis in the bacteroid. Several mutants of the lower EMP pathway showed Fix – 928
phenotypes, including enolase, G3P dehydrogenase and 3PG kinase (Finan et al. 1988). 929
Moving Forward/ Conclusion 930 931 The understanding of the transport and catabolism of many carbon sources in S. meliloti 932
is well understood and can be used as a model for carbon metabolism in other bacteria that do 933
not fit the E. coli or B. subtilis paradigms. The majority of the information presented in this 934
review has been elucidated from an integration of biochemical and genetic evidence. The former 935
being primarily elucidated during the 70s and 80s, and the latter coincident with the genomic era 936
of the 90s and early 2000s. Global studies of carbon metabolism have been invaluable in 937
enlightening our understanding of the metabolism of many different carbon sources en masse. 938
These include studies such as the global mapping of expression of expression patterns of ABC 939
and TRAP transport systems in response to carbon sources in S. meliloti presented by 940
Maucheline et al. (2006), and the characterization of carbon utilization phenotypes following 941
large-scale mutagenesis of genes encoding short-chain dehydrogenases (Jacob et al. 2008). It is 942
clear that more broad scale studies such as these could drive expansion in our understanding of 943
carbon metabolism in S. meliloti in the future; to date most global approaches have been carried 944
out using a limited number of growth conditions. 945
It is also important to consider the need for continued directed functional characterization 946
arising from the growing trend to assign function based on homology. As shown here many 947
Page 41 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
paradigm differences exist between S. meliloti and other model organisms. Therefore 948
establishing models for catabolism in different bacterial lineages will help refine the accuracy of 949
functional association from bacteria in similar lineages. The large metabolic potential of S. 950
meliloti also has the advantage of allowing the characterization of pathways of peripheral 951
metabolism that may not be present in model bacteria with smaller metabolic capacity. 952
From a biological standpoint several key questions need to be addressed in the future 953
with respect to carbon metabolism. Transcriptome studies such as those performed by 954
Ramachandran et al. (2011) should be performed on S. meliloti during interaction with its hosts 955
to help push our understanding of the carbon experienced in the rhizosphere by S. meliloti 956
beyond a number of disparate phenotypes. As studies like this move further it is clear that it is 957
also critical to consider the rhizosphere in a special and temporal manner. The profile of 958
compounds present in root exudates of Arabidopsis plants has recently been shown to change 959
dramatically during plant development, based on both direct quantitation of compounds and 960
global expression profiles of bacteria in the rhizosphere (Chaparro et al. 2013). Exudate profile 961
is also variable spatially along the axis of the root (Herron et al. 2013). This seems an important 962
consideration, particularly since rhizobia are biased towards invasion of specific zones of the 963
plant root (Bauer 1981). Biosensors for specific carbon sources are a new and promising 964
technology that may help with studies such as this (Bourdès et al. 2012; Herron et al. 2013). 965
Moving further into the future, the influence of the microbial community in the soil environment 966
on exudate profiles and the response of rhizobia to these changes may prove insightful. 967
Acknowledgments 968
Page 42 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Work by IJO was funded by an NSERC Discovery grant. BAG acknowledges funding from an 969
NSERC CGS-D award. The authors are grateful T. Charles and T. Finan for comments and 970
suggestions. 971
972 973
Page 43 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
974 Literature Cited 975 976 Alvarez-Martinez, M.-T., Machold, J., Weise, C., Schmidt-Eisenlohr, H., Baron, C., and Rouot, 977 B. 2001. The Brucella suis homologue of the Agrobacterium tumefaciens chromosomal 978 virulence operon chvE is essential for sugar utilization but not for survival in macrophages. 979 J Bacteriol 183(18): 5343-5351. 980 981 Ampomah, O.Y., A., A., Hansen, E., Svenson, J., Huser, T., Jensen, J.B., and Bhuvaneswari, T.V. 982 2013. The thuEFGKAB operon of Rhizobia and Agrobacterium tumefaciens codes for 983 transport of trehalose, maltitiol, and isomers of sucrose and their assimilation through the 984 formation of their 3-keto derivatives. J Bacteriol 195: 3797-3807. 985 986 Aneja, P., and Charles, T.C. 1999. Poly-3-hydroxybutyrate degradation in Rhizobium 987 (Sinorhizobium) meliloti: isolation and characterization of a gene encoding 3-988 hydroxybutyrate dehydrogenase. J Bacteriol 181: 849-857. 989 990 Aneja, P., Zachertowska, A., and Charles, T. 2005. Comparison of the symbiotic and 991 competition phenotypes of Sinorhizobium meliloti PHB synthesis and degradation pathway 992 mutants. Can J Microbiol 51(7): 599-604. 993 994 Arias, A., and Cerveñansky, C. 1982. Transport and catabolism of D-mannose in Rhizobium 995 meliloti. J Bacteriol 151: 1069–1072. 996 997 Arias, A., and Cerveñansky, C. 1986. Galactose Metabolism in Rhizobium meliloti L5-30. J 998 Bacteriol 167(3): 1092-1094. 999 1000 Arias, A., Cerveñansky, C., Gardiol, A., and Martinez de Drets, G. 1979. Phosphoglucose 1001 isomerase mutant of Rhizobium meliloti. J Bacteriol 137: 409-414. 1002 1003 Arias, A., and Martinez de Drets, G. 1976. Glycerol metabolism in Rhizobium. Can J Microbiol 1004 22: 150-153. 1005 1006 Avetisyan, A., Jensen, J.B., and Huser, T. 2013. Monitoring trehalose uptake and conversion 1007 by single bacteria using laser tweezers Raman spectroscopy. Anal Chem(85): 7624-7270. 1008 1009 Barnett, M.J., Fisher, R.F., Jones, T., Komp, C., Abola, A.P., Barloy-Hubler, F., Bowser, L., 1010 Capela, D., Galibert, F., Gouzy, J., Gurjal, M., Hong, A., Huizar, L., Hyman, R.W., Kahn, D., Kahn, 1011 M.L., Kalman, S., Keating, D.H., Palm, C., Peck, M.C., Surzycki, R., Wells, D.H., Yeh, K.C., Davis, 1012 R.W., Federspiel, N.A., and Long, S.R. 2001. Nucleotide sequence and predicted functions of 1013 the entire Sinorhizobium meliloti pSymA megaplasmid. Proc Natl Acad Sci USA 98: 9883-1014 9888. 1015 1016 Barnett, M.J., Toman, C.J., Fisher, R.F., and Long, S.R. 2004. A dual-genome Symbiosis-Chip 1017 for coordinate study of signal exchange and development in a prokaryote-host interaction 1018 Proc Natl Acad Sci USA 101: 16636-16641. 1019
Page 44 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
1020 Barra, L., Bowser, L., Pica, N., Gouffi, K., Walker, G.C., Blanco, C., and Trautwetter, A. 2003. 1021 Glucose 6-phosphate dehydrogenase is required for sucrose and trehalose to be efficient 1022 osmoprotectants in Sinorhizobium meliloti. FEMS Microbiol Lett 229: 183-188. 1023 1024 Bauer, W.D. 1981. Infection of legumes by rhizobia. Ann Rev Plant Physiol 32(1): 407-449. 1025 1026 Bélanger, L., and Charles, T.C. 2013. Members of the Sinorhizobium meliloti ChvI regulon 1027 identified by a DNA binding screen. BMC Microbiology 13(1): 132. 1028 1029 Bélanger, L., Dimmick, K.A., Fleming, J.S., and Charles, T.C. 2009. Null mutations in 1030
Sinorhizobium meliloti exoS and chvI demonstrate the importance of this two‐component 1031
regulatory system for symbiosis. Mol Micro 74(5): 1223-1237. 1032 1033 Bernardelli, C.E., Luna, M.F., Galar, M.L., and Boiardi, J.L. 2001. Periplasmic PQQ-dependent 1034 glucose oxidation in free-living and symbiotic rhizobia. Curr Microbiol 42: 310-315. 1035 1036 Bernardelli, C.E., Luna, M.F., Galar, M.L., and Boiardi, J.L. 2008. Symbiotic phenotype of a 1037 membrane-bound glucose dehydrogenase mutant of Sinorhizobium meliloti. Plant and Soil 1038 313(1-2): 217-225. 1039 1040 Bolton, E., Higgisson, B., Harrington, A., and O'Gara, F. 1986. Dicarboxylic acid transport in 1041 Rhizobium meliloti: isolation of mutants and cloning of dicarboxylic acid transport genes. 1042 Arch Microbiol 144: 142-146. 1043 1044 Bourdès, A., Rudder, S., East, A.K., and Poole, P.S. 2012. Mining the Sinorhizobium meliloti 1045 transportome to develop FRET biosensors for sugars, dicarboxylates and cyclic polyols. 1046 PloS one 7(9): e43578. 1047 1048 Boutte, C.C., Srinivasan, B.S., Flannick, J.A., Novak, A.F., Martens, A.T., Batzoglou, S., Viollier, 1049 P.H., and Crosson, S. 2008. Genetic and computational identification of a conserved 1050 bacterial metabolic module. PLoS genetics 4(12): e1000310. 1051 1052 Breedveld, M.W., Dijkema, C., Zevenhuizen, L.P., and Zehnder, A.J. 1993. Response of 1053 intracellular carbohydrates to a NaCl shock in Rhizobium leguminosarum biovar trifolii TA-1054 1 and Rhizobium meliloti SU-47. J Gen Microbiol 139(12): 3157-3163. 1055 1056 Bringhurst, R.M., Cardon, Z.G., and Gage, D.J. 2001. Galactosides in the rhizosphere: 1057 Utilization by Sinorhizobium meliloti and development as a biosensor. Proc Natl Acad Sci 1058 USA 98(8): 4540-4545. 1059 1060 Bringhurst, R.M., and Gage, D.G. 2000. An AraC-like transcriptional activator is required for 1061 induction of genes needed for α-galactoside utilization in Sinorhizobium meliloti. FEMS 1062 Microbiol Lett 188: 23-27. 1063 1064
Page 45 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Bringhurst, R.M., and Gage, D.J. 2002. Control of inducer accumulation plays a key role in 1065 succinate-mediated catabolite repression in Sinorhizobium meliloti. J Bacteriol 184(19): 1066 5385-5392. 1067 1068 Brom, S., de los Santos, A.G., Stepkowsky, T., Flores, M., Davila, G., Romero, D., and Palacios, 1069 R. 1992. Different plasmides of Rhizobium leguminosarum bv. phaseoli are required for 1070 optimal symbiotic performance. J Bacteriol 174: 5183-5189. 1071 1072 Cabanes, D., Boistard, P., and Batut, J. 2000. Symbiotic induction of pyruvate dehydrogenase 1073 genes from Sinorhizobium meliloti. Mol Plant Microbe Interact 13: 483-493. 1074 1075 Capela, D., Barloy-Hubler, F., Gouzy, J., Bothe, G., Ampe, F., Batut, J., Boistard, P., Becker, A., 1076 Boutry, M., Cadieu, E., Dréano, S., Gloux, S., Godrie, T., Goffeau, A., Kahn, D., Kiss, E., Lelaure, 1077 V., Masuy, D., Pohl, T., Portetelle, D., Pühler, A., Purnelle, B., Ramsperger, U., Renard, C., 1078 Thébault, P., Vandenbol, M., Weidner, S., and Galibert, F. 2001. Analysis of the chromosome 1079 sequence of the legume symbiont Sinorhizobium meliloti strain 1021. Proc Natl Acad Sci 1080 USA 98: 9877-9882. 1081 1082 Cerveñansky, C., and Arias, A. 1984. Glucsoe-6-phsophate dehydrogenase deficiency in 1083 pleiotropic carbohydrate-negative mutant strains of Rhizobium meliloti. J Bacteriol 160: 1084 1027-1030. 1085 1086 Chaparro, J.M., Badri, D.V., Bakker, M.G., Sugiyama, A., Manter, D.K., and Vivanco, J.M. 2013. 1087 Root exudation of phytochemicals in Arabidopsis follows specific patterns that are 1088 developmentally programmed and correlate with soil microbial functions. PloS one 8(2): 1089 e55731. 1090 1091 Charles, T.C., Cai, G.-q., and Aneja, P. 1997. Megaplasmid and chromosomal loci for the PHB 1092 degradation pathway in Rhizobium (Sinorhizobium) meliloti. Genetics 146(4): 1211-1220. 1093 1094 Charles, T.C., and Finan, T.M. 1991. Analysis of a 1600-kilobase Rhizobium meliloti 1095 megaplasmid using defined deletions generated in vivo. Genetics 127: 5-20. 1096 1097 Charles, T.C., Singh, R.S., and Finan, T.M. 1990. Lactose utilization and enzymes encoded in 1098 Rhizobium meliloti: implications for population studies. J Gen Microbiol 136: 2497-2502. 1099 1100 Cheng, H.-P., and Walker, G.C. 1998. Succinoglycan production by Rhizobium meliloti is 1101 regulated through the ExoS-ChvI two-component regulatory system. J Bacteriol 180(1): 20-1102 26. 1103 1104 Conway, T. 1992. The Entner-Doudoroff pathway: history, physiology and moecular 1105 biology. FEMS Microbiol Rev 9: 1-27. 1106 1107 Dahms, A.S., and Anderson, R.L. 1972a. D-Fucose metabolism in a Pseudomonad: I. 1108 Oxidation of D-fucose to γ-fuconolactone by a D-aldohexose dehydrogenase. J Biol Chem 1109 247: 2222-2227. 1110
Page 46 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
1111 Dahms, A.S., and Anderson, R.L. 1972b. D-Fucose metabolism in a Pseudomonad: II. 1112 Oxidation of D-fucose to D-fucono-γ lactone by an L-aranino-aldolase dehydrogenase and 1113 hydrolysis of the lactone by a lactonase. J Biol Chem 247: 2228-2232. 1114 1115 Dahms, A.S., and Anderson, R.L. 1972c. D-Fucose metabolism in a Pseudomonad: III. 1116 Conversion of D-fuconate to 2-keto-3-deoxy-D-fuconate by a dehydratase. J Biol Chem 247: 1117 2233-2237. 1118 1119 Dahms, A.S., and Anderson, R.L. 1972d. D-Fucose metabolism in a Pseudomonad: IV. 1120 Cleavage of 2-keto-3-deoxy-D-fuconate to pyruvate and d-lactate by 2-keto-3-deoxy-1121 arabonate aldolase. J Biol Chem 247: 2238-2241. 1122 1123 De Ley, J., and Doudoroff, M. 1957. The metabolism of D-galactose in Pseudomonas 1124 saccharophila. J Biol Chem 227: 745-757. 1125 1126 de Vries, G.E., van Brussel, A., and Quispel, A. 1982. Mechanism of regulation of glucose 1127 transport in Rhizobium leguminosarum. J Bacteriol 149(3): 872-879. 1128 1129 de Werra, P., Péchy-Tarr, M., Keel, C., and Maurhofer, M. 2009. Role of gluconic acid 1130 production in the regulation of biocontrol traits of Pseudomonas fluorescens CHA0. Appl 1131 Environ Microbiol 75(12): 4162-4174. 1132 1133 Ding, H., Yip, C.B., Geddes, B.A., Oresnik, I.J., and Hynes, M.F. 2012. Glycerol utilization by 1134 Rhizobium leguminosarum requires an ABC transporter and affects competition for 1135 nodulation. Microbiology 158: 1369-1378. 1136 1137 Djordjevic, M.A. 2004. Sinorhizobium meliloti metabolism in the root nodule: a proteomic 1138 perspective. Proteomics 4: 1859–1872. 1139 1140 Domínguez-Ferreras, A., Soto, M.J., Pérez-Arnedo, R., Olivares, J., and Sanjuán, J. 2009. 1141 Importance of trehalose biosynthesis for Sinorhizobium meliloti osmotolerance and 1142 nodulation of alfalfa roots. J Bacteriol 191(24): 7490-7499. 1143 1144 Driscoll, B.T., and Finan, T.M. 1993. NAD+-dependent malic enzyme of Rhizobium meliloti is 1145 required for symbiotic nitrogen fixation. Mol Microbiol 7: 865-873. 1146 1147 Driscoll, B.T., and Finan, T.M. 1996. NADP+-dependent malic enzyme of Rhizobium meliloti. J 1148 Bacteriol 178: 2224-2231. 1149 1150 Driscoll, B.T., and Finan, T.M. 1997. Properties of NAD+- and NADP+-dependent malic 1151 enzymes of Rhizobium (Sinorhizobium) meliloti and differential expression of their genes in 1152 nitrogen-fixing bacteroids. Microbiology 143: 489-498. 1153 1154 Duncan, M.J. 1981. Properties of Tn5-induced carbohydrate mutants in Rhizobium meliloti. J 1155 Gen Microbiol 122: 61-67. 1156
Page 47 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Duncan, M.J., and Fraenkel, D.G. 1979. α-Ketoglutarate dehydrogenase mutant of Rhizobium 1157 meliloti. J Bacteriol 137(1): 415-419. 1158 1159 Dunn, M.F. 1998. Tricarboxylic acid cycle and anaplerotic enzymes in rhizobia. FEMS 1160 Microbiol Rev 22: 105-123. 1161 1162 Dunn, M.F. 2014. Key roles of microsymbiont amino acid metabolism in rhizobia-legume 1163 interactions. Crit Rev Microb(0): 1-41. 1164 1165 Dunn, M.F., Araíza, G., and Finan, T.M. 2001. Cloning and characterization of the pyruvate 1166 carboxylase from Sinorhizobium meliloti Rm1021. Arch Microbiol(176): 355-363. 1167 1168 Dylan, T., Helinski, D., and Ditta, G.S. 1990a. Hypoosmotic adaptation in Rhizobium meliloti 1169 requires β-(1-2)-glucan. J Bacteriol 172(3): 1400-1408. 1170 1171 Dylan, T., Nagpal, P., Helinski, D., and Ditta, G.S. 1990b. Symbiotic pseudorevertants of 1172 Rhizobium meliloti ndv mutants. J Bacteriol 172(3): 1409-1417. 1173 1174 Dymov, S.I., Meek, D.J., Steven, B., and Driscoll, B.T. 2004. Insetion of transposon Tn5tac1 in 1175 the Sinorhizobium meliloti malate dehydrogenase (mdh) gene results in conditional polar 1176 effects on downstream TCA cycle genes. Mol Plant Microbe Interact 17: 1318-1327. 1177 1178 Engelke, T., Jagadish, M.N., and Puhler, A. 1987. Biochemcial and genetical analysis of 1179 Rhizobium meliloti mutants defective in C4-dicarboxylate transport. J Gen Microbiol 133: 1180 3019-3029. 1181 1182 Fennington, G.J., and Hughes, T.A. 1996. The fructokinase from Rhizobium leguminosarum 1183 biovar trifolii belongs to group I fructokinase enzymes and is encoded separately from 1184 other carbohydrate metabolism enzymes. Microbiology 142: 321-330. 1185 1186 Finan, T.M., McWhinne, E., Driscoll, B., and Watson, R.J. 1991. Complex symbiotic 1187 phenotypes result from gluconeogenic mutations in Rhizobium meliloti. Mol Plant Microbe 1188 Interact 4: 386-392. 1189 1190 Finan, T.M., Oresnik, I., and Bottacin, A. 1988. Mutants of Rhizobium meliloti defective in 1191 succinate metabolism. J Bacteriol 170(8): 3396-3403. 1192 1193 Finan, T.M., Weidner, S., Wong, K., Buhrmester, J., Chain, P., Vorholter, F.J., Hernández-Lucas, 1194 I., Becker, A., Cowie, A., Gouzy, J., Golding, B., and Pühler, A. 2001. The complete sequence of 1195 the 1,683-kb pSymB megaplasmid from the N2-fixing endosymbiont S. meliloti. Proc Natl 1196 Acad Sci USA 98: 9889-9894. 1197 1198 Finan, T.M., Wood, J.M., and Jordan, D.C. 1983. Symbiotic properties of C4-dicarboxylic acid 1199 transport mutants of Rhizobium leguminosarum. J Bacteriol 154: 1403-1413. 1200 1201
Page 48 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Fry, J., Wood, M., and Poole, P.S. 2001. Investigation of myo-inositol catabolism in 1202 Rhizobium leguminosarum bv. viciae and its effect on nodulation competitiveness. Mol Plant 1203 Microbe Interact 14: 1016-1025. 1204 1205 Fuhrer, T., Fisher, E., and Sauer, U. 2005. Experimental identification and quantification of 1206 glucose metabolism in seven bacterial species. J Bacteriol(187): 1581-1590. 1207 1208 Gage, D.J., and Long, S.R. 1998. α-galactoside uptake in Rhizobium meliloti: isolation and 1209 characterization of agpA, a gene encoding a periplasmic binding protein required for 1210 melibiose and raffinose utilization. J Bacteriol 180(21): 5739-5748. 1211 1212 Galbraith, M.P., Feng, S.F., Borneman, J., Triplett, E.W., de Bruijn, F.J., and Rossbachl, S. 1998. 1213 A functional myo-inositol catabolism pathway is essential for rhizopine utilization by 1214 Sinorhizobium meliloti. Microbiology 144(10): 2915-2924. 1215 1216 Galibert, F., Finan, T.M., Long, S.R., Pühler, A., Abola, P., Ampe, F., Barloy-Hubler, F., Barnett, 1217 M.J., Becker, A., Boistard, P., Bothe, G., Boutry, M., Bowser, L., Buhrmester, J., Cadieu, E., 1218 Capela, D., Chain, P., Cowie, A., Davis, R.W., Dréano, S., Federspiel, N.A., Fisher, R.F., Gloux, S., 1219 Godrie, T., Goffeau, A., Golding, B., Gouzy, J., Gurjal, M., Hernández-Lucas, I., Hong, A., Huizar, 1220 L., Hyman, R.W., Jones, T., Kahn, D., Kahn, M.L., Kalman, S., Keating, D.H., Kiss, E., Komp, C., 1221 Lelaure, V., Masuy, D., Palm, C., Peck, M.C., Pohl, T.M., Portetelle, D., Purnelle, B., 1222 Ramsperger, U., Surzycki, R., Thébault, P., Vandenbol, M., Vorhölter, F.J., Weidner, S., Wells, 1223 D.H., Wong, K., Yeh, K.C., and Batut, J. 2001. The composite genome of the legume symbiont 1224 Sinorhizobium meliloti. Science 293: 668-672. 1225 1226 Gao, M., Barnett, M.J., Long, S.R., and Teplitski, M. 2010. Role of the Sinorhizobium meliloti 1227 global regulator Hfq in gene regulation and symbiosis. Mol Plant Microbe Interact 23(4): 1228 355-365. 1229 1230 Garcia, P.P., Bringhurst, R.M., Pinedo, C.A., and Gage, D.J. 2010. Characterization of a two-1231 component regulatory system that regulates succinate-mediated catabolite repression in 1232 Sinorhizobium meliloti. J Bacteriol 192(21): 5725-5735. 1233 1234 Gardiol, A., Arias, A., Cerveñanský, C., Gaggero, C., and Martínez-Drets, G. 1980. Biochemical 1235 characterization of a fructokinase mutant of Rhizobium meliloti. J Bacteriol 151: 1621–1236 1623. 1237 1238 Gardiol, A., Arias, A., Cerveñansky, C., and Martinez de Drets, G. 1982. Succinate 1239 dehydrogenase mutant of Rhizobium meliloti. J Bacteriol 151: 1621-1623. 1240 1241 Geddes, B.A., Hausner, G., and Oresnik, I.J. 2013. Phylogenetic analysis of erythritol 1242 catabolic loci within the Rhizobiales and Proteobacteria. BMC Microbiology 13(1): 46. 1243 1244 Geddes, B.A., and Oresnik, I.J. 2012a. Genetic characterization of a complex locus necessary 1245 for the transport and catabolism of erythritol, adonitol, and L-arabitol in Sinorhizobium 1246 meliloti. Microbiology 158(8): 2180-2191. doi: 10.1099/mic.0.057877-0. 1247
Page 49 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
1248 Geddes, B.A., Pickering, B.S., Poysti, N.J., Yudistira, H., Collins, H., and Oresnik, I.J. 2010. A 1249 locus necessary for the transport and catabolism of erythritol in Sinorhizobium meliloti 1250 Microbiology 156: 2970-2981. 1251 1252 Geddes, G.A., and Oresnik, I.J. 2012b. Inability to catabolize galactose leads to increased 1253 ability to compete for nodule occupancy in Sinorhizobium meliloti. J. Bacteriol 194: 5044-1254 5505. 1255 1256 Geddes, G.A., and Oresnik, I.J. 2012c. Inability to catabolize galactose leads to increased 1257 ability to compete for nodule occupancy in Sinorhizobium meliloti. J Bacteriol 194: 5044-1258 5505. 1259 1260 Glenn, A.R., and Dilworth, M.J. 1981. The uptake and hydrolysis of disaccharides by fast- 1261 and slow-growing species of Rhizobium. Arch Microbiol 129: 233-239. 1262 1263 Goodwin, R.A., and Gage, D.J. 2014. Biochemical Characterization of a Nitrogen-Type 1264 Phosphotransferase System Reveals that Enzyme EINtr Integrates Carbon and Nitrogen 1265 Signaling in Sinorhizobium meliloti. J Bacteriol 196(10): 1901-1907. 1266 1267 Gordon, D.M., Ryder, M.H., Heinrich, K., and Murphy, P.J. 1996. An experimental test of the 1268 rhizopine concept in Rhizobium meliloti. Appl Environ Microbiol 62(11): 3991-3996. 1269 1270 Gosselin, I., Wattraint, O., Riboul, D., Barbotin, J., and Portais, J. 2001. A deeper investigation 1271 on carbohydrate cycling in Sinorhizobium meliloti. FEBS Lett(499): 45-49. 1272 1273 Hanus, F.J., Maier, R.J., and Evans, H.J. 1979. Autotrophic growth of H2-uptake-positive 1274 strains of Rhizobium japonicum in an atmosphere supplied with hydrogen gas. Proc Natl 1275 Acad Sci USA 76: 1788-1792. 1276 1277 He, F., Nair, G.R., Soto, C.S., Chang, Y., Hsu, L., Ronzone, E., DeGrado, W.F., and Binns, A.N. 1278 2009. Molecular basis of ChvE function in sugar binding, sugar utilization, and virulence in 1279 Agrobacterium tumefaciens. J Bacteriol 191(18): 5802-5813. 1280 1281 Herron, P.M., Gage, D.J., Pinedo, C.A., Haider, Z.K., and Cardon, Z.G. 2013. Better to light a 1282 candle than curse the darkness: illuminating spatial localization and temporal dynamics of 1283 rapid microbial growth in the rhizosphere. Front Plant Sci 4(323). 1284 1285 Hynes, M.F., and McGregor, N.F. 1990. Two plasmids other than the nodulation plasmid are 1286 necessary for formation of nitrogen-fixing nodules by Rhizobium leguminosarum. Mol 1287 Microbiol 4: 567-574. 1288 1289 Hynes, M.F., and O' Connell, M.P. 1990. Host plant effect on competition among strains of 1290 Rhizobium leguminosarum. Can J Microbiol 36: 864-890. 1291 1292
Page 50 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Irigoyen, J.J., Sanchez-Diaz, M., and Emerich, D.W. 1990. Carbon metabolism enzymes of 1293 Rhizobium meliloti cultures and bacteroids and their distribution within alfalfa nodules. 1294 Appl Environ Microbiol 56: 2587-2589. 1295 1296 Jacob, A.I., Adhamn, S.A.I., Capstick, D.S., Clark, S.R.D., Spence, T., and Charles, T.C. 2008. 1297 Mutational analysis of the Sinorhizobium meliloti short-chain dehydrogenase/reductase 1298 family reveals substantial contribution to symbiosis and catabolic diversity. Mol Plant 1299 Microbe Interact 21: 979-989. 1300 1301 Jelesko, J.G., and Leigh, J.A. 1994. Genetic characterization of a Rhizobium meliloti lactose 1302 utilization locus. Mol Microbiol 11: 165-173. 1303 1304 Jensen, J.B., Ampomah, O.Y., Darrah, R., Peters, N.K., and Bhuvaneswari, T.V. 2005. Role of 1305 trehalose transport and utilization in Sinorhizobium meliloti-alfalfa interactions. Mol Plant 1306 Microbe Interact 18: 694-702. 1307 1308 Jensen, J.B., Peters, N.K., and Bhuvaneswari, T.V. 2002. Redundancy in periplasmic binding 1309 protein-dependent transport systems for trehalose, sucrose, and maltose in Sinorhizobium 1310 meliloti. J Bacteriol 184: 2978-2986. 1311 1312 Jiang, G., Krishnan, A.H., Kim, Y.-W., Wacek, T.J., and Krishnan, H.B. 2001. A functional myo-1313 inositol dehydrogenase gene is required for efficient nitrogen fixation and competitiveness 1314 of Sinorhizobium fredii USDA191 to nodulate soybean (Glycine max [L.] Merr.). J Bacteriol 1315 183(8): 2595-2604. 1316 1317 Jiang, J., Gu, B.H., Albright, L.M., and Nixon, T.B. 1989. Conservation between coding and 1318 regulatory elements of Rhizobiu meliloti and Rhizobium leguminosarum dct genes. J 1319 Bacteriol 171: 5244-5253. 1320 1321 Jiménez-Zurdo, J.I., van Dillewijn, P., Soto, M.J., de Filipe, M.R., Olivares, J., and Toro, N. 1995. 1322 Characterization of a Rhizobium meliloti proline dehydrogenase mutant altered in 1323 nodulation efficiency and competitiveness on alfalfa roots. Mol Plant Microbe Interact 8: 1324 492-498. 1325 1326 Jones, K.M., Kobayashi, H., Davies, B.W., Taga, M.E., and Walker, G.C. 2007. How rhizobial 1327 symbionts invade plants: the Sinorhizobium–Medicago model. Nat Rev Microbiol 5(8): 619-1328 633. 1329 1330 Jordan, D.C. 1984. Rhizobiaceae. In Bergey's Manual of Systematic Bacteriology. Edited by 1331 N.R. Kreig. Williams & Wilkins, Baltimore. pp. 234-241. 1332 1333 Koch, M., Delmotte, N., Ahrens, C.H., Omasits, U., Schneider, K., Danza, F., Padhi, B., Murset, 1334 V., Braissant, O., and Vorholt, J.A. 2014. A link between arabinose utilization and 1335 oxalotrophy in Bradyrhizobium japonicum. Appl Environ Microbiol 80(7): 2094-2101. 1336 1337
Page 51 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Kohler, P.R., Choong, E.-L., and Rossbach, S. 2011. The RpiR-like repressor IolR regulates 1338 inositol catabolism in Sinorhizobium meliloti. J Bacteriol 193(19): 5155-5163. 1339 1340 Kohler, P.R.A., Zheng, J.Y., Schoffers, E., and Rossbach, S. 2010. Inositol catabolism, a key 1341 pathway in Sinorhizobium meliloti for competitive host nodulation. Appl Environ Microbiol 1342 76: 7972-7980. 1343 1344 Koziol, U., Hannibal, L., Rodríguez, M.C., Fabiano, E., Kahn, M.L., and Noyz, F. 2009. Deletion 1345 of citrate synthase restores growth of Sinorhizobium meliloti 1021 aconitase mutants. J 1346 Bacteriol 191: 7581-7586. 1347 1348 Lambert, A., Østeräs, M., Mandon, K., Poggi, M.C., and Rudulier, D.L. 2001. Fructose uptake 1349 in Sinorhizobium meliloti is mediated by a high-affinity ATP-binding cassette transport 1350 system. J Bacteriol 183: 4709–4717. 1351 1352 Lessie, T.G., and Phibbs, P.V. 1984. Alternative pathways of carbohydrate utilization in 1353 pseudomonads. Annu Rev Microbiol 38: 359-388. 1354 1355 Lin, E.C.C. 1996. Dissimilatory pathways for sugars, polyols, and carboxylates. In Eschericia 1356 coli and Salmonella. Cellular and Molecular Biology. Edited by F.C. Neidhardt. ASM Press, 1357 Washington, D. C. pp. 307-341. 1358 1359 Lodwig, E., Leonard, M., Marroqui, S., Wheeler, T., Findlay, K., Downie, J., and Poole, P. 2005. 1360 Role of polyhydroxybutyrate and glycogen as carbon storage compounds in pea and bean 1361 bacteroids. Mol Plant Microbe Interact 18(1): 67-74. 1362 1363 Lodwig, E., and Poole, P. 2003. Metabolism of Rhizobium bacteroids. Crit Rev Plant Sci 22: 1364 37-78. 1365 1366 Manian, S.S., and O'Gara, F. 1982a. Derepression of ribulose bisphosphate carboxylase 1367 activity in Rhizobium meliloti. FEMS Microbiol Lett 14: 95-99. 1368 1369 Manian, S.S., and O'Gara, F. 1982b. Induction and regulation of ribulose bisphosphate 1370 carboxylase activity in Rhizobium japonicum during formate-dependent growth. Arch 1371 Microbiol 131: 51-54. 1372 1373 Martinez de Drets, G., and Arias, A. 1970. Metabolism of some polyols by Rhizobium meliloti. 1374 J Bacteriol 103: 97-103. 1375 1376 Martínez de Drets, G., and Arias, A. 1972. Enzymatic basis for differentiation of Rhizobium 1377 into fast- and slow-growing groups. J Bacteriol 109: 467-470. 1378 1379 Martinez de Drets, G., Arias, A., Rovira, A.D., and Cutinella, M. 1974. Fast- and slow growing 1380 rhizobia: differences in sucrose utilization and invertase activity. Can J Microbiol 20: 605-1381 609. 1382 1383
Page 52 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Mattevi, A., de Kok, A., and Perham, R.N. 1992. The pyruvate dehydrogenase multienzyme 1384 complex. Curr Opin Strct Biol 2: 877-887. 1385 1386 Mauchline, T.H., Fowler, J.E., East, A.K., Sartor, A.L., Zaheer, R., Hosie, A.H.F., Poole, P.S., and 1387 Finan, T.M. 2006. Mapping the Sinorhizobium meliloti 1021 solute-binding protein-1388 dependent transportome. Proc Natl Acad Sci USA 103(47): 17933-17938. 1389 1390 McDermott, T.R., and Kahn, M.L. 1992. Cloning and mutageneis of the Rhizobium meliloti 1391 isocitrate dehydrogenase gene. J Bacteriol 174(14): 4790-4797. 1392 1393 Milunovic, B., Morton, R.A., and Finan, T.M. 2014. Cell growth inhibition upon deletion of 1394 four toxin-antitoxin loci from the megaplasmids of Sinorhizobium meliloti. J Bacteriol 1395 196(4): 811-824. 1396 1397 Mitsch, M.J., Cowie, A., and Finan, T.M. 2007. Malic enzyme cofactor and domain 1398 requirements for symbiotic N2 fixationby Sinorhizobium meliloti. J Bacteriol 189: 160-168. 1399 1400 Mortimer, M.W., McDermott, T.R., York, G.M., Walker, G.C., and Kahn, M.L. 1999. Citrate 1401 synthase mutants of Sinorhizobium meliloti are ineffective and have altered cell surface 1402 polysaccharides. J Bacteriol 181: 7608-7613. 1403 1404 Murphy, P., Wexler, W., Grzemski, W., Rao, J., and Gordon, D. 1995. Rhizopines—their role 1405 in symbiosis and competition. Soil Biol Biochem 27(4): 525-529. 1406 1407 Niel, C., Guillaume, J.B., and Bechet, M. 1977. Mise en évidence de deux enzymes presentant 1408 une activité β-galactosidasique chez Rhizobium meliloti. Can J Microbiol 23: 1178-1181. 1409 1410 Oresnik, I.J., Pacarynuk, L.A., O'Brien, S.A.P., Yost, C.K., and Hynes, M.F. 1998. Plasmid 1411 encoded catabolic genes in Rhizobium leguminosarum bv. trifolii: evidence for a plant-1412 inducible rhamnose locus involved in competition for nodulation. Mol Plant Microbe 1413 Interact 11: 1175-1185. 1414 1415 Østeräs, M., Driscoll, B.T., and Finan, T.M. 1995. Molecular and expression analysis of the 1416 Rhizobium meliloti phosphoenolpyruvate carboxykinase (pckA) gene. J Bacteriol 177: 1417 1639-1648. 1418 1419 Østeräs, M., Driscoll, B.T., and Finan, T.M. 1997. Increased pyruvate orthophosphate 1420 dikinase activity results in an alternative gluconeogenic pathway in Rhizobium 1421 (Sinorhizobium) meliloti. Microbiology 143: 1639-1648. 1422 1423 Pedrosa, F.O., and Zancan, G.T. 1974. L-Arabinose metabolism in Rhizobium japonicum. J 1424 Bacteriol 119(1): 336-338. 1425 1426 Pickering, B.S., and Oresnik, I.J. 2008. Formate-dependent autotrophic growth in S. meliloti. 1427 J Bacteriol 190: 6409-6418. 1428 1429
Page 53 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Pinedo, C.A., Bringhurst, R.M., and Gage, D.J. 2008. Sinorhizobium meliloti mutants lacking 1430 PTS enzymes HPr or EIIA are altered in diverse processes including carbon metabolism, 1431 cobalt requirements and succinoglycan production. J Bacteriol 190(8): 2947-2956. 1432 1433 Pinedo, C.A., and Gage, D.J. 2009. HPrK Regulates succinate-mediated catabolite repression 1434 in the Gram-Negative symbiont Sinorhizobium meliloti. J Bacteriol 191(1): 298-309. 1435 1436 Poole, P.S., Blyth, A., Reid, C.J., and Walters, K. 1994. myo-Inositol catabolism and catabolite 1437 regulation in Rhizobium leguminosarum bv. viciae. Microbiology 140(10): 2787-2795. 1438 1439 Portais, J., Tavernier, P., Gosselin, I., and Barbotin, J. 1999. Cyclic organization of the 1440 carbohydrate metabolism in Sinorhizobium meliloti. Eur J Biochem 265: 473-480. 1441 1442 Poysti, N.J., Loewen, E.D., Wang, Z., and Oresnik, I.J. 2007. Sinorhizobium meliloti pSymB 1443 carries genes necessary for arabinose transport and catabolism. Microbiology 153: 727-1444 736. 1445 1446 Poysti, N.J., and Oresnik, I.J. 2007. Characterization of Sinorhizobium meliloti triose 1447 phosphate isomerase genes. J Bacteriol 189: 3445-3451. 1448 1449 Prell, J., Mulley, G., Haufe, F., White, J., Williams, A., Karunakaran, R., Downie, J., and Poole, P. 1450
2012. The PTSNtr system globally regulates ATP‐dependent transporters in Rhizobium 1451
leguminosarum. Mol Micro 84(1): 117-129. 1452 1453 Prell, J., and Poole, P. 2006. Metabolic changes of rhizobia in legume nodules. Trends 1454 Microbiol 14(4): 161-168. 1455 1456 Primrose, S.B., and Ronson, C.W. 1980. Polyol metabolism by Rhizobium trifolii. J Bacteriol 1457 141: 1109-1114. 1458 1459 Ramachandran, V.K., East, A.K., Karunakaran, R., Downie, J.A., and Poole, P. 2011. 1460 Adaptation of Rhizobium leguminosarum to peas, alfalfa, and sugar beet rhizospheres 1461 investigated by comparative transcriptomics. Genome Biol 12: R106. 1462 1463 Ramírez-Trujillo, J., Encarnación, S., Salazar, E., de los Santos, A.G., Dunn, M., Emerich, D., 1464 Calva, E., and Hernández-Lucas, I. 2007. Functional characterization of the Sinorhizobium 1465 meliloti acetate metabolism genes aceA, SMc00767, and glcB. J Bacteriol 189(16): 5875-1466 5884. 1467 1468 Richardson, J.S., Carpena, X., Switalta, J., Perez-Luque, R., Donald, L.J., Loewen, P.C., and 1469 Oresnik, I.J. 2008. RhaU of Rhizobium leguminosarum is a rhamnose mutarotase. J Bacteriol 1470 190: 2903-2910. 1471 1472 Richardson, J.S., Hynes, M.F., and Oresnik, I.J. 2004. A genetic locus necessary for rhamnose 1473 uptake and catabolism in Rhizobium leguminosarum bv. trifolii. J Bacteriol 186: 8433-8442. 1474
Page 54 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
1475 Richardson, J.S., and Oresnik, I.J. 2007. L-rhamnose transport in Rhizobium leguminosarum 1476 is dependent upon RhaK, a sugar kinase. J Bacteriol 189: 8437-8446. 1477 1478 Rivers, D., and Oresnik, I.J. 2013. RhaK dependant ABC-transport of rhamnose in R. 1479 leguminosarum: genetic separation of kinase and transport activities J Bacteriol 195: 3424-1480 3432. 1481 1482 Ronson, C.W., Lyttleton, P., and Robertson, J.G. 1981. C4-dicarboxylate transport mutants of 1483 Rhizobium trifolii form ineffectivie nodules on Trifolium repens. Proc Natl Acad Sci USA 78: 1484 4284-4288. 1485 1486 Ronson, C.W., and Primrose, S.B. 1979. Effect of glucose on polyol metabolism by Rhizobium 1487 trifolii. J Bacteriol 139: 1075-1078. 1488 1489 Sangari, F.J., Agüero, J., and García-Lobo, J.M. 2000. The genes for erythritol catabolism are 1490 organized as an inducible operon in Brucella abortus. Microbioloy 146: 487-495. 1491 1492 Soto, M.J., Sanjuan, J., and Olivares, J. 2001. The disruption of a gene encoding a putative 1493 arylesterase impairs pyruvate dehydrogenase complex activity and nitrogen fixation in 1494 Sinorhizobium meliloti. Mol Plant Microbe Interact 14: 811-815. 1495 1496 Sperry, J.F., and Robertson, D.C. 1975a. Erythritol catabolism by Brucella abortus. J Bacteriol 1497 121: 619-630. 1498 1499 Sperry, J.F., and Robertson, D.C. 1975b. Inhibition of growth by erythritol catabolism in 1500 Brucella abortus. J Bacteriol 124: 391-397. 1501 1502 Steele, T.T., Fowler, C.W., and Griffitts, J.S. 2009. Control of gluconate utilization in 1503 Sinorhizobium meliloti. J Bacteriol 191(4): 1355-1358. 1504 1505 Stowers, M.D. 1985. Carbon metabolism in Rhizobium species. Annu Rev Microbiol 39: 89-1506 108. 1507 1508 Torres-Quesada, O., Millán, V., Nisa-Martínez, R., Bardou, F., Crespi, M., Toro, N., and 1509 Jiménez-Zurdo, J.I. 2013. Independent activity of the homologous small regulatory RNAs 1510 AbcR1 and AbcR2 in the legume symbiont Sinorhizobium meliloti. PloS one 8(7): e68147. 1511 1512 Torres-Quesada, O., Oruezabal, R.I., Peregrina, A., Jofré, E., Lloret, J., Rivilla, R., Toro, N., and 1513 Jiménez-Zurdo, J.I. 2010. The Sinorhizobium meliloti RNA chaperone Hfq influences central 1514 carbon metabolism and the symbiotic interaction with alfalfa. BMC Microbiology 10(1): 71. 1515 1516 Torres-Quesada, O., Reinkensmeier, J., Schlüter, J.-P., Robledo, M., Peregrina, A., Giegerich, 1517 R., Toro, N., Becker, A., and Jiménez-Zurdo, J.I. 2014. Genome-wide profiling of Hfq-binding 1518 RNAs uncovers extensive post-transcriptional rewiring of major stress response and 1519 symbiotic regulons in Sinorhizobium meliloti. RNA biology 11(4): 0--1. 1520
Page 55 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
1521 Triplett, E.W., and Sadowsky, M. 1992. Genetics of competition for nodulation of legumes. 1522 Annu Rev Micorbiol 46: 399-428. 1523 1524 Ucker, D.S., and Signer, S.R. 1978. Catabolite-repression-like phenomenon Rhizobium 1525 meliloti. J Bacteriol 136: 1197-1200. 1526 1527 Udvardi, M., and Poole, P.S. 2013. Transport and metabolism in legume-rhizobia symbioses. 1528 Annu Rev Plant Biol 64: 781-805. 1529 1530 Untiet, V., Karunakaran, R., Krämer, M., Poole, P., Priefer, U., and Prell, J. 2013. ABC 1531 Transport is inactivated by the PTSNtr under potassium limitation in Rhizobium 1532 leguminosarum 3841. PloS one 8(5): e64682. 1533 1534 van Dillewijn, P., Soto, M.a.J., Villadas, P.J., and Toro, N. 2001. Construction and 1535 environmental release of a Sinorhizobium meliloti Strain genetically modified to be more 1536 competitive for alfalfa nodulation. Appl Environ Microbiol 67(9): 3860-3865. 1537 1538 Vanderlinde, E.M., Hynes, M.F., and Yost, C.K. 2013. Homoserine catabolism by Rhizobium 1539
leguminosarum bv. viciae 3841 requires a plasmid‐borne gene cluster that also affects 1540
competitiveness for nodulation. Environmental Microbiology 16(1): 205-217. 1541 1542 Vincent, J.M. 1970. A manual for the practical study of root-nodule bacteria. Blackwell 1543 Scientific Publications, Oxford, England. 1544 1545 Wang, C., Saldanha, M., Sheng, X., Shelswell, K.J., Walsh, K.T., Sobral, B.W., and Charles, T.C. 1546 2007. Roles of poly-3-hydroxybutyrate (PHB) and glycogen in symbiosis of Sinorhizobium 1547 meliloti with Medicago sp. Microbiology 153(2): 388-398. 1548 1549 Watson, R.J., Chan, Y.K., Wheatcroft, R., Yang, A.-F., and S., H. 1988. Rhizobium meliloti genes 1550 required for C4-dicarboxylate transport and symbiotic nitrogen fixation located on a 1551 megaplasmid. J Bacteriol 170: 927-934. 1552 1553 White, C., Gavina, J.M., Morton, R., Britz-McKibbin, P., and Finan, T.M. 2012. Control of 1554 hydroxyproline catabolism in Sinorhizobium meliloti. Mol Microbiol 85: 1133-1147. 1555 1556 Wiebo, J., Marek-Kozaczuk, M., Kubik-Komar, A., and Skorupska, A. 2007. Increased 1557 metabolic potential of Rhizobium spp. is associated with bacterial competitiveness. Can J 1558 Microbiol 53: 957-967. 1559 1560 Willis, L.B., and Walker, G.C. 1998. The phbC (poly-β-hydroxybutyrate synthase) gene of 1561 Rhizobium (Sinorhizobium) meliloti and characterization of phbC mutants. Can J Microbiol 1562 44: 554-564. 1563 1564
Page 56 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Willis, L.B., and Walker, G.C. 1999. A novel Sinorhizobium meliloti operon encodes an α-1565 glucosidase and a periplasmic-binding-protein dependent transport system for α-1566 glucosides. J Bacteriol 181: 4176-4184. 1567 1568 Wood, M., and Stanway, A. 2001. Myo-inositol catabolism by Rhizobium in soil: HPLC and 1569 enzymatic studies. Soil Biol Biochem 33(3): 375-379. 1570 1571 Yarosh, O.K., Charles, T.C., and Finan, T.M. 1989. Analysis of C4-dicarboxylate transport 1572 genes in Rhizobium meliloti. Mol. Microbiol. 3: 813-823. 1573 1574 Yoshida, K.-i., Yamaguchi, M., Morinaga, T., Kinehara, M., Ikeuchi, M., Ashida, H., and Fujita, 1575 Y. 2008. myo-Inositol catabolism in Bacillus subtilis. J Biol Chem 283(16): 10415-10424. 1576 1577 Yost, C.K., Rath, A.M., Noel, T.C., and Hynes, M.F. 2006. Characterization of genes involved in 1578 erythritol catabolism in Rhizobium leguminosarum bv. viciae. Microbiology 152: 2061-1579 2074. 1580 1581
1582
Page 57 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
1583
Figure Legends 1584 Figure 1. Schematic of the S. meliloti ED pathway. Yellow arrows correspond to the 1585
enzymatic reactions of the ED pathway. Enzymes are represented by circled numbers that 1586
correspond to: 1) glucose-6-phosphate dehydrogenase; 2) 6-phosphogluconolactonase; 3) 1587
6-phosphogluconate dehydratase; 4) 2-keto-3-deoxy-6-phosphogluconate aldolase. Bold 1588
auxiliary arrows are used to represent interactions with the EMP pathway (blue), PP 1589
pathway (orange) and the TCA cycle (green). Arrows indicate the direction of carbon flow 1590
as described by the literature. See text for details. 1591
1592
Figure 2. Schematic of the S. meliloti EMP pathway. Blue arrows correspond to the 1593
enzymatic reactions of the EMP pathway. Enzymes are represented by circled numbers 1594
that correspond to: 1) phosphoglucose isomerase; 2) fructose bisphosphatase; 3) fructose-1595
bisphosphate aldolase; 4) triose-phosphate isomerase; 5) glyceraldehyde-3-phosphate 1596
dehydrogenase; 6) phosphoglycerate kinase; 7) phosphoglycerate mutase; 8) enolase; 9) 1597
pyruvate kinase. The red arrow represents the absence of phosphofructokinase in S. 1598
meliloti. Bold auxiliary arrows are used to represent interactions with the ED pathway 1599
(yellow), PP pathway (orange) and the TCA cycle (green). Arrows indicate the direction of 1600
carbon flow as described by the literature. See text for details. 1601
1602
Figure 3: Schematic of S. meliloti PP pathway. Orange arrows correspond to the enzymatic 1603
reactions of the PP pathway. Enzymes are represented by circled numbers that correspond 1604
to: 1) glucose-6-phosphate dehydrogenase; 2) 6-phosphogluconolactonase; 3) 6-1605
phosphogluconate dehydrogenase; 4) ribose-5-phosphate isomerase; 5) ribulose-5-1606
phosphate epimerase; 6) transketolase; 7) transaldolase; 8) transketolase. Bold auxiliary 1607
arrows are used to represent interactions with the ED pathway (yellow), EMP pathway 1608
(blue) and the TCA cycle (green). Arrows indicate the direction of carbon flow as described 1609
by the literature. See text for details. 1610
1611
Figure 4: Schematic of S. meliloti TCA cycle, gluconeogenesis and anaplerotic pathways. . 1612
Green arrows correspond to the enzymatic reactions of the TCA cycle. Purple arrows 1613
correspond to key enzymatic reactions of gluconeogenesis. Red arrows correspond to 1614
enzymatic reactions in anaplerotic pathways. Enzymes are represented by circled numbers 1615
that correspond to: 1) pyruvate dehydrogenase; 2) citrate synthase; 3) aconitase; 4) 1616
isocitrate dehydrogenase; 5) α-ketoglutarate dehydrogenase; 6) succinyl-CoA synthetase; 1617
7) succinate dehydrogenase; 8) fumarase; 9) malate dehydrogenase; 10) malic enzyme; 11) 1618
phosphoenolpyruvate carboxykinase; 12) pyruvate orthophosphate dikinase; 13) pyruvate 1619
carboxylase; 14) isocitrate lyase; 15) malate synthase. Bold auxiliary arrows are used to 1620
represent interactions with the ED pathway (yellow), EMP pathway (blue) and and the PP 1621
Page 58 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
pahway (orange). Arrows indicate the direction of carbon flow as described by the 1622
literature. See text for details. 1623
1624
Figure 5: Schematic of entry-points for carbon sources into S. meliloti central metabolism. 1625
Colored arrows represent the cyclic central metabolism of S. meliloti. Different colors 1626
represent the EMP pathway (blue), ED pathway (yellow), PP pathway (orange), TCA cycle 1627
(green) and gluconeogenic enzymes (purple). Grey arrows indicate entry-points for 1628
different sugars and polyols into central metabolism as defined by the literature. Arrows 1629
indicate the direction of carbon flow as described by the literature. See text for details. 1630
1631
1632 1633
Page 59 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Table 1. Genes and enzymes of central metabolism in S. meliloti
Enzyme (EC number) Gene Sys Id Num Pathway Reference(s)
Phosphoglucose isomerase
(5.3.1.9)
pgi p SMc02163 EMP; GNG Arias et al. 1979
Fructose bisphosphatase
(3.1.3.11)
cbbF P SMc20202 GNG; CBB Arias et al. 1979;
Pickering and
Oresnik, 2008
Fructose-bisphosphate
aldolase (4.1.2.13)
cbbA P
cbbA2 P
fbaB P
SMc20199
SMb21192
SMc03983
EMP; GNG;
CBB
Pickering and
Oresnik, 2008
Triose-phosphate isomerase
(5.3.1.1)
tpiA D
tpiB D
SMc01023
SMc01614
EMP; GNG;
CBB
Poysti and Oresnik,
2007
Glyceraldehyde-3-
phosphate dehydrogenase
(1.2.1.12)
gap D SMc03979 EMP; GNG;
CBB
Finan et al. 1988,
1991
Phosphoglycerate kinase
(2.7.2.3)
pgk D SMc03981 EMP; GNG;
CBB
Finan et al. 1991
Phosphoglycerate mutase
(5.4.2.1)
gpmA P
gpmB P
SMc02838
SMc00006
EMP; GNG;
CBB
Enolase (4.2.1.11) eno D SMc01028 EMP; GNG Finan et al. 1988,
1991
Pyruvate kinase (2.7.1.40) pykA P SMc04005 EMP; GNG Mulley et al. 2010
(R. leguminosarum)
Pyruvate dehydrogenase
complex (1.2.4.1),
(2.3.1.12)
(1.8.1.4)
pdhA D
pdhB D
pdhC D
lpdA1 P
SMc01030
SMc01031
SMc01032
SMc01035
Cabanes et al. 2000;
Soto et al. 2001
Glucose-6-phosphate
dehydrogenase (1.1.1.49)
zwf D SMc03070 ED; PP Cerveñanský and
Arias 1984; Barra et
al. 2003; Willis and
Walker, 1999
6-Phosphogluconolactonase
(3.1.1.31)
pgl P SMc03069 ED; PP Willis and Walker,
1999
6-Phosphogluconate
dehydratase (4.2.1.12)
edd P SMc03068 ED Willis and Walker,
1999
2-Keto-3-deoxy-6-
phosphogluconate aldolase
(4.1.3.16/4.1.2.14),
eda1 P
eda2 P
SMc02043
SMc03153
ED
6-Phosphogluconate
dehydrogenase (1.1.1.44)
gnd P SMc04262 PP Martinez-De Drets
and Arias, 1972,
Irigoyen et al. 1990.
Ribose-5-phosphate
isomerase (5.3.1.6)
rpiA P
rpiB P
PP Poysti and Oresnik,
2007
Ribulose-5-phosphate
epimerase (5.1.3.1)
rpe D
ppe P
SMc00511
SMb20195
PP Pickering and
Oresnik, 2008
Transketolase (2.2.1.1) tkt1 P SMc02342 PP; CBB
Page 60 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
tkt2 P
cbbT P
SMc03978
SMb20200
Transaldolase (2.2.1.2) tal P SMc02495 PP
Phosphoribulokinase
(2.7.1.19)
cbbP P SMc20201 CBB Pickering and
Oresnik, 2008
Ribulose-1,5-bisphosphate
carboxylase (4.1.1.39)
cbbL P
cbbS P
SMc20198
SMc20197
CBB Pickering and
Oresnik, 2008
Citrate synthase (2.3.3.1) gltA D SMc02087 TCA Mortimer et al. 1999
Aconitase (4.2.1.3) acnA D SMc03896 TCA Koziol et al. 2009
Isocitrate dehydrogenase
(1.1.1.42)
icd D SMc00480 TCA McDermott and
Kahn, 1992
α-Ketoglutarate
dehydrogenase complex
(1.2.4.1), (2.3.1.61),
(1.8.1.4)
sucA D
sucB D
lpdA2 P
SMc02482
SMc02483
SMc02487
TCA Duncan and
Fraenkel, 1979;
Dymov et al. 2004,
Soto et al. 2001
Succinyl-CoA synthetase
complex (6.2.1.5),
sucC D
sucD D
SMc02480
SMc02481
TCA Dymov et al. 2004.
Succinic dehydrogenase
complex (1.3.99.1),
sdhA P
sdhB P
sdhC P
sdhD P
SMc02466
SMc02465
SMc02464
SMc02463
TCA Gardiol et al. 1982
Fumarase (4.2.1.2) fumC P SMc00149 TCA
Malate dehydrogenase
(1.1.1.37)
mdh D SMc02479 TCA Dymov et al. 2004
Phosphoenolpyruvate
Carboxykinase (4.1.1.49)
pckA D SMc02562 GNG Østeräs et al. 1995,
1997
NAD+ Malic enzyme
(1.1.1.39)
dme D SMc00169 GNG Driscoll and Finan,
1993, 1997
NADP+
Malic enzyme
(1.1.1.40)
tme D SMc01126 GNG Driscoll and Finan,
1996, 1997
Pyruvate orthophosphate
dikinase (2.7.9.1)
ppdK
(pod) D
SMc00025 GNG Driscoll and Finan
1997, Østeräs et al.
1997
Pyruvate carboxylase
(6.4.1.1)
pyc D SMc03895 TCA Dunn et al. 2001
Isocitrate lyase (4.1.3.1) aceA D SMc00768 TCA Duncan and
Fraenkel, 1979;
Ramírez-Trujillo et
al. 2007
Malate synthase (2.3.3.9) glcB D SMc02581 TCA Duncan and
Fraenkel, 1979;
Ramírez-Trujillo et
al. 2007
Enzymes (Enzyme Commission (EC) number ) are presented that carry out the following
key pathways of central metabolism in S. meliloti: Embden-Meyerhof-Pernas pathway
Page 61 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
(EMP), Entner-Doudoroff pathway (ED), Calvin-Benson-Bassham pathway (CBB),
Gluconeogenesis (GNG), Tricarboxylic acid cycle (TCA). Genes either predicted to
encode these enzymes based on sequence homology (P), or that have been shown to
encode them based on carbon utilization, and or enzyme activity assays (D) are included
with systematic identifier numbers (Sys Id Num).
Page 62 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Schematic of the S. meliloti ED pathway. Yellow arrows correspond to the enzymatic reactions of the ED pathway. Enzymes are represented by circled numbers that correspond to: 1) glucose-6-phosphate
dehydrogenase; 2) 6-phosphogluconolactonase; 3) 6-phosphogluconate dehydratase; 4) 2-keto-3-deoxy-6-phosphogluconate aldolase. Bold auxiliary arrows are used to represent interactions with the EMP pathway (blue), PP pathway (orange) and the TCA cycle (green). Arrows indicate the direction of carbon flow as
described by the literature. See text for details. 158x179mm (300 x 300 DPI)
Page 63 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Schematic of the S. meliloti EMP pathway. Blue arrows correspond to the enzymatic reactions of the EMP pathway. Enzymes are represented by circled numbers that correspond to: 1) phosphoglucose isomerase;
2) fructose bisphosphatase; 3) fructose-bisphosphate aldolase; 4) triose-phosphate isomerase; 5)
glyceraldehyde-3-phosphate dehydrogenase; 6) phosphoglycerate kinase; 7) phosphoglycerate mutase; 8) enolase; 9) pyruvate kinase. The red arrow represents the absence of phosphofructokinase in S.
meliloti. Bold auxiliary arrows are used to represent interactions with the ED pathway (yellow), PP pathway (orange) and the TCA cycle (green). Arrows indicate the direction of carbon flow as described by the
literature. See text for details. 284x416mm (300 x 300 DPI)
Page 64 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Schematic of S. meliloti PP pathway. Orange arrows correspond to the enzymatic reactions of the PP pathway. Enzymes are represented by circled numbers that correspond to: 1) glucose-6-phosphate dehydrogenase; 2) 6-phosphogluconolactonase; 3) 6-phosphogluconate dehydrogenase; 4) ribose-5-
phosphate isomerase; 5) ribulose-5-phosphate epimerase; 6) transketolase; 7) transaldolase; 8) transketolase. Bold auxiliary arrows are used to represent interactions with the ED pathway (yellow), EMP pathway (blue) and the TCA cycle (green). Arrows indicate the direction of carbon flow as described by the
literature. See text for details. 268x325mm (300 x 300 DPI)
Page 65 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Schematic of S. meliloti TCA cycle, gluconeogenesis and anaplerotic pathways. . Green arrows correspond to the enzymatic reactions of the TCA cycle. Purple arrows correspond to key enzymatic reactions of
gluconeogenesis. Red arrows correspond to enzymatic reactions in anaplerotic pathways. Enzymes are
represented by circled numbers that correspond to: 1) pyruvate dehydrogenase; 2) citrate synthase; 3) aconitase; 4) isocitrate dehydrogenase; 5) α-ketoglutarate dehydrogenase; 6) succinyl-CoA synthetase; 7)
succinate dehydrogenase; 8) fumarase; 9) malate dehydrogenase; 10) malic enzyme; 11) phosphoenolpyruvate carboxykinase; 12) pyruvate orthophosphate dikinase; 13) pyruvate carboxylase; 14) isocitrate lyase; 15) malate synthase. Bold auxiliary arrows are used to represent interactions with the ED pathway (yellow), EMP pathway (blue) and and the PP pahway (orange). Arrows indicate the direction of
carbon flow as described by the literature. See text for details. 214x349mm (300 x 300 DPI)
Page 66 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.
Schematic of entry-points for carbon sources into S. meliloti central metabolism. Colored arrows represent the cyclic central metabolism of S. meliloti. Different colors represent the EMP pathway (blue), ED pathway
(yellow), PP pathway (orange), TCA cycle (green) and gluconeogenic enzymes (purple). Grey arrows
indicate entry-points for different sugars and polyols into central metabolism as defined by the literature. Arrows indicate the direction of carbon flow as described by the literature. See text for details.
165x114mm (300 x 300 DPI)
Page 67 of 67C
an. J
. Mic
robi
ol. D
ownl
oade
d fr
om w
ww
.nrc
rese
arch
pres
s.co
m b
y U
nive
rsité
Lav
al B
iblio
theq
ue o
n 07
/14/
14Fo
r pe
rson
al u
se o
nly.
Thi
s Ju
st-I
N m
anus
crip
t is
the
acce
pted
man
uscr
ipt p
rior
to c
opy
editi
ng a
nd p
age
com
posi
tion.
It m
ay d
iffe
r fr
om th
e fi
nal o
ffic
ial v
ersi
on o
f re
cord
.