Journal of Experimental Botany, Vol. 56, No. 412, pp. 577–585, February 2005
doi:10.1093/jxb/eri046 Advance Access publication 20 December, 2004
This paper is available online free of all access charges (see http://jxb.oupjournals.org/open_access.html for further details)
RESEARCH PAPER
Fatty acid synthesis and the oxidative pentose
phosphate pathway in developing embryos of
oilseed rape (Brassica napus L.)
David Hutchings1, Stephen Rawsthorne2,* and Michael J. Emes1,†
1
School of Biological Sciences, University of Manchester, 3.614 Stopford Building, Oxford Street, Manchester M13
9PT, UK
2
Department of Metabolic Biology, John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, UK
Received 31 August 2004; Accepted 6 October 2004
Abstract
The potential role of the plastidial oxidative pentose
phosphate pathway (OPPP) in providing the NADPH for
fatty acid synthesis in plastids from developing embryos
of Brassica napus (L.) has been investigated. Measurements of distributions of enzyme activities in fractions
obtained from homogenates of isolated embryos have
revealed that the glucose 6-phosphate and 6-phosphogluconate dehydrogenases are present in both cytosol
and plastid, as is ribose 5-phosphate isomerase. However, transketolase and transaldolase are most probably
confined to the plastid, while ribulose 5-phosphate
epimerase is essentially cytosolic, although a very small
proportion of plastid-localized activity cannot be ruled
out. The activity of the OPPP in intact plastids was
measured by the release of 14CO2 from [1-14C]glucose 6phosphate. Activity was detectable in the absence of
electron sinks created by the addition of metabolites to
the incubation media and was stimulated 1.3-, 3.2-, and
7.9-fold by the respective additions of glutamine plus 2oxoglutarate, cofactors and substrates for fatty acid
synthesis, or methyl viologen. An increase in OPPP
activity in response to additions that are absolutely
required for fatty acid synthesis in these isolated plastids provides direct evidence that these two processes
are connected, most probably by NADP/NADPH metabolism. The OPPP activity with methyl viologen was
more than twice that during fatty acid synthesis, suggesting that the latter is not limited by OPPP capacity.
Light energy may also contribute to reductant provision
and, consistent with the possibility of maintenance of
a balance of NADPH from light and the OPPP, glucose
6-phosphate dehydrogenase activity in the isolated
plastids was decreased by light or by DTT.
Key words: Brassica napus, embryo, fatty acid synthesis,
oxidative pentose phosphate pathway, plastid.
Introduction
In oil seed crops, seeds store a large proportion of their
reserves of energy and reduced carbon as lipid. The production of fatty acids in the plastids of plants requires a supply of reducing power which could, in non-photosynthetic
plastids, be supplied by the oxidative pentose phosphate
pathway (OPPP) and/or glycolysis and the pyruvate dehydrogenase complex (PDC). Plastids from embryos of
Brassica napus (L.) have a full complement of glycolytic enzymes and a PDC (Kang and Rawsthorne, 1994;
Eastmond and Rawsthorne, 2000) as well as possessing
isozymes of the oxidative reactions of the OPPP catalysed
by the glucose 6-phosphate- and 6-phosphogluconate dehydrogenases (Eastmond and Rawsthorne, 2000). Studies
with plastids isolated from B. napus embryos at different
stages of development have shown that in the pre- and
early-oil embryo stages (defined as A and B, at approximately 1.5 mg and 2.5 mg FW per embryo, respectively;
Eastmond and Rawsthorne, 2000), glucose 6-phosphate
(Glc6P) supports the highest rate of fatty acid synthesis
in vitro, when compared with triose phosphate, malate,
pyruvate, and acetate. During the main period of oil synthesis (stage C, 3.5 mg FW per embryo), pyruvate supplied
* To whom correspondence should be addressed. Fax: +44 (0)1603 450014. E-mail: [email protected]
y
Present address: College of Biological Sciences, University of Guelph, Guelph, Ontario, N1G 2W1 Canada.
Journal of Experimental Botany, Vol. 56, No. 412, ª Society for Experimental Biology 2004; all rights reserved
578 Hutchings et al.
in vitro to embryo plastids sustains higher rates of fatty acid
synthesis than Glc6P, and this is accompanied by the
induction of a specific pyruvate transporter (Eastmond and
Rawsthorne, 2000). When the activity of the OPPP was
measured as the evolution of 14CO2 from labelled [1-14C]
Glc6P in the same plastid preparations, the amount and
proportion of Glc6P metabolized through the OPPP also
increased through to stage C (Eastmond and Rawsthorne,
2000). The oxidation of Glc6P may, therefore, serve two
functions depending on the stage of development. First,
acetyl units can be generated as a result of glycolytic flux
coupled to the activity of a plastidial PDC. Second, reducing power may be generated as a result of glycolysis and
PDC activity as NADH, or by metabolism through the
OPPP which generates NADPH.
In plants, the best characterized linkage of OPPP activity
to reductant utilization in the plastid is that of reductant
supply for nitrite reduction and glutamate synthesis in nongreen plastids (Bowsher et al., 1989, 1992) where OPPP
activity is coupled to the reduction of a non-photosynthetic
ferredoxin. Indirect evidence for a linkage between reductant provision from the OPPP and fatty acid synthesis in
heterotrophic tissues of plants can be drawn from the effect
of Glc6P metabolism on pyruvate-dependent fatty acid
synthesis by plastids isolated from B. napus embryos.
Metabolism of Glc6P through the OPPP increases the rate
of fatty acid synthesis from pyruvate, and when import of
Glc6P is reduced by inhibition of the Glc6P translocator,
this stimulus is removed (Eastmond and Rawsthorne, 2000;
Johnson et al., 2000; Fox et al., 2000). More direct
evidence can be drawn from recent experiments in which
13
C-labelled sugars were supplied in vitro to isolated whole
B. napus embryos (Schwender and Ohlrogge, 2002;
Schwender et al., 2003). These experiments have shown
that the OPPP contributes an estimated 38% of the reducing
power required for fatty acid synthesis under the imposed
conditions, with the remainder derived through the photosynthetic activity of the plastids in these embryos
(Schwender et al., 2003). The supply of reducing equivalents for fatty acid synthesis from the OPPP relative to that
from photosynthetic electron transport in the developing
B. napus embryo could be influenced by (i) the sensitivity
of NADP-dependent Glc6P dehydrogenase (G6PDH) enzymes to the redox balance of the NADPH/NADP+ pool,
and (ii) by a reversible ferredoxin/thioredoxin-dependent
inactivation of G6PDH through dithiol-disulphide interconversion of regulatory cysteines that can be mimicked
by DTT in vitro (Kruger and von Schaewen, 2003). The
isoform of G6PDH that is present in chloroplasts has
much greater sensitivity to these modulations than the isoforms that predominate in plastids of heterotrophic tissues
(Wenderoth et al., 1997; Wendt et al., 2000).
As a result of recent studies there has been a reappraisal
of the subcellular localization of the OPPP (Kruger and von
Schaewen, 2003). Whereas previously it had been thought
that the entire sequence of reactions operated in both
cytosol and plastids, there is increasing evidence based on
studies with several species that some of the non-oxidative
enzymes are restricted to plastids. Initially, Schnarrenberger
et al. (1995) showed that, in spinach leaves, only the dehydrogenases could be detected in the cytosol, and that
none of the other steps could be detected outside of the
plastids. Subsequent cellular fractionation studies with
leaves and roots of maize, pea, and spinach provided
evidence that the non-oxidative reactions were confined to
the plastids, although in tobacco roots and leaves, ribulose
5-phosphate 3-epimerase (RPE), ribose 5-phosphate isomerase (RPI), transketolase (TK), and transaldolase (TA)
activities could not be accounted for as solely plastidial
(Debnam and Emes, 1999). A more recent study has indicated that the vast majority of TK in tobacco leaves is
associated with the plastid (Henkes et al., 2001). A recent
bioinformatics analysis, based on homology searching and
investigation of putative transit peptide sequences, indicated that Arabidopsis thaliana possesses genes coding for
cytosolic and plastidial forms of RPI and RPE, but that TK
and TA were likely to be plastidial only (Eicks et al., 2002).
While informative, the latter approach requires biochemical
data to corroborate the findings based on genome mining.
Although it has been proposed that activity of the OPPP
is linked to that of fatty acid synthesis in developing B.
napus embryos, the evidence to date is indirect. Moreover,
the distribution of OPPP enzyme activities between the
cytosol and plastid can vary between species and is not
understood in B. napus, although the Arabidopsis genome
provides important clues. Understanding their distribution
is vital to understanding their function in vivo in the
developing embryo. In addition, the relationship between
the demand for reductant by fatty acid synthesis and its
provision by the OPPP and light, and how light could affect
OPPP activity through redox regulation is not fully understood. To address these points for B. napus embryos, (i) the
subcellular localization of all the reactions of the OPPP has
been investigated, (ii) a detailed study has been carried out
of the ability of the OPPP to support fatty acid synthesis
in vitro, and (iii) a preliminary study of the redox sensitivity
of the plastidial NADP-G6PDH has been made.
Materials and methods
Plant growth, chemicals and reagents, and preparation of
plastids
Oilseed rape (Brassica napus L.) cultivar Topas was grown in
a glasshouse in John Innes number 3 compost, with supplementary
illumination from sodium lamps giving a 16 h day length at average
day/night temperatures of 20–25 8C and 10–15 8C, respectively, and
Phostrogen fertilizer fed as recommended by the manufacturer (pbi
Home & Garden Limited, Waltham Cross, Herts, UK).
Substrates, cofactors, and coupling enzymes were purchased from
either Sigma Chemical Company (Poole, Dorset, UK) or Boehringer
Mannheim (Lewes, Sussex, UK). Radioisotopes were from Amersham
Reductant provision for fatty acid synthesis
International (Amersham, Bucks. UK) and NEN DuPont (Herts, UK).
Harvesting of embryos from siliques and purification of embryo
plastids from early-oil-stage embryos was as described previously
(Kang and Rawsthorne, 1994). Preparation of chloroplasts from
leaves was as described by Eastmond et al. (1996).
Enzyme assays
All enzymes were assayed spectrophotometrically at 340 nm and
25 8C.
Glucose 6-phosphate dehydrogenase (EC 1.1.1.49) was assayed
without 5 mM DTT in the initial extraction medium. The reaction
contained 50 mM HEPES/KOH (pH 7.7), 10 mM MgCl2, 0.3 mM
NADP+, 5 mM Glc6P, and 0.12 units 6-phosphogluconate dehydrogenase. The assay for 6-phosphogluconate dehydrogenase (EC
1.1.1.44) was the same, except that 5 mM 6-phosphogluconate replaced glucose 6-phosphate and coupling enzyme was not included.
Ribulose 5-phosphate 3-epimerase (EC 5.1.3.1) was assayed by
two methods. In both assays the substrate, ribulose 5-phosphate, is
generated in the assay, as commercial preparations of ribulose 5phosphate were found to contain large quantities of the reaction
product, xylulose 5-phosphate. This contamination leads to a measureable rate that is independent of RPE when TK acts upon this
xylulose 5-phosphate and the added ribose 5-phosphate. Assay A
included, 50 mM HEPES/KOH (pH 7.8), 5 mM MgCl2, 0.5 mM
cocarboxylase, 0.2 mM NADH, 0.5 mM erythrose 4-phosphate,
0.5 mM ribose 5-phosphate, 10 units ribose 5-phosphate isomerase,
0.5 units transketolase, 10 units triose phosphate isomerase, and 1 unit
glycerol 3-phosphate dehydrogenase. In assay B, commercial ribose
5-phosphate isomerase was omitted, ribulose 5-phosphate being generated by endogenous ribose 5-phosphate isomerase.
Ribose 5-phosphate isomerase (EC 5.3.1.6), transaldolase (EC
2.2.1.2), and transketolase (EC 2.2.1.1) were assayed according to
Schnarrenberger et al. (1995). Activities of marker enzymes were
assayed as follows; NADP-glyceraldehyde 3-phosphate dehydrogenase (EC 1.2.1.13) according to Denyer and Smith (1988), UDPglucose pyrophosphorylase (EC 2.7.7.9) as in Tetlow et al. (1993),
and pyrophosphate:fructose 6-phosphate 1-phosphotransferase (EC
2.7.1.90) as in ap Rees et al. (1985).
Metabolite feeding to isolated plastids
Feeding of 14C-labelled metabolites, determination of incorporation
of carbon into fatty acids, and evolution of CO2 was carried out according to Kang and Rawsthorne (1996). [6-14C] Glc6P was prepared by incubating 740 kBq [6-14C]-glucose, 10 mM ATP, 10 mM
MgCl2, and 2 units of hexokinase for 2 h at 25 8C. Samples were
boiled for 5 min and denatured protein removed by centrifugation.
Further purification and verification of the product was achieved by
thin layer chromatography (Sigma Cell Type 100 polyester on
cellulose) in methanol:25% (v/v) ammonia:water (7:1:2 by vol.).
Bands containing [6-14C] Glc6P were identified using a standard of
[U-14C] Glc6P. [6-14C] Glc6P was removed by scraping the bands
from the TLC plate, washing and vortexing in 0.5 ml distilled water,
and removal of the polyester by centrifugation. The specific activity
of [6-14C] Glc6P prepared this way was approximately 2 kBq mol1.
Incubation of plastids and assay of Glc6P dehydrogenase
activity in the presence of thiols or illumination
Embryo plastids or leaf chloroplasts (Eastmond et al., 1996) were
prepared as described earlier, except all procedures from the
homogenization step onwards were carried out in green light or the
total absence of light, and DTT was omitted from the plastid isolation
medium (Kang and Rawsthorne, 1994). Purified plastids or chloroplasts were resuspended inside a glass centrifuge tube and were
illuminated, with three 250 W tungsten lamps, at a photon flux
2
579
1
density of approximately 300–500 lmol m s , at 25 8C. The
centrifuge tubes were rotated mechanically and gently. A control set
of plastids was covered in aluminium foil to maintain darkness. After
known periods of time, samples of illuminated and non-illuminated
plastids were removed and the activity of glucose 6-phosphate
dehydrogenase determined.
Alternatively, embryo plastids and leaf chloroplasts were prepared
as described, but without illumination. Samples of intact plastids
were then supplied with the oxidizing agents sodium tetrathionate
(5 mM) or hydrogen peroxide (10 mM) (Farr et al., 1994), or with the
reducing agent DTT (5 mM) (Johnson, 1972; Farr et al., 1994) for
known periods of time. Samples of plastids were removed and the
activity of glucose 6-phosphate dehydrogenase determined.
Results
In order to study the subcellular distribution of the OPPP
enzymes, the distribution of their activity between plastidenriched and supernatant fractions of homogenates prepared
from embryos at the early-oil stage of development was
measured. The total recovery of activity of each enzyme with
respect to the activity in crude homogenates was also
measured and was between 90% and 115% (data not shown).
Based on the marker enzymes, approximately 10% of the
plastids were recovered in the final, washed plastid preparation with up to 0.5% contamination from cytosolic marker
enzymes (Table 1). The intactness of the final plastid
preparation was 60.462.0% (mean 6SE, n = 42) as determined by latency of NADP-GAPDH activity, a value in
keeping with previous reports (Kang and Rawsthorne,
1994). Based on the activities recovered in the washed
plastid pellet, with one exception, all the enzymes of the
OPPP were found to be present to some extent within
plastids (Table 1). The dehydrogenases have previously
been shown to be present in B. napus embryo plastids. Their
activities and subcellular distribution are similar to that seen
in other species/organs in which there is proportionately
more of the total 6PGDH activity in the plastids than for
G6PDH (cf. data in Debnam and Emes, 1999), but that the
greater part of each total activity is cytosolic. The proportions of transaldolase and transketolase activities (6.9% and
6.2%, respectively) that were present in the washed plastid
fraction were not significantly different from that of the
plastid marker, consistent with the proposal that their major
location is plastidial. Interestingly, the enzymes involved
in the interconversion of pentose phosphates, ribose 5phosphate isomerase (RPI), and ribulose 5-phosphate 3epimerase (RPE) show markedly different distribution
profiles. Whereas approximately half of the RPI activity
was estimated to be plastidial, the distribution of RPE
activity (as measured by either of the assays used) was not
significantly different from that of the cytosolic marker
enzymes, UGPase and PFP. Given that overall recovery of
RPE in all fractions (data not shown) was 97%, there is no
indication that the lack of RPE activity in plastids is due to
loss of enzyme activity during the fractionation process.
580 Hutchings et al.
Table 1. Subcellular localization of each enzyme of the oxidative pentose phosphate pathway
The measured activities of each enzyme in the washed plastid pellet (P) and the proportion of activity that this represents with respect to the total activity
in the homogenate (H) are given. Values are the mean 6SE of measurements made on three independent plastid preparations. The proportion of each
enzyme activity that is plastidial in whole extracts is calculated as ((% activity of enzyme X in P–average % cytosolic marker activity in P)/(% plastid
marker in P–average % cytosolic marker activity in P))3100 (Thorbjørnsen et al., 1996).
Enzyme
Marker enzymes
NADP-GAPDH
UDP-glucose pyrophosphorylase
Pyrophosphate: fructose
6-phosphate 1-phosphotransferase
OPPP enzymes
Glucose 6-phosphate dehydrogenase
6-Phosphogluconate dehydrogenase
Ribulose 5-phosphate epimerase
Ribose 5-phosphate isomerase
Transaldolase
Transketolase
a
b
Enzyme activity
in homogenate
(H) (lmol min1)
Enzyme activity
in fraction
P/activity in H (%)
Proportion of total
activity that is
plastidial (%)
Actual/deduced
subcellular location
1.4260.99
17.2162.61
4.1660.39
9.461.2
0.460.1
0.560.1
100
0
0
Plastid
Cytosol
Cytosol
2.5760.29
3.1360.37
18.9163.59
5.8861.83
0.7760.08
3.1060.27
1.860.4
3.260.4
0.360.2a
4.560.9
6.961.3b
6.261.1b
15
31
0
45
72
64
Cytosol > Plastid
Cytosol > Plastid
Cytosol
Plastid Cytosol
Plastid
Plastid
Indicates value that does not differ from those of the cytosol markers (P > 0.05).
Indicates values that do not differ from that of the plastid marker (P > 0.05).
Having investigated the localization of the OPPP reactions, the metabolism of Glc6P by this pathway was then
studied in intact organelles in the presence of a range of
substrates whose metabolism would require generation of
reductant. In these experiments, flux through the OPPP was
measured as evolution of 14CO2 from [1-14C]Glc6P, a technique which has been used previously to establish the link
between the plastidial OPPP and pathways which require
reducing power (Bowsher et al., 1992). In all cases,
controls were included in which osmotically lysed plastids
were used in place of intact organelles. All the values
presented are net of control values and all values are
expressed per unit of the plastidial marker enzyme, NADPGAPDH. The evolution of 14CO2 was measured under four
sets of conditions. Glc6P was provided alone or in
combination with either: glutamine plus 2-oxoglutarate,
substrates for the ferredoxin-dependent glutamate synthase
reaction; HCO
3 , ATP, coenzyme A, and thiamine pyrophosphate, substrates and cofactors for fatty acid synthesis;
or methyl viologen, a herbicide which also acts as an
electron acceptor from the OPPP (Tetlow et al., 1994).
Figure 1 shows that flux through the OPPP was linear with
time up to 60 min and was measurable even when plastids
were supplied only with Glc6P, although the rate of
evolution of 14CO2 was lowest in this case. There was
a modest, but significant (P <0.05 after 60 min) stimulation
of 14CO2 evolution when glutamine and 2-oxoglutarate
were supplied simultaneously. A greater than 3-fold stimulation was seen in the presence of the substrates and
cofactors required for fatty acid synthesis. Previous experiments have shown that, in the absence of these cofactors
and substrates, plastids isolated from B. napus embryos are
unable to carry out fatty acid synthesis (Kang, 1995). In the
present experiment, the carbon skeletons for fatty acid
Fig. 1. Oxidation of Glc6P by B. napus embryo plastids in the presence
of various substrates. Time-dependent evolution of 14CO2 from plastids
supplied with 5 mM [1-14C]Glc6P (1.48 kBq mol1) alone; and in the
presence of 5 mM glutamine plus 5 mM 2-oxoglutarate; 10 mM
NaHCO3, 7 mM MgCl2, 4 mM ATP, 0.2 mM thiamine pyrophosphate,
0.2 mM coenzyme A; or 5 mM methyl viologen. The data are corrected
for 14CO2 evolution from osmotically-lysed plastids and are expressed
relative to the maximum value with methyl viologen (1.1260.07 lmol
CO2 unit1 GAPDH). Values are the mean 6SE of measurements made
with at least three separate plastid preparations.
synthesis would have been supplied by the oxidation of
Glc6P by plastidial glycolysis, since no other carbon
precursor was present. However, by far the largest stimulation of OPPP flux was seen when methyl viologen was
supplied to the organelles where the 14CO2 released after
60 min was almost 8-fold greater than when Glc6P was
supplied alone.
These interactions were studied in more detail to determine the degree to which OPPP flux would be limited by
the supply of Glc6P as opposed to the sink for reductant.
Figure 2 demonstrates the increase in CO2 evolution as
a function of Glc6P concentration, in the absence or
Reductant provision for fatty acid synthesis
581
Table 2. Synthesis of fatty acids and evolution of CO2 by
isolated plastids
Isolated plastids were incubated with 5 mM [1-14C], [U-14C], [6-14C]
glucose 6-phosphate or [6-14C] glucose (1.48 kBq mol1) in an
incubation medium containing cofactors for fatty acid synthesis and
5 mM DTT. Incorporation of 14C into fatty acids and the evolution of
14
CO2 were determined during 1 h incubation. Values are corrected for
14
C incorporation into fatty acids or 14CO2 evolution by lysed plastids
and are the mean 6standard error of at least three independent plastid
preparations or the mean of two preparations (asterisk). ND, not
determined.
Measurement
14
14
[1- C]
Fig. 2. Effect of Glc6P concentration on CO2 evolution from B. napus
embryo plastids supplied with various substrates. Evolution of 14CO2 was
measured from plastids supplied with [1-14C]Glc6P (1.48 kBq mol1)
alone, and with previously described metabolites (see Fig. 1 for details)
over a 60 min period. The data are corrected for 14CO2 evolution from
osmotically-lysed plastids and are expressed relative to the maximum
value with methyl viologen (1.4360.09 lmol CO2 unit1 GAPDH).
Values are the mean 6SE of measurements made with at least three
separate plastid preparations.
presence of methyl viologen or the substrates and cofactors
for fatty acid synthesis. The relationship between the three
conditions remains the same at all concentrations of Glc6P,
with methyl viologen providing by far the greatest stimulation. The data presented in Fig. 2 also allow a comparison
of the apparent affinity for Glc6P when it is being
metabolized solely by the OPPP (e.g. when methyl viologen acts as a terminal electron acceptor) or when there is
an expected demand for reductant through fatty acid synthesis. In both cases, the calculated apparent Km for Glc6P
was approximately 1 mM. This is far higher than the Km of
130 lM observed for the embryo plastid Glc6P transporter
(Eastmond and Rawsthorne, 1998).
The relationship between fatty acid synthesis and OPPP
flux when Glc6P was the sole source of acetyl units was
investigated by simultaneously measuring the incorporation
of 14C into fatty acids and the release of 14CO2 (Table 2).
The relative rates of carbon flux into fatty acids and into CO2
were 2.6:1 (mol:mol) for plastids supplied with [U-14C]
Glc6P and 0.8:1 for those supplied with [1-14C]Glc6P. The
latter ratio was very similar to that shown previously for
plastids isolated from B. napus embryos at the same developmental stage (Eastmond and Rawsthorne, 2000). The
amount of carbon incorporated into fatty acids from all
atoms of the hexose molecule when plastids were fed
[U-14C]Glc6P was almost 8-fold greater than that from the
C-1 position when [1-14C]Glc6P was supplied (Table 2).
This value is larger than the expected 4-fold increase based
upon the difference in conservation of 14C in the actetate that
is derived, via glycolysis and decarboxylation of pyruvate by
PDC, from 1- or U-labelled Glc6P. Almost 2.5 times as
much 14CO2 was evolved from [U-14C]Glc6P as from
[1-14C]Glc6P (Table 2). While release of 14CO2 from the
14
C-glucose 6-phosphate
14
[U- C]
C-glucose
14
[6- C] [6-14C]
120632 9496146 ND
Fatty acid synthesis
(nmol C unit1 GAPDH)
Evolution of CO2
152662 367622 4*
(nmol C unit1 GAPDH)
ND
16*
C-l position of Glc6P is diagnostic of flux through the OPPP,
a greater release of 14CO2 would be expected with the Ulabelled Glc6P as CO2 is released from carbon atoms 3 and 4
during fatty acid synthesis due to glycolysis/PDC activity as
described above. Release of CO2 from carbon atom 6 on the
other hand would be suggestive of recycling of carbon via
glycolysis/gluconeogenesis. When plastids were supplied
with [6-14C]Glc6P, or with [6-14C]Glc added as a control for
any contamination of the in vitro-synthesized [6-14C]Glc6P
by [6-14C]Glc, in the presence of the cofactors necessary for
fatty acid synthesis, the evolution of 14CO2 was negligible
(Table 2). This indicates that no recycling through glycolysis/gluconeogenesis occurs.
To determine whether light energy and the redox status
could influence the activity of NADP-G6PDH in B. napus
embryo plastids, and therefore the entry of carbon into the
oxidative reactions of the pentose phosphate pathway, the
effect of incubating isolated, intact plastids in the light or
dark on extractable enzyme activity was studied. When
embryo plastids were incubated in darkness, G6PDH
activity in lysates from them was not significantly changed
over 90 min (Fig. 3). However, when plastids were incubated in the light approximately 15% of the activity was
lost within 2 min, followed by a slower loss, to give less
than 50% of the starting activity after 90 min (Fig. 3).
Addition of 5 mM DTT to plastid incubations in the dark
led to a 20% loss of G6PDH activity within 1 min and this
activity was unchanged over a 10 min period (Fig. 4). This
DTT-dependent loss of activity closely resembled that
caused by light (Fig. 4). To provide a comparison of the
sensitivity of the embryo plastid G6PDH activity with that
of the chloroplast enzyme, chloroplasts isolated from
B. napus leaves were incubated in the presence or absence
of 5 mM DTT. The chloroplast enzyme was much more
sensitive to plastid redox change caused by DTT, and more
than 60% of its activity in the dark was lost within 1 min of
incubation in DTT (data not shown).
582 Hutchings et al.
Fig. 3. Effect of light on the activity of glucose 6-phosphate dehydrogenase in isolated plastids. Plastids were incubated in isolation medium
in the light or dark, lysed and the enzyme activity was expressed relative
to that at zero time (45.362.4 nmol min1 ml1 plastid preparation).
Fig. 4. Effect of DTT on the activity of glucose 6-phosphate dehydrogenase in isolated plastids. Plastids were incubated in the dark in isolation
medium with or without 5 mM DTT, lysed and the enzyme activity was
expressed relative to that at zero time (43.662.4 nmol min1 ml1 plastid
preparation).
Discussion
These studies of the subcellular localization of all of the
OPPP enzymes in the developing embryos of Brassica
napus have added to other studies that have questioned
a dual localization of the complete OPPP in both the cytosol
and the plastid (Schnarrenberger et al., 1995; Debnam and
Emes, 1999; Eicks et al., 2002). For example, biochemical
evidence for the confinement of the TK and TA activities to
the plastid fraction is provided here. This observation for
B. napus supports the predictions from the genome sequence
of the closely related species Arabidopsis that the TK and
TA genes encode plastid isoforms (Eicks et al., 2002). The
ratio of TK to TA activity in the isolated B. napus embryo
plastids was 3.6:1 reflecting a more chloroplast-like profile
than that of a plastid from a heterotrophic tissue. Thom
et al. (1998) have demonstrated that the ratio of TK to TA
activity declines from 3.1:1 in chloroplasts of pepper fruits
to 0.09:1 in the chromoplasts as the fruits mature, a change
caused by decreased TK and increased TA activities.
While TA and TK are plastid localized, it is predicted
that very little, if any, RPE activity is present in the plastids
isolated from B. napus embryos (Table 1), despite the
presence of genes encoding cytosolic and plastidial isoforms of both RPE and RPI in Arabidopsis (Eicks et al.,
2002). Although cDNAs for both plastidic and cytosolic
isoforms of RPE have been identified (Kopriva et al.,
2000), there is very limited information on the expression
patterns of the genes that encode them across a range of
organs and tissues. Two assays were used to determine RPE
activity. They gave identical results and 97% of the total
activity in the homogenate was recovered in all fractions, so
it is likely that the results are robust. Although the
percentage recovery of RPE activity in the plastid fraction
was no higher than that of contaminating cytosolic enzymes, which implies a cytosolic localization of RPE
activity, the total RPE activity in homogenates was, on
average, an order of magnitude greater than that of the other
OPPP enzymes (Table 1). If, as the data imply, this activity
is mostly cytosolic, it would be difficult to detect a small
proportion of the total RPE that was within the plastids
because of cytosolic contamination. Therefore, although
the large majority of RPE is localized in the cytosol, the
possibility that there is RPE activity in the plastid cannot be
ruled out. A low activity (or an absence) of plastid RPE
relative to that of other plastid OPPP enzymes, would mean
that further metabolism of Ru5P via the non-oxidative
OPPP reactions within plastids could occur only after its
export to the cytosol, conversion to X5P by the cytosolic
RPE, and import of the X5P back into the plastid. Ensuring
export of Ru5P to the cytosol would allow a balance
between the production of Xu5P for re-entry into the plastid
and the cytosolic synthesis of R5P. This model of partitioning of pentose phosphates for different purposes is
reinforced if TK and TA are confined to the plastid. Eicks
et al. (2002) have identified a pentose phosphate translocator with a high affinity for xylulose 5-phosphate, but
which is also able to transport ribulose 5-phosphate as well
as triose phosphates. This translocator would enable metabolic interaction between cytosolic oxidative and plastidial
non-oxidative reactions of the OPPP.
Whilst the dehydrogenases of the OPPP are generally
accepted as having a dual location in both cytosol and plastids,
as reported here for B. napus embryos, there seems to be no
concensus-model for the localization of the non-oxidative
steps. This disparity may reflect genuine species differences
and also the possibility that, within a species, there is tissuespecific expression of particular isozymes in different subcellular locations. In leaves and roots of spinach, pea, and
maize, the non-oxidative steps in roots and leaves are largely,
if not entirely, plastid localized (Schnarrenberger et al., 1995;
Debnam and Emes, 1999; Henkes et al., 2001), while in the
same organs of tobacco, distribution of activities suggest
Reductant provision for fatty acid synthesis
a cytosolic and plastidial location (Debnam and Emes, 1999).
The reasons for these different distributions of OPPP enzymes
remain to be elucidated.
As the major function of the OPPP in plastids is likely to
be the generation of reducing power for biosynthesis, the
ability of different metabolic sinks for electrons to stimulate
flux through the pathway were examined. The first point to
note is that it appears that even when Glc6P is supplied
without any additional supplement to intact plastids, it can
be metabolized via the OPPP (Fig. 1). This was achieved
consistently and at rates well above any background decarboxylation of Glc6P arising from the presence of
contaminating G6PDH and 6PGDH. It must be concluded
therefore that metabolic sinks for reductant are retained
within the preparations allowing continuous turnover of
NADP and NADPH. The small stimulation of OPPP flux
by the addition of glutamine and 2-oxoglutarate suggests
that there is a functional glutamate synthase reaction within
the plastids of B. napus embryos. A much greater stimulation of the OPPP was seen when cofactors for fatty acid
synthesis were supplied, enabling the utilization of Glc6P
as the source of acetyl units as well as for the generation of
reductant. Previous studies (Eastmond and Rawsthorne,
2000) have shown that (i) Glc6P is the cytosolic precursor
best able to supply carbon skeletons for fatty acid synthesis
by plastids isolated from embryos at the same developmental stage used in this present study and (ii) that fatty
acid synthesis by these plastids absolutely requires the
added cofactors and substrates (Kang, 1995). An increase in
flux through the OPPP in response to the initiation of fatty
acid synthesis provides direct evidence for a link between
the two pathways, most likely explained by coupling
through NADP/NADPH metabolism.
When a non-physiological electron acceptor, methyl
viologen, was provided to the isolated plastids metabolizing
Glc6P, the stimulation of the OPPP was more than 2-fold
greater than when the substrates for fatty acid synthesis
were supplied. This suggests that there is ample capacity for
increasing the flux through the OPPP and that the generation of reductant via this route is unlikely to be limiting
with respect to fatty acid synthesis. The apparent affinity for
Glc6P of Glc6P-dependent fatty acid synthesis and Glc6Pdependent methyl viologen reduction (Km of 1 mM) in
embryo plastids is close to the estimated concentration of
Glc6P in whole embryos (Eastmond and Rawsthorne,
2000). As the Km for the Glc6P transporter is approximately
0.1 mM, and the Km for most isoforms of plastidial G6PDH
is in the region of 0.5–1.5 mM, the apparent Km values
more likely reflect the ability of the OPPP to utilize
substrate once inside the organelle, rather than a limitation
by the transporter itself which is probably saturated in vivo.
These observations suggest that generation of acetyl units
and their subsequent utilization is more of a limitation to the
production of fatty acids than the uptake of Glc6P or its
oxidation in the OPPP.
583
The relationship between fatty acid synthesis and the
activity of the OPPP was addressed in greater detail by
supplying 1-14C and U-14C-Glc6P to isolated plastids and
measuring the 14C incorporated into fatty acids and released
as 14CO2. The amount of 14C incorporated into fatty acids
from [U-14C]Glc6P was much greater (almost 8-fold) than
that from [1-14C]Glc6P. This is a rather higher ratio than
would be expected if the following empirical equation for
fatty acid synthesis is considered:
4Glc6P + 7NADPH + 9NAD + 5ADP + Pi + H +
! 16C + 7NADP + + 9NADH + 5ATP + 14H2 O + 8CO2
Given that CO2 in this equation arises from carbon atoms
3 and 4, as a result of pyruvate dehydrogenase activity, this
would imply that for every 16 carbon atoms synthesized as
fatty acids (palmitate), 4 would arise from carbon atom 1.
Therefore, when feeding [U-14C]Glc6P it would be expected that the incorporation of labelled carbon into fatty
acid would be approximately 4-fold that from [1-14C]
Glc6P, not 8-fold. However, this prediction is based on the
assumptions that the only carbon entering fatty acid
synthesis is derived from the supplied, labelled source
and that there is no recycling of carbon through the OPPP.
For several reasons it is difficult to make accurate predictions of the stoichiometry of 14C incorporation into CO2
and fatty acids. First, the actual contribution that the OPPP
makes to NADPH provision for fatty acid synthesis cannot
be certain because NADP-GAPDH is also present in the
plastid. Second, there are sinks for reductant in the plastids
apart from fatty acid synthesis (Fig. 1) which could influence the evolution of CO2 independently from that
coupled to fatty acid synthesis. Third, the [1-14C]Glc6P
that enters the OPPP may be diluted by recycling through
the non-oxidative reactions. Although B. napus embryo
plastids have photosynthetic properties (Eastmond et al.,
1996), the possibility that photosynthetic CO2 fixation may
also have affected the measured CO2 release and subsequent calculations can be excluded, as incubations were
carried out in the dark. Given that the observed stoichiometries of fatty acid synthesis and CO2 release are not
simple to interpret, it was concluded that metabolism in the
isolated plastids is more complex than the above equation
implies, and that more detailed flux measurements using,
for example, 13C-labelled carbon sources as described
by Schwender and colleagues (Schwender and Ohlrogge,
2002; Schwender et al., 2003) would be required. The
linkage between fatty acid synthesis and the OPPP that is
reported here supports the studies of Schwender and
colleagues, who incubated isolated B. napus embryos with
13
C-sugars under conditions designed to simulate the
in vivo environment. They concluded from their flux
analyses that the OPPP does provide a source of NADPH
for fatty acid synthesis, contributing an estimated 38% of
the total required (with a confidence range of 22–45%), and
584 Hutchings et al.
that it does not make a major contribution to carbon flux to
fatty acids.
A problem in interpreting either of the experimental
approaches described above, in the context of in vivo
metabolism, is that neither adequately addresses the contribution that light energy makes to reductant and ATP
provision in vivo. Relatively short-term studies (1 h) with
isolated plastids have shown that light does not stimulate
the rate of incorporation of carbon from Glc6P (at a saturating concentration) into fatty acids when exogenous ATP
is provided (Eastmond and Rawsthorne 1998). This implies
that the capacity of the OPPP is sufficient, under the
incubation conditions, for the provision of enough reductant to sustain in vitro rates of fatty acid synthesis
comparable with those in vivo. In the longer term (15 d)
experiments with whole embryos, light with a photosynthetic photon flux density (PPFD) of 50–100 lmol m2 s1
was provided through fluorescent tubes with a 16 h daylength (Schwender and Ohlrogge, 2002) or continuously
throughout the experiment (Schwender et al., 2003). These
low-light incubation conditions were used to allow for
the fact that ;80% of the incident light is absorbed by the
silique wall in vivo (Eastmond et al., 1996). However, the
light that an embryo receives in vivo is attenuated green
light with a much lower efficiency for driving photosynthetic electron transport than an equivalent PPFD of white
light. It is therefore possible that the experiments of
Schwender and colleagues (Schwender and Ohlrogge,
2002; Schwender et al., 2003) may well have provided
a PPFD, of a light quality and over an extended duration,
that is greater than that normally experienced in vivo,
thereby overestimating the contribution that light energy
might play in providing reductant for fatty acid synthesis in
B. napus embryos. It is clear that addressing the relative
contributions of the OPPP and light energy for the provision of reductant for fatty acid synthesis will require
careful experimental design and/or the use of non-invasive,
in vivo, methods that are based on gas exchange and isotope
discrimination (Willms et al., 2000). The latter method has
shown that light energy is likely to contribute to the supply
of reductant and or ATP in developing soybean embryos
and more recently, Borisjuk et al. (2003) have demonstrated that the presence of photosynthetic characteristics in
developing Vicia faba embryos was spatially correlated
with the elevation of ATP levels within the cotyledons.
These measurements of NADP-G6PDH activity in plastids from B. napus embryos have revealed that the isoform(s) show some sensitivity to redox regulation by light
and DTT (Kruger and von Schaewen, 2003). A reduction in
NADP-G6PDH activity in the light would be consistent
with maintaining a balance between the OPPP and light
energy as sources of reducing power for fatty acid synthesis in developing embryos depending on the light environment around the silique. However, the inhibition of
NADP-G6PDH activity in the embryo plastids was less
than that of chloroplasts, implying that this activity would
be less regulated by redox than that in the leaf. Further
studies of which isoforms of the enzyme are present in
plastids of the developing embryo, and what their properties are, would be beneficial in helping interpret their role,
and that of the plastid OPPP, in the provision of reducing
power to fatty acid synthesis in Brassicaceous oilseeds.
Acknowledgements
The authors thank Professor Alison Smith and Drs Kay Denyer and
Matthew Hills for their discussions and for comments on the
manuscript. This work was supported by the Biotechnology and
Biological Sciences Research Council through a Special studentship
(DH) and by the Competitive Strategic Grant to JIC (SR).
References
ap Rees T, Green JH, Wilson PM. 1985. Pyrophosphate:
fructose 6-phosphate 1-phosphotransferase and glycolysis in nonphotosynthetic tissues of higher plants. Biochemical Journal 227,
299–304.
Bowsher CG, Boulton EL, Rose J, Nayagam S, Emes MJ. 1992.
Reductant for glutamate synthase is generated by the oxidative
pentose phosphate pathway in non-photosynthetic root plastids.
The Plant Journal 2, 893–898.
Bowsher CG, Hucklesby DP, Emes MJ. 1989. Nitrite reduction and
carbohydrate metabolism in plastids purified from roots of Pisum
sativum L. Planta 177, 359–366.
Borisjuk L, Rolletschek H, Walenta S, Panitz R, Wobus U, Weber
H. 2003. Energy status and its control on embryogenesis of
legumes: ATP distribution within Vicia faba embryos is developmentally regulated and correlated with photosynthetic capacity.
The Plant Journal 36, 318–329.
Debnam PM, Emes MJ. 1999. Subcellular distribution of enzymes
of the oxidative pentose phosphate pathway in root and leaf tissues.
Journal of Experimental Botany 50, 1653–1661.
Denyer K, Smith AM. 1988. The capacity of plastids from developing pea cotyledons to synthesize acetyl CoA. Planta 173,
172–182.
Eastmond PJ, Koláčná L, Rawsthorne S. 1996. Photosynthetic
characteristics of developing embryos of oilseed rape (Brassica
napus L.). Journal of Experimental Botany 47, 1763–1769.
Eastmond PJ, Rawsthorne S. 1998. Comparison of the metabolic
properties of plastids isolated from developing leaves and embryos
of Brassica napus L. Journal of Experimental Botany 49,
1105–1111.
Eastmond PJ, Rawsthorne S. 2000. Coordinate changes in carbon
partitioning and plastidial metabolism during the development of
oilseed rape embryos. Plant Physiology 122, 767–774.
Eicks M, Maurino V, Knappe S, Flugge UI, Fischer K. 2002. The
plastidic pentose phosphate translocator represents a link between
the cytosolic and the plastidic pentose phosphate pathways in
plants. Plant Physiology 128, 512–522.
Farr TJ, Huppe HC, Turpin DH. 1994. Coordination of chloroplastic metabolism in N-limited Chlamydomonas reinhardii by
redox modulation. I. The activation of phosphoribulokinase and
glucose 6-phosphate dehydrogenase is relative to the photosynthetic supply of electrons. Plant Physiology 105, 1037–1042.
Fox SR, Hill LM, Rawsthorne S, Hills MJ. 2000. Inhibition of the
glucose-6-phosphate transporter in oilseed rape (Brassica napus L.)
Reductant provision for fatty acid synthesis
plastids by acyl-CoA thioesters reduces fatty acid synthesis. Biochemical Journal 352, 525–532.
Henkes S, Sonnewald U, Badur R, Flachmann R, Stitt M. 2001.
A small decrease of plastid transketolase activity in antisense
tobacco transformants has dramatic effects on photosynthesis and
phenylpropanoid metabolism. The Plant Cell 13, 535–551.
Johnson HS. 1972. Dithiothreitol: an inhibitor of glucose 6phosphate dehydrogenase activity in leaf extracts and isolated
chloroplasts. Planta 106, 273–277.
Johnson PE, Fox SR, Hills MJ, Rawsthorne S. 2000. Inhibition by
long chain acyl-CoAs of glucose-6-phosphate metabolism in
plastids isolated from developing embryos of oilseed rape
(Brassica napus L.). Biochemical Journal 348, 145–150.
Kang F. 1995. Carbon sources for starch and fatty acid synthesis in
plastids from developing embryos of oilseed rape. PhD thesis,
University of East Anglia, UK.
Kang F, Rawsthorne S. 1994. Starch and fatty acid synthesis in
plastids from developing embryos of oilseed rape (Brassica napus
L.). The Plant Journal 6, 795–805.
Kang F, Rawsthorne S. 1996. Metabolism of glucose-6-phosphate
and utilization of multiple metabolites for fatty acid synthesis by
plastids from developing oilseed rape embryos. Planta 199,
321–327.
Kopriva S, Koprivova A, Süss K-H. 2000. Identification, cloning,
and properties of cytosolic D-ribulose-5-phosphate 3-epimerase
from higher plants. Journal of Biological Chemistry 275,
1294–1299.
Kruger NJ, von Schaewen A. 2003. The oxidative pentose
phosphate pathway: structure and organization. Current Opinion
in Plant Biology 6, 236–246.
Schnarrenberger C, Flechner A, Martin W. 1995. Enzymic
evidence for a complete oxidative pentose phosphate pathway in
chloroplasts and an incomplete pathway in the cytosol of spinach
leaves. Plant Physiology 108, 609–614.
585
Schwender J, Ohlrogge JB. 2002. Probing in vivo metabolism
by stable isotope labelling of storage lipids and proteins in
developing Brassica napus embryos. Plant Physiology 130,
347–361.
Schwender J, Ohlrogge JB, Shacher-Hill Y. 2003. A flux model of
glycolysis and the oxidative pentose phosphate pathway in developing Brassica napus embryos. Journal of Biological Chemistry 278, 29442–29453.
Tetlow IJ, Blissett KJ, Emes MJ. 1993. A rapid method for the
isolation of purified amyloplasts from wheat endosperm. Planta
189, 597–600.
Tetlow IJ, Blissett KJ, Emes MJ. 1994. Starch synthesis and
carbohydrate oxidation in amyloplasts from developing wheat
endosperm. Planta 194, 454–460.
Thom E, Möhlmann T, Quick WP, Camara B, Neuhaus H-E
1998. Sweet pepper plastids: enzymic equipment, characterization
of the plastidic oxidative pentose-phosphate pathway, and transport of phosphorylated intermediates across the envelope membrane. Planta 204, 226–233.
Thorbjørnsen T, Villand P, Denyer K, Olsen OA, Smith AM.
1996. Distinct isoforms of ADPglucose pyrophosphorylase occur
inside and outside the amyloplasts in barley endosperm. The Plant
Journal 10, 243–250.
Wenderoth I, Scheibe R, von Schaewen A. 1997. Identification
of the cysteine residues involved in redox modification of
plant plastidic glucose-6-phosphate dehydrogenase. Journal of
Biological Chemistry 272, 26985–26990.
Wendt UA, Wenderoth I, Tegeler A, von Schaewen A. 2000.
Molecular characterisation of a novel glucose-6-phosphate dehydrogenase from potato (Solanum tuberosum L.). The Plant
Journal 23, 723–733.
Willms JR, Salon C, Layzell DB. 2000. Evidence for lightstimulated fatty acid synthesis in soybean fruit. Plant Physiology
120, 1117–1127.