• Open Access

Doubled sugar content in sugarcane plants modified to produce a sucrose isomer

Authors


* Correspondence (fax +61 7 3365 1699; e-mail r.birch@uq.edu.au)

Summary

Sucrose is the feedstock for more than half of the world's fuel ethanol production and a major human food. It is harvested primarily from sugarcane and beet. Despite attempts through conventional and molecular breeding, the stored sugar concentration in elite sugarcane cultivars has not been increased for several decades. Recently, genes have been cloned for bacterial isomerase enzymes that convert sucrose into sugars which are not metabolized by plants, but which are digested by humans, with health benefits over sucrose. We hypothesized that an appropriate sucrose isomerase (SI) expression pattern might simultaneously provide a valuable source of beneficial sugars and overcome the sugar yield ceiling in plants. The introduction of an SI gene tailored for vacuolar compartmentation resulted in sugarcane lines with remarkable increases in total stored sugar levels. The high-value sugar isomaltulose was accumulated in storage tissues without any decrease in stored sucrose concentration, resulting in up to doubled total sugar concentrations in harvested juice. The lines with enhanced sugar accumulation also showed increased photosynthesis, sucrose transport and sink strength. This remarkable step above the former ceiling in stored sugar concentration provides a new perspective into plant source–sink relationships, and has substantial potential for enhanced food and biofuel production.

Introduction

Sugarcane and sugarbeet are unusual amongst plants in storing sucrose (α-d-glucopyranosyl-1,2-d-fructofuranose) – rather than polymeric compounds such as starch, proteins or lipids – as the primary carbon and energy reserve. Sugarcane is one of the world's major food crops, providing about 75% of the sugar harvested for human consumption [Food and Agriculture Organization (FAO) statistics]. It is also the industrial crop closest to sustainability as a renewable energy source (De Oliveira et al., 2005; Patzek and Pimentel, 2005). Sugar yield is a key determinant of both economic and environmental sustainability. During recent decades, sugarcane improvement has been achieved almost entirely through increased cane yield, despite higher commercial value and higher heritability of increased sugar content (Jackson, 2005). Attempts to increase sugar content in elite sugarcane cultivars through the modification of plant genes involved in sugar metabolism have also been unsuccessful (Lakshmanan et al., 2005). There is evidently sufficient control redundancy in elite cultivars to buffer against increases in stored sucrose levels through the manipulation of single sugarcane genes, as in other plant systems that have been subject to strong selection for the yield of other stored products (Capell and Christou, 2004). Mechanisms contributing to the apparent sugar concentration ceiling may include feedback regulation from broad (e.g. osmotic) or specific (e.g. sucrose) sensors, thermodynamic limitations, such as leakage of sucrose through storage compartment membranes, or energetic limitations from the continuous ‘futile’ cycle of sucrose cleavage and synthesis within the storage pool (Moore, 1995; Grof and Campbell, 2001; Bindon and Botha, 2002).

Some bacteria are able to sequester sucrose by conversion into isomaltulose (IM) (α-d-glucopyranosyl-1,6-d-fructofuranose). This sucrose isomer is resistant to invertases and is not metabolized by many microbes, including the predominant oral microflora, conferring an advantage in many foods as an acariogenic sweetener. IM is digested by humans with the same primary products (glucose, fructose) and ultimate caloric value as sucrose. However, because the first step involves an intestinal disaccharidase rather than salivary invertase, IM is digested more slowly, with advantages of a lower fluctuation in blood glucose and insulin concentrations. Consequently, IM has a growing market as a stable, slowly digested, acariogenic, non-hygroscopic sugar (Takazoe, 1989; Lina et al., 2002). Its accessible carbonyl group also makes an attractive renewable starting material for the manufacture of biomaterials as eventual petrochemical replacements (Lichtenthaler and Peters, 2004), but use is currently limited as a result of the high cost of IM production through fermentation (Schiweck et al., 1991).

IM biosynthesis occurs via a sucrose isomerase (SI) with no cofactor or substrate activation requirements (Wu and Birch, 2005), indicating feasibility for engineered production in plants. IM is apparently not metabolized or transported in plants, but exogenous application triggers some plant sugar-sensing mechanisms and changes gene expression profiles differently from sucrose (Loreti et al., 2000; Sinha et al., 2002). In growing plant tissues, efficient conversion of sucrose into the non-metabolized isomer is lethal or disruptive (Börnke et al., 2002a). Tuber-specific expression of an apoplasm-targeted SI allowed the partial conversion of the low soluble sucrose levels in potato tubers to IM [∼15 µmol/g fresh weight (FW)] without affecting plant appearance, but with a substantial decrease in total non-structural carbohydrate content (Börnke et al., 2002b; Hajirezaei et al., 2003).

We set out to determine the effect of directing SI activity into the sugar storage compartment of a sucrose-accumulating plant. In the storage parenchyma cells of mature sugarcane stems, the sugar storage vacuole occupies about 90% of the symplast and 80% of the total tissue space. The vacuole stores a correspondingly large proportion of sucrose, which can accumulate to 500 µmol/g FW (Moore, 1995). Previous results have indicated that the N-terminal pro-peptide (NTPP) sequence from sweet potato sporamin can target diverse proteins to the sugarcane vacuole, but low pH and high protease activity make this a hostile environment to introduced proteins (Gnanasambandam and Birch, 2004).

Here, we show that vacuolar targeting of a highly efficient SI allows high IM yields (up to 440 µmol/g FW) in sugarcane stems. Remarkably, IM can be accumulated without a commensurable reduction in sucrose, resulting in twice the total sugar concentration in juice from selected transgenic lines relative to their elite parent cultivar.

Results

Substantial IM accumulation in healthy plants through expression of a vacuolar SI

Of 31 independent transgenic lines positive by polymerase chain reaction (PCR) analysis for the transgene encoding a vacuole-targeted SI, 80% showed high-performance liquid chromatography (HPLC)-detectable IM in stalk tissues. There were substantial differences in IM concentration between lines (Figure 1), consistent with the expected effects of variable transgene copy number and integration position amongst independent transformants from particle bombardment (Bower et al., 1996). As expected from the high product specificity of the UQ68J SI (Wu and Birch, 2005), trehalulose concentrations were below 4% of the IM concentrations in corresponding internodes.

Figure 1.

Screening transgenic sugarcane lines for the presence of isomaltulose in juice. (A, B) High-performance liquid chromatography (HPLC) profiles from the separation of sugar standards (dotted line) and 4000-fold diluted juice from the basal internode (full lines) of Q117 control (A) and transgenic line N3.2H (B). Sugar standards in order of elution were glucose (G, 75.0 µm), fructose (F, 37.5 µm), sucrose (S, 37.5 µm), trehalulose (T, 37.5 µm) and isomaltulose (I, 75.0 µm). (C) Isomaltulose concentrations in juice from the basal internodes of transgenic lines. The plants had 14–31 internodes when sampled, after growth in a containment glasshouse for 6–10 months in the first vegetative generation after regeneration from tissue culture. Only a single plant exists per transgenic line in this generation.

Transgenic lines with the vacuole-targeted NTPP-68J SI (Gnanasambandam and Birch, 2004), expressed from a constitutive ubiquitin (Ubi) promoter (Christensen and Quail, 1996), were morphologically similar and equivalent in measured growth parameters to non-transformed controls of the same background genotype (elite sugarcane cultivar Q117) throughout observation periods of up to 16 months in a containment glasshouse. In contrast, expression of a cytosol-targeted form of the same SI from the same promoter was very damaging to sugarcane plant development, with low sugar contents and severe stunting in surviving plants.

Developmental profile of sugar accumulation in high-level IM producers

Two NTPP-68J lines, designated N3.2 and N3.2H, with high IM yields were selected for further characterization over several vegetative generations under containment glasshouse conditions. These lines were derived from separate gene transfer experiments. The IM concentration was below 1 mm in roots, increased with age in leaves to a maximum of about 1.5 mm, and increased down the stalk developmental profile reaching 300–520 mm in juice from mature internodes (Figure 2). Sucrose concentration in juice from mature internodes of Q117 control plants ranged from 430 to 530 mm (equivalent to c. 380–460 µmol/g FW, or 13–16% FW in mature stalks). This is similar to the sucrose content achieved under commercial field conditions. Surprisingly, the transgenic lines with high yields of IM did not show a commensurable reduction in stored sucrose concentrations. Indeed, there was no significant decrease in sucrose content in most internodes, and there was a consistent trend of higher sucrose content in internodes of intermediate age, relative to Q117 control plants (Figure 2).

Figure 2.

Sugar accumulation profiles of sugarcane cultivar Q117 (A) and transgenic lines N3.2 (B) and N3.2H (C) showing substantial isomaltulose accumulation and enhanced total sugar concentration in juice. The plants were 16 months old with 59 internodes in the third vegetative generation. Results are means of three replicates, with standard errors shown for sucrose results. Analysis of variance (anova) with Bonferroni post-tests showed significant differences (P < 0.001) in total sugar concentration between Q117 (A) and N3.2 (B) or N3.2H (C) in juice from internodes older than #10. These developmental profiles of sugar accumulation were stable over three tested vegetative generations. TVD, top visual dewlap.

Hexoses followed the typical pattern for sugarcane, declining from a substantial fraction of total sugars in the youngest internodes to very low levels in mature internodes. The net effect was an increase of 15–115% in total sugar concentration in the juice from internodes typically harvested commercially for sugar extraction (older than internode 10), with IM comprising 39%−53% of total sugar in the most mature internodes (Figure 2).

Enhanced sugar accumulation occurs without a decrease in structural carbohydrates

For convenience, we refer to lines N3.2 and N3.2H as SugarBooster lines. The relationship between juice and whole-tissue effects was tested in plants at an earlier harvest date, when the sugar concentration in juice was 50%−80% higher in SugarBooster lines relative to Q117 control plants. There was correspondingly more sugar per unit FW, with no significant change in the insoluble (fibre) content of 9%−10% FW in mature internode tissues (Figure 3). This indicates increased photosynthate storage as sugar, rather than altered partitioning between sugar and fibre. In high-sugar cultivars of sugarcane, the water content typically decreases down the stalk with increased sucrose content, to a minimum of about 70% moisture in mature internodes (Bull and Glasziou, 1963). This effect continued further down the stalk developmental profile in SugarBooster lines, to about 60% moisture in the oldest internodes (Figure 3). Therefore, the commercially important traits of higher sugar yield and higher sugar concentration in juice from SugarBooster lines are underpinned by an increased storage of photosynthate as sugars, and a decrease in mature stalk water content (Figure 3).

Figure 3.

Total sugar, water and fibre contents per unit fresh weight in internodes of sugarcane cultivar Q117 control (squares, full line) and SugarBooster transgenic lines N3.2 (triangles, dotted line) and N3.2H (circles, broken line). The plants were 12 months old with 42–43 internodes in the third vegetative generation. Internodes were numbered from #1 for the top visible dewlap (TVD) leaf. Results are means with standard errors from four replicate plants. Analysis of variance (anova) with Bonferroni post-tests showed that, compared with the Q117 control, both SugarBooster lines had significantly increased sugar content and reduced water content (P < 0.001) in internodes older than #20.

Enhanced sugar accumulation is accompanied by increases in photosynthesis, sugar transport and sink strength

Key physiological characteristics were examined to help to understand the underlying mechanisms. The SugarBooster lines were visually indistinguishable from Q117 control plants, except that leaf senescence was typically delayed by 15–20 days, resulting in an additional one to three leaves functional in photosynthesis per stalk for most of the growth period.

Electron transport rates and carbon dioxide (CO2) fixation rates were higher for all tested light intensities and leaves of the SugarBooster lines relative to Q117 controls. The increases in electron transport rates measured by chlorophyll fluorescence (reflecting photosynthetic efficiency in photosystem II) were in the range 20%−40% relative to Q117 controls at a photosynthetically active radiation (PAR) level of 400–4000 µmol photons/m2/s. The increases in CO2 assimilation rates were typically about 20% at PAR = 80–1250 µmol photons/m2/s in young (TVD2; TVD, top visual dewlap) and old (TVD12) leaves, and 5%−15% in leaves of intermediate age. Light response curves from fully expanded leaf 2 are shown as an example (Figure 4A,B). Proton gradient-dependent sucrose transport into plasma membrane vesicles (PMVs) isolated from leaf 2 of the SugarBooster lines was 20%−30% higher than in Q117 controls (Figure 4C).

Figure 4.

Enhanced photosynthesis and sucrose transport in the second fully emerged leaf (TVD2; TVD, top visible dewlap) of SugarBooster transgenic lines N3.2 and N3.2H, relative to sugarcane cultivar Q117 controls. (A) Photosynthetic electron transport rate. (B) CO2 assimilation rate. (C) Sucrose transport rate into plasma membrane vesicles. Results are means with standard errors from at least three replicates. Analysis of variance (anova) with Bonferroni post-tests showed significant differences (P < 0.001) in all three characteristics between Q117 and N3.2 or N3.2H. These physiological indices were stable over tested second and third vegetative generations. CCCP, carbonyl cyanide m-chlorophenylhydrazone.

In internode 7 and older, cell wall acid invertase (CWAI) activities were 50%−80% higher in SugarBooster lines than in Q117 controls in the central storage parenchyma-rich zone (Figure 5A), but not in the peripheral fibre- and vascular-rich zone (Figure 5B). When vascular bundles were dissected from the storage parenchyma cells in the central zone of internode 7 and assayed separately, the increased CWAI activity was clearly confined to the storage parenchyma (Figure 5C).

Figure 5.

Increased cell wall acid invertase (CWAI) activity in storage parenchyma of SugarBooster transgenic lines N3.2 and N3.2H, relative to sugarcane cultivar Q117 controls. CWAI activity in the central parenchyma-rich zone (A), the peripheral vascular-rich zone (B) and separated vascular bundles and parenchyma tissue from the central zone of internode 7 (C). Results are means with standard errors from three replicates. Analysis of variance (anova) with Bonferroni post-tests showed significant differences between Q117 controls and SugarBooster lines in CWAI activity of parenchyma cells in the central zone. *P < 0.10; **P < 0.05.

SI transgene is expressed in sugarcane internodes, and the enzyme is sensitive to vacuolar proteases

As expected from the constitutive Ubi promoter, and given the IM accumulation shown in Figure 2, expression of the SI transgene was readily detectable by Northern blot analysis in internodes of IM-accumulating lines, but not in controls transformed without the SI gene (Figure 6).

Figure 6.

Northern blot analysis for sucrose isomerase (SI) gene expression in transgenic sugarcane lines. Lane 1, N3.2H internode 20; lane 2, N3.2H internode 31; lane 3, N3.2 internode 31; lane 4, control line transformed using selection marker only, internode 31. Total RNA (30 µg per lane) from internodal tissues was probed with a 32P-labelled full-length UQ68J SI DNA.

Even in mature stems with substantial IM accumulation, SI enzyme activity was below the detection limit in cell extracts, consistent with a short half-life of this protein after delivery into the sucrose storage vacuoles. All attempts to isolate vacuoles from mature sugarcane stem parenchyma in yields and purity sufficient for biochemical characterization have been unsuccessful (Moore, 1995; Rae et al., 2005; Gnanasambandam and Birch, 2006). However, vacuoles occupy about 80% of this tissue volume and are expected to make the dominant contribution to sugar and protease content in the fresh juice from crushed stems. As indicated by sodium dodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and SI enzyme assays, purified UQ68J SI was degraded within minutes in juice from sugarcane stem parenchyma, and this process was slowed by the addition of a cysteine protease inhibitor (Table 1).

Table 1.  Reduction in sucrose isomerase (SI) activity by sugarcane cell extract, and protection by protease inhibitors
SI reaction medium*Protease targetedSucrose isomerized (%)
  • EDTA-Na2, sodium ethylenediaminetetraacetate; PMSF, phenylmethylsulphonylfluoride.

  • *

    Purified UQ68J SI (150 µg/mL) was incubated in the specified medium for 15 min at 30 °C, followed by measurement of isomaltulose (IM) concentration.

  • Results are means with standard errors from three replicates.

270 mm sucrose in citrate–phosphate buffer pH 5.293.0 ± 1.7
Sugarcane juice (containing 270 mm sucrose at pH 5.2)58.2 ± 3.3
Juice + EDTA-Na2 (1 mm)Metalloprotease62.4 ± 6.1
Juice + Pepstatin A (0.0014 mm)Aspartic acid protease62.5 ± 4.4
Juice + PMSF (0.57 mm)Serine and cysteine protease74.2 ± 4.6
Juice + E64d (0.05 mm)Cysteine protease75.3 ± 5.1

Discussion

Our results show that high yields of the high-value sugar IM can be obtained by vacuolar targeting of an efficient SI, without disruption of sugarcane growth and development. In contrast, expression of a cytosolic form of the same enzyme from the same constitutive promoter caused stunting with reduced sugar accumulation. The sugarcane sucrose storage vacuoles are highly acidic and proteolytic (Gnanasambandam and Birch, 2004), and the vacuole-targeted SI enzyme was undetectable in cell extracts. Rapid degradation of vacuole-targeted SI most likely protects against rapid sucrose depletion in growing tissues. IM accumulates gradually during development (Figure 2), probably because of the following: (i) continuous delivery of SI expressed from the constitutive Ubi promoter (Zhang et al., 1999); (ii) high catalytic efficiency allowing occasional IM production before SI inactivation (Wu and Birch, 2005); and (iii) absence of plant enzymes for IM metabolism (Loreti et al., 2000). For efficient commercial production of this valued sugar, it will be useful to achieve patterns of developmental expression, compartmentation and enzyme stability yielding high IM content further up the harvested stalk profile.

High-level IM accumulation without a decrease in sucrose concentration is an unexpected outcome with important scientific and biotechnological implications. Doubling the total sugar content in mature internodes of an elite high-sugar cultivar eliminates osmotic limits and osmotic sensing as primary constraints behind the previous concentration ceiling in sugarcane (Moore, 1995; Jackson, 2005). Our results also challenge the long-standing assumption that high-sugar genotypes require at least 70% moisture content in mature internodes (Bull and Glasziou, 1963). This shifts the focus for work towards enhanced sucrose content from increased sugar : fibre partitioning to increased sugar : water ratio, with substantial scope for gains, as some Saccharum accessions have only 50%−60% moisture content in mature internodes (Bull and Glasziou, 1963).

SugarBooster lines will allow new insights into the mechanisms by which plants regulate sugar accumulation, a pivotal question in plant biology (Rolland et al., 2002; Koch, 2004; Fernie et al., 2005). Their high total sugar phenotype is accompanied by delayed leaf senescence, increased photosynthetic activity and enhanced sucrose loading rates in source tissues, as well as higher activity in stalk storage parenchyma of CWAI, which has multiple roles in sink tissues (Koch, 2004). Each of these activities can contribute to the observed high sugar yields. Further comparative analysis of the SugarBooster lines and their parent cultivar will help to reveal key molecular and physiological control points in plant source–sink flux.

It will be very interesting to test the effects of a vacuolar SI gene in other plants. Sweetness is an important commercial trait in many food crops. Enhanced sweetness through a slowly digested, acariogenic sugar, such as IM, can bring direct health benefits for consumers, an important consideration in the development of public support for plant improvement through biotechnology. Fermentable carbohydrate content is also a key determinant of the economic and environmental feasibility of renewable biofuel production (De Oliveira et al., 2005). In the longer term, because sugars ultimately underpin all other biosynthesis in plants, the SugarBooster effect may be a foundation for higher yields of many other biomaterials.

Experimental procedures

Constructs and gene transfer

The UQ68J SI gene without periplasmic leader sequence (Wu and Birch, 2005) was amplified by PCR using high-fidelity PfuTurbo DNA polymerase (Stratagene, Cedar Creek, TX), BglII forward primer (GTA GAT CTC GCA ACG AAT ATA CAA AAG TCC G) and reverse primers introducing either a SacI site (AAG AGC TCA GTT CAG CTT ATA GAT CCC) or a 6 × histidine (6 × His) tag followed by a SacI site (AAG AGC TCA GTG GTG GTG GTG GTG GTG GTT CAG CTT ATA GAT CCC); it was then cloned in frame into a vector providing the maize Ubi-1 promoter, a 21-amino-acid endoplasmic reticulum (ER) leader peptide, the 16-amino-acid NTPP of sweet potato sporamin and an Agrobacterium nos polyadenylation region (Gnanasambandam and Birch, 2004).

Previous results have indicated that the NTPP of sweet potato sporamin can target diverse proteins to the sugarcane vacuole, but the failure to accumulate any of the tested reporter proteins highlights protein stability as a major challenge for efficient metabolic engineering of the sugar storage compartment in sugarcane (Gnanasambandam and Birch, 2004).

The SI construct and selectable marker construct pEmuKN were co-precipitated on to tungsten microprojectiles and introduced into sugarcane embryogenic callus, followed by selection for geneticin resistance and regeneration of transgenic plants, essentially as described previously (Bower et al., 1996).

Sugarcane growth conditions and plants analysed

Sugarcane cultivar Q117 is a current elite commercial variety selected for high sucrose yield. Sugarcane cultivars are highly heterozygous, complex polyploid interspecific hybrids of Saccharum species. They have generally low fertility and are propagated vegetatively for both commercial and experimental purposes. Plants were grown in a containment glasshouse under natural light intensity at 28 °C with watering twice a day. Each plant was grown as a single stalk in a 20-cm-diameter pot (pot volume, 4 L) and a density of 18 pots/m2, and fertilized with Osmocote® at 5 g/month for the first and second months, followed by 10 g/month. Leaves were numbered from one for the TVD, with higher numbers for older leaves. Internodes were numbered according to the leaf attached to the node immediately above. The measured growth parameters were the height from the soil surface to the TVD, stalk diameter at the lowest above-ground internode, number of nodes and FW. These were recorded fortnightly over three vegetative generations reproduced by stalk cuttings and in one set of ratoons.

Unless stated otherwise, physiological data are provided from plants in the second vegetative generation after regeneration from tissue culture, growing from single-node stalk cuttings for 8 months to produce stalks with approximately 25 internodes. Unless stated otherwise, statistical analysis was performed using GraphPad Prism Software (V4.0; San Diego, CA).

Determination of sugar concentrations by high-performance liquid chromatography-electrochemical detection (HPLC-ED)

For stalk samples, a transverse tissue slice was taken at the mid-point of each designated internode and cut into radial sectors that were proportionately representative of the different stalk tissues by area. Sectors (about 0.15 g FW) were placed on a support screen (Promega Spin Basket, Madison, WI) within a 1.5-mL microfuge tube, liquid nitrogen frozen, thawed on ice and centrifuged at 10 000 g for 15 min at 4 °C to collect the juice. After boiling for 5 min to inactivate enzymes, insoluble material was removed by centrifugation at 16 000 g for 20 min. In comparative tests conducted on internodes of intermediate maturity (equivalent to internode 15 in Figure 2), this procedure gave sugar concentrations equivalent to the manual crushing of stalk samples to express the juice, and it was adaptable to larger numbers of samples. FWs were recorded before and after juice extraction, and residual dry weights (DWs) were measured after 48 h at 75 °C for tissues, or 90 °C for juice samples. Water contents were measured in alternate subsamples to those used for juice extraction and analysis.

The resolution and quantification of IM, trehalulose, sucrose, glucose and fructose were achieved by isocratic HPLC at high pH (120 mm NaOH), using a Dionex BioLC system (Sunnyvale, CA) with PA20 analytical anion exchange column and quad waveform pulsed ED, with calibration against a dilution series of sugar standards for every sample batch (Wu and Birch, 2004). Sugar concentrations were corrected for dilutions in the procedure and presented as sucrose equivalents in juice. Total sugar contents were calculated on an FW and DW basis, taking account of the residual juice in internode tissues after centrifugation (up to 60% of total juice) and assuming 10% reduction in solute concentration in residual juice relative to first expressed juice, as typically observed in the sugarcane industry (Hugot, 1986).

For leaf samples, about 1 g FW of leaf blade without midrib was taken at one-third of the distance from the dewlap to the leaf tip. For root samples, about 0.5 g FW of young roots was taken from the interface between the soil and pot. Fluids were extracted and assayed by the freeze–thaw–centrifuge–HPLC method described above for stalk samples.

Gas exchange and chlorophyll fluorescence measurements

The photosynthetic electron transport rate was estimated from the fluorescence light curve generated using a fibre-optic MINI-PAM/F (Heinz Waltz GmbH, Effeltrich, Germany) and leaf-clip holder 2030-B positioned at one-tenth of the distance from the dewlap to the leaf tip. The MINI-PAM light intensity, saturation pulse intensity, saturation pulse width, leaf absorption factor and illumination time were set at 680 µmol/m2/s, 680 µmol/m2/s, 0.8 s, 0.84 and 10 s, respectively. The internal temperature of the MINI-PAM was controlled between 25 and 30 °C during measurement. An LI-6400 portable photosynthesis system (LI-COR, Lincoln, NE, USA) was used to measure CO2 fixation rates in the same leaves. Measurements were made on at least three replicate plants per line.

Plasmalemma vesicle (PMV) isolation and transport assays

TVD2 leaf blades without midribs (25 g FW) were homogenized in 100 mL of 240 mm sorbitol, 50 mm N-2-hydroxyethylpiperazine-N′-2-ethanesulphonic acid (HEPES), 3 mm ethyleneglycol-bis(β-aminoethylether)-N,N′-tetraacetic acid (EGTA), 3 mm dithiothreitol (DTT), 10 mm KCl, 0.5% bovine serum albumin (BSA), 0.6% polyvinylpyrrolidone (PVP) and 2 mm phenylmethylsulphonylfluoride (PMSF) (adjusted to pH 8.0 using solid Bistris propane) at 4 °C. The homogenate was filtered through four layers of cheese cloth to remove tissue debris, and then centrifuged at 10 000 g for 10 min to remove mitochondria and chloroplasts. Microsomal membranes were pelleted by centrifugation at 50 000 g for 60 min. PMVs were purified from the microsomal fraction by phase partitioning (Bush et al., 1996), washed in 25 mL of sorbitol-based resuspension buffer (SBRB) (330 mm sorbitol, 2 mm HEPES, 0.1 mm DTT, 10 mm KCl, pH 8.0 with solid Bistris propane), repelleted by centrifugation at 50 000 g for 60 min and resuspended at 3–5 mg FW/mL of resuspension buffer.

The phase-purified PMVs were layered over a 20%−50% sucrose gradient in 2 mm HEPES, 1 mm HCl and 1 mm DTT (pH 8.0 with solid Bistris propane), centrifuged for 15 h at 100 000 g and collected in 1-mL fractions. The fractions were washed in 11 mL SBRB and pelleted by centrifugation at 100 000 g for 60 min. The pellet was suspended in 0.4 mL of SBRB, checked for purity using routine tests for enzymatic activities characteristic of other cellular membrane types, and used for transport experiments.

Transport assays were conducted at 12 °C using three replicate reactions per treatment (Bush et al., 1996). Briefly, for each reaction mixture, 20 µL of resuspended PMVs were diluted into 400 µL of assay buffer [as for SBRB, except adjusted to pH 6.0 with solid 2-(N-morpholino)ethanesulphonic acid (MES)] containing 0.2 µCi [14C]sucrose and unlabelled sucrose to the desired concentration. At each time point, vesicles from one reaction mixture were collected on 0.45-µm filters and rinsed three times with 0.6 mL of assay buffer containing only unlabelled sucrose (1 mm). The accumulated radioactivity was measured by scintillation spectrometry. The difference between samples with and without 5 µm carbonyl cyanide m-chlorophenylhydrazone (CCCP) was defined as ΔpH-dependent sucrose transport.

Internode tissue fractionation and enzyme assays

Transverse sections of internode were divided into the outer rind of 3 mm thickness and three internal concentric cylinders at equal distances along the stalk radius. Of these, the central parenchyma-rich zone and the peripheral vascular-rich zone were examined for invertase activity. In addition, vascular bundles were separated by dissection from parenchyma tissue in the central zone for separate assays. The separated tissues were frozen immediately in liquid nitrogen for enzyme extraction, followed by the determination of CWAI activity, using three replicate plants or dissected tissue subsamples per assay (Albertson et al., 2001). SI activity was measured by incubating enzyme extract with 292 mm sucrose solution in 0.1 m citrate–phosphate buffer (pH 6.0) at 30 °C, and testing for IM accumulation over 80 min by HPLC-ED as described above.

The effect of proteases from sugarcane stem tissue on SI activity was measured by incubating 15 µg of purified UQ68J SI (Wu and Birch, 2005) with 90 µL of juice crushed from Q117 internode 9 and selected protease inhibitors in a total volume of 100 µL. The juice contained 300 mm sucrose at pH 5.2. Protease inhibitors were sodium ethylenediaminetetraacetate (EDTA-Na2) (Sigma, St Louis, MO), PMSF (Sigma), E64d (Roche, Castle Hill, NSW, Australia) and Pepstatin A (Roche). Controls included sugarcane juice without protease inhibitor, and 300 mm sucrose in 100 mm citrate–phosphate buffer (pH 5.2) instead of sugarcane juice. After incubation at 30 °C for 15 min, the IM concentration was determined by HPLC-ED as described above.

Northern blot analysis

Total RNA was isolated from the internodes of sugarcane lines, and 30 µg of total RNA per lane was fractionated by 2.2 m formaldehyde and 1.0% agarose gel electrophoresis, blotted on to Hybond™-N+ nylon membrane (Amersham Pharmacia Biotech) and hybridized as described previously (Tsai et al., 1998) using randomly primed 32P-labelled probes from UQ68J SI DNA.

Acknowledgements

This work was supported by the Australian Research Council, Sugar Research and Development Corporation, CSR Limited and the University of Queensland. We thank Frikkie Botha (University of Stellenbosch and South African Sugar Research Institute) and Christa Critchley (University of Queensland) for helpful discussions, Pan Yunrong and Simon Hansom for expert research assistance, and John Manners (Commonwealth Scientific and Industrial Research Organization), Darren Schliebs and Peter Collins (CSR Limited) for critical review of the manuscript.

Ancillary