Lactate metabolism in potato tubers deficient in lactate dehydrogenase activity
ABSTRACT
The aim of this work was to investigate the effect of decreased activity of lactate dehydrogenase (EC 1.1.1.27; LDH) on lactate metabolism in potato tubers. By expressing a cDNA-encoding potato tuber LDH in the antisense orientation, we generated transgenic potato plants with a preferential decrease in two of the five isozymes of LDH. Surprisingly, transgenic tubers grown under normoxic conditions did not contain less lactate, but rather instead contained approximately two-fold more lactate than control tubers. This result is explicable if the decreased isozymes of LDH are responsible for the oxidation of lactate to pyruvate in vivo. This was confirmed by measurements of the rate of metabolism of lactate supplied to tuber discs: the rate in transgenic tubers was approximately half that of control tubers. The decrease in LDH activity had no measurable effect on the accumulation of lactate in cold-stored tubers under anoxia, nor during the subsequent utilization of this lactate upon return to normoxia. In both control and transgenic tubers, the accumulation of lactate during anoxia was not accompanied by an induction of LDH activity or a change in isozyme distribution. In contrast, the metabolism of lactate after a period of anoxia was accompanied by a two-fold increase in LDH activity and the induction of two isozymes that were distinct from those which had been decreased in the transgenic plants.
INTRODUCTION
Lactate dehydrogenase (LDH; EC 1.1.1.27) provides a mechanism for recycling the NADH produced by glycolysis in the absence of oxygen, and the resulting accumulation of lactate is a characteristic feature of the metabolic response of many plant tissues to anoxia (for review, see Ratcliffe 1995). Analysing the contribution of LDH to this response is complicated by several factors. First, the LDH holoenzyme is a tetramer with complex pH-dependent kinetic properties ( Mulcahy & O’Carra 1997). Secondly, in many plant species (but not all ( O’Carra & Mulcahy 1997)) the subunits of the holoenzyme are encoded by two different genes ( Germain & Ricard 1997) giving rise to a set of five possible isozymes ( Asker & Davies 1984; Hoffman, Bent & Hanson 1986; Hoffman & Hanson 1986). This situation is analagous to that found in mammalian tissues, where one subunit predominates in liver and skeletal muscle and the other in heart tissue ( Everse & Kaplan 1973), and it raises the question of the functional significance of the different forms. Thirdly, LDH activity is inducible under anoxia ( Hoffman et al. 1986 ; Good & Paetkau 1992; Germain & Ricard 1997) and oxygen deprivation has a differential effect on the two genes. Thus, in tomato, ldh1 is inducible by oxygen deficit whereas ldh2 is not ( Germain, Raymond & Ricard 1997; Germain & Ricard 1997).
The role of the different LDH isozymes in lactate metabolism is not yet fully understood. One view ( Mulcahy & O'Carra 1997) is that the absence of isozymes in some species, for example leek, onion and soybean, argues against any functional importance for the multiple forms found in other species. However, this view is not widely accepted and on the assumption that different forms of LDH may be responsible for lactate metabolism under different physiological conditions various attempts have been made to assign particular functions to specific isozymes. Initially, protein purification and kinetic analysis were used to infer possible functions from the kinetic properties of particular isozymes, and on this basis it was argued that particular isozymes in potato tubers were best suited for either an oxidative role or a reductive role ( Asker & Davies 1984). More recently, on the basis of correlations between gene expression, LDH activity and lactate production in tomato roots, it was argued that the constitutively expressed gene ldh2 is responsible for the transient production of lactate following the onset of anoxia, and that the inducible gene ldh1 could be important in the oxidation of lactate under anoxia ( Germain et al. 1997 ). No general picture has emerged from these studies, and the extent to which different isozymes contribute to lactate metabolism during normoxia, hypoxia, anoxia and the return to normoxia remains unclear.
The successful cloning and identification of a selection of plant LDH genes ( Hondred & Hanson 1990; Good & Paetkau 1992; Germain et al. 1997 ; Germain & Ricard 1997) opened up the possibility of using transgenic plants to investigate the role of LDH in lactate metabolism. Over-expression of a barley LDH in tomato root cultures was attempted first, but this had no effect on fermentative metabolism, leading to the conclusion that LDH had a very low flux control coefficient for lactate accumulation in the chosen tissue ( Rivoal & Hanson 1994). Subsequently there have been various attempts to introduce antisense LDH constructs into plants (e.g. Germain, Saglio & Ricard 1995; Dennis et al. 2000), and we have now applied this strategy to potato, using a cDNA for a constitutively expressed potato LDH gene in the antisense orientation to produce plants with reduced LDH activities. This article reports experiments in which the activity of a particular LDH subunit was decreased by preferentially decreasing the expression of one of two closely related ldh genes. The effect of decreasing this LDH on lactate metabolism in tubers was investigated under three conditions: (i) normoxia; (ii) anoxia; and (iii) return to normoxia after a period of anoxia. The function of the LDH subunit was considered and its contribution to LDH metabolism under different conditions was assessed.
MATERIALS AND METHODS
Materials
All enzymes and chemicals were from Roche Molecular Biochemicals (Lewes, Sussex, UK) and Sigma-Aldrich (Poole, Dorset, UK). Radiochemicals were from Amersham International (Aylesbury, Bucks, UK). Compost was from EA Goundry & Son Ltd. (Duns Tew, Oxford, UK).
Plant growth
Potato plants (Solanum tuberosum L. cv. Desirée) were grown by planting sprouted tubers in 150-mm-diameter pots containing a mixture of sand and compost (1 : 2). The plants were maintained in a greenhouse at 16–25 °C with a 16 h photoperiod of natural daylight supplemented to give a minimum irradiance of 200 μmol photons m−2 s−1.
Transformation of potato plants
The DNA sequence used to generate the antisense-LDH gene construct was derived from a previously isolated potato tuber cDNA clone encoding LDH (Genbank accession no. AF067859). An EcoRI restriction fragment containing 26-bp of 5′ untranslated sequence, the full coding region, and 99 bp of the 3′ untranslated region was cloned into the vector pJIT62K in the antisense orientation. Plasmid pJIT62K was formed by inserting a linker containing a KpnI restriction site into EcoRV-digested pJIT62 (P. Mollineaux, John Innes Centre, Norwich). The modified plasmid contained the CaMV 35S promoter and terminator separated by a multiple cloning site and flanked by KpnI restriction sites. The resulting chimeric gene construct was then excised by digestion with KpnI and subcloned into the binary vector pBINPLUS ( Van Engelen et al. 1995 ) to yield the plasmid pLDH2122. This plasmid was introduced into Agrobacterium tumefaciens strain LBA4404, containing pAL4404, by direct transformation ( Höfgen & Willmitzer 1988). The gene construct was transferred from pLDH2122 to potato cv. Desirée by Agrobacterium-mediated transformation of stem segments as described in Twell & Ooms (1987), with the exception that after 14 d incubation on LRN medium, the stem segments were transferred onto 3C5ZR medium to induce shoot formation ( Sheerman & Bevan 1988). Comparable stem segments were infected with Agrobacterium containing unmodified pBINPLUS to generate transformed control plants.
Enzyme assays
Tuber tissue was processed and extracted for assay of enzymes exactly as described in Sweetlove, Burrell & ap Rees (1996). Unless otherwise stated, enzymes were assayed at 25 °C in 0·2 mL of the following reaction mixtures by measuring the change in absorbance at 340 nm. Lactate dehydrogenase (EC 1.1.1.27): 100 m M Tris (pH 7·1), 0·2 m M NADH, 10 m M pyruvate. Pyruvate decarboxylase (EC 4.1.1.1): 100 m M Tris (pH 8·5), 0·2 m M NADH, 30 m M pyruvate, 1 unit alcohol dehydrogenase. Alcohol dehydrogenase (EC 1.1.1.1): 100 m M Tris (pH 8·9), 0·2 m M NAD, 100 m M ethanol. 6-phosphofructokinase (EC 2.7.1.11) and pyrophosphate fructose-6-phosphate 1-phosphotransferase (EC 2.7.1.90) were assayed as described in Kruger, Hammond & Burrell (1988). Fructose-1,6-bisphosphate aldolase (EC 4.1.2.13), enolase (EC 4.2.1.11) and pyruvate kinase (EC 2.7.1.40) were assayed as described in Burrell et al. (1994) . Sucrose synthase (EC 2.4.1.13) was assayed as described in Sweetlove et al. (1996) .
Metabolite assays
Frozen tuber powder was extracted with trichloroacetic acid ( Weiner, Stitt & Heldt 1987). Metabolites were assayed spectrophotometrically by the methods described in the following references: lactate ( Gutman & Wahlesfeld 1974), ethanol ( Witt 1974), phosphoenolpyruvate and pyruvate ( Burrell et al. 1994 ), ATP and ADP ( Hatzfeld & Stitt 1990), NADH and NAD+ ( Passoneau & Lowry 1993). Malate, citrate and succinate were measured by HPLC using an Aquasep C8, 5 μm pore size, 25 × 0·46 cm column (ES Industries, Berlin, Germany). Carboxylic acids were eluted isocratically in a mobile phase of 50 m M KH2PO4 at a flow rate of 1 mL min−1 and detected by absorbance at 210 nm. Malate, citrate and succinate were eluted with the following retention times: malate (5·0 min), citrate (8·2 min) and succinate (10·0 min).
Measurement of LDH isozymes
LDH isozymes were separated by non-denaturing polyacrylamide gel electrophoresis (PAGE) and visualized by activity staining as described in Hanson & Jacobsen (1984). Frozen potato tissue was extracted and desalted exactly as described in the section ‘Enzyme assays’. The extract was diluted two-fold by the addition of an equal volume of 50% glycerol (v/v), 0·01% (w/v) bromophenol blue and applied to a 7% polyacrylamide gel containing no sodium dodecyl sulphate. The LDH activity on the gel was visualized using a nitro-blue tetrazolium-linked stain.
Incubation of tuber discs with [U-14C]glucose
Incubation of potato tuber discs with [U-14C]glucose and subsequent analysis of the distribution of label was exactly as described by Sweetlove et al. (1999) .
Measurement of the rate of lactate metabolism
Potato tuber discs were incubated with 37 kBq [U-14C]lactate (6·44 MBq mmol−1) under the conditions described in Sweetlove et al. (1999) but replacing [U-14C]glucose with [U-14C]lactate. After 2 h, the tissue was extracted and the distribution of label was determined as described in Sweetlove et al. (1999) .
Determination of the specific activity of the lactate pool
The labelling of the internal tuber lactate pool was determined as follows: the acidic fraction obtained after ion exchange chromatography of the trichloroacetic acid-soluble fraction was reduced to dryness in vacuo and resuspended in 1 mL of 10 m M Tris (pH 8·0), 2 m M NAD, 2 m M glutamate, 10 units of lactate dehydrogenase and 10 units of glutamate-pyruvate transaminase. After incubating at 37 °C for 2 h, the pH was adjusted by the addition of 60 μL 1 M sodium acetate buffer (pH 5·0) and the basic compounds were separated by ion exchange chromatography ( Quick et al. 1989 ). The amount of 14C in the basic fraction was determined by liquid scintillation counting and ascribed to [14C]lactate. The amount of lactate was measured in the trichloroacetic acid-soluble fraction as described in the section ‘Metabolite assays’.
RESULTS AND DISCUSSION
Characterization of transgenic plant lines
To decrease the activity of LDH, the complete coding region from one of two closely related potato LDH cDNA clones (Genbank accession no. AF067859) was placed in the antisense orientation under the control of the CaMV 35S promoter and used to transform potato cv. Desirée through Agrobacterium-mediated infection of stem segments. This cDNA showed 96% identity to the constitutively expressed ldh2 and 81% identity to the inducible ldh1 from tomato. Forty-four independent kanamycin-resistant shoots were selected. Microtubers from 10 of the lines derived from these shoots showed LDH activities that were significantly lower than those in control lines transformed with unmodified pBINPLUS ( Fig. 1). The assay for LDH was fully optimized and the extraction procedure used has previously been shown to give good recovery for a range of enzyme activities ( Sweetlove et al. 1996 ). Two LDH-antisense (LDH-311 and LDH-312) and two pBINPLUS control lines were selected for further analysis. Tubers from the primary transformants were propagated to obtain second generation plants. Subsequent generations were obtained by propagation from tubers of greenhouse plants grown in soil. The decrease in LDH activity in the two antisense lines (44–66% of the control values) was retained in developing tubers from soil-grown plants ( Table 1). This decrease in activity was associated with a large decrease in two of the three isozymes of LDH resolved in extracts of developing tubers by nondenaturing PAGE and activity staining ( Fig. 2). The five possible isozymes of LDH have been numbered 1–5, with isozyme 1 having the lowest molecular mass and isozyme 5 the highest ( Hoffman et al. 1986 ). The three isozymes in normoxic tubers correspond to isozymes 1, 2 and 3 (see Fig. 2). Isozymes 1 and 2 were preferentially decreased in the transgenic lines. We also assayed the activities of a range of enzymes associated with carbohydrate metabolism and respiration ( Table 1). All assays were optimized and an extraction procedure was used that had previously been shown to give good recovery of a range of enzyme activities ( Sweetlove et al. 1996 ). None of the activities of these enzymes in developing tubers of the transgenic lines were significantly different from those of control lines. This suggests that any changes in metabolism are likely to be the result of the changes in LDH rather than of changes in other enzymes, and these lines are therefore suitable for further analysis of the effect of reduced LDH on lactate metabolism.
Maximum catalytic activity of lactate dehydrogenase in microtubers of control and transgenic plants. Microtubers were harvested from plants grown in tissue culture. Within 30 min of their removal from the plant, the microtubers were immersed in liquid nitrogen and ground to a fine powder. The frozen powder was stored at − 80 °C until required. Samples of the frozen powder were extracted, clarified by centrifugation, desalted and then assayed for enzyme activities. Each value represents the activity in an individual microtuber and is expressed relative to that in control lines grown under the same conditions. The dotted lines indicate the 95% confidence limits for LDH activity in eight independently transformed control plants. , lines LDH-311 and LDH-312 which were used in subsequent metabolic analysis.
| Enzyme activity (nmol min−1 g−1 FW) in the following lines | ||||
|---|---|---|---|---|
| Control-113 | Control-311 | LDH-311 | LDH-312 | |
| Lactate dehydrogenase | 196 ± 6 | 171 ± 9 | 82 ± 7 * | 122 ± 20 * |
| Pyruvate decarboxylase | 13 ± 5 | 7 ± 2 | 9 ± 3 | 6 ± 1 |
| Alcohol dehydrogenase | 529 ± 27 | 575 ± 19 | 486 ± 36 | 480 ± 20 |
| Phosphofructokinase | 60 ± 6 | 57 ± 1 | 53 ± 4 | 59 ± 1 |
| PFP | 274 ± 19 | 317 ± 19 | 298 ± 10 | 331 ± 7 |
| Aldolase | 216 ± 22 | 202 ± 8 | 201 ± 8 | 198 ± 12 |
| Enolase | 1500 ± 113 | 1539 ± 120 | 1443 ± 67 | 1541 ± 61 |
| Pyruvate kinase | 726 ± 63 | 661 ± 23 | 687 ± 47 | 661 ± 16 |
| Sucrose synthase | 1369 ± 267 | 1548 ± 253 | 1440 ± 81 | 1536 ± 34 |
- * Significantly different from mean of pooled control values (t-test, P < 0·05).
Isozymes of LDH in developing tubers of control and transgenic plants. Proteins were extracted from developing, normoxic control and transgenic tubers and separated by nondenaturing polyacrylamide gel electrophoresis. LDH proteins were visualized by staining for LDH activity using nitro blue tetrazolium. Lanes 1–4 contain equivalent amounts of tuber tissue (1 mg) and lanes 5–8 contain equal amounts of LDH activity (0·4 nmol min−1). Lanes 1 and 5, control-113; lanes 2 and 6, control-311; lanes 3 and 7, LDH-311; lanes 4 and 8, LDH-312.
Effect of reduced LDH activity on tuber lactate metabolism under normoxic conditions
Developing, normoxic potato tubers contain appreciable amounts of lactate (approximately 100 nmol g−1 FW; data not shown). If isozymes 1 and 2 of LDH are responsible for this lactate accumulation, one would predict that there would be a decreased amount of lactate in the transgenic tubers. We assayed lactate content and LDH activity in replicate samples of control and transgenic tubers ( Fig. 3). Surprisingly, we found that the transgenic tubers contained up to two-fold more lactate than control tubers, and that there was a strong inverse correlation between lactate content and LDH activity. This suggests that either isozyme 1 or 2 of LDH, or both, are responsible for the oxidation of lactate to pyruvate rather than the expected reduction of pyruvate to lactate.
The effect of decreased LDH activity on lactate content in developing normoxic tubers. Samples from developing normoxic control and transgenic tubers were frozen in liquid nitrogen and ground to a fine powder under liquid nitrogen. Replicate samples of the frozen powder were extracted and assayed either for LDH activity or lactate content. Each point on the graph represents the LDH activity and lactate content of a single tuber. ▪, control-113; ●, control-311; □, LDH-312; ○, LDH-311.A line was fitted to the points by linear regression analysis.
Potato tuber discs can metabolize externally supplied lactate in a manner that is consistent with lactate entering metabolism via oxidation to pyruvate by LDH (R. Dunford, A. Roscher, R.G. Ratcliffe and N.J. Kruger, unpublished results). If isozymes 1 and 2 of LDH catalyse this process, discs from the transgenic tubers should show a lower rate of metabolism of externally supplied lactate. To test this, we supplied [U-14C]lactate to tuber discs cut from developing, normoxic control and transgenic tubers. The tuber discs were also supplied with 200 m M glucose to prevent substrate starvation, since the rate of lactate metabolism is insufficient to fuel normal metabolism. After 2 h, metabolism was quenched by the addition of 16% trichloroacetic acid and the distribution of 14C was determined ( Table 2). We were able to recover more than 90% of the 14C at each stage of the analysis (data not shown). In the control discs, most of the label (approximately 70%) was recovered either as CO2 or in the acidic and basic fractions, a pattern consistent with oxidation of lactate to pyruvate, entry of this pyruvate into the TCA cycle and the subsequent production of organic acids and amino acids. We found no significant difference between the distribution of 14C in the transgenic and control discs. Crucially, however, the total amount of label metabolized in the transgenic discs was significantly lower than in the control discs. We measured the specific activity of the internal lactate pool at the end of the experiment and used this information, together with the total metabolized label, to calculate the rate of lactate oxidation. The following estimates (nmol min−1 g−1 FW; mean of three experiments ± SEM) were obtained: control-113, 20 ± 2; control-311, 20 ± 3; LDH-311, 12 ± 1; LDH-312, 10 ± 3. Thus, the rate of oxidation of lactate to pyruvate in transgenic discs was approximately 50% of that in control discs. We therefore conclude that isozymes 1 and 2 of LDH are responsible for the oxidation of lactate to pyruvate and that these isozymes have a high degree of control over this flux in normoxic conditions.
| 14C recovered per fraction (as percentage of totalmetabolized) from discs of the following lines | ||||
|---|---|---|---|---|
| Control-113 | Control-311 | LDH-311 | LDH-312 | |
| CO2 | 32 ± 1 | 29 ± 3 | 34 ± 10 | 36 ± 6 |
| Water-soluble substances | 59 ± 0 | 66 ± 3 | 63 ± 9 | 59 ± 7 |
| Neutral components | 22 ± 1 | 28 ± 1 | 20 ± 3 | 22 ± 2 |
| Acidic components | 15 ± 1 | 16 ± 2 | 15 ± 3 | 20 ± 3 |
| Basic components | 22 ± 1 | 22 ± 1 | 18 ± 3 | 13 ± 2 * |
| Water-insoluble substances | 9 ± 1 | 5 ± 2 | 3 ± 1 | 5 ± 1 |
| Starch | ND | ND | ND | ND |
| Protein | ND | ND | ND | ND |
| 14C metabolized (Bq g−1 FW) | 1278 ± 14 | 1134 ± 36 | 741 ± 155 * | 781 ± 110 * |
- * Significantly different from mean of pooled control values (t-test, P < 0·05). ND, Not detected.
Consequences of elevated lactate content on tuber metabolism under normoxic conditions
The transgenic tubers contained up to twice as much lactate as control tubers. To investigate the effect of increased lactate content on metabolism, we supplied [U-14C]glucose to discs of developing, normoxic control and transgenic tubers and determined the distribution of 14C after 2 h ( Table 3). More than 90% of the 14C was recovered at each stage of the analysis (data not shown). The distribution of 14C was similar to that previously reported in comparable studies of potato tuber discs ( Geiger, Stitt & Geigenberger 1998; Sweetlove et al. 1999 ). We found no significant differences in the distribution of label between control and transgenic discs suggesting that the fate of metabolized glucose is not affected by an increased lactate content. However, the total amount of 14C uptake by the transgenic discs was significantly lower than that by control discs (in kBq: control-113, 12·6 ± 1·1; control-311, 11·3 ± 0·4; LDH-311, 8·7 ± 0·1; LDH-312 8·3 ± 0·5). It is not clear why this should be the case, but the decrease in glucose uptake is sufficient to account for a significant decrease in the amount of [14C]glucose metabolized in the transgenic discs.
| 14C recovered per fraction (as percentage of totalmetabolized) from discs of the following lines | ||||
|---|---|---|---|---|
| Control-113 | Control-311 | LDH-311 | LDH-312 | |
| CO2 | 1·5 ± 0·4 | 1·5 ± 0·1 | 1·1 ± 0·1 | 1·7 ± 0·2 |
| Water-soluble substances | 51·7 ± 7·0 | 61·5 ± 5·1 | 49·3 ± 4·9 | 55·8 ± 3·1 |
| Neutral components | 34·8 ± 7·3 | 50·2 ± 11·1 | 32·4 ± 6·7 | 39·6 ± 5·8 |
| Sucrose | 16·5 ± 5·6 | 28·2 ± 9·0 | 10·7 ± 5·5 | 17·2 ± 3·6 |
| Fructose | 5·1 ± 0·4 | 8·7 ± 0·8 | 7·1 ± 0·4 | 5·7 ± 1·3 |
| Acidic components | 3·6 ± 0·6 | 3·9 ± 0·1 | 3·4 ± 0·0 | 4·2 ± 0·1 |
| Basic components | 3·5 ± 0·2 | 3·8 ± 0·4 | 3·4 ± 0·3 | 3·7 ± 0·1 |
| Water-insoluble substances | 46·7 ± 6·6 | 37·0 ± 5·0 | 49·6 ± 5·0 | 42·6 ± 2·9 |
| Starch | 39·0 ± 9·6 | 30·6 ± 2·4 | 44·5 ± 2·5 | 37·8 ± 1·3 |
| Protein | ND | ND | ND | ND |
| Metabolized 14C (KBq) | 8·9 ± 1·1 | 7·2 ± 0·4 | 5·9 ± 0·2 * | 5·3 ± 0·3 * |
- * Significantly different from mean of pooled control values (t-test, P < 0·05). ND, Not detected.
Although we were able to find no evidence of major changes in the fate of metabolized glucose in the transgenic tubers, it is possible that there are more subtle changes in specific metabolite pools. For example, one might expect perturbations in the amounts of the other metabolites that participate in the reaction catalysed by LDH. However, we could find no significant differences in the steady-state levels of pyruvate, NAD+ or NADH between developing normoxic control and transgenic tubers ( Table 4). Nor were we able detect any changes in the amounts of other glycolytic intermediates such as phosphoenolpyruvate, in the intermediates of the TCA cycle such as malate, citrate and succinate, or in the adenylate pools ( Table 4).
| Metabolite amount (nmol g−1 FW) in the following lines: | ||||
|---|---|---|---|---|
| Control-113 | Control-311 | LDH-311 | LDH-312 | |
| Pyruvate | 145 ± 15 | 152 ± 16 | 153 ± 15 | 159 ± 35 |
| Phosphoenolpyruvate | 275 ± 49 | 275 ± 36 | 258 ± 24 | 341 ± 24 |
| Malate | 4839 ± 385 | 6473 ± 735 | 5812 ± 735 | 5909 ± 857 |
| Citrate | 2055 ± 293 | 1565 ± 256 | 1615 ± 171 | 1247 ± 62 |
| Succinate | 177 ± 14 | 194 ± 52 | 119 ± 23 | 187 ± 5 |
| ATP | 26 ± 7 | 35 ± 5 | 32 ± 2 | 36 ± 11 |
| ADP | 20 ± 1 | 16 ± 3 | 22 ± 1 | 14 ± 1 |
| ATP/ADP | 1·4 ± 0·4 | 2·2 ± 0·2 | 1·5 ± 0·2 | 2·5 ± 0·6 |
| NADH | 3·6 ± 0·3 | 4·1 ± 0·7 | 4·7 ± 0·8 | 3·9 ± 0·1 |
| NAD | 17 ± 1 | 22 ± 2 | 15 ± 1 | 19 ± 2 |
| NADH/NAD | 0·22 ± 0·02 | 0·20 ± 0·04 | 0·32 ± 0·07 | 0·21 ± 0·02 |
Effect of reduced activity of LDH on the accumulation of lactate and ethanol during anoxia
Having demonstrated that isozymes 1 and 2 of LDH are responsible for the oxidation of lactate, and that a reduction in these isozymes leads to an increase in lactate content in normoxic tubers, we were interested to see whether a reduction in these isozymes also affected the accumulation of lactate in tubers under anoxia and particularly the subsequent metabolism of this lactate upon return to normoxic conditions. We investigated this by measuring the lactate content of tubers that were exposed to anoxia and subsequently returned to normoxia. Initial measurements were made on discs cut from developing tubers. The discs were incubated in an anoxic medium and sampled at time intervals for lactate determinations. However, we found that less than 10% of the accumulated lactate was recovered in the tuber disc samples, with the remaining 90% present in the incubation medium, presumably as a result of excretion from the tuber tissue (data not shown). This is in sharp contrast to the large amounts of lactate that accumulate in whole tubers under anoxic conditions ( Sieber & Brändle 1991) and raises the possibility that the metabolism of lactate in the tuber disc system may differ from that in the whole tuber.
We therefore decided to switch our attention to lactate metabolism in intact tubers. Developing tubers were removed from control and transgenic plants and stored at 4 °C for 10 weeks. The tubers were then placed in an atmosphere containing less than 0·05% oxygen (henceforth referred to as ‘anoxia’) and maintained at 25 °C. Longitudinal cores of 1 cm diameter were taken from tubers after 7 d anoxia and then at intervals during a subsequent 48 h period of normoxia. The content of both lactate and ethanol were measured ( Fig. 4). In keeping with previously published work ( Sieber & Brändle 1991) both lactate and ethanol accumulated to high levels in the control tubers during anoxia. The reduction of LDH activity in the transgenic lines appeared to have little effect on this accumulation of fermentative end products. When the tubers were returned to normoxic conditions, ethanol ceased to accumulate and remained at a constant level. A similar pattern was observed in both control and transgenic plants. In contrast, lactate content fell rapidly upon return to normoxic conditions, with approximately 75% of the accumulated lactate being metabolized in 48 h. This corresponds to a rate of lactate metabolism of approximately 5 nmol min−1 g−1 FW, a rate well below the maximum catalytic activity of LDH ( Fig. 5). We found that this metabolism of lactate was not reduced in the transgenic tubers with decreased activity of isozymes 1 and 2 of LDH. In fact, the rate of metabolism of lactate in the transgenic tubers was slightly faster, with the lactate content falling to between 2 and 6 μmol (g FW)−1 after 48 h of normoxia in comparison to 10 μmol (g FW)−1for the control tubers. This was confirmed by fitting the data for each line to a two-parameter single exponential decay curve, yielding the following first order rate constants (h−1): control-113, 0·027 ± 0·012; control-311, 0·027 ± 0·008; LDH-311, 0·058 ± 0·014; LDH-312, 0·076 ± 0·035). Given that we have shown that isozymes 1 and 2 of LDH are responsible for the oxidation of lactate in developing tubers, it was surprising to find a similar rate of mobilization of lactate that had accumulated during anoxia in control and transgenic tubers. One might expect that a decrease in isozymes 1 and 2 would reduce the rate of oxidation of lactate. There are three possible reasons why this prediction has not been borne out. First, the oxidation of lactate after a period of anoxia may not be catalysed by LDH. The observation, however, that the decrease in lactate content that occurs when anoxic potato tubers are returned to normoxic conditions is accompanied by a transient increase in pyruvate content would strongly suggest that the oxidation of lactate is catalysed by LDH ( Barker & Mapson 1963). Secondly, it is possible that LDH activity is not decreased in transgenic tubers that had been cold-stored and is therefore not decreased during this experiment. Thirdly, it is conceivable that isozymes 1 and 2 of LDH are responsible for the oxidation of LDH during normoxic conditions but not during anoxia or during a subsequent 48 h normoxic period.
Lactate and ethanol contents of control and transgenic tubers during anoxia and a subsequent period of normoxia. Control and transgenic tubers that had been removed from the plant and stored at 4 °C for 10 weeks were placed in an anaerobic workbench (with an atmosphere containing less than 0·05% O2 at 25 °C). After 7 d, the tubers were removed and placed in air. Tubers were taken for sampling after 7 d anoxia and then at intervals during a subsequent 48 h period of normoxia. The contents of lactate and ethanol in the samples were determined. Each point represents a single tuber. ▪, control-113; ●, control-311; □ LDH-312; ○, LDH-311.
LDH activity of control and transgenic tubers during anoxia and a subsequent period of normoxia. Experimental material was exactly as detailed in the legend to Fig. 4. Tuber material was extracted, clarified by centrifugation and desalted prior to assay of LDH.
In order to test whether LDH activity was decreased in the transgenic tubers under the conditions of the above experiment, we measured LDH activity in the same tubers in which lactate and ethanol were assayed ( Fig. 5). LDH activity was significantly lower in the transgenic tubers and remained so throughout the experiment. We found no evidence that LDH activity in either the control or transgenic lines increased during the anoxic treatment. Thus, large amounts of lactate are produced during anoxia without an increase in the activity of LDH, suggesting that LDH activity exerts no measurable control over this process. However, there was a large increase in LDH activity when the anoxically treated tubers were returned to normoxia ( Fig. 5). This increase in LDH activity was of a similar magnitude (approximately 200 nmol min−1 g−1 FW) in both the control and the transgenic lines. We conclude that a reduction in isozymes 1 and 2 of LDH does not influence the accumulation of lactate during anoxia. The re-oxidation of this lactate during a subsequent period of normoxia is associated with an induction of LDH activity. This induction of LDH activity and re-oxidation of lactate are unaffected by the reduction of isozymes 1 and 2 of LDH in the transgenic tubers.
Which isozymes of LDH are responsible for the oxidation of lactate after a period of anoxia?
The induction of LDH activity when anoxic tubers are returned to normoxia is likely to be associated with isozymes other than 1 and 2, since the induction occurs equally in transgenic tubers, which contain decreased amounts of isozymes 1 and 2 ( Fig. 5). We investigated the isozyme composition of LDH during anoxia and a subsequent period of normoxia by activity staining of native PAGE ( Fig. 6). In keeping with the measurement of total LDH activity ( Fig. 5), the isozyme composition of LDH did not change during anoxia. This contrasts with the induction of LDH activity and change in isozyme composition in cereal roots during anoxia ( Hoffman et al. 1986 ) and demonstrates that induction of LDH by anoxia is not universal in all plant species. We have found that LDH activity increases in potato tubers during flooding (R. Dunford, A. Roscher, R.G. Ratcliffe and N.J. Kruger, unpublished results) consistent with the idea that LDH is induced during hypoxia, but not during complete absence of oxygen. It is evident that an increase in LDH activity is not a prerequisite for an increase in lactate accumulation. During a subsequent period of normoxia, however, two new isozymes of LDH were observed ( Fig. 6). These correspond to isozymes 4 and 5 of LDH and their appearance paralleled the increase in total LDH activity and decrease in lactate content. In the transgenic lines, isozymes 1 and 2 remained low throughout. As in the control tubers, the isozyme pattern did not change during anoxia, but there was a clear induction of isozymes 4 and 5 during a subsequent period of normoxia (although perhaps not to the same extent as occurred in the control tubers). We suggest that isozymes 4 and 5 are induced after a period of anoxia and function to oxidize the accumulated lactate to pyruvate. This proposal is consistent with the kinetic properties of these isozymes ( Asker & Davies 1984). The oxidation of the lactate that accumulated in cold-stored tubers under anoxia was unaffected by the decrease in isozymes 1 and 2 in the transgenic tubers, even though we have previously demonstrated the oxidation of lactate by these isozymes during normoxic conditions. We propose that, although isozymes 1 and 2 can function to oxidize lactate in vivo, their contribution to the metabolism of lactate after a period of anoxia is negligible.
Isozymes of LDH in developing tubers of control and transgenic plants during anoxia and a subsequent period of normoxia. Experimental material was exactly as detailed in the legend to Fig. 4. Proteins were extracted from tuber tissue and separated by nondenaturing polyacrylamide gel electrophoresis. LDH proteins were visualized by staining for LDH activity using nitro blue tetrazolium. All lanes contain equivalent amounts of tuber protein. Lanes 1–4, control-113; lanes 5–8, LDH-311. Lanes 1 and 5, normoxic control. Lanes 2 and 6, 7 d anoxia. Lanes 3 and 7, 7 d anoxia + 1 d normoxia. Lanes 4 and 8, 7 d anoxia + 2 d normoxia.
CONCLUSION
We have demonstrated that isozymes 1 and 2 of LDH function in normoxic potato tubers to oxidize lactate to pyruvate. Thus, a specific decrease of these isozymes in transgenic tubers leads to an increase in lactate content but has no major effect on tuber metabolism. In cold-stored tubers these isozymes of LDH appear to have no essential role in the accumulation of lactate during anoxia, nor in the oxidation of this lactate during a subsequent period of normoxia. Instead, the anoxic accumulation of lactate occurs without any change in LDH activity, whereas the oxidation of this lactate upon return to normoxia is facilitated by an induction in the activity of isozymes 4 and 5 of LDH. This work demonstrates the complexity of regulation of lactate metabolism by LDH and illustrates the different roles of the different isozymes of LDH.
ACKNOWLEDGMENTS
This research was supported by the Biotechnology and Biological Sciences Research Council, UK (grant numbers GR/J73612 and 43/P09460). We thank Dr P. Mullineaux, John Innes Centre, Norwich, UK for generously providing the original plasmid pJIT62.
