Mitochondrial AOX (alternative oxidase) is the terminal oxidase of the CN (cyanide)-resistant alternative respiratory pathway in plants. To investigate the role of the tobacco AOX gene (NtAOX1a) (where Nt is Nicotiana tabacum) under deleterious conditions which could induce ROS (reactive oxygen species) accumulation, we generated and characterized a number of independent transgenic tobacco (N. tabacum) lines with altered NtAOX1a gene expression and AP (alternative pathway) capacity. AOX efficiently inhibited the production of low-temperature-induced H2O2 and might be a major enzyme for scavenging H2O2 at low temperature. Furthermore, NtAOX1a may act as a regulator of KCN-induced resistance to TMV (tobacco mosaic virus) through the regulation of H2O2. Notably, a moderate accumulation of H2O2 under the control of NtAOX1a was crucial in viral resistance. Analysis of seed germination indicated an important role for NtAOX1a in germination under H2O2-induced oxidative stress when the CP (cytochrome pathway) was inhibited. These results demonstrate that NtAOX1a is necessary for plants to survive low temperature, pathogen attack and oxidative stress by scavenging ROS under these adverse conditions when the CP is restricted.
The mitochondrial respiratory chain in higher plants consists of the CN (cyanide)-sensitive CP (cytochrome pathway) and the CN-resistant AP (alternative pathway) . AOX (alternative oxidase) is the terminal oxidase of the AP and is encoded by a small family of nuclear genes . As a part of the electron transport chain, AOX can catalyse oxygen-dependent oxidation of ubiquinol . The expression of AOX genes is induced in response to diverse biotic and abiotic stresses . It has been suggested that AOX plays a crucial role in maintaining homoeostasis under varying growth conditions  and in protecting plants against the lethal effects of ROS (reactive oxygen species) [5,6].
All organisms produce a range of ROS and mitochondria are a major source of ROS in eukaryotic cells. It is known that ROS play a dual role depending on their accumulation levels. High intracellular concentration of ROS leads to extensive cell injury or death. A moderate accumulation of ROS functions as a key inducer for secondary programmed metabolism, defence signals and activation of MAPKs (mitogen-activated protein kinases), leading to environmental stress tolerance . Several biotic and abiotic stresses increase ROS production in various tissues, and frequently result in a concomitant increase in AOX expression in higher plants . The function of AOX in preventing ROS production has been shown in tobacco cells in culture  and isolated mitochondria . Also, in transgenic Arabidopsis plants with antisense AOX cDNA, there is an increase in the level of ROS when the CP is inhibited by KCN treatment .
AOX is thought to contribute to the acclimation of respiration to low temperature . The enhancement of the AP capacity in higher plants by exposure to low temperature is partly due to enhanced expression of the AOX genes at the transcriptional level . Exposure of plants to low temperature impaired electron flow through the CP and caused an increase in H2O2 production . A down-regulation of the CP is accompanied by an increase in AOX capacity .
AOX has been implicated in plant defence pathway against viruses. SA (salicylic acid) induces increased viral resistance and AOX expression [14,15]. It has been reported that SA-induced TMV (tobacco mosaic virus) resistance resulted from the activation of multiple mechanisms, a subset of which are inducible by AA (antimycin A) and influenced by AOX . TMV vector-driven transient high-level expression of AOX enhanced virus spread and symptoms of infection in plants . Although some results suggested that AOX was not a critical component of plant viral resistance, it might play a role in the HR (hypersensitive response) . In spite of these previous studies showing that AOX is involved in viral resistance, to date, the underlying mechanism of the involvement of AOX in signalling during pathogen resistance has not been fully elucidated.
Germination of plant seeds after imbibition is a dynamic process requiring the activation of a number of metabolic enzymes, which is accompanied by a rapid increase in oxygen consumption through mitochondrial respiration. It has been reported recently that the expression profiles of respiratory components are associated with mitochondrial biogenesis during germination and early seedling growth in wheat . Furthermore, AOX can support embryo germination and early seedling growth in conjunction with Complex I in wheat when the CP is restricted .
To elucidate the relationship of AOX, ROS and adverse stresses, we generated and characterized a series of transgenic tobacco lines with altered NtAOX1a (where Nt is Nicotiana tabacum) expression. The results of the present study indicate that NtAOX1a plays a crucial role in protecting plants against a variety of abiotic and biotic stresses which lead to ROS accumulation.
MATERIALS AND METHODS
Plant materials and growth conditions
Seeds from the tobacco cultivar Samsun (NN genotype, TMV resistant) and from Samsun (NN genotype) transgenic lines were germinated in soil and maintained under the following growth conditions: 16 h light/8 h dark photoperiod (light intensity of approx. 200 μmol·m−2·s−1), constant air temperature of 26°C and relative humidity of 60–75%.
Vector construction and plant transformation
The NtAOX1a coding sequence was obtained from NN-type Samsun tobacco by RT-PCR (reverse transcription-PCR) using primers based on the published AOX1a sequence from Bright Yellow tobacco .
The cDNA was inserted either in sense or antisense orientation under the control of CaMV35S (35S cauliflower mosaic virus promoter) in the expression cassette of pBI121. An empty transformation vector control was also generated. The expression cassettes were introduced into Agrobacterium tumefaciens (strain LBA4404) and then used for tobacco leaf disc transformation . Primary transformants were selected on kanamycin (100 mg/l) plates before they were transferred into soil.
Northern blot analysis
NtAOX1a transcript levels were assessed by Northern blot analysis using methods previously described . These experiments were repeated twice using independent samples and representative results are shown.
Mitochondria used for Western blot analysis were isolated from 50 g (fresh weight) of tobacco leaf tissue. After harvest, the leaves were homogenized using a mortar and pestle in 120 ml of grinding medium [20 mM Hepes/Tris (pH 7.6), 0.4 M sucrose, 5 mM EDTA, 0.6% (w/v) insoluble PVPP (polyvinylpolypyrrolidone), 0.3% (w/v) BSA and 25 mM potassium metabisulfite]. The subsequent material was then filtered through eight layers of gauze. The method was performed according to standard protocols described previously .
Western blot analysis
The proteins isolated from mitochondria of wild-type and transgenic plants were separated by SDS/PAGE according to the method of Laemmli , and were subsequently electrotransferred on to a PVDF membrane. Immunoblot analysis was performed as previously described . A monoclonal antibody against AOX was developed previously in our laboratory and used at 1:100 (v/v) dilution. The anti-AOX antibody was detected using a goat anti-rat IgG (Dingguo, Beijing, China) conjugated to HRP (horseradish peroxidase) and was used at 1:5000 (v/v) dilution. The binding was visualized using a chemiluminescent HRP substrate.
Measurement of respiratory capacity
The respiration capacity was measured by obtaining cells (with cell walls still intact) from leaf strips using 0.5% (w/v) macerase (Macerozyme R-10; Yakult, Tokyo, Japan) in 0.7 M mannitol at 26°C, using the method developed by Gilliland et al.  and authenticated as valid by Pasqualini et al. . This cell production procedure was based on the first step of a published protoplast purification method . Cells were added to the reaction medium containing 50 mM KH2PO4 (pH 7.2), 0.4 M mannitol, 10 mM KCl, 10 mM MgCl2 and 0.1% (w/v) BSA [FA (fatty acid)-free]. This mixture was then transferred to an oxygen electrode (Oxytherm; Hansatech, King's Lynn, U.K.). Measurements of oxygen consumption were performed in the absence or presence of 2 mM KCN, 2 mM SHAM (salicylhydroxamic acid, an AOX inhibitor), 1 μM FCCP (carbonyl cyanide p-trifluoromethoxyphenylhydrazone, an uncoupler) or in the presence of both types of inhibitors. As residual respiration (oxygen uptake in the presence of KCN and SHAM) was often not detectable, it was assumed to be equal to zero. Measurements were carried out in the dark to prevent photosynthetic oxygen production. For each cell line examined, at least five sets of measurements were carried out to obtain statistically valid comparisons of AP capacity between wild-type and transgenic lines.
TMV inoculation of plants pre-treated with or without KCN
Non-transgenic and transgenic plants (8-weeks-old) were used for experiments. The foliage was sprayed with 2 mM KCN or water daily for 2 days before inoculation with TMV (strain U1). TMV (2 μg/ml in water) was inoculated on to one lower leaf per plant using a cotton bud soaked in the virus suspension. The leaves destined for inoculation with TMV were sprinkled with carborundum before application of the virus suspension to enhance the efficiency of infection.
Histochemical detection of H2O2
H2O2 was visually detected in the leaves of plants by using DAB (diaminobenzidine) . After treatment, leaves were sampled from plants and infiltrated with 1 mg/ml DAB (Sigma) solution (pH 3.8) for 6 h under light at 25°C and then treated with 95% ethanol to remove chlorophyll. This treatment decolorized the leaves except for the deep brown polymerization product produced by the reaction of DAB with H2O2.
Measurement of H2O2 and antioxidative indices
H2O2 was extracted from 1 g of leaf ground in 5 ml of acetone. Enzymes were extracted using 50 mM phosphate buffer (pH 7.8). All of these indices were measured with kits produced by the Nanjing Jiancheng Bioengineering Institute (Nanjing City, People's Republic of China).
Generation and identification of transgenic lines
A number of independent transgenic tobacco lines harbouring the NtAOX1a cDNA either in a sense or antisense orientation as well as the cell line expressing the empty vector were produced. Primary transformants (T0) were isolated on MS medium (Murashige and Skoog medium) containing kanamycin. Seeds of eight T0 plants, including three sense lines [S2 (sense line 2), S3 and S6], three antisense lines [AS1 (antisense line 1), AS4 and AS5] and two empty vector control lines [E1 (empty vector line 1) and E7], were chosen for germination. T-DNA (transfer DNA) inheritance was scored by kanamycin segregation analysis in the T1 generation. Resistance of T1 seedlings was segregated as a single dominant gene, and results are shown in Table 1.
|Lines .||Resistant seedlings .||Sensitive seedlings .||Ratio .||χ2 .||P .|
|Lines .||Resistant seedlings .||Sensitive seedlings .||Ratio .||χ2 .||P .|
NtAOX1a transcript accumulation of T1 lines was analysed using Northern blotting (Figure 1). Compared with wild-type plants, the expression levels of NtAOX1a were significantly up-regulated in S2, S3 and S6, especially in S2 and S6. In contrast with a slight reduction in AS4, the transcript accumulation in AS1 and AS5 was reduced to an almost invisible level. Moreover, NtAOX1a expression in E7 was similar to that in wild-type and much higher than that in E1. Taken together, we selected S2, S6, AS1, AS5 and E7 for further analysis, because they displayed representative phenotypes of their transgenic lines. T2 plants were also selected by kanamycin segregation. Similarly, homozygous T3 individuals were used for further characterization.
Characterization of transcript accumulation of
NtAOX1a in T1 plants using Northern blot analysis
Characterization of T3 lines
Analysis of NtAOX1a gene expression and protein accumulation was carried out by Northern and Western blotting respectively. The five T3 plants mentioned above were used, including E7, S2, S6, AS1 and AS5. It was observed that high levels of NtAOX1a transcript and protein accumulated in the two sense lines and low levels of NtAOX1a were observed in the two antisense lines (Figures 2A and 2B), which is consistent with that in T1 plants. Accordingly, NtAOX1a expression was inherited stably from the T1 to the T3 generation.
Characterization of NtAOX transcript and protein levels and AP respiratory capacity in T3 lines
Since AP capacity is an indicator of the potential activity of AOX , the respiratory characteristics of T3 plants were examined to determine whether transformation with NtAOX1a cDNA constructs affected their AP capacities. As expected, AP capacity was enhanced in the sense lines and reduced in the antisense lines (Figure 2C). In addition, we noticed that there were no significant differences in NtAOX1a expression and AP capacity between the two sense lines (S2 and S6), as well as the two antisense lines (AS1 and AS5), so we only selected S6, AS5 and E7 for further analysis.
Effects of low temperature on H2O2 accumulation and antioxidant enzymes
To analyse the effects of low temperature on transgenic and wild-type plants, tobacco seedlings were grown under normal conditions for 4 weeks and then transferred to low temperature conditions (12°C) for 24 h. Proteins were isolated from leaf mitochondria at the end of treatment (24 h). Western blot analysis indicated that NtAOX protein was induced by low temperature treatment in all lines except AS5 (Figure 3).
Immunoblot analysis of NtAOX protein levels at low temperature
Exposure to low temperature always impairs electron flow through the CP and results in an increase in H2O2 production . Thus we examined H2O2 accumulation in wild-type and transgenic plants under this condition. Leaves were collected from plants kept at 12°C for 0, 2, 6, 12 and 24 h. The DAB staining assay showed the appearance of H2O2 at different levels in the leaves of four genotypes at low temperature (Figure 4A). Specifically, the level of H2O2 in S6 was significantly lower than that in wild-type, E7 and AS5 from 2 to 24 h. AS5 contained the highest level of H2O2 at all time points except for 0 h, suggesting that the NtAOX1a overexpression efficiently inhibited low-temperature-induced ROS production (mainly H2O2). For an accurate quantification of the H2O2 level across the treatment times, H2O2 levels were measured. The result was in accordance with that from the DAB coloration assay (Figure 4B).
Analysis of H2O2 accumulation in wild-type and transgenic plants at low temperature
Since cellular ROS level is dependent upon both the rate at which ROS are generated by various cellular processes and the rate at which ROS are scavenged by the various antioxidant defences , we assumed that expression of the antioxidant enzymes might change in the leaves kept at 12°C treatment, and thus examined the activity of MSD (manganese superoxide dismutase) and CAT (catalase) during this period. MSD is known to be localized to mitochondria. CAT is targeted to the peroxisome or mitochondria. Strikingly, both the MSD and CAT activities were decreased in all genotypes under low temperature conditions, but they were at the highest level in AS5 throughout (Figure 5), accompanied by the highest H2O2 accumulation. Thus the higher activity of the antioxidant enzymes in AS5 did not effectively reduce the ROS formation caused by low temperature.
Changes of antioxidant enzyme activities in the leaves of wild-type and transgenic plants at low temperature
Analysis of viral resistance responses in transgenic plants
To analyse viral resistance responses in wild-type and transgenic tobaccos, 8-week-old plants were used for inoculation with TMV. HR was expected to be induced in the NN-type Samsun tobacco.
Initial experiments were carried out to examine the effect of AOX expression on resistance responses under normal conditions. It was observed that all S6, wild-type, E7 and AS5 plants generated well-defined circular HR lesions approx. 3 days after TMV inoculation, but no difference was observed between plant lines in the number and size of lesions produced.
Resistance of plants to TMV can be activated using CN , and thus we investigated KCN-induced viral resistance in transgenic tobacco. Plants were sprayed with 2 mM KCN for 2 days and then inoculated with TMV. As a result, different HR lesions appeared in the four lines. Specifically, yellow haloes revealed the appearance of much more lesions in S6 than in wild-type or E7. Also, S6 lesions were significantly larger than the lesions observed on the other lines. Compared with the above three lines, AS5 displayed the least severe disease symptoms (Figure 6A). To quantify the lesions, we counted the number of lesions and measured their diameter at 7 days post-inoculation (Figures 6B and 6C). The results are consistent with that defined visually. In addition, all of the lines treated with KCN displayed smaller and fewer lesions in comparison with the corresponding untreated plants (Figures 6B and 6C), suggesting an enhanced resistance to TMV in response to KCN treatment.
Different resistance responses to TMV in wild-type (WT) and transgenic tobacco
Given the different HR lesions of the four lines and the function of moderate ROS accumulation as a defence signal, the H2O2 level in each KCN-treated line after TMV inoculation was measured (Figure 7). We observed that accumulation of H2O2 in AS5 was significantly higher than in wild-type and E7. S6 contained the lowest level of H2O2. This may account for the different HR lesions observed in the four lines. When plants were inoculated with TMV, rapid generation of H2O2 necessitated the activation of additional defences. Thus the lower concentration of H2O2 in S6 made it less capable of activating defence mechanisms than AS5. In addition, MSD and CAT activities were also examined in the four lines, but there was no notable difference (results not shown).
Analysis of H2O2 accumulation in KCN-treated plants with or without TMV injection
Germination of transgenic plant seeds under H2O2 stress
Germination is a critical period in the life cycle of a plant and can be affected by a number of stresses, including oxidative stress. To analyse NtAOX1a function during seed germination under specific conditions, T3 seeds were germinated and grown on MS medium supplemented with 20 mM H2O2. All of the genotypes showed delayed germination by 1 day. But the different seeds did not differ in their germination percentages, with all of them exhibiting >90% germination (Figure 8A).
Germination of wild-type (WT) and transgenic tobacco seeds
Considering the function of KCN in inhibiting the CP, we germinated the seeds on MS medium containing 20 mM H2O2 and 2 mM KCN. In response to the treatment, germination was further delayed. More importantly, the four genotypes exhibited distinct germination percentages. S6 germination did not plateau until 8 days and was reduced to 70% germination relative to that on H2O2-containing MS medium. Subjecting AS5 seeds to the same treatment resulted in a >70% reduction in germination (Figure 8B). Meanwhile, as a negative control, the germination test was also performed under conditions using KCN only, with no notable difference observed among the four lines (results not shown). These results clearly demonstrate an important role for NtAOX1a in germination under H2O2-induced oxidative stress when the CP is inhibited.
AOX expression is responsive to low temperature and has also been implicated in resistance to viruses. The issues of whether AOX affects ROS accumulation at low temperature and pathogen attack, and whether there exist additional functions for AOX, have yet to be determined. In the present study, we used transgenic plants with altered NtAOX1a expression to further verify the possible correlation of AOX and ROS under these conditions and the potential role for AOX in seed germination.
A lower growth temperature increases AP capacity and AOX protein in tobacco . Also, low temperature treatment impairs electron flow through the CP and results in an increase in H2O2 formation . Nonetheless, information on the interrelationships of low temperature, ROS and AOX is quite limited. In the present study, analysis of H2O2 levels in NtAOX1a transgenic plants has conclusively demonstrated the correlation of AOX and ROS accumulation at low temperature. NtAOX1a overexpression efficiently inhibited the low-temperature-induced ROS (mainly H2O2) production (Figure 4). The result supports the hypothesis that AOX may alleviate oxidative stress under low temperature conditions .
Besides AOX, antioxidant enzymes might also affect ROS accumulation at low temperature, and thus we investigated the expression of MSD and CAT in this condition. Results indicated that AS5 had the highest activity levels of MSD and CAT at all time points except for 0 h among the four lines. However, the higher level of enzyme activity could not offset the higher rate of ROS generation caused by a lack of NtAOX1a, resulting in a higher level of ROS in AS5 than was found in S6 and wild-type leaves. Accordingly, AOX may be a main enzyme to reduce ROS production at low temperature.
Several lines of evidence implicate AOX in signalling during defence against viruses . But the molecular mechanism underlying AOX-mediated viral resistance remains to be determined. In the present study, the results obtained with the transgenic plants indicate that AS5 had far better viral resistance than S6 after KCN treatment. However, it should be noted that an important physiological role for AOX resides in the reduction of ROS levels in mitochondria [5,6]. Furthermore, ROS have been suggested to participate in the plant defence system in a variety of ways, including acting as signalling agents . Increasing evidence suggests that the ROS network is essential to induce disease resistance . Thus we hypothesize that ROS accumulation in mitochondria is most probably the resistance-inducing signal under the control of AOX. Measurement of H2O2 accumulation supports our hypothesis and provides us with new insight into the appearance of different HR lesions in the four KCN-treated lines. Although ROS are potentially harmful to cells, plants can use them as secondary messengers in signal transduction regulating defence mechanisms. When leaves are inoculated with TMV, rapid generation of H2O2 may be essential to activate pathogenesis-related proteins and to provide enough protection against the virus. In this process, AOX may function as a regulator of ROS-mediated signalling in the mitochondria and this potential AOX-regulated signalling mechanism appears to be involved in the activation of a subset of CN-inducible antiviral defence. Therefore, it may be premature to conclude, as Ordog et al.  have, that AOX does not play a role in the induction of viral resistance. However, the level of AOX did not influence TMV infection under non-KCN conditions. A possible explanation for this result is that, under this condition, the electron did not flow to AOX, and thus the accumulation of the signalling agent H2O2 may be mainly affected by some other factors rather than AOX expression, which results in almost identical HR lesions being produced in the four lines. Further investigation is of great importance to uncover a more comprehensive picture of the functional roles and action mechanism of AOX in viral resistance.
AOX function in seed germination was determined by subjecting the seeds of different genotypes to oxidative stress in the presence or absence of KCN. The obvious difference in germination profiles among the four lines was observed only with medium supplemented with both H2O2 and KCN (Figure 8). However, in this situation, it is also possible that application of KCN, not AOX, leads to a different respiratory rate, which determines the final differences in germination profiles. So we performed a negative control, a germination test using KCN only. But we found no difference among the four lines and eliminated the possibility. Hence the up-regulation of NtAOX1a expression in S6 did contribute to the enhancement of oxidative stress tolerance in seed germination.
In conclusion, the present study supports the idea that AOX plays a role in regulating defence responses against abiotic and biotic stresses by preventing the production of ROS when the CP is restricted.
- AS1 etc.
antisense line 1 etc.
- E1 etc.
empty vector line 1 etc.
manganese superoxide dismutase
- MS medium
Murashige and Skoog medium
reactive oxygen species
- S2 etc.
sense line 2 etc.
tobacco mosaic virus
We thank Dr Xueshui Guo (Department of Biology and Microbiology, South Dakota State University, Brookings, SD, U.S.A.) and Professor Hailong An (College of Life Sciences, Shandong Agricultural University, Taian, Shandong, China) for reading the manuscript and providing suggestions and revisions prior to acceptance.
This work was partly supported by the National Natural Science Foundation of China [grant numbers 30370928, 30571215].
These authors have contributed equally to this work.