Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Heterologous Expression of ATG8c from Soybean Confers Tolerance to Nitrogen Deficiency and Increases Yield in Arabidopsis

  • Tongmei Xia ,

    Contributed equally to this work with: Tongmei Xia, Dong Xiao

    Affiliation Department of Plant Biology and Ecology, College of Life Sciences, Nankai University, Tianjin, China

  • Dong Xiao ,

    Contributed equally to this work with: Tongmei Xia, Dong Xiao

    Affiliation Department of Plant Biology and Ecology, College of Life Sciences, Nankai University, Tianjin, China

  • Dong Liu,

    Affiliation Department of Plant Biology and Ecology, College of Life Sciences, Nankai University, Tianjin, China

  • Wenting Chai,

    Affiliation Department of Plant Biology and Ecology, College of Life Sciences, Nankai University, Tianjin, China

  • Qingqiu Gong , (NNW); (QG)

    Affiliation Department of Plant Biology and Ecology, College of Life Sciences, Nankai University, Tianjin, China

  • Ning Ning Wang (NNW); (QG)

    Affiliation Department of Plant Biology and Ecology, College of Life Sciences, Nankai University, Tianjin, China

Heterologous Expression of ATG8c from Soybean Confers Tolerance to Nitrogen Deficiency and Increases Yield in Arabidopsis

  • Tongmei Xia, 
  • Dong Xiao, 
  • Dong Liu, 
  • Wenting Chai, 
  • Qingqiu Gong, 
  • Ning Ning Wang


Nitrogen is an essential element for plant growth and yield. Improving Nitrogen Use Efficiency (NUE) of crops could potentially reduce the application of chemical fertilizer and alleviate environmental damage. To identify new NUE genes is therefore an important task in molecular breeding. Macroautophagy (autophagy) is an intracellular process in which damaged or obsolete cytoplasmic components are encapsulated in double membraned vesicles termed autophagosomes, then delivered to the vacuole for degradation and nutrient recycling. One of the core components of autophagosome formation, ATG8, has been shown to directly mediate autophagosome expansion, and the transcript of which is highly inducible upon starvation. Therefore, we postulated that certain homologs of Saccharomyces cerevisiae ATG8 (ScATG8) from crop species could have potential for NUE crop breeding. A soybean (Glycine max, cv. Zhonghuang-13) ATG8, GmATG8c, was selected from the 11 family members based on transcript analysis upon nitrogen deprivation. GmATG8c could partially complement the yeast atg8 mutant. Constitutive expression of GmATG8c in soybean callus cells not only enhanced nitrogen starvation tolerance of the cells but accelerated the growth of the calli. Transgenic Arabidopsis over-expressing GmATG8c performed better under extended nitrogen and carbon starvation conditions. Meanwhile, under optimum growth conditions, the transgenic plants grew faster, bolted earlier, produced larger primary and axillary inflorescences, eventually produced more seeds than the wild-type. In average, the yield was improved by 12.9%. We conclude that GmATG8c may serve as an excellent candidate for breeding crops with enhanced NUE and better yield.


The vigor of plants is largely dependent on inorganic nitrogen (N), principally in the form of NO3 and NH4+, and 85–90 million tons of nitrogenous fertilizers are applied worldwide annually [1], [2]. To maintain crop production, more and more commercial fertilizers are applied, generating enormous costs and resulting in severe environmental damage [3], [4], [5]. Therefore, breeding crops with high nitrogen use efficiency (NUE) and better yield has always been a major goal for breeders, and the identification of new potential NUE genes is important for scientists. Soybean is a major crop for protein and oil production globally. Though capable of biological N2 fixation when associated with Rhizobia bacteria, soybean acquires a large portion (from 40% to 75%) of nitrogen from soil, depending on the inorganic nitrogen content of soil [6]. It is estimated that, to produce 1,000 kg of soybean seeds, 70 to 90 kg of nitrogenous fertilizer needs to be applied [7]. Thus, it is critical to identify useful soybean genes for the development of new transgenic soybean cultivars with high NUE and better yield.

The nitrogen utilization of plants involves uptake, assimilation, translocation, and remobilization, in which nitrogen remobilization being the key step during seed maturation [8], [9], [10]. In Arabidopsis, for instance, nitrogen concentration (N%) in the dry remains was reported to be 4-fold lower at low nitrate supplies compared with high nitrate supplies, while nitrogen concentration in seeds barely changed [11]. In wheat (Triticum aestivum), remobilization contributes 69.8% to 88.8% to grain N content depending on cultivars [12]. The primary control for N filling in seeds is in the source organs [13], and N import into developing seeds is mainly derived from the amino acids produced by proteolysis of proteins synthesized before the onset of the reproductive phase [14]. RuBisCO, the most abundant soluble protein in green plants, as well as other photosynthesis-related proteins, is the major N source for remobilization. The main pathways for protein degradation include 26S proteasome/ubiquitin system, endoplasmic reticulum-associated degradation (ERAD), senescence-associated vesicles (SAV) and autophagy, leading to proteolytic degradation in the vacuole [9], [15]. Among them, autophagy has the largest capacity to degrade and recycle organelles and cytosolic macromolecules [16], [17], and could be an important source of N for seed production.

So far, most of our knowledge on autophagy has come from genetic studies in yeast (Saccharomyces cerevisiae), in which more than thirty AuTophaGy-related (ATG) genes have been identified [18], [19], [20]. Central to autophagy is the formation of the autophagosome, a double-membrane vesicle containing cargos ranging from damaged proteins to obsolete organelles to be degraded in the vacuole/lysosome. ATG8, a lipid-conjugated ubiquitin-like protein, whose transcript level is known to be greatly induced by starvation [21], serves as a scaffold for membrane expansion during autophagosome formation, and is the determinant of autophagosome size [22], [23]. ATG8 also participates in the cytoplasm-to-vacuole targeting (Cvt) pathway, a constitutive and specific form of autophagy, in which vacuolar hydrolases such as aminopeptidase 1 (APE1) and alpha-mannosidase (AMS1p) are selectively transported into the vacuole under normal growth condition [24], [25]. In both pathways, ATG8 binds to the cargo receptor ATG19, facilitating its localization to the pre-autophagosomal structure/phagophore assembly site (PAS), where the autophagic/Cvt transport vesicles are assembled [25], [26], [27].

In higher plants, homologous autophagic systems have been identified. In vitro reconstitution of the two conjugation systems (ATG5-ATG12 and ATG8-PE) has been successfully done, indicating that the mechanism of autophagy is conserved from yeasts to plants [28]. Analyses on the loss-of-function atg mutants revealed the involvement of autophagy in many physiological processes, including nutrient deficiency adaptation, disease resistance, innate immune response, and stress resistance [15], [29], [30], [31], [32], [33], [34], [35], [36]. An intriguing fact is that plants generally have many ATG8s; there are nine in Arabidopsis [29], [30], presumably five in maize [37], and five in rice [37]. The distinct expression patterns of individual ATG8s in Arabidopsis suggest that, apart from possible redundancy, at least some ATG8s may possess unique functions [33], [38]. Accordantly, Honig et al had found two plant specific AtATG8-interacting proteins (ATI1 and ATI2) that participate in the formation of a new type of compartment which was induced while exposure to carbon starvation [39]. They also showed that the constitutive expression of AtATG8f led to an increment of plant size, and that the transgenic plants became more tolerant to nitrogen and carbon starvation [40]. Given the specific function of ATG8 and the expression characteristics of the plant ATG8 gene families, we postulated that it is possible for us to identify at least one specific ATG8 for the development of NUE crops. Furthermore, it has been shown that, under normal growth conditions, loss-of-function autophagy mutants such as atg7, atg9, and atg10 produced less flowers, siliques, and seeds [29], [33], [38]. Whether constitutive expression of an ATG could improve seed yield is therefore worthy of exploring.

In order to screen for potential soybean NUE genes, our lab has previously set up a fast and effective soybean calli transformation system (patent pending), through which large amounts of uniformly transformed soybean cells could be obtained and analyzed on plates. Here, we firstly identified a soybean ATG8 gene, namely GmATG8c, which is strongly induced by nitrogen starvation. We then confirmed its function as an ATG8 by yeast complementation. Constitutive expression of GmATG8c in soybean calli not only led to better tolerance towards nitrogen starvation, but accelerated calli growth under optimum growth conditions. Heterologous expression of GmATG8c in Arabidopsis led to better performance of the transgenic lines under both starvation conditions and normal growth condition. The transgenic lines had a better vegetative growth, entered the reproductive stage slightly earlier, and produced more flowers, siliques, and seeds. Our results implied that GmATG8c functions in soybean autophagy process, and that it may be an important candidate for breeding crops with enhanced NUE and better yield.


Identification of GmATG8c as a Nitrogen Deficiency-inducible/responsive ATG8 of Soybean

Based on their sequence homology to the yeast (Saccharomyces cerevisiae) ATG8, eleven GmATG8s (see Table S1 for accession numbers) were identified from the soybean genome (Phytozome, and named GmATG8a to GmATG8k, respectively (Fig. 1A). All GmATG8s possess a conserved Gly residue near their C-termini, which corresponds to the PE-acceptor site of yeast ATG8 (Fig. 1B). The GmATG8s share 42% to 94% sequence similarity with one another, and are all at least moderately similar to ScATG8, with 48% (GmATG8i) to 72% (GmATG8d) identical amino acid sequences (Table. S1). When a phylogenetic tree was built with ATG8 homologs from Arabidopsis, several major crop species including rice and maize, and lower green plants, the soybean ATG8s fall into three sub-groups (Fig. 1A). The first one is relatively close to ScATG8, and contains GmATG8a, b, c, d, and e. The second and third ones contain GmATG8f, g, j, and k; and GmATG8h and i, separately (Fig. 1A). The fern Selaginella moellendorffii possesses three ATG8s; each appeared in a sub-group (Fig. 1A).

Figure 1. Soybean has eleven homologs of yeast ATG8.

A, Phylogenetic tree of ATG8s in Glycine max (Gm), Arabidopsis thaliana (At), Saccharomyces cerevisiae (Sc), Selaginella moellendorffii (Sm), Physcomitrella patens (Pp), Ostreococcus lucimarinus (Ol), Oryza sativa (Os) and Zea mays (Zm). Deduced amino acid sequences were aligned by CLUSTALX and the phylogenetic tree was generated by the neighbour-joining method and displayed using MEGA5. GmATG8s were underlined. B, Deduced amino acid sequences of GmATG8c and alignment with ScATG8. Closed arrowhead indicates the C-terminal Glycine residue which is processed by Atg4 cysteine protease. C, Fold Changes of the transcript levels of soybean ATG8s in the primary leaves after nitrogen starvation. The average expression levels for the three biological replicates are shown. Significant differences between +N and –N for 3 days or 6 days, and between –N for 3 days and 6 days are labeled. All significant levels were calculated by t-test. *, significant (p<0.05); **, very significant (p<0.01).

Table 1. Bolting and flowering times of the wild-type and 35S:GmATG8c transgenic Arabidopsis.

Similar to their homolog in yeast, transcript levels of some specific Arabidopsis ATG8s have been shown to be induced by nutrient deprivation [30], [41], [42]. We examined whether the transcript levels of GmATG8s could be induced by nitrogen starvation by real-time RT-PCR (Fig. 1C). When the soybean seedlings (8-day-old) grown hydroponically were deprived of nitrogen for 3 days, the transcript levels of all GmATG8s in the primary leaves were all slightly induced (p<0.05) (Fig. 1C). When the time of N-starvation was prolonged to 6 days, expression of some GmATG8s was further induced, with GmATG8c peaked at 12 fold (Fig. 1C). The results indicated that the GmATG8c was a potential soybean functional ATG8 gene in response to nitrogen starvation. GmATG8c encodes a soluble protein of 119 amino acids, with a predicted pI of 9.16 ( [43]. A 3D model for GmATG8c was constructed with Phyre 2 ( [44] to reveal its striking resemblance to the yeast, protist and mamalian ATG8 homologs; the protein is composed of an N-terminal domain containing two α-helices and a C-terminal ubiquitin-like domain (Fig. S1).

Partial Complementation of Saccharomyces cerevisiae atg8 Mutant by GmATG8c

In yeast, as a cargo of Cvt/autophagy pathway, the precursor amino-peptidase 1 (prAPE1) is incorporated into Cvt vesicles/autophagic bodies and delivered to the vacuole for processing into mature APE1 (mAPE1), making it a marker for Cvt and autophagy pathways [45], [46]. To verify the function of GmATG8c, we carried out a functional complementation experiment, in which the protein levels of both prAPE1 and mAPE1 were monitored. Upon nitrogen starvation, mAPE1 accumulated in wild-type yeast cells which suggested pAPE1 protein has been delivered to the vacuole, and that processing of pAPE1 into mAPE1 occurred (Fig. 2). atg8 cells transformed with pADH:GmATG8c also accumulated mAPE1, albeit to a lesser extent than the wild-type (Fig. 2). In contrast, in atg8 cells or atg8 cells transformed with the vector containing the ADH promoter only, no mAPE1 was detected under nitrogen rich condition, and mAPE1 were barely detectable under nitrogen starvation condition (Fig. 2). The result hence confirmed the role of GmATG8c as an ATG8.

Figure 2. Functional complementation of a yeast atg8 mutant by GmATG8c.

Cultures grown to mid-log phase in YPD were harvested or shifted to YPD without nitrogen for 4 hours and then harvested for protein extraction (3OD600 cells each). Proteins were then resolved by SDS-PAGE followed by immunoblotting with anti-APE1 antibody. Intensities of band signals of mAPE1 were quantified using the software Quantity One (Bio-Rad). Five independent replicates were done to give the typical results shown here. WT: TN124; atg8: TN124 atg8Δ:KAN URA3 TRP1; GmATG8c: atg8 carrying pADH1-GmATG8c; pESC in atg8: atg8 carrying pADH1 only (on pESC).

Constitutive Expression of GmATG8c Enhanced the Tolerance of Soybean Calli to Nitrogen Limitation

We then evaluated the NUE potential of GmATG8c at the cellular level. GmATG8c was constitutively expressed under the control of the CaMV 35S promoter in soybean calli. The over-expression of GmATG8c in three homozygous callus lines was confirmed by semi-quantitative RT-PCR (Fig. 3A). Transgenic and wild-type calli of similar size were transferred to MS medium with 60 mM N (sufficient nitrogen, as control), 10 mM N and 1 mM N (low nitrogen) or without nitrogen. The status of the calli was evaluated by their weight and browning rate [47]. On day 7, the wild-type calli grown on medium with 1 mM N or no N began to turn brown, while the transgenic calli remained transparent (Fig. 3B). On day 17, the wild-type calli grown on 1 mM N had turned brown more severely, while the transgenic calli remained healthy (Fig. 3B). Furthermore, nitrogen deficiency resulted in growth retardation in the wild-type earlier than in the transgenic calli (Fig. 3B). A typical transgenic line, together with the wild-type, was weighed daily to track the growth of the calli. Under all conditions, the GmATG8c over-expressing calli grew much faster and became significantly heavier than the wild-type (Fig. 3C). The results further confirmed that over-expression of GmATG8c confers nitrogen deficiency tolerance at the cellular level.

Figure 3. Constitutive expression of GmATG8c enhanced the tolerance of soybean calli to nitrogen limitation conditions.

A, The constitutive expression of GmATG8c in transgenic calli lines was confirmed by semi-quantitative RT-PCR. WT: wild type calli. L-1 to L-3: three 35S:GmATG8c transgenic calli lines. B, Morphology of GmATG8c over-expressing calli and the wild-type cultivated on MS medium with indicated nitrogen concentration over indicated period of time. The calli were placed on the medium all in the same order as marked in the upper left panel (0 mM, Day 0). C, Growth curves of GmATG8c over-expressing calli and those of the wild-type on medium with full (60 mM), low (1 mM), and no (0 mM) nitrogen over 14 days. Results are the mean ± SE for three biological replicates.*, p<0.05 (t-test); significant difference from the wild-type (WT).

GmATG8c Enhanced Tolerance to Nitrogen Starvation in Transgenic Arabidopsis

We then evaluated the NUE potential of GmATG8c in the model plant Arabidopsis. 35S:GmATG8c was introduced into Arabidopsis by floral dipping. The heterologous expression of GmATG8c in three T3 homozygous transgenic lines was confirmed through both quantitative RT-PCR (Fig. 4A) and western blotting (Fig. 4B, S2). Real-time RT-PCR showed that the transcript levels of the endogenous AtATGs were unchanged in the transgenic lines (Fig. 4C).

Figure 4. Heterologous expression of GmATG8c in Arabidopsis.

A, Quantitative RT-PCR analysis of GmATG8c transcript in 5-day-old seedlings of the wild-type (WT) and 35S:GmATG8c lines (L-2 to L-4). B, Immunoblot analysis of the accumulation of GmATG8c in 5-day-old seedlings of the wild-type (WT) and 35S:GmATG8c lines (L-2 to L-4) with a polyclonal anti-GmATG8c antibody. Equal protein loads were confirmed by immunoblot analysis with an anti-Actin antibody. C, Real-time RT-PCR analysis of the transcript levels of AtATGs in 5-day-old seedlings of the wild-type and 35S:GmATG8c lines. The average expression levels for the three biological replicates are shown (p>0.05). Heterologous expression of GmATG8c appeared to have no effect on the transcript levels of the endogenous AtATG8s.

Nitrogen starvation tolerance was evaluated in both seedlings and adult plants. Five-day-old seedlings of the wild-type and transgenic lines, similar in sizes, were transferred to 1/2 MS medium without nitrogen. Within three weeks, the difference in sizes became clearly visible. Besides, all transgenic lines produced in average two more true leaves than the wild-type (p<0.05) (Fig. 5A). When grown under continuous light in liquid medium free of nitrogen, the wild-type plants ceased growing, whereas the transgenic plants continued to grow, only at a reduced rate (Fig. 5B). The phenotypes suggested that the transgenic seedlings did endure better under the nitrogen-deficient condition.

Figure 5. 35S:GmATG8c seedlings were more tolerant to nitrogen starvation than the wild-type.

A, Five-day-old seedlings of the wild-type (WT) and 35S:GmATG8c lines were transferred to 1/2MS medium without nitrogen and starved for 21 days under a long-day photoperiod, then photographed. The white arrowheads indicate the newly emerged true leaves. Scale bar  = 50 µm. B, Seedlings (n = 10) were grown for 1 week on 1/2MS+N liquid medium and then transferred to N-rich (+N) or N-deficient (–N) media for 4 days. Scale bar  = 1 cm. Results from one out of five biological replicates were shown.

To evaluate the tolerance of nitrogen starvation in the adult plants, we set up a hydroponic system similar to a previous study [48]. Nine-day-old seedlings grown on 1/2 MS medium were transferred to vermiculite and perlite (1∶1, v/v) supplied with 1/2 Hoagland’s solution and let grown to bolting (23 days after sowing, DAS), then transferred to either 1/2 Hoagland’s solution or nitrogen-free solution and let grown for another 5 days. On 28 DAS, all three transgenic lines had slightly, yet significantly larger rosettes than the wild-type under both nitrogen-sufficient (p  = 0.022, 0.029, 0.012) and nitrogen-deficient condition (p = 0.010, 0.003, 0.001) (Fig. 6A and Table. S3). Meanwhile, all plants were harvested and weighed. Under both the conditions with or without nitrogen starvation, all three GmATG8c over-expressing lines weighed significantly more than the wild-type (Fig. 6B). Contents of nitrogen, protein, and soluble sugar were measured in control and starved plants. As expected, nitrogen contents and protein levels in all materials decreased in the nitrogen-deprived wild-type and transgenic lines (Fig. 6C, D). The transgenic lines accumulated more nitrogen in their rosette leaves and stems (Fig. S3). Specifically, in the juvenile rosette leaves (leaves 3 to 6), the nitrogen concentration (mg/g of fresh weight) in the transgenic lines was higher than that of the wild-type (Fig. 6C). The adult rosette leaves (leaves 7 to 10) and the stems of the transgenic plants had similar nitrogen contents with the wild-type (Fig. 6C).

Figure 6. GmATG8c confers nitrogen starvation tolerance to adult plants.

Wild-type and 35S:GmATG8c plants were grown hydroponically in a mixture of pearlite/vermiculite irrigated every other day with half Hoagland’s until bolting (23 d after sowing), and then maintained in either the same nitrogen-rich solution (N sufficient) or transferred to nitrogen-free solution (N deficient) for another 5 days. A, Top view of the wild-type and 35S:GmATG8c lines without (upper panel) and with (lower panel) 5 days of nitrogen starvation. B, The fresh weights of the wild-type and transgenic lines. C, Nitrogen contents in juvenile/adult rosette leaves and stems of the wild-type and 35S:GmATG8c lines. D, Protein contents in the materials. E, Soluble sugar contents in the materials. Representative data from one out of three biological replicates are shown. NS: N sufficient, ND: N deficient. Data shown are the means ±SD of one representative biological replicate (n = 21) out of three.*, p<0.05 (t-test); significant difference from the wild-type (WT).

Under the nitrogen-sufficient condition, the protein concentrations in the juvenile rosette leaves of the transgenic plants seemed lower than those of the wild-type, but showed no significant difference by student’ s t-test (p>0.05) (Fig. 6D). In both juvenile and adult rosette leaves, the protein concentrations were similar between the wild-type and transgenic plants under the nitrogen-deficient condition (Fig. 6D).

As expected, more soluble sugar accumulated in the nitrogen-deprived wild-type and transgenic plants (Fig. 6E). The soluble sugar concentrations in the transgenic plants generally exhibited no difference from those of the wild-type in all the materials under the same condition (Fig. 6E).

GmATG8c Over-expressing Arabidopsis Adapted Better to Carbon (C-) Starvation than the Wild-type

Previous studies have shown that the Arabidopsis atg mutants were generally more sensitive to carbon limitation conditions [32], [33], [38]. To see if GmATG8c over-expressing Arabidopsis could better survive carbon limitation, seedlings of a typical transgenic line (L-4) and the wild-type were grown on 1/2 MS to stage 1.02 [49], transferred to soil and let grown for 33 more days under short-day condition [50], [51], kept in dark for 9 days, then allowed to recover under the same short-day condition (Fig. 7A). A larger portion of 35S:GmATG8c transgenic plants survived, and resumed growth earlier than the wild-type (Fig. 7B). Therefore, GmATG8c indeed conferred carbon starvation tolerance to Arabidopsis.

Figure 7. Enhanced tolerance of 35S:GmATG8c to carbon-limiting conditions induced by extended darkness.

A, Diagram of treatment. Six-week-old plants were grown under a short-day (SD) photoperiod, transferred to darkness for 9 days, and then transferred back to the short-day photoperiod for recovery. B, The upper panel shows plants immediately after dark treatment; the middle panel shows plants after 7-day recovery; and the bottom panel shows plants after 25-day recovery. Data from one representative line (L-4) out of four lines analyzed are shown.

Constitutive Expression of GmATG8c Promotes Growth and Increases Yield in Arabidopsis

We have noticed from the nitrogen starvation experiments that the transgenic plants grew faster than the wild-type (Fig. 6A). Therefore, the growth parameters of the transgenic plants and the wild-type under normal growth condition under long-day photoperiods were recorded daily over a period of 8 weeks (Fig. 8). The rosettes of GmATG8c over-expressing plants reached larger sizes upon bolting (Fig. 8A) and grew faster than the wild-type (Fig. 8B). The plant heights of transgenic plants were higher than the wild-type plants at the time points we observed (Fig. 8C, D). All the transgenic lines bolted and flowered slightly earlier than the wild-type (Table 1). These data indicated that the over-expression of GmATG8c could promote vegetative growth of Arabidopsis, and facilitate the transition into reproductive growth.

Figure 8. Heterologous expression of GmATG8c promotes growth in Arabidopsis.

Nine-day-old seedlings were transferred to soil and cultivated under a long-day photoperiod. The radius of rosette and the plant height were recorded daily. A, Top view of 4-week-old representative wild-type (WT) and 35S:GmATG8c lines (L-2, L-3, L-4) grown under a long-day photoperiod. B, Growth curves of the rosette of the wild-type and one representative transgenic line (L-4). C, Side view of 5-week-old representative wild-type (WT) and 35S:GmATG8c lines (L-2, L-3, L-4) grown under a long-day photoperiod. D, The plant height of the wild-type and transgenic lines. Data shown are the means ±SE of one representative biological replicate (n = 24) out of three. X-axis represents the days after sowing (DAS).

We then analyzed the reproductive phase of the plants in more details (Fig. 9). Branching patterns were dissected in seven-week-old wild-type and transgenic lines, and it became clear that although the plant architecture has not changed, the transgenic lines did have larger primary and axillary inflorescences that looked more robust (Fig. 9A, S4). The primary inflorescence (without the lateral branches) of the transgenic lines produced significantly more fertile siliques than the wild-type (Fig. 9B), which is consistent with our observation that the transgenic lines entered the reproductive phase slightly earlier (Table 1). The numbers of siliques of transgenic plants on the primary inflorescence and the axillary inflorescences were larger than the wild-type (Fig. 9C, D). Overall, all three transgenic lines produced significantly more siliques than the wild-type (Fig. 9E).

Figure 9. Phenotypic characterization and yield-related characters of the 35S:GmATG8c transgenic Arabidopsis.

A, Schematic diagrams of branching patterns of wild-type plants (WT) and GmATG8c over-expressing lines (L-2 to L-4). Bar  = 5 cm. B, The number of siliques on the primary inflorescence of the wild-type and 35S:GmATG8c transgenic lines. C, The number of siliques on lateral branches generated from the primary inflorescence. D, The number of siliques on axillary inflorescence of the wild-type and 35S:GmATG8c transgenic lines. E, Total number of siliques of the wild-type and 35S:GmATG8c transgenic lines. Data shown are the means ±SE of one representative biological replicate (n = 10) out of two. *, p<0.05 (t-test); significant difference from the wild-type (WT).

To further evaluate the potential of GmATG8c in yield increment, four transgenic lines (L1-4) were grown to stage 9.70 [49] for seed harvest. All seeds from individual plants were harvested and weighed. All four transgenic lines evaluated had higher yields than the wild-type (Table 2). One of the lines was subjected to further analysis on the reason for the increment of yield. The thousand grain weight was not different from that of the wild-type (Table 3). The total number of seeds per silique, however, differed slightly (by two seeds) but significantly from that of the wild-type (Table 3). The total number of siliques was 12.7% more than the wild-type, which may have a major contribution to the increment of yield in GmATG8c over-expressing plants (Table 3).

Table 2. The yields of 35S:GmATG8c transgenic lines and those of the wild-type.

Table 3. Yield related characteristics of the wild-type and 35S:GmATG8c transgenic Arabidopsis.

To further explore the yield increment potential of GmATG8c, we also transformed tomato (cv. Micro-Tom), a model plant for fruit bearing, with 35S:GmATG8c. Similar to our observations in transgenic Arabidopsis, 35S:GmATG8c transgenic tomato also grew faster and entered reproductive phase earlier than the wild-type, and produced more fruits (Fig. S5, Table S4).

Based on our observations, we concluded that GmATG8c not only has NUE potential but also has potential for yield increment, and may be an excellent target gene for breeding better crops.


Though autophagy has been established as a ubiquitous intracellular degradation pathway of eukaryotic cells, much of the detailed process, and even components of autophagy, remains elusive in plants, especially in crop species. We identified eleven soybean ATG8 genes through sequence analysis. It is well known that a number of ATG8 isoforms exist in plants and mammals, and the existence of an ATG8 gene family may reflect the existence of specific functions of individual ATG8s, as have been suggested by other studies [29], [30], [38], [52]. When making a phylogenetic tree for ATG8 homologs in crops and lower plants, we noticed two interesting facts. Firstly, all six ATG8s of the moss Physcomitrella patens were clustered together to form a branch relatively close to the AtATG8a and GmATG8c (Fig. 1A). Secondly, all other ATG8s formed two more sub-groups represented by AtATG8f and the longer ATG8s, separately (Fig. 1A). The branching pattern of the phylogenetic tree suggests that a major functional specialization event of plant ATG8s may have happened when tracheophytes firstly emerged.

Based on the hypothesis that, similar to the yeast ATG8, the transcript level of a functional plant ATG8 may be highly induced by nitrogen deprivation, we searched for such an ATG8 homolog from soybean. Out of the eleven ATG8s, GmATG8c showed the distinctively high level of induction (more than ten folds) under nitrogen deprivation (Fig. 1C). We then confirmed the biochemical function of GmATG8c as an ortholog of ScATG8 by yeast complementation in a similar fashion to a previous study, in which AtATG6 was confirmed as a functional ortholog of ScATG6 [53] (Fig. 2).

We further verified the physiological function and NUE potential of GmATG8c at both cellular and whole-plant level. It has been shown in maize that, during nitrogen and carbon limitations, ATG transcripts and ATG8-PE accumulates [37]. Such observation indicates that autophagy may play a key role in nutrient remobilization. Our hypothesis was that an extra dose of a nitrogen limitation-inducible ATG8 may confer nitrogen-limitation tolerance by promoting autophagy, and thus nitrogen remobilization, more efficiently when necessary.

Indeed, the transgenic lines excelled in many ways. Not only did they survive extended period of carbon or nitrogen starvation, but promptly recovered and resumed growth when moved back to normal growth conditions. It is noteworthy that the transgenic lines generally produced more leaves when starved (Fig. 5A, 6A) or moved back to optimum conditions (Fig. 7B), suggesting that the ectopic expression of GmATG8c contributes to the fitness and survival of the shoot apical meristem. Two previous studies had suggested the involvement of autophagy in the activity and maintenance of root apical meristem: over-expression of AtATG8f resulted in elevated sensitivity of root growth towards exogenous cytokinin in Arabidopsis [40], and an atg5 mutant of Arabidopsis failed to keep its root apical meristem alive upon phosphorus limitation [54]. Therefore, the positive function of autophagy in the maintenance of meristem activity could be essential for the adaptation of plants towards nutrient limitation.

Interestingly, under optimum growth conditions, no change in nitrogen concentrations was observed in the adult rosette leaves or stems of the transgenic lines from those of the wild-type (Fig. 6C), whereas the growth of the plants appeared accelerated [Fig. 6A, 6B, 8A, 8B, Table S3]. Such observations were consistent with our hypothesis that nitrogen remobilization was more efficient in 35S:GmATG8c. Possibly a result of better vegetative growth, the transgenic lines also had a longer period of reproductive growth, and produced more siliques and seeds. The yield of transgenic lines were improved (Table 2) not for the thousand grain weight, but for the total seed number, in which the increment of siliques contributed more compared with the increment of number of seeds per silique (Table 3). It is known that grain yield is based both on the nitrogen uptake from soil and on the nitrogen remobilization during seed maturation [12]. The organic nitrogen supply from source leaves during grain filling is of particular importance [55], [56], since it contributes to plant N economy and limits exogenous N demand after flowering [9]. Since autophagic recycling has been suggested to be required for N and C mobilization [9], [57], the yield increment in the transgenic lines thus could be due to an enhanced autophagic recycling ability which led to more efficient nutrient remobilization from the rosette into the reproductive organs.

In order to improve the NUE and yield of crops, many genes have been analyzed mainly through transgenic (over-expression) studies. Traditional approach has focused on the essential genes in nitrogen uptake and assimilation, including NRTs (nitrate transporters), NR (nitrate reductase), NiR (nitrite reductase), GS (glutamine synthesis), and GOGAT (glutamate synthesis) [58]. Despite the great efforts, the corresponding transgenic plants have shown little improvement in NUE and yield [58], except in one case: the over-expression of NADH-GOGAT (NADH-dependent glutamate synthesis ) of rice enhanced grain filling and increased grain weight [59]. Therefore, new pathways should be explored for transgenic purposes. Since autophagy is one of the main pathways for protein degradation, and very likely contributes to the nitrogen pool for seed production, autophagy-related genes may have NUE potential and could probably improve the yield. Previous reports and our study have seen the reduced fitness and fertility of several atg mutants, including atg5 (data not shown), atg7, atg9, atg10 [29], [33], [38], and our study showed that the over-expression of a specific ATG8 could make soybean cells and Arabidopsis plants grow better under both nitrogen-rich and nitrogen-poor conditions. Clearly, autophagy is a pathway with NUE potentials. It is noteworthy that our strategy accord with one of the strategies proposed in [60], which is to modify C- and N- allocation and to elevate the strength of the sink [60].

GFP-ATG8 has been widely used as a marker of autophagy [30], [32], [61], [62]. Nevertheless, the current work, together with a previous study [40], demonstrates that the over-expression of plant ATG8s could have impact on the response to nutrient starvation and growth at both the cellular and the whole plant level. Therefore, we suggest that caution should be taken when using GFP-ATG8 driven by strong promoters as a marker for autophagy in plants, especially when studying plant growth and development.

In summary, our study has linked autophagy to plant nitrogen homeostasis/distribution as well as yield formation, and has identified a new node with great potential to be targeted in molecular/transgenic breeding to enhance NUE and improve yield in crop species. Further study will be carried out to elucidate the detailed molecular mechanisms through which GmATG8c promotes NUE and increases yield.

Materials and Methods

Sequence Analysis of ATG8 Proteins

Deduced ATG8 amino acid sequences of Glycine max (Gm), Physcomitrella patens (Pp) and Selaginella moellendorffii (Sm) were identified by searching the corresponding genomic sequence ( using Saccharomyces cerevisiae ATG8 as the query. The ATG8 amino acid sequence of Ostreococcus lucimarinus (Ol) was identified by searching the Ostreococcus lucimarinus Assembled RefSeq Genomes ( using ScATG8 as the query. The ATG8s serial numbers of Oryza sativa (Os) and Zea mays (Zm) were used as described [37]. OsATG8s and ZmATG8s were identified as follows, with the corresponding loci in parentheses; OsATG8a (NP_001059767), OsATG8b (NP_001053929), OsATG8c (NP_001061171), OsATG8d (NP_001065220), OsATG8e (NP_001065220), ZmATG8a (NP_001137492), ZmATG8b (NP_001137493), and ZmATG8c (NP_001137496). The serial numbers of GmATG8s and SmATG8s were made according to their relationship with Arabidopsis ATG8s. The Phytozome ID and GenBank accession numbers of the corresponding GmATG8s and SmATG8s can be found in Table S1.

Amino acid sequences were aligned using CLUSTALX2 ( [63]. The phylogenetic tree was generated in CLUSTALX2 by the neighbour-joining method and a thousand replicates and displayed using MEGA5 ( [64].

The 3D model for GmATG8c was constructed with Phyre 2 ( ) [44] and compared with the model of Homo sapiens GABARAP (HsGABARAP), Saccharomyces cerevisiae ATG8 (ScATG8) and Trypanosoma brucei ATG8 (TbATG8).

Expression Analysis of GmATG8 Genes Responsive to N Limitation

Real-time RT-PCR was used to compare the expression of GmATG8s induced by nitrogen starvation. Four-day-old soybean seedlings were transferred to half Hoagland solution, and then transferred to N-deficient half Hoagland with the cotyledons cut off, when the primary leaves unfolded. After 3 d and 6 d N-limitation treatment, the primary leaves were harvested for gene expression analysis. The half Hoagland hydroponic Solution (pH 5.7), contained 3 mM KNO3, 0.5 mM sodium biphosphate buffer (pH 5.7), 2.0 mM Ca(NO3)2, 1.0 mM MgSO4, 0.25 mM (NH4)2SO4, 50 µM Fe-EDTA, 0.7 mM H3BO4, 170 µM MnCl2, 2.0 µM Na2MoO4, 100 µM NaCl, 5.0 µM CuSO4, 10 µM ZnSO4, and 0.1 µM CoCl2. The nitrogen-free solution was prepared by replacing KNO3 and Ca(NO3)2 with KCl and CaCl2, respectively. Three biological replicates, represented by three batches of plants grown at different time, were analyzed as follows.

RNA extraction and cDNA synthesis were done as described [65]. Total RNA was extracted, and residual genomic DNA was digested with RNase-free DNase I. The absence of genomic DNA was confirmed with a control PCR. First strand cDNA was synthesized from 2.0 µg of total RNA using AMV reverse transcriptase (Promega, Madison, WI, USA). cDNA samples diluted for 10 folds were used as templates.

Real-time RT-PCR analysis was performed using SYBR Green Perfect mix (TaKaRa, Dalian, China) on an iQ5 (Bio-Rad, California, USA). All reactions were performed under the following conditions: 95°C for 2 min; 40 cycles of 95°C for 10 s and 60°C for 30 s. All reactions were done at least in triplicates. Actin11 was used as an internal control. All primers used are listed in Table S2. Results from one out of three biological replicates were shown.

Functional Complementary Test in Yeast atg8 Mutant

The plasmid pESC-URA was constructed by ligating the corresponding GmATG8c ORF fragments and ScADH1 promoter fragments into Spe I and Bgl II?sites of pESC-URA and EcoR I?and Spe I sites of pESC-URA respectively. In addition, the plasmid pESC-URA was constructed by ligating the corresponding ScADH1 promoter fragments into EcoR I?and Spe I sites of pESC-URA alone for control. The primers used are listed in Table S2.

The plasmids pESC-URA with ADH1:GmATG8c or ADH1 promoter alone were introduced in the atg8 mutant (TN124 atg8Δ:KAN URA3 TRP1) respectively. Three OD600 cell extracts from mid-log phase growing cells or 4 hours nitrogen-starved cells were resolved by SDS-PAGE followed by immunoblotting with anti-APE1 antibody, and equal transfer to the membrane was confirmed by Ponceau S staining (Sigma, USA). Preparation of cell extracts and immunoblot analyses were performed as described [66]. Intensities of band signals of mAPE1 were quantified using the software Quantity One (Bio-Rad). Five independent replicates were done to give the typical results shown here. The wild-type yeast (strain TN124) [67], atg8 yeast mutant (TN124 atg8Δ:KAN URA3 TRP1) and anti-APE1 antibody were gifts from Dr. Zhiping Xie (Nankai University, China).

Construction of Binary Over-expression Vector of GmATG8c Gene and Plant Transformation

The complete coding region of GmATG8c was amplified by PCR using a pair of primers F (Nco I site was introduced) and R (BstE II site was introduced) with full-length cDNA of soybean leaves as the template. The specific PCR fragment was then digested and ligated into binary vector pCAMBIA1301 between Nco I and BstE II sites, resulting in the 35S:GmATG8c fusion gene. The authenticity of the fusion construct was confirmed by DNA sequencing. Agrobacterium-mediated floral dipping method [68] was used for Arabidopsis transformation. The primers used are listed in Table S2.

T1 transformants of Arabidopsis were screened on half-strength Murashige and Skoog (1/2 MS) medium containing 30 µg/ml hygromycin. Within 2 weeks, seedlings with green true leaves were identified as transformants and transferred to soil, then verified by PCR and RT-PCR. Seeds collected from individual T1 tranformants were scored for segregation on hygromycin to determine the number of T-DNA insertion. Five homozygous T2 lines were obtained, and the ectopic expression of GmATG8c were verified by real-time PCR (data not shown). For this study, three homozygous lines were analyzed in details. The transcript levels of the endogenous autophagy-related genes were analyzed through real-time RT-PCR. Primers used are listed in Table S2 or as described [30], [69].

The soybean hypocotyl segments were inoculated with Agrobacterium, co-cultivated in darkness for 3 days, and then transferred to callus induction medium (Liu, et al, unpublished) for calli induction. The callus tissues that were induced at both the acropetal and basipetal ends were excised from the hypocotyl segment explants and transferred to callus screening medium (Liu, et al, unpublished) for further selection. Over-expression of GmATG8c was confirmed by RT-PCR. Three independent transgenic lines were analyzed.

The method for generation of 35S:GmATG8c transgenic tomato was described in Methods S1.

Plant Protein Isolation and Immunoblot Analysis

Total proteins were extracted in extraction buffer as described [65], and the homogenates were centrifuged at 12,000 rpm for 15 min at 4°C. Protein concentration was determined with the Bradford method [70]. The full length GmATG8c was used to raise a polyclonal antibody (Abmart, Shanghai China). 20 µg protein extracts were resolved by Tricine-SDS-PAGE followed by immunoblotting with anti-GmATG8c antibody and anti-Actin (Santa Cruz Biotechnology, USA). Five independent replicates were done to give the typical results shown here.

Plant Growth Conditions and Phenotypic Analysis

Arabidopsis seeds (Columbia-0) were surface-sterilized in 1% (v/v) sodium hypochlorite for 2 min, washed with sterilized water for 10 times, stratified in water at 4°C for 2 days in the dark, and grown on half-strength Murashige and Skoog (1/2 MS) medium (0.8% agar (w:v), pH 5.7, 1% sucrose). The plates were incubated in a plant growth chamber (22°C/19°C, 16 h light/8 h dark, with a photosynthetic photon flux density (PPFD) of 80 µmol m−2 s−1). Plants were transferred to soil after 9 days to analyze the basal phenotype in the sufficient nutrient condition, and the maximum rosette radius and the plant height were recorded every day. Seeds were harvested individually at the growth stage 9.7 [49] and weighed. Three biological repeats were done to give the typical results shown here.

To analyze and illustrate the plant architecture, the numbers of branches of 7-week-old wild-type and 35S:GmATG8c lines were counted, and the lengths of each branch were measured. To properly describe the branching patterns, individual branches that have been produced by over 70% of plants were accepted as present. Then the average lengths of individual branches were calculated for illustration using Adobe Illustrator (CS5).

To study the responses to low nitrogen treatment, wild-type and GmATG8c constitutive-expressing calli were grown on MS medium with 0 mM, 1 mM, 5 mM and 60 mM nitrogen for up to a month under a long-day photoperiod (16 h/8 h) at 25°C. Three biological repeats of three transgenic lines with different culture runs were done.

To study the response of the wild-type and 35S:GmATG8c transgenic Arabidopsis towards nitrogen starvation, young seedlings and adult plants were tested separately. Five-day-old seedlings grown in the long-day photoperiod (16 h/8 h)were transferred to N-deficient agar medium with 1% Suc: Murashige and Skoog micronutrient salts (Sigma), 3 mM CaCl2, 1.5 mM MgSO4, 1.25 mM KH2PO4, 5 mM KCl, 2 mM MES, pH 5.7. Effects of N limitation on plant growth in liquid condition were done as described [71]. For nitrogen-limitation treatment of adult plants, plants were grown hydroponically on vermiculite and perlite (1∶1, v/v) with the half Hoagland hydroponic Solution. When the plants started bolting at about 23 DAS, the solution was replaced with either the same nitrogen-rich solution as the control or nitrogen-free solution, and plants were grown for 5 days before the rosette leaves and stems were harvest respectively.

For C-limiting experiments, seedlings grown in a short-day photoperiod (8 h/16 h) for 9 days were transferred to soil and grown for 33 more days. The plants were then transferred to continuous darkness for 9 days, and then returned to the short-day photoperiod for a 25-day recovery. Two independent experiments were done; results from one were shown here.

Quantification of Nitrogen, Soluble Sugar and Proteins

The juvenile and adult rosette leaves and the stems of the seedlings after 5-day N-limitation treatment were collected separately and rapidly frozen in liquid N2 and then stored at −80°C until analysis. Total leaf nitrogen was determined by Kjeldahl method [72]. Total proteins were extracted in extraction buffer as described [65], and the homogenates were centrifuged at 12,000 rpm for 15 min at 4°C. Protein concentrations of the supernatants were determined using the Bradford method [70]. Soluble sugar contents were determined as described [73]. Data shown are representative of at least three independent experiments; each includes three technical repeats with ten seedlings.

Supporting Information

Figure S1.

Overall structure of GmATG8c, HsGABARAP, ScATG8 and TbATG8. Image colored by rainbow N → C terminus.


Figure S2.

Immunoblot analysis of the purified GmATG8c and protein extracts from one representative transgenic line (L-4).


Figure S3.

Total contents of nitrogen, protein, and soluble sugar in rosette leaves and stems. Wild-type and 35S:GmATG8c plants were grown hydroponically in half Hoagland’s until bolting (23 days after sowing), and then maintained in either the same nitrogen-rich solution (+N) or transferred to nitrogen-free solution (−N) for another 5 days. Total nitrogen (A), protein (B) and soluble sugar contents (C) in the rosette leaves and stem were measured. NS: N sufficient, ND: N deficient.*, p<0.05 (t-test); significant difference from the wild-type (WT).


Figure S4.

Inflorescences of 7-week-old wild-type and transgenic plants. Ectopic expression of GmATG8c extended the period of flowering in Arabidopsis. White arrows indicate the flowers. WT: the wild-type; OE: 35S:GmATG8c transgenic plants.


Figure S5.

Heterologous expression of GmATG8c promotes growth and fruit setting in Tomato. A, Six-week-old wild-type (WT) and 35S:GmATG8c transgenic tomato. B, Growth curves of the plant height of the wild-type (WT) and 35S:GmATG8c tomato. C and D, Side views of eleven-week-old (C) and twelve-week-old (D) wild-type (WT) and 35S:GmATG8c tomato.


Table S1.

Genbank Accession Numbers Associated with soybean (Glycine max, Gm), Physcomitrella patens (Pp), Selaginella moellendorffii (Sm) and Ostreococcus lucimarinus (Ol) ATG8s.


Table S3.

Rosette radius of the wild-type and 35S:GmATG8c lines without (+N) and with (−N) 5 days of nitrogen starvation. Data shown are the means ±SD of one representative biological replicate (n = 21) out of three. *, p<0.05 (t-test); significant difference from the wild-type (WT).


Table S4.

The flowering time and the fruit setting of the wild-type and 35S:GmATG8c transgenic tomato. *, p<0.05 (t-test); significant difference from the wild-type (WT), n = 21.


Methods S1.

Description of the generation of the 35S:GmATG8c transgenic tomato in this study.



We wish to thank Dr. Zhiping Xie for the yeast TN124 and atg8 mutant strain and anti-APE1 antibody, and our lab members Mr. Dasong Chang for soybean calli preparation, Mr. Min Zhang for helping with nitrogen and sugar content measurements, and Mr. Haoqing Ao for phenotype recording of Arabidopsis.

Author Contributions

Conceived and designed the experiments: TX DX NNW QG. Performed the experiments: TX DX DL WC. Analyzed the data: TX DX QG. Wrote the paper: TX DX QG.


  1. 1. Good AG, Shrawat AK, Muench DG (2004) Can less yield more? Is reducing nutrient input into the environment compatible with maintaining crop production? Trends in Plant Science 9: 597–605.
  2. 2. Lam HM, Coschigano KT, Oliveira IC, Melo-Oliveira R, Coruzzi GM (1996) The Molecular-Genetics of Nitrogen Assimilation into Amino Acids in Higher Plants. Annu Rev Plant Physiol Plant Mol Biol 47: 569–593.
  3. 3. Vitousek C, John Aber, Robert W Howarth, Gene E Likens, Pamela A Matson, et al. (1997) Human alteration of the global nitrogen cycle causes and Consequences. Ecol Appl. pp. 737–750.
  4. 4. Woods J, Williams A, Hughes JK, Black M, Murphy R (2010) Energy and the food system. Philos Trans R Soc Lond B Biol Sci 365: 2991–3006.
  5. 5. Mulvaney RL, Khan SA, Ellsworth TR (2009) Synthetic nitrogen fertilizers deplete soil nitrogen: a global dilemma for sustainable cereal production. J Environ Qual 38: 2295–2314.
  6. 6. Shibles RM (1998) Soybean nitrogen acquisition and utilization. In Proc of the 28th North Central Extension–Industry Soil Fertility Conf. St. Louis, MO. pp. 5–11.
  7. 7. K Tewari TS, H Fujikake, N Ohtake, K Sueyoshi, Y Takahashi, et al. (2004) Effect of Deep Placement of N Fertilizers and Different Inoculation Methods of Bradyrhizobia on Growth, N2 Fixation Activity and N Absorption Rate of Field-grown Soybean Plants. Journal of Agronomy and Crop Science 190: 46–58.
  8. 8. Mickelson S, See D, Meyer FD, Garner JP, Foster CR, et al. (2003) Mapping of QTL associated with nitrogen storage and remobilization in barley (Hordeum vulgare L.) leaves. Journal of Experimental Botany 54: 801–812.
  9. 9. Masclaux-Daubresse C, Reisdorf-Cren M, Orsel M (2008) Leaf nitrogen remobilisation for plant development and grain filling. Plant Biol (Stuttg) 10: Suppl 123–36.
  10. 10. Masclaux-Daubresse C, Daniel-Vedele F, Dechorgnat J, Chardon F, Gaufichon L, et al. (2010) Nitrogen uptake, assimilation and remobilization in plants: challenges for sustainable and productive agriculture. Annals of Botany 105: 1141–1157.
  11. 11. Masclaux-Daubresse C, Chardon F (2011) Exploring nitrogen remobilization for seed filling using natural variation in Arabidopsis thaliana. Journal of Experimental Botany 62: 2131–2142.
  12. 12. Le Gouis J, Kichey T, Hirel B, Heumez E, Dubois F (2007) In winter wheat (Triticum aestivum L.), post-anthesis nitrogen uptake and remobilisation to the grain correlates with agronomic traits and nitrogen physiological markers. Field Crops Research 102: 22–32.
  13. 13. Zhang WH, Zhou YC, Dibley KE, Tyerman SD, Furbank RT, et al. (2007) Nutrient loading of developing seeds. Functional Plant Biology 34: 314–331.
  14. 14. Patrick JW, Offler CE (2001) Compartmentation of transport and transfer events in developing seeds. Journal of Experimental Botany 52: 551–564.
  15. 15. Liu Y, Schiff M, Czymmek K, Talloczy Z, Levine B, et al. (2005) Autophagy regulates programmed cell death during the plant innate immune response. Cell 121: 567–577.
  16. 16. Xie Z, Klionsky DJ (2007) Autophagosome formation: core machinery and adaptations. Nat Cell Biol 9: 1102–1109.
  17. 17. Suzuki K, Ohsumi Y (2007) Molecular machinery of autophagosome formation in yeast, Saccharomyces cerevisiae. FEBS Lett 581: 2156–2161.
  18. 18. Okamoto K, Kondo-Okamoto N, Ohsumi Y (2009) Mitochondria-anchored receptor Atg32 mediates degradation of mitochondria via selective autophagy. Dev Cell 17: 87–97.
  19. 19. Yoshimoto K, Takano Y, Sakai Y (2010) Autophagy in plants and phytopathogens. Febs Letters 584: 1350–1358.
  20. 20. Ohsumi Y, Suzuki K, Kondo C, Morimoto M (2010) Selective Transport of alpha-Mannosidase by Autophagic Pathways IDENTIFICATION OF A NOVEL RECEPTOR, Atg34p. Journal of Biological Chemistry 285: 30019–30025.
  21. 21. Kirisako T, Baba M, Ishihara N, Miyazawa K, Ohsumi M, et al. (1999) Formation process of autophagosome is traced with Apg8/Aut7p in yeast. Journal of Cell Biology 147: 435–446.
  22. 22. Xie Z, Nair U, Klionsky DJ (2008) Atg8 controls phagophore expansion during autophagosome formation. Mol Biol Cell 19: 3290–3298.
  23. 23. Nakatogawa H, Ichimura Y, Ohsumi Y (2007) Atg8, a ubiquitin-like protein required for autophagosome formation, mediates membrane tethering and hemifusion. Cell 130: 165–178.
  24. 24. Hutchins MU, Klionsky DJ (2001) Vacuolar localization of oligomeric alpha-mannosidase requires the cytoplasm to vacuole targeting and autophagy pathway components in Saccharomyces cerevisiae. J Biol Chem 276: 20491–20498.
  25. 25. Yamaguchi M, Noda NN, Nakatogawa H, Kumeta H, Ohsumi Y, et al. (2010) Autophagy-related protein 8 (Atg8) family interacting motif in Atg3 mediates the Atg3-Atg8 interaction and is crucial for the cytoplasm-to-vacuole targeting pathway. J Biol Chem 285: 29599–29607.
  26. 26. Yorimitsu T, Klionsky DJ (2005) Atg11 links cargo to the vesicle-forming machinery in the cytoplasm to vacuole targeting pathway. Molecular Biology of the Cell 16: 1593–1605.
  27. 27. Chang CY, Huang WP (2007) Atg19 mediates a dual interaction cargo sorting mechanism in selective autophagy. Mol Biol Cell 18: 919–929.
  28. 28. Fujioka Y, Noda NN, Fujii K, Yoshimoto K, Ohsumi Y, et al. (2008) In vitro reconstitution of plant Atg8 and Atg12 conjugation systems essential for autophagy. J Biol Chem 283: 1921–1928.
  29. 29. Doelling JH, Walker JM, Friedman EM, Thompson AR, Vierstra RD (2002) The APG8/12-activating enzyme APG7 is required for proper nutrient recycling and senescence in Arabidopsis thaliana. J Biol Chem 277: 33105–33114.
  30. 30. Yoshimoto K, Hanaoka H, Sato S, Kato T, Tabata S, et al. (2004) Processing of ATG8s, ubiquitin-like proteins, and their deconjugation by ATG4s are essential for plant autophagy. Plant Cell 16: 2967–2983.
  31. 31. Yoshimoto K, Jikumaru Y, Kamiya Y, Kusano M, Consonni C, et al. (2009) Autophagy negatively regulates cell death by controlling NPR1-dependent salicylic acid signaling during senescence and the innate immune response in Arabidopsis. Plant Cell 21: 2914–2927.
  32. 32. Thompson AR (2005) Autophagic Nutrient Recycling in Arabidopsis Directed by the ATG8 and ATG12 Conjugation Pathways. Plant Physiology 138: 2097–2110.
  33. 33. Phillips AR, Suttangkakul A, Vierstra RD (2008) The ATG12-Conjugating Enzyme ATG10 Is Essential for Autophagic Vesicle Formation in Arabidopsis thaliana. Genetics 178: 1339–1353.
  34. 34. Lenz HD, Haller E, Melzer E, Kober K, Wurster K, et al. (2011) Autophagy differentially controls plant basal immunity to biotrophic and necrotrophic pathogens. Plant Journal.
  35. 35. Lai Z, Wang F, Zheng Z, Fan B, Chen Z (2011) A Critical Role of Autophagy in Plant Resistance to Necrotrophic Fungal Pathogens. Plant Journal.
  36. 36. Liu TB, Liu XH, Lu JP, Zhang L, Min H, et al. (2010) The cysteine protease MoAtg4 interacts with MoAtg8 and is required for differentiation and pathogenesis in Magnaporthe oryzae. Autophagy 6: 74–85.
  37. 37. Chung T, Suttangkakul A, Vierstra RD (2009) The ATG autophagic conjugation system in maize: ATG transcripts and abundance of the ATG8-lipid adduct are regulated by development and nutrient availability. Plant Physiol 149: 220–234.
  38. 38. Hanaoka H, Noda T, Shirano Y, Kato T, Hayashi H, et al. (2002) Leaf senescence and starvation-induced chlorosis are accelerated by the disruption of an Arabidopsis autophagy gene. Plant Physiol 129: 1181–1193.
  39. 39. Honig A, Avin-Wittenberg T, Ufaz S, Galili G (2012) A New Type of Compartment, Defined by Plant-Specific Atg8-Interacting Proteins, Is Induced upon Exposure of Arabidopsis Plants to Carbon Starvation. Plant Cell 24: 288–303.
  40. 40. Slavikova S, Ufaz S, Avin-Wittenberg T, Levanony H, Galili G (2008) An autophagy-associated Atg8 protein is involved in the responses of Arabidopsis seedlings to hormonal controls and abiotic stresses. Journal of Experimental Botany 59: 4029–4043.
  41. 41. Xiong Y, Contento AL, Bassham DC (2005) AtATG18a is required for the formation of autophagosomes during nutrient stress and senescence in Arabidopsis thaliana. The Plant Journal 42: 535–546.
  42. 42. Slavikova S, Shy G, Yao Y, Glozman R, Levanony H, et al. (2005) The autophagy-associated Atg8 gene family operates both under favourable growth conditions and under starvation stresses in Arabidopsis plants. Journal of Experimental Botany 56: 2839–2849.
  43. 43. Wilkins MR, Gasteiger E, Bairoch A, Sanchez JC, Williams KL, et al. (1999) Protein identification and analysis tools in the ExPASy server. Methods Mol Biol 112: 531–552.
  44. 44. Kelley LA, Sternberg MJE (2009) Protein structure prediction on the Web: a case study using the Phyre server. Nature Protocols 4: 363–371.
  45. 45. Shintani T, Klionsky DJ (2004) Cargo proteins facilitate the formation of transport vesicles in the cytoplasm to vacuole targeting pathway. J Biol Chem 279: 29889–29894.
  46. 46. Wang CW, Stromhaug PE, Shima J, Klionsky DJ (2002) The Ccz1-Mon1 protein complex is required for the late step of multiple vacuole delivery pathways. Journal of Biological Chemistry 277: 47917–47927.
  47. 47. He Y, Guo XL, Lu R, Niu B, Pasapula V, et al. (2009) Changes in morphology and biochemical indices in browning callus derived from Jatropha curcas hypocotyls. Plant Cell Tissue and Organ Culture 98: 11–17.
  48. 48. Izumi M, Wada S, Makino A, Ishida H (2010) The autophagic degradation of chloroplasts via rubisco-containing bodies is specifically linked to leaf carbon status but not nitrogen status in Arabidopsis. Plant Physiol 154: 1196–1209.
  49. 49. Boyes DC, Zayed AM, Ascenzi R, McCaskill AJ, Hoffman NE, et al. (2001) Growth stage-based phenotypic analysis of Arabidopsis: a model for high throughput functional genomics in plants. Plant Cell 13: 1499–1510.
  50. 50. Brouquisse R, Gaudillere JP, Raymond P (1998) Induction of a carbon-starvation-related proteolysis in whole maize plants submitted to light/dark cycles and to extended darkness. Plant Physiology 117: 1281–1291.
  51. 51. Matt P, Schurr U, Klein D, Krapp A, Stitt M (1998) Growth of tobacco in short-day conditions leads to high starch, low sugars, altered diurnal changes in the Nia transcript and low nitrate reductase activity, and inhibition of amino acid synthesis. Planta 207: 27–41.
  52. 52. Behrends C, Sowa ME, Gygi SP, Harper JW (2010) Network organization of the human autophagy system. Nature 466: 68–76.
  53. 53. Fujiki Y, Yoshimoto K, Ohsumi Y (2007) An Arabidopsis homolog of yeast ATG6/VPS30 is essential for pollen germination. Plant Physiol 143: 1132–1139.
  54. 54. Sakhonwasee S, Abel S (2009) Autophagy Sustains the Arabidopsis Root Meristem during Phosphate Starvation.
  55. 55. Hirel B, Bertin P, Quillere I, Bourdoncle W, Attagnant C, et al. (2001) Towards a better understanding of the genetic and physiological basis for nitrogen use efficiency in maize. Plant Physiol 125: 1258–1270.
  56. 56. Masclaux C, Quillere I, Gallais A, Hirel B (2001) The challenge of remobilisation in plant nitrogen economy. A survey of physio-agronomic and molecular approaches. Annals of Applied Biology 138: 69–81.
  57. 57. Bassham DC, Laporte M, Marty F, Moriyasu Y, Ohsumi Y, et al. (2006) Autophagy in development and stress responses of plants. Autophagy 2: 2–11.
  58. 58. Pathak RR, Altaf Ahmad, Sunila Lochab, Nandula Raghuram (2008) Molecular physiology of plant nitrogen use efficiency and biotechnological options for its enhancement. Current 94:
  59. 59. Yamaya T, Obara M, Nakajima H, Sasaki S, Hayakawa T, et al. (2002) Genetic manipulation and quantitative-trait loci mapping for nitrogen recycling in rice. Journal of Experimental Botany 53: 917–925.
  60. 60. Ainsworth EA, Yendrek CR, Skoneczka JA, Long SP (2012) Accelerating yield potential in soybean: potential targets for biotechnological improvement. Plant Cell and Environment 35: 38–52.
  61. 61. Contento AL, Xiong Y, Bassham DC (2005) Visualization of autophagy in Arabidopsis using the fluorescent dye monodansylcadaverine and a GFP-AtATG8e fusion protein. Plant Journal 42: 598–608.
  62. 62. Ishida H, Yoshimoto K, Izumi M, Reisen D, Yano Y, et al. (2008) Mobilization of Rubisco and Stroma-Localized Fluorescent Proteins of Chloroplasts to the Vacuole by an ATG Gene-Dependent Autophagic Process. Plant Physiology 148: 142–155.
  63. 63. Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 4673–4680.
  64. 64. Kumar S, Tamura K, Nei M (1994) MEGA: Molecular Evolutionary Genetics Analysis software for microcomputers. Comput Appl Biosci 10: 189–191.
  65. 65. Liu D, Gong Q, Ma Y, Li P, Li J, et al. (2010) cpSecA, a thylakoid protein translocase subunit, is essential for photosynthetic development in Arabidopsis. Journal of Experimental Botany 61: 1655–1669.
  66. 66. Hamasaki M, Noda T, Baba M, Ohsumi Y (2005) Starvation triggers the delivery of the endoplasmic reticulum to the vacuole via autophagy in yeast. Traffic 6: 56–65.
  67. 67. Noda T, Matsuura A, Wada Y, Ohsumi Y (1995) Novel system for monitoring autophagy in the yeast Saccharomyces cerevisiae. Biochem Biophys Res Commun 210: 126–132.
  68. 68. Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant Journal 16: 735–743.
  69. 69. Rose TL, Bonneau L, Der C, Marty-Mazars D, Marty F (2006) Starvation-induced expression of autophagy-related genes in Arabidopsis. Biol Cell 98: 53–67.
  70. 70. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254.
  71. 71. Suttangkakul A, Li F, Chung T, Vierstra RD (2011) The ATG1/ATG13 protein kinase complex is both a regulator and a target of autophagic recycling in Arabidopsis. Plant Cell 23: 3761–3779.
  72. 72. Kjeldahl J (1883) A new method for the determination of nitrogen in organic matter. Z Anal Chem 22: 336.
  73. 73. Geiger M, Walch-Liu P, Engels C, Harnecker J, Schulze ED, et al. (1998) Enhanced carbon dioxide leads to a modified diurnal rhythm of nitrate reductase activity in older plants, and a large stimulation of nitrate reductase activity and higher levels of amino acids in young tobacco plants. Plant Cell and Environment 21: 253–268.