Pepper asparagine synthetase 1 (CaAS1) is required for plant nitrogen assimilation and defense responses to microbial pathogens


  • In Sun Hwang,

    1. Laboratory of Molecular Plant Pathology, School of Life Sciences and Biotechnology, Korea University, Anam-dong, Sungbuk-ku, Seoul 136–713, Korea
    Search for more papers by this author
  • Soo Hyun An,

    1. Breeding Research Institute, Dongbu Hannong Co., Ltd., Deokbong-ri 481–3, Yangseong-myeon, Anseong-si, Gyeonggi-do 456–933, Korea
    Search for more papers by this author
  • Byung Kook Hwang

    Corresponding author
    1. Laboratory of Molecular Plant Pathology, School of Life Sciences and Biotechnology, Korea University, Anam-dong, Sungbuk-ku, Seoul 136–713, Korea
      (fax +82 2 921 1715; e-mail
    Search for more papers by this author

(fax +82 2 921 1715; e-mail


Asparagine synthetase is a key enzyme in the production of the nitrogen-rich amino acid asparagine, which is crucial to primary nitrogen metabolism. Despite its importance physiologically, the roles that asparagine synthetase plays during plant defense responses remain unknown. Here, we determined that pepper (Capsicum annuum) asparagine synthetase 1 (CaAS1) is essential for plant defense to microbial pathogens. Infection with Xanthomonas campestris pv. vesicatoria (Xcv) induced early and strong CaAS1 expression in pepper leaves and silencing of this gene resulted in enhanced susceptibility to Xcv infection. Transgenic Arabidopsis (Arabidopsis thaliana) plants that overexpressed CaAS1 exhibited enhanced resistance to Pseudomonas syringae pv. tomato DC3000 and Hyaloperonospora arabidopsidis. Increased CaAS1 expression influenced early defense responses in diseased leaves, including increased electrolyte leakage, reactive oxygen species and nitric oxide bursts. In plants, increased conversion of aspartate to asparagine appears to be associated with enhanced resistance to bacterial and oomycete pathogens. In CaAS1-silenced pepper and/or CaAS1-overexpressing Arabidopsis, CaAS1-dependent changes in asparagine levels correlated with increased susceptibility or defense responses to microbial pathogens, respectively. Linking transcriptional and targeted metabolite studies, our results suggest that CaAS1 is required for asparagine synthesis and disease resistance in plants.


In plants, nitrogen storage and remobilization is an important metabolic process that occurs continually throughout the normal course of plant growth and development. Nitrogen is crucial for the nutrient relationship between plants and pathogens. Nitrogen mobilization is promoted by pathogen infection and it represents a strategy that deprives the plant of nutrients. Nutritional and metabolic changes also occur in the defense response against pathogens (Hammond-Kosack and Parker, 2003). In compatible interactions between plants and pathogens, pathogen attack leads to modifications in plant nitrogen levels and important metabolic changes in the infected plants. During these nutritional and metabolic shifts, plant pathogens may become adapted to modifications in plant nitrogen content and, ultimately, these metabolic changes may lead to the benefit of the pathogen (Tavernier et al., 2007).

Similar metabolic pathways are activated during pathogen infection and abiotic stress, as in the plant senescence process (Pageau et al., 2006). Biotic and abiotic stresses up-regulate the expression of genes encoding proteins involved in the nitrogen mobilization process, including asparagine synthetase, glutamine synthetase and glutamate dehydrogenase (Olea et al., 2004; Zurbriggen et al., 2009). In many higher plants, asparagine (ASN) and glutamine (GLN) represent central intermediates in nitrogen metabolism and they contribute to nitrogen transport. High levels of ASN are observed in infected leaves and GLN may also be used as a vehicle for nitrogen transport (Zurbriggen et al., 2009). ASN is involved in the transport of nitrogen from infected plant tissues during Pseudomonas syringae infection (Pérez-García et al., 1998a,b), and it is also preferred for nitrogen transport and/or transient storage of nitrogen, since it has a higher nitrogen/carbon ratio than glutamine or other amino acids (Lam et al., 1994; Potel et al., 2009). In addition, ASN is one of the major nitrogenous compounds in the phloem of several legumes (Atkins et al., 1975). During infection with Xanthomonas campestris pv. vesicatoria (Xcv), pfld tobacco plants expressing cyanobacterial flavodoxin (Fld) in chloroplasts showed significantly increased levels of ASN and GLN (Zurbriggen et al., 2009). It is generally accepted that in plants, the main route for ASN biosynthesis is the conversion of aspartate (ASP) to ASN by a GLN-dependent asparagine synthetase and that this is the primary enzyme for the production of ASN.

Asparagine synthetase is found mainly in the roots, where it is involved in glutamine and ammonia pathways. In tomato leaves, pathogen infection induces several genes encoding asparagine synthetases (Olea et al., 2004) and full-length cDNA clones encoding this enzyme have been characterized from many plant species including pea, asparagus, Arabidopsis, rice and soybean (Tsai and Coruzzi, 1990; Davies and King, 1993; Lam et al., 1994; Watanabe et al., 1996; Hughes et al., 1997). Significant progress has been made in understanding the molecular roles played by asparagine synthetase in these plants. Asparagine synthetases is comprised of 579–591 amino acid residues (approximately 65 kDa) and they contain a glutamine-binding site and an AsnB domain (Shi et al., 1997). However, the role that these enzymes play in plant defense remains poorly understood. High levels of asparagine synthetase activity are induced in tomato leaves infected by the bacterial pathogen P. syringae pv. tomato (Pst; Olea et al., 2004). In common bean, the promoter PVAS3 is induced by environmental stresses and pathogen infection, and a number of different regulatory elements could be involved in its regulation (Parra-Peralbo et al., 2009). ASN accumulation appears to be induced by abiotic and biotic stresses, such as mineral deficiencies, drought, salt, toxic metals and pathogen attack (Lea et al., 2007). The hemibiotrophic fungus Colletotrichum lindemuthianum, which causes an anthracnose disease on common bean, can utilize a wide range of nitrogen sources. When the preferentially used primary sources, such as ammonia and glutamine, are absent or present at low concentrations, ASP and ASN can represent valuable secondary sources of nitrogen (Pellier et al., 2003).

Nitric oxide (NO) is a widespread intra- and intercellular messenger. It functions in a diverse range of physiological processes in plants, including growth and development, as well as in innate immunity (Hausladen and Stamler, 1998). In plants, NO formation is closely linked to nitrogen assimilation and metabolism (Grün et al., 2006; Stöhr, 2007) and it functions as an essential negative regulator in senescence signaling cascades (Ma et al., 2010). In particular, NO acts in synergy with reactive oxygen species (ROS) and this relationship has been best-characterized for the defense response against pathogen attack (Besson-Bard et al., 2008; Perchepied et al., 2010). Defense responses, such as hypersensitive response (HR) cell death, require the balanced production of NO and ROS (Asai et al., 2010). Recently, plants with altered NO levels have been used to obtain genetic evidence for the importance of NO in gene induction (Yoshioka et al., 2009).

Here, the full-length cDNA encoding asparagine synthetase 1, CaAS1, was isolated from pepper plants (Capsicum annuum L.). Expression of CaAS1 in defense response was evaluated in plants infected by microbial pathogens. The effects of CaAS1 loss- or gain-of-function were examined using a virus-induced gene silencing (VIGS) system and constitutively activated transgenic CaAS1 overexpression (OX) plants. In this study, CaAS1-silenced plants showed enhanced susceptibility to virulent and avirulent Xcv strains (Ds1 and Bv5–4a, respectively). Transgenic CaAS1-OX Arabidopsis plants exhibited enhanced resistance to the bacterial pathogens P. syringae pv. tomato DC3000 (Pst DC3000) and DC3000 (avrRpm1), as well as the biotrophic oomycete Hyaloperonospora arabidopsidis. Analysis of individual amino acids revealed that CaAS1 is essential for the biosynthesis of ASN from ASP in plants. Taken together, these findings provide convincing evidence that CaAS1 functions as a key regulator in a signal transduction pathway promoting disease resistance in pepper and Arabidopsis plants.


Isolation and identification of CaAS1 cDNA

A pepper cDNA library constructed from Xcv-infected leaves was screened using a macro cDNA array method. DNA sequencing revealed a positively hybridizing clone, designated CaAS1 (2230 bp in length), with a 1773 bp open reading frame that encodes an asparagine synthetase of 591 amino acids (Figure S1). Using the GenBank database, BLASTX analyses indicated that CaAS1 shared amino acid sequence identity with asparagine synthetases from diverse sources including Triphysaria versicolor (accession no. AAD05035, Delavault et al., 1998), Elaeagnus umbellata (accession no. AAC16325, Kim et al., 1999), garden asparagus (Asparagus officinalis, accession no. CAA67889, Moyle et al., 1996), soybean (Glycine max, accession no. AAC09952, Yamagata et al., 1998), Arabidopsis ASN 1 (A. thaliana, accession no. NP_190318) and rice (Oryza sativa, accession no. BAD54377) (Figure S2). Sequence comparisons revealed that CaAS1 and its related plant proteins all contain an AsnB domain and an ASN synthetase domain.

CaAS1 expression in pepper plants

To determine whether CaAS1 is induced during normal plant growth and development as well as pathogen infection, transcription of this gene was investigated using RNA gel blot analysis. CaBPR1 expression was used as a defense-gene marker. Although CaAS1 transcripts were not detected in the leaves, stems, roots or green and red fruits of pepper plants, high levels of CaAS1 and CaBPR1 transcripts were found in the flower (Figure 1a). Consistent with our results, the tomato asparagine synthetase gene was reported to be expressed constitutively in flowers of tomato (Olea et al., 2004). However, the function of asparagine synthetases in flowers remains largely unknown.

Figure 1.

 RNA gel blot analyses of CaAS1 and CaBPR1 expression in pepper.
RNA gel blots were hybridized with a 32P-labeled probe against the CaAS1 3′-UTR or the full-length cDNA clone of CaBPR1. Equal loading (20 μg) was verified by visualizing RNA on gels stained with ethidium bromide. (a) Expression of CaAS1 in various organs of pepper plants. (b) Expression of CaAS1 and CaBPR1 in pepper leaves inoculated with Xanthomonas campestris pv. vesicatoria (Xcv) strains Ds1 (virulent, compatible) and Bv5-4a (avirulent, incompatible). H, healthy leaves; M, mock-treated with 10 mm MgCl2. (c) Expression of CaAS1 and CaBPR1 in pepper leaves at different time points after treatment with salicylic acid (SA, 5 mm) and methyl jasmonate (MeJA, 100 μm).

Infection with the virulent Ds1 (compatible) and avirulent Bv5–4a (incompatible) strains enhanced CaAS1 transcript levels in pepper leaves (Figure 1b). In compatible interactions, CaAS1 transcripts could not be detected during the first 5 h after inoculation; however, CaAS1 transcripts increased gradually thereafter. In contrast, CaAS1 transcripts were induced early and rapidly in response to infection with the avirulent Bv5-4a strain, peaking 5–15 h after inoculation (Figure 1b). CaAS1 expression was also examined with respect to abiotic stress responses in pepper plants. Significant induction of CaAS1 expression occurred in pepper leaves following treatment with salicylic acid (SA), methyl jasmonate (MeJA) or wounding (Figure 1c). MeJA treatment induced earlier (1–5 h) CaAS1 expression in pepper leaves than SA, which strongly induced transcription 25 h after treatment.

CaAS1-silenced plants exhibit enhanced susceptibility to Xcv infection

The VIGS system was used to knock down CaAS1 expression in pepper plants. CaAS1 was silenced in pepper plants using the TRV2 vector that contained CaAS1 (TRV:CaAS1). The empty TRV2 vector (TRV:00) was used as a negative control. No morphological difference was observed between TRV:00 and TRV:CaAS1 plants.

A reverse transcription polymerase chain reaction (RT-PCR) analysis showed that the level of CaAS1 transcripts was reduced drastically in CaAS1-silenced leaves during Xcv infection (Figure S3a), which indicated that CaAS1 was silenced effectively in pepper. The role that CaAS1 plays in the defense response during Xcv infection was investigated further in CaAS1-silenced plants that had been inoculated with virulent (Ds1; compatible) and avirulent (Bv5–4a; incompatible) strains of Xcv (105 cfu ml−1). In TRV:CaAS1 leaves, the virulent strain Ds1 caused severe disease symptoms, which were accompanied by increased bacterial growth (Figure 2a,b); however, infection with the avirulent strain Bv5-4a (5 × 104 cfu ml−1) did not cause any cell death or disease symptoms. Higher levels of bacterial growth were observed in CaAS1-silenced leaves compared to un-silenced leaves. When examined 2 and 4 days after inoculation, it was clear that both strains of Xcv grew faster in TRV:CaAS1 leaves than in leaves that bore the empty control. These data indicate that CaAS1-silenced plants are more susceptible to Xcv infection than plants that contained the empty vector. In particular, the avirulent Xcv strain Bv5-4a (incompatible) exhibited a much stronger increase in growth in CaAS1-silenced leaves than in control leaves, although an HR was not induced.

Figure 2.

 Enhanced susceptibility of CaAS1-silenced (TRV:CaAS1) pepper leaves to Xanthomonas campestris pv. vesicatoria (Xcv).
(a) Bacterial growth in leaves of empty vector control (TRV:00) and silenced plants at different time points after Xcv inoculation (5 × 104 cfu ml−1). (b) Disease symptoms on empty vector control and silenced leaves infected with Xcv (106 cfu ml−1). Infected leaves were stained with trypan blue 1 day after inoculation. (c) Quantification of ion leakage from empty vector control and silenced leaves infected with Xcv. (d) Quantification of H2O2 in leaves of empty and silenced pepper plants inoculated with Xcv. (e) Effect of CaAS1 silencing on the nitric oxide (NO) burst in empty vector control and silenced leaves 24 h after infection with Xcv. Sodium nitroprusside (SNP; 500 μm) was used as a positive control. Fluorescence was visualized by confocal laser scanning microscopy. Signal intensities were quantified by color histogram analysis. Values are presented as means ± standard deviations from three independent experiments. Asterisks indicate significant differences from empty vector control leaves (Student’s t-test, < 0.05).

To determine whether the silencing of CaAS1 affects electrolyte leakage from plant cells, ion conductivity was measured in pepper leaves infiltrated with Xcv (Figure 2c). Enhanced electrolyte leakage was compromised significantly in the leaves of CaAS1-silenced plants infected with strain Bv5–4a. This finding indicates that CaAS1 expression is required for hypersensitive cell death in pepper leaves. Xcv infection also promoted H2O2 production in pepper leaves (Figure 2d) and CaAS1 silencing attenuated H2O2 production during virulent and avirulent Xcv infection, especially at 24 h after Xcv inoculation.

NO is a product of plant nitrogen assimilation and it functions as an important signaling molecule in the defense responses of plants (Neill et al., 2003). During avirulent Xcv infection, NO generation was significantly reduced in CaAS1-silenced leaves compared with the empty vector control leaves (Figure 2e). This link between CaAS1 silencing and NO generation supports the hypothesis that CaAS1 is involved in the NO burst.

Next, the expression profiles of CaAS1 and the defense-related genes CaBPR1 (PR1) and CaDEF1 (defensin) were examined in empty vector control and CaAS1-silenced pepper plants during Xcv infection (Figure 3a). During infection with Xcv, CaAS1 expression was suppressed significantly in TRV:CaAS1 plants compared with TRV:00 control plants. CaAS1 silencing also reduced significantly the induction of CaBPR1 and CaDEF1 expression, especially in incompatible interactions. Notably, this effect was most apparent at relatively late time points after inoculation with the avirulent Xcv Bv5-4a strain. Similar expression patterns could also be observed for other defense-related genes, including CaSAR82A (SAR8.2), CaMBL1 (mannose binding lectin), CaPR10 and CaPOA1 (ascorbic peroxidase), in CaAS1-silenced plants (Figure S3b). SA signaling plays a crucial role in defense against hemibiotrophic pathogens such as Pst DC3000 (Glazebrook, 2005). Lower levels of free and total SA (free SA plus Glc-conjugated SA) accumulated in CaAS1-silenced leaves infected with the avirulent (incompatible) Xcv strain than in those of the empty vector control (Figure 3b). However, no significant differences in SA levels were detected between empty vector control and silenced leaves during virulent Xcv infection.

Figure 3.

 Effects of CaAS1 silencing on gene expression and SA accumulation in pepper leaves infected by Xanthomonas campestris pv. vesicatoria (Xcv).
(a) Real-time quantitative PCR analysis showing expression of CaAS1 and pepper defense-related genes in empty vector control leaves (TRV:00) and CaAS1-silenced (TRV:CaAS1) leaves inoculated with Xcv. Transcript levels were normalized using expression of pepper 18S rRNA. (b) Induction of free SA and total SA synthesis by Xcv infection. Values are presented as means ± standard deviations. Asterisks indicate significant differences from empty vector control leaves (Student’s t-test, < 0.05).

Lower ASN levels in CaAS1-silenced plants infected with Xcv

To determine whether or not the silencing of CaAS1 negatively regulates ASN biosynthesis, free amino acids levels were compared in healthy CaAS1-silenced, wild-type and empty vector control (TRV:00) leaves (Figure 4a). No significant differences were observed in the levels of ASN (ASN), glutamate (GLU) or glutamine (GLN) between these healthy, un-inoculated pepper plants. However, CaAS1-silenced leaves showed slightly higher ASP levels compared to wild-type or empty vector (TRV:00) control leaves, which indicated that CaAS1 did not play a crucial role in ASN pool production during normal plant development.

Figure 4.

 Effect of CaAS1 silencing on amino acid levels in pepper leaves.
(a) Amino acid contents of healthy wild-type, empty vector control (TRV:00) and silenced (TRV:CaAS1) leaves. (b) Changes in amino acid contents in empty vector control and silenced leaves inoculated with Xanthomonas campestris pv. vesicatoria (Xcv) strains Ds1 (virulent, compatible) and Bv5–4a (avirulent, incompatible). Values are presented as means ± standard deviations. Asterisks indicate significant differences from empty vector control leaves (Student’s t-test, < 0.05). ASP, aspartate; ASN, asparagine; GLU, glutamate; GLN, glutamine.

Amino acid levels were also determined in leaves from empty vector control (TRV:00) and silenced (TRV:CaAS1) pepper plants at 0 (immediately), 18 and 36 h after Xcv inoculation (Figures 4b and S4b). Total free amino acid levels were significantly elevated during the entire Xcv infection process. During the infection of TRV:00 plants with the virulent (compatible) Xcv strain Ds1, the ASP level rose slightly 36 h after inoculation and the ASN content rose up also albeit remaining at low levels. In the leaves of CaAS1-silenced plants, the enhancement of ASN production by Xcv infection was suppressed and the suppression was more drastic in incompatible than compatible interactions. More importantly, significantly enhanced levels of ASP, ASN, GLU and GLN were evident in empty vector control leaves during infection with the avirulent (incompatible) Xcv strain Bv5-4a but were not detected during virulent (compatible) Xcv infection. In contrast, ASN biosynthesis did not occur in CaAS1-silenced leaves during avirulent Xcv infection, possibly due to the knock down of CaAS1. In the incompatible interactions, no significant differences were found in the levels of ASP, GLU or GLN between empty vector control and silenced leaves. Collectively, these data indicate that the silencing of CaAS1 may impair the conversion of ASP to ASN during Xcv infection, but that the conversion of GLU to GLN remains unaffected. This impairment to ASN synthesis may be substantially greater in incompatible interactions.

Profiles of other free amino acids were examined in un-inoculated or Xcv-inoculated pepper leaves (Figure S4). With the exception of serine (SER), which accumulated to a higher level in silenced leaves, no significant differences were observed in other amino acid levels between wild-type and CaAS1-silenced pepper leaves (Figure S4a). However, significantly higher levels of some amino acids, including valine (VAL), isoleucine (ILE), phenylalanine (PHE) and tryptophan (TRP), accumulated in empty vector control leaves infected with Xcv. These results suggest that VAL, ILE, PHE and TRP may contribute to the plant resistance to Xcv infection.

CaAS1-OX Arabidopsis plants exhibit reduced susceptibility to infection with P. syringae pv. tomato and Hyaloperonospora arabidopsidis

Transgenic Arabidopsis plants that constitutively express CaAS1 were generated to examine CaAS1 gain-of-function effects. Constitutive expression of PR1 was significantly lower in CaAS1 overexpression (OX) transgenic lines than in wild-type plants (Figure S5a). In contrast, the expression of the jasmonic acid-dependent gene PDF1.2 was upregulated in the three CaAS1-OX transgenic lines.

To determine whether CaAS1 overexpression affects bacterial disease responses in Arabidopsis, CaAS1-OX transgenic plants were infected with Pst DC3000 and Pst DC3000 avrRpm1. With both pathogens, the leaves of wild-type plants (Col-0) developed more severe disease symptoms, which supported higher levels of bacterial growth than the CaAS1-OX transgenic plants (Figure 5a,b). Specifically, the overexpression of CaAS1 significantly inhibited the growth of Pst DC3000 and Pst DC3000 avrRpm1 in transgenic plants 3 days after inoculation.

Figure 5.

 Enhanced resistance of CaAS1-OX Arabidopsis transgenic lines to Pseudomonas syringae pv. tomato (Pst) infection.
(a) Disease symptoms in wild-type (WT) and transgenic leaves infected with Pst DC3000 or DC3000 (avrRpm1) (106 cfu ml−1). (b) Bacterial growth in WT and transgenic leaves. Infected with virulent Pst DC3000 or DC3000 (avrRpm1) (105 cfu ml−1). Different letters indicate significant differences from three independent experiments, as analyzed using the least significant difference (LSD) test (< 0.05). (c) Visualization of cell death and H2O2 accumulation in infected leaves stained with DAB and trypan blue, respectively. Bars = 0.2 mm. (d) Quantification of ion leakage from WT and transgenic leaves infected with Pst. (e) Quantification of H2O2 in leaves of WT and transgenic Arabidopsis plants after inoculation with Pst. (f) Effect of CaAS1 expression on the nitric oxide (NO) burst in leaves of wild-type and transgenic plants 24 h after infection with Pst. Sodium nitroprusside (SNP; 500 μm) was used as a positive control. Fluorescence was visualized by confocal laser scanning microscopy. Signal intensities were quantified by color histogram analysis. Values are presented as means ± standard deviations. Asterisks indicate significant differences from empty vector control leaves (Student’s t-test, < 0.05).

To examine the necrotizing process at the microscopic level, the extent of HR-like cell death was assessed by trypan blue staining (dark blue) of leaves infected with Pst (Figure 5c). CaAS1-OX plants exhibited increased HR-like cell death (dark blue) zones compared with wild-type plants and, consistent with these results, showed significantly higher levels of electrolyte leakage in leaf tissues during the early infection (Figure 5d). H2O2 levels were examined using diaminobenzidine (DAB) staining (dark brown). Increased H2O2 production was observed in infected CaAS1-OX transgenic leaves and the highest levels were detected in leaves infected with the avirulent Pst DC3000 avrRpm1 (Figure 5c). H2O2 production was quantified using the xylenol orange assay and these findings confirm that Pst infection promotes a significantly greater ROS burst in CaAS1-OX leaves than in wild-type leaves (Figure 5e). However, NO production was only induced in CaAS1-OX transgenic leaves 24 h after inoculation with virulent Pst DC3000 (Figure 5f).

Next, we investigated the effect of CaAS1 overexpression on the expression of some defense-related genes, which are associated with defense signaling pathways in Arabidopsis (Figure 6). Interestingly, ectopic expression of CaAS1 induced significant expression of PDF1.2, RD29a, NPR1 and RboHD in transgenic Arabidopsis leaves. Moreover, infection with virulent Pst DC3000, but not avirulent Pst DC3000 avrRpm1, activated significantly greater expression of PR1, PDF1.2, RD29a, NPR1, RboHD and GST in CaAS1-OX plants than in wild-type plants. This finding supports the hypothesis that the induction of these defense-related genes is critical for basal defense against Pst DC3000 infection. Notably, induction of RbBoHD and GST, which are known to be involved in the ROS burst (Torres et al., 2002), appears to contribute to the enhanced disease resistance of CaAS1-OX Arabidopsis.

Figure 6.

 Expression of Arabidopsis defense-related genes in wild-type (WT) and CaAS1-OX transgenic plants (#11) inoculated with Pseudomonas syringae pv. tomato (Pst) DC3000 or DC3000 (avrRpm1) (106 cfu ml−1).
H, healthy. Actin was used as an internal control for normalization of data. Values are presented as means ± standard deviations. Asterisks indicate significant differences from empty vector control plants (Student’s t-test, < 0.05).

To determine the role played by CaAS1 during biotrophic oomycete infection, CaAS1-OX transgenic plants were analyzed with respect to their response to H. arabidopsidis. During infection with H. arabidopsidis isolate Noco2, cotyledons from wild-type plants developed high levels of mycelial growth, sporulation and sporangiophores, as observed by trypan blue staining (Figure 7a). In contrast, H. arabidopsidis infection of the CaAS1-OX transgenic plants caused only mild disease symptoms, showing lower levels of mycelial growth, sporulation and sporangiophores. CaAS1-OX transgenic plants exhibited less sporangiophore formation and sporulation on cotyledons than wild-type plants (Figure 7b,c). These data indicate that the overexpression of CaAS1 confers reduced susceptibility against H. arabidopsidis isolate Noco2. However, the putative Arabidopsis ortholog of CaAS1 may not contribute to the basal defense and R gene-mediated resistance to Pst and H. arabidopsidis infection (Figure S6 and Appendix S1).

Figure 7.

 Quantification of infection phenotypes on cotyledons of wild-type (WT) and CaAS1-OX Arabidopsis transgenic plants (#1) inoculated with Hyaloperonospora arabidopsidis isolate Noco2.
(a) Disease symptoms and cellular responses in infected cotyledons. Photographs were taken 6 days after inoculation with 5 × 104 conidiospores ml−1 (left panel). Diseased cotyledons were stained with trypan blue at 1 and 7 days after inoculation (right panel). (b) Spore formation on infected cotyledons. (c) Quantification of sporangiophores produced on 50 cotyledons. Production of sporangiophores was examined 6 days after inoculation. Different letters indicate significant differences from three independent experiments, as analyzed by the LSD test (< 0.05).

ASN levels increase in CaAS1-OX Arabidopsis plants infected with P. syringae pv. tomato

The effect of ectopic CaAS1 expression in Arabidopsis was examined with respect to the biosynthesis of nitrogen-rich amino acids, such as ASP, ASN, GLU and GLN. The leaves of CaAS1-OX transgenic plants exhibited higher levels of ASN and GLN than wild-type plants (Figure 8a). However, no significant differences in the levels of ASP or GLU were detected between the wild-type and CaAS1-OX transgenic plants. These findings indicate that CaAS1 overexpression confers enhanced biosynthesis of ASN and GLN in Arabidopsis plants. The levels of these amino acids were also determined in the leaves of wild-type and CaAS1-OX transgenic plants during infection with P. syringae pv. tomato (Pst) (Figure 8b). Virulent Pst DC3000 infection did not consistently induce the accumulation of ASN and GLN in wild-type or transgenic Arabidopsis leaves. In contrast, avirulent Pst DC3000 (avrRpm1) infection induced significantly greater synthesis of ASP and ASN in CaAS1-OX leaves compared with wild-type leaves. However, no significant changes were observed in the levels of other amino acids during Pst infection of wild-type and CaAS1-OX transgenic plants (Figure S7a,b). These data indicate that the overexpression of CaAS1 in Arabidopsis contributes to the increase in ASP and ASN biosynthesis.

Figure 8.

 Effect of CaAS1 overexpression on amino acid levels in Arabidopsis.
(a) Amino acid contents in leaves of wild-type (WT) and CaAS1-OX transgenic plants (lines #1, #11 and #12). Values are presented as means ± standard deviations. Different letters indicate significant differences from three independent experiments, as analyzed by the LSD test (< 0.05). (b) Changes in amino acid contents in wild-type and transgenic plants (line #11) during infection with Pst DC3000 or DC3000 (avrRpm1) (106 cfu ml−1). Values are presented as means ± standard deviations. Asterisks indicate significant differences from empty vector control plants (Student’s t-test, < 0.05). ASP, aspartate; ASN, asparagine; GLU, glutamate; GLN, glutamine.


Microbial infection triggers a wide range of metabolic changes in plants. The size of the lesions that develop on pathogen-infected plants differs with respect to plant nutrition (Long et al., 2000). Successful invasion of plants depends on the ability of pathogens to utilize available nutrient sources (Solomon et al., 2003). In particular, nitrogen metabolism plays a pivotal role in the establishment of plant disease and the form of nitrogen available can also affect the disease severity.

In higher plants, asparagines synthetase catalyzes the formation of the nitrogen-rich ASN from transamination between ASP and GLN or directly from ASP and ammonium condensation (Gaufichon et al., 2010; Figure 9). During seed development, asparagine synthetase is involved in the conversion of ASP to ASN, which is used in nitrogen assimilation, transport and storage (Horst et al., 2010). ASN is also an important carrier for long-range nitrogen transport and storage in higher plants. Asparagine synthetase genes are regulated negatively by light and sugars, but induced in dark-adapted plants (Tsai and Coruzzi, 1990; Herrera-Rodriguez et al., 2004). Expression of the three Soybean Asparagine Synthetase (SAS) genes (SAS1, SAS2 and SAS3) was observed in the roots of non-nodulated soybean plants cultivated on nitrate (Antunes et al., 2008). However, it remains unclear whether or not asparagine synthetase plays a direct role during pathogen infection.

Figure 9.

 Working model of CaAS1-mediated disease resistance in the Xanthomonas campestris pv. vesicatoria (Xcv)–pepper interaction.
AS, asparagine synthetase; ASN, asparagine; ASP, aspartate; GLN, glutamine; GLU, glutamate; GS, glutamine synthetase; GOGAT, glutamine:2-oxoglutarate amidotransferase or glutamine synthetase.

In this study, we show that Capsicum annuum asparagine synthetase1 (CaAS1) is induced rapidly and strongly in pepper leaves by X. campestris pv. vesicatoria (Xcv) infection. In particular, the avirulent Xcv strain Bv5–4a caused earlier and stronger induction of CaAS1 expression in pepper leaves than the virulent Xcv strain Ds1. SA and MeJA treatment also induced high levels of CaAS1 expression in pepper leaves. These findings suggest that asparagine synthetase may be involved in the defense responses of pepper plants and that, during Xcv infection, CaAS1 expression may be regulated via SA and MeJA signaling pathways.

The tobacco rattle virus (TRV)-induced gene silencing system was used to investigate CaAS1 loss-of-function in pepper plants, as it is a fast and highly effective tool for the downregulation of gene expression (Liu et al., 2002; Sarowar et al., 2007). CaAS1-silenced plants were significantly more susceptible to Xcv infection than empty vector control plants. ROS (including H2O2 production) are important signaling molecules in the complex network of signal transduction pathways (Asai et al., 2010). During Xcv infection, the increased production of H2O2 was compromised in CaAS1-silenced plants, which suggests that CaAS1 expression may be involved in triggering the ROS burst during the early response to Xcv.

NO is believed to play a pivotal role in plant defense signaling and senescence signaling cascades (Stöhr, 2007; Ma et al., 2010). Together with ROS, NO is rapidly generated during many plant-pathogen interactions (Besson-Bard et al., 2008) and this study has shown that CaAS1 silencing compromises the pathogen-induced NO burst and reduces ROS accumulation and electrolyte leakage. This finding raises the possibility that CaAS1 may be involved in the NO generation process, as well as in alterations to available nitrogen-rich amino acids, such as ASP, ASN, GLU and GLN.

Some asparagine synthetase gene expression is affected by sugar treatments (Bläsing et al., 2005) and expression may be enhanced when plants are subject to sugar starvation (Kawachi et al., 2002). Arabidopsis asparagine synthetase 1 (ASN1) and glutamate dehydrogenase 1 (GDH1) are known to be low sugar-induced genes. When sugar levels are high, ASN1 expression is repressed, leading to reduced levels of ASN (Lam et al., 1994). However, during pathogen infection, endogenous sugar levels are reduced in tomato (Berger et al., 2004). In common beans (Phaseolus vulgaris), the presence of metabolizable sugars triggers PvNAS2 induction and concomitant ASN production (Silvente et al., 2008). These findings suggest that carbohydrate metabolism and the ratio of organic nitrogen/carbon during pathogen infection may control the expression of asparagine synthetase genes in plants. However, whether carbohydrate metabolism affects CaAS1 expression in pepper remains to be determined.

ASN and GLN are major forms of nitrogen in the phloem sap of plants such as rice (Tanaka et al., 2009). ASN serves as an important nitrogen storage and transport compound, and it is synthesized at night and during low-sugar conditions (Lam et al., 1994, 2003). In Arabidopsis, ASN1 expression levels and ASN availability are regulated tightly by environmental factors and metabolites (Lam et al., 2003). In plants, asparagine metabolism is also likely to be affected by pathogen infection. Xcv infection of pepper leaves affected levels of the nitrogen-rich amino acids ASN and GLN, as well as ASP and GLU. These findings suggest that when leaves are challenged by Xcv, the infection sites may represent strong local metabolic sinks that drain nutrients from uninfected healthy tissues. Similar to the findings in this study, changes in the content of nitrogen-rich amino acids were observed during the infection of the potato (Solanum tuberosum) with Phytophthora infestans (Grenville-Briggs et al., 2005). In tomato leaves, induction of nitrogen-rich amino acids also occurred during P. syringae infection (Olea et al., 2004). VIGS analysis of CaAS1 silencing during Xcv infection revealed that CaAS1 is crucial for the biosynthesis of ASN from ASP. These data support the notion that CaAS1 silencing may interrupt this biosynthesis process and increase susceptibility to Xcv infection. Thus, it is likely that the silencing of CaAS1 expression leads to compromised resistance to Xcv infection and this finding may be due to the reduced conversion of ASP to ASN. CaAS1-dependent changes to ASN levels correlate with susceptible or defense responses to infectious microbial pathogens in CaAS1-silenced pepper and/or CaAS1-overexpressing Arabidopsis plants, respectively. These changes in nitrogen-rich amino acids also suggest that CaAS1 expression may be triggered in response to infection.

CaAS1-OX transgenic Arabidopsis plants exhibited enhanced resistance to Pst DC3000, Pst DC3000 (avrRpm1) and H. arabidopsidis. Infection of CaAS1-OX plants by Pst resulted in a ROS burst, and H2O2 accumulation was observed in infected leaves. This ROS burst was accompanied by the development of necrotic disease symptoms, which are typical of the resistance symptoms induced by infection with Pst. The enhanced resistance that CaAS1-OX plants exhibited against these bacterial and oomycete diseases may be conferred by ectopic CaAS1 expression in Arabidopsis. Infection with C. lindemuthianum causes increased ASP levels in Phaseolus leaves (Tavernier et al., 2007) and asparagine synthetase also accumulates during the late growth of P. syringae in tomato (Olea et al., 2004). Ectopic expression of CaAS1 led to the enhanced biosynthesis of ASN and GLN in healthy leaves of CaAS1-OX Arabidopsis plants. This finding raises the possibility that CaAS1 functions as a positive regulator for the formation of the nitrogen-rich amino acids ASN and GLN. The higher levels of ASN observed in transgenic Arabidopsis plants may also contribute to enhanced resistance to Pst infection, especially by avirulent Pst avrRpm1 DC 3000.

The results presented here provide compelling evidence that CaAS1 is required for ASN synthesis associated with plant disease resistance. Virus-induced gene silencing and ectopic expression experiments revealed that the asparagine synthase encoded by CaAS1 functions as a positive regulator for enhancing basal defense against microbial pathogens in pepper and Arabidopsis. The disease resistance triggered by CaAS1 is accompanied by early defense responses, such as defense-marker gene expression, increased electrolyte leakage, as well as ROS and NO bursts. Taken together, we propose a working model that strong induction of pepper asparagine synthetase 1, CaAS1, by Xcv infection promotes not only ASN biosynthesis, but also SA, ROS, NO production and defense-related gene expression leading to disease resistance in pepper plants (Figure 9).

Experimental Procedures

Plant materials and growth conditions

Pepper plants (Capsicum annuum L., cv. Nockwang) were grown in pots containing a soil mix (peat moss: vermiculite: perlite; 3:3:1; v/v/v) at 26 ± 2°C under a 16 h light/8 h dark cycle. Arabidopsis (Arabidopsis thaliana) ecotype Col-0 (wild-type), the CaAS1-overexpression (OX) line, and the homozygous mutants asn1-1 (Salk_144656) and asn1-2 (Salk_043167) were grown in a growth chamber at 24°C with a 14 h light/10 h dark cycle. Prior to sowing on soil (peat moss: vermiculite: perlite; 1:1:0.5; v/v/v), seeds were vernalized at 4°C under low light conditions for 2 days.

Pathogen inoculation procedures

The virulent Ds1 and avirulent Bv5-4a strains of Xanthomonas campestris pv. vesicatoria (Xcv) were grown overnight in yeast-nutrient (YN) broth (5 g yeast extract, 8 g nutrient broth in 1 L H2O) at 28°C. Pepper plant leaves at the six-leaf stage were inoculated by infiltration with a bacterial suspension (5 × 108 cfu ml−1) or mock-infiltrated with 10 mm MgCl2. Inoculated plants were incubated for 18 h in a moist chamber at 28°C and then transferred to a growth room at 26 ± 2°C.

Arabidopsis leaves were infiltrated with P. syringae pv. tomato (Pst) DC3000 or DC3000 expressing avrRpm1 (DC3000 avrRpm1). Avirulent and virulent strains of Pst were grown in YN broth containing kanamycin (50 ml−1) and rifampicin (50 ml−1) at 28°C. To examine bacterial growth, infected leaves were harvested at different time points after inoculation.

Hyaloperonospora arabidopsidis isolate Noco2 was maintained in Arabidopsis Col-0 seedlings. To produce a large quantity of inocula, 7- to 10-day-old seedlings were inoculated with H. arabidopsidis and the spores produced on cotyledons were collected in water. The seedlings were spray-inoculated with a suspension of asexual inoculum (5 × 104 conidiosporangia ml−1). The inoculated seedlings were covered with a transparent dome to maintain high humidity (80–100%) and were incubated at 17°C for 7 days. Asexual sporulation of H. arabidopsidis was assessed visually by counting the number of sporangiophores on both sides of the cotyledons 5 days after inoculation. A visual disease rating was performed using five classes, which were based on the number of sporangiophores, i.e., 0–5, 6–10, 11–15, 16–20, and >50 sporangiophores per cotyledon. Seven days after inoculation, infected cotyledons from each line were excised and shaken vigorously for a spore-count assay. Spores were harvested from 20 cotyledons per replicate and counted using a hemocytometer.

Treatment with abiotic elicitors and environmental stresses

At the six-leaf stage, pepper plant leaves were treated with SA (5 mm) and MeJA (100 μm). For wounding, leaves were pricked with a needle. Treated pepper plants were examined at various time intervals after treatment, and samples were frozen in liquid nitrogen and stored at −70°C for RNA isolation.

Isolation of CaAS1 from a pepper cDNA library

A macro cDNA array method (Jung and Hwang, 2000) was used to isolate cDNA for a pathogen-induced pepper asparagine synthase gene (CaAS1). Total RNA was extracted from pepper leaves 18 h after inoculation with avirulent Xcv strain Bv5-4a. The pathogen-induced cDNA library was prepared using a λ ZAPII-cDNA library synthesis kit (Jung and Hwang, 2000). DNA sequencing was performed with an ABI 310 DNA sequencer (PE Biosystems, and PRISM BigDYE Terminator sequencing ready reaction kits.

RNA gel blot and RT-PCR analyses

Total RNA was obtained from different pepper tissues, as well as Arabidopsis leaves. RNA was extracted using TRIzol (Invitrogen,, according to the manufacturer’s instructions. For RNA gel blot analysis, total RNA (10 μg) was separated electrophoretically on a 1.2% formaldehyde-agarose gel and then transferred onto a nylon membrane (Hybond+, Amersham Pharmacia Biosciences, Membranes were hybridized with 32P-labeled probes against CaAS1 and CABPR1 (Capsicum annuum basic pathogenesis-related protein; Kim and Hwang, 2000) and then exposed to X-ray film. Transcript abundance and silencing analyses were performed by RT-PCR. First strand cDNAs were synthesized using 1 μg total RNA, oligo (dT) primers and avian myeloblastosis virus reverse transcriptase (BIO BASIC, PCR was carried out using Ex Taq DNA polymerase (TaKaRa, and the following gene-specific primers: CaAS1 forward 5′-AAACAAGGCAACATCAGCATGGGGCTTAG-3′, and reverse 5′-TTAGCTCCTTATTGTGAGCTCC-3′; and CaBPR1 forward 5′-CAGGATGCAACACTCTGGTGG-3′, and reverse 5′-ATCAAAGGCCGGTTGGTC-3′.

Virus-induced gene silencing (VIGS) of CaAS1 in pepper

Agrobacterium tumefaciens GV3101 was used for VIGS in pepper plants, according to the protocols described by Liu et al. (2002). To silence CaAS1 specifically, a 686-bp fragment of CaAS1 cDNA was amplified by PCR and cloned into pTRV2 to generate pTRV2:CaAS1. Transformed Agrobacterium containing pTRV1, pTRV2 or pTRV2:CaAS1 were selected using kanamycin and rifampicin (50 μl ml−1 each). For agro-infiltration of pepper leaves, Agrobacterium tumefaciens GV3101 containing pTRV1 was mixed in a 1:1 ratio with strains containing pTRV2 or pTRV2:CaAS1. Agrobacterium transformants containing pTRV1 and pTRV2 or pTRV2:CaAS1 were coinfiltrated into cotyledons. The binary clone pTRV2-PDS, which contains phytoene desaturase sequence, was used as a positive control to check that pTRV1 was active. Inoculated plants were maintained at 17°C for 2 days in 60% relative humidity, and then placed in a growth room for 3–4 weeks at 24 ± 2°C under a 16 h light/8 h dark photoperiod. Un-silenced (TRV:00) and CaAS1-silenced plants (TRV:CaAS1) were inoculated with Xcv strains to examine gene expression (10cfu ml−1) and bacterial growth (5 × 104 cfu ml−1).

Amino acid analysis

Amino acids were extracted from pepper leaves with 3% 5-sulfosalicylic acid in 70% ethanol. Extracts were centrifuged at 13 000 g for 15 min. The pellet was re-extracted and supernatants were pooled. Detection and quantification of amino acids were performed using the PICO-Tag system (, as described by Heinrikson and Meredith (1984). Free amino acids were separated on a Nova-Pak C18 column (4 μm, 3.9 × 3300 mm, Waters) at 46°C. The sample was injected and eluted with a linear gradient of 0–100% (v/v) acetonitrile. Fluorescence was monitored using an HP 1100 Series (Agilent, at 254 nm. l-norleucine (Sigma, was used as an internal standard, since it is an amino acid that is not commonly found in proteins.

Staining with trypan blue and DAB

To monitor plant cell death and fungal growth, infected or non-infected leaves were stained with 1 mg ml−1 3, 3′-diaminobenzidine (DAB; Sigma) and lactophenol-trypan blue (10 ml lactic acid, 10 ml glycerol, 10 g phenol and 10 mg trypan blue, dissolved in 10 ml distilled water). After staining overnight, the leaves were cleared by boiling in ethanol for 10 min and de-stained overnight in ethanol. For trypan blue staining, infected leaves were boiled for 5 min in the staining solution and de-stained overnight in 2.5 g ml−1 chloral hydrate. De-stained leaf tissues were then mounted in 70% glycerol.

Generation of transgenic CaAS1-OX Arabidopsis plants

The open reading frame (ORF) of CaAS1 was amplified by PCR and then subcloned into pCR2.1-TOPO (Invitrogen). The CaAS1 fragment was digested with XbaI and BamHI and ligated into the binary expression vector pBIN35S. The primers used to generate the XbaI and BamHI sites were 5′-TCTAGAATGTGTGGAATATTGGCTTTGTTG-3′ (forward) and 5′-GGATCCTTAGCTCCTTATTGTGAGCTCC-3′ (reverse). Constructs were transformed into Agrobacterium tumefaciens strain GV3101 via electroporation. The floral dip method (Clough and Bent, 1998) was used to generate transgenic Arabidopsis plants (ecotype Col-0) with A. tumefaciens strain GV3101 expressing CaAS1. To select transgenic CaAS1-overexpressing (OX) plants, transgenic plant seeds were germinated on Murashige and Skoog (MS) media (Duchefa, containing 50 μg L−1 kanamycin. Kanamycin-resistant transgenic plants were identified by PCR amplification of inserted cDNAs.

Measurement of ion leakage

To determine ion leakage, leaves from empty vector control plants and CaAS1-silenced plants were infiltrated with the Xcv (107 cfu ml−1). Leaf discs (1.2 cm in diameter) were washed in double-distilled water for 30 min and then incubated in 20 ml double-distilled water. The electrolyte leakage from leaf discs was measured with a conductivity meter (Model sensION7; Hach,

Measurement of SA

SA and its glycoside were extracted and quantified according to the method described by Verberne et al. (2002). Leaf tissue samples (0.5 g) were frozen in liquid nitrogen, ground to a fine power, and extracted sequentially with 90 and 100% methanol. As an internal standard, 3-hydroxybenzoic acid (Sigma) was added at a mass ratio of 50 mg g−1 leaf fresh weight. Following separation on a C18 reverse-phase HPLC (high pressure liquid chromatography) column (Waters), SA levels were determined by fluorescence (excitation 305 nm, emission 405 nm).

Measurement of NO

Nitric oxide (NO) accumulation was monitored using the NO-sensitive dye 4,5-diaminofluorescein 2-diacetate (DAF-2DA) (Sigma), as previously described (Kojima et al., 1998). For NO detection, leaves were infiltrated with 200 mm sodium phosphate buffer (pH 7.4) supplemented with 12.5 μm DAF-2DA and then incubated for 1 h in the dark at room temperature. As a positive control, sodium nitroprusside (SNP; 500 μm) was infiltrated into leaves with DAF-2 DA. The product of the reaction between DAF-2DA and NO was observed using a confocal laser scanning microscope, with excitation at 470 nm. Emission images at 525 nm were obtained using a constant acquisition time and fluorescence intensity was determined via color histogram analysis.


This work was supported by the following: a grant (CG1133) from the Crop Functional Genomics Center of the 21st Century, Frontier Research Program, funded by the Ministry of Education, Science and Technology, Korea; and a grant from the Next Generation BioGreen21 Program (Plant Molecular Breeding Center), Rural Development Administration, Korea.