• Open Access

Durable field resistance to wheat yellow mosaic virus in transgenic wheat containing the antisense virus polymerase gene

Authors

  • Ming Chen,

    1. Institute of Crop Sciences, Chinese Academy of Agricultural Sciences/National Key Facility for Crop Gene Resources and Genetic Improvement, Key Laboratory of Biology and Genetic Improvement of Triticeae Crops, Ministry of Agriculture, Beijing, China
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Liying Sun,

    1. State Key Laboratory Breeding Base for Zhejiang Sustainable Pest and Disease Control, MoA Key Laboratory for Plant Protection and Biotechnology, Zhejiang Provincial Key Laboratory of Plant Virology, Institute of Virology and Biotechnology, Zhejiang Academy of Agricultural Sciences, Hangzhou, China
    Search for more papers by this author
    • These authors contributed equally to this work.
  • Hongya Wu,

    1. Institute of Agricultural Sciences of Lixiahe Districts, Jiangsu, China
    Search for more papers by this author
  • Jiong Chen,

    1. State Key Laboratory Breeding Base for Zhejiang Sustainable Pest and Disease Control, MoA Key Laboratory for Plant Protection and Biotechnology, Zhejiang Provincial Key Laboratory of Plant Virology, Institute of Virology and Biotechnology, Zhejiang Academy of Agricultural Sciences, Hangzhou, China
    Search for more papers by this author
  • Youzhi Ma,

    1. Institute of Crop Sciences, Chinese Academy of Agricultural Sciences/National Key Facility for Crop Gene Resources and Genetic Improvement, Key Laboratory of Biology and Genetic Improvement of Triticeae Crops, Ministry of Agriculture, Beijing, China
    Search for more papers by this author
  • Xiaoxiang Zhang,

    1. Institute of Agricultural Sciences of Lixiahe Districts, Jiangsu, China
    Search for more papers by this author
  • Lipu Du,

    1. Institute of Crop Sciences, Chinese Academy of Agricultural Sciences/National Key Facility for Crop Gene Resources and Genetic Improvement, Key Laboratory of Biology and Genetic Improvement of Triticeae Crops, Ministry of Agriculture, Beijing, China
    Search for more papers by this author
  • Shunhe Cheng,

    1. Institute of Agricultural Sciences of Lixiahe Districts, Jiangsu, China
    Search for more papers by this author
  • Boqiao Zhang,

    1. Institute of Agricultural Sciences of Lixiahe Districts, Jiangsu, China
    Search for more papers by this author
  • Xingguo Ye,

    1. Institute of Crop Sciences, Chinese Academy of Agricultural Sciences/National Key Facility for Crop Gene Resources and Genetic Improvement, Key Laboratory of Biology and Genetic Improvement of Triticeae Crops, Ministry of Agriculture, Beijing, China
    Search for more papers by this author
  • Junlan Pang,

    1. Institute of Crop Sciences, Chinese Academy of Agricultural Sciences/National Key Facility for Crop Gene Resources and Genetic Improvement, Key Laboratory of Biology and Genetic Improvement of Triticeae Crops, Ministry of Agriculture, Beijing, China
    Search for more papers by this author
  • Xinmei Zhang,

    1. Institute of Crop Sciences, Chinese Academy of Agricultural Sciences/National Key Facility for Crop Gene Resources and Genetic Improvement, Key Laboratory of Biology and Genetic Improvement of Triticeae Crops, Ministry of Agriculture, Beijing, China
    Search for more papers by this author
  • Liancheng Li,

    1. Institute of Crop Sciences, Chinese Academy of Agricultural Sciences/National Key Facility for Crop Gene Resources and Genetic Improvement, Key Laboratory of Biology and Genetic Improvement of Triticeae Crops, Ministry of Agriculture, Beijing, China
    Search for more papers by this author
  • Ida B. Andika,

    1. State Key Laboratory Breeding Base for Zhejiang Sustainable Pest and Disease Control, MoA Key Laboratory for Plant Protection and Biotechnology, Zhejiang Provincial Key Laboratory of Plant Virology, Institute of Virology and Biotechnology, Zhejiang Academy of Agricultural Sciences, Hangzhou, China
    Search for more papers by this author
  • Jianping Chen,

    Corresponding author
    1. State Key Laboratory Breeding Base for Zhejiang Sustainable Pest and Disease Control, MoA Key Laboratory for Plant Protection and Biotechnology, Zhejiang Provincial Key Laboratory of Plant Virology, Institute of Virology and Biotechnology, Zhejiang Academy of Agricultural Sciences, Hangzhou, China
    Search for more papers by this author
  • Huijun Xu

    Corresponding author
    1. Institute of Crop Sciences, Chinese Academy of Agricultural Sciences/National Key Facility for Crop Gene Resources and Genetic Improvement, Key Laboratory of Biology and Genetic Improvement of Triticeae Crops, Ministry of Agriculture, Beijing, China
    Search for more papers by this author

Summary

Wheat yellow mosaic virus (WYMV) has spread rapidly and causes serious yield losses in the major wheat-growing areas in China. Because it is vectored by the fungus-like organism Polymyxa graminis that survives for long periods in soil, it is difficult to eliminate by conventional crop management or fungicides. There is also only limited resistance in commercial cultivars. In this research, fourteen independent transgenic events were obtained by co-transformation with the antisense NIb8 gene (the NIb replicase of WYMV) and a selectable gene bar. Four original transgenic lines (N12, N13, N14 and N15) and an offspring line (N12-1) showed high and durable resistance to WYMV in the field. Four resistant lines were shown to have segregated and only contain NIb8 (without bar) by PCR and herbicide resistance testing in the later generations. Line N12-1 showed broad-spectrum resistance to WYMV isolates from different sites in China. After growing in the infested soil, WYMV could not be detected by tissue printing and Western blot assays of transgenic wheat. The grain yield of transgenic wheat was about 10% greater than the wild-type susceptible control. Northern blot and small RNA deep sequencing analyses showed that there was no accumulation of small interfering RNAs targeting the NIb8 gene in transgenic wheat plants, suggesting that transgene RNA silencing, a common mechanism of virus-derived disease resistance, is not involved in the process of WYMV resistance. This durable and broad-spectrum resistance to WYMV in transgenic wheat will be useful for alleviating the damage caused by WYMV.

Introduction

Wheat yellow mosaic disease is characterized by mosaic or yellow-striped leaves and plant stunting and is one of the major threats to wheat production in Europe, North America and East Asia (Kühne, 2009). In China, this disease first occurred in the middle and lower reaches of the Changjiang River in 1970s and since the 1990s has spread to several provinces (Chen et al., 1989; Han et al., 2000). It results in reduction in grain yield by 20–70% in some individual fields (Liu et al., 2005b), and in China, it is predominantly caused by wheat yellow mosaic virus (WYMV) (Han et al., 2000). The virus is transmitted by an obligate soil-inhabiting fungus-like organism Polymyxa graminis (order Plasmodiophorales) and is protected inside the thick-walled resting spores and zoospores of the vector (Kühne, 2009). Because dormant spores of P. graminis can survive for long periods in the soil, the inoculum in contaminated fields is difficult to eliminate by conventional crop management or fungicides. The best effective method to control the disease is to grow resistant wheat varieties, but there are few known resistance genes in current wheat cultivars. It has been reported that WYMV resistance in wheat is controlled by one to three genes (Qin et al., 1986) with one major gene being associated with homologous group 2 chromosome (Liu et al., 2005a; Nishio et al., 2010; Zhu et al., 2012). Furthermore, genetic analyses showed that the inheritance of this resistance is complex and influenced by many factors (Zhou et al., 2000). It therefore seems that genetic engineering may be more promising than conventional breeding for producing plants resistant to WYMV.

WYMV belongs to the genus Bymovirus (family Potyviridae). Its genome is encapsidated in filamentous particles and consists of two positive sense single-stranded RNA segments (RNA1, 7.5 kb and RNA2, 3.6 kb), each of which encodes a polyprotein that undergoes post-translational cleavage (Namba et al., 1998). The polyprotein encoded by RNA1 (269 kDa) produces eight proteins including the coat protein (CP) and a nuclear inclusion ‘b’ protein (NIb) that functions as an RNA dependent RNA polymerase (RdRp) and is important for virus replication (Figure 1a). RNA2 encodes a polyprotein (101 kDa) that give rise to two proteins of 28 kDa (P1) and 73 kDa (P2) (Namba et al., 1998).

Figure 1.

Structure of the wheat yellow mosaic virus genome (RNA1) and transformation plasmids. (a) The wheat yellow mosaic virus genome (RNA1). P3, P3 protein; 7k, 7 kDa protein; CI, cylindrical inclusion protein; 14k, 14 kDa protein; NIa-VPg, NIa-VPg protein; NIa-Pro, NIa-Pro protease; NIb, NIb polymerase; CP, coat protein. (b) The structure of vector pUbi-NIb. NIb8, Reverse complement sequence of NIb polymerase gene (5284-6495 bp region); 35S polyA, Terminator. (c) Diagram of vector pAHC20 containing bar gene for plant selection in the co-transformation experiments. bar, herbicide-resistant gene; Nos 3′, Terminator.

Over the past two decades, several strategies have been developed based on the concept of pathogen-derived resistance (Sanford and Johnston, 1985). The integration of various viral sequences into their host plant genomes has proved to be effective in preventing or reducing the infection of many viruses (Gottula and Fuchs, 2009). The virus resistance in transgenic plants can be conferred by the expression of viral protein (protein-mediated) or viral RNA sequences alone (RNA-mediated) (Prins et al., 2008). RNA-mediated resistance can be achieved by transformation of plant with cDNA of the viral genome that is nontranslatable, in antisense orientation or arranged as an inverted repeat to produce hairpin RNA (Tenllado et al., 2004). RNA silencing, also called RNA interference (RNAi), has been demonstrated as a major mechanism of RNA-mediated resistance in transgenic plants carrying viral sequence (Simón-Mateo and García, 2011). Transgenic constructs capable of forming dsRNA (double-strand RNAs) transcripts were proved to be more effective in yielding high-level virus-resistant plants than the constructs producing either sense or antisense RNA alone (Pinto et al., 1999; Waterhouse et al., 1998). RNA silencing is an evolutionarily conserved process in a wide variety of eukaryotic organisms that is initiated when dsRNAs are processed by a ribonuclease III-like enzyme called Dicer into 21- to 24-nucleotide (nt) small interfering RNAs (siRNAs) (Baulcombe, 2004; Ding, 2010; Ding et al., 2004). The siRNAs are then incorporated into an RNA-induced silencing complex to guide the degradation or translational repression of homologous RNA targets in a sequence-specific manner (Ding and Voinnet, 2007).

Pathogen-derived resistance to viruses has been successfully engineered into a number of crop plants including wheat. Potyviridae is one of the largest plant virus families, and many of its members cause economically important crop diseases (Adams et al., 2011). The infection of many viruses belonging to the family Potyviridae has been successfully inhibited by viral-derived transgenes, and most of this engineered resistance is achieved through the RNA-mediated mechanism (Collinge et al., 2010). For examples, transgenic wheat plants carrying the complementary DNA of coat protein of wheat streak mosaic virus (WSMV, genus Tritimovirus) or carrying an RNAi construct that was designed from the nuclear inclusion protein ‘a’ (Nla) gene of WSMV to produce hairpin RNA were highly resistant to WSMV infection (Fahim et al., 2010; Sivamani et al., 2002). Likewise, RNA-mediated mechanism was implicated in the resistance of transgenic papaya against papaya ringspot virus (genus Potyvirus) (Gonsalves, 1998). There remain some problems hindering the use of pathogen-derived resistance in crop molecular breeding programmes. For example, most reported transgenic resistance has only been tested in the greenhouse, and there is little information from the field. The stability of transgenic resistance is not guaranteed, and only a few long-term tests have so far been reported.

In this article, we chose the NIb8 region (1212 bp) as target sequence because it is one of the most conserved portions in WYMV genome and is therefore a good candidate for broad-spectrum resistance. The sequences of eight WYMV isolates sequenced in this part of their genome (one from Japan and seven from different parts of China) are >97% identity to one another (Figure S1). We report the production of wheat plants containing most of the antisense coding region of WYMV NIb and the results of field testing for disease resistance. Field testing from 2000 to 2010 indicated that four original transgenic lines and a derived offspring line were highly resistant to WYMV infection and that the resistance was stably inherited, suggesting that this durable field resistance meets the requirements for practical application and could be therefore be used in breeding programmes for wheat disease resistance. Moreover, siRNAs analysis suggested that transgene RNA silencing, which is involved in virus resistance to other members of the family Potyviridae, is not part of the process of WYMV, suggesting that a novel transgenic virus resistance mechanism exists in these transgenic wheat plants.

Results

Selection and molecular identification of WYMV-resistant transgenic lines

The vector pubi-NIb (Figure 1b) and assistant plasmid pAHC20 containing the selectable marker gene bar (Figure 1c) were co-transformed into wheat. After transformation, callus differentiation and regeneration of plants on the selective medium, 14 positive plants were identified by PCR from the T0 generation plants. The T1 generation plants derived from these 14 PCR-positive lines were tracked using PCR (Figure 2a and data not shown), and only PCR-positive plants were retained. The NIb8 segment amplified from the transgenic wheat was sequenced, confirming that the NIb8 inserted in the wheat genome was identical to that in the vector. RT-PCR showed that the NIb8 gene was transcribed in some T2 generation transgenic lines including N12, N13, N14 and N15 (Figure 2b). At same time, WYMV resistance testing of the T2 transgenic wheat lines was performed in the disease nursery in 2000. Among the 14 T2 transgenic lines, four lines showed high resistance to WYMV (Figure 3a): N12, N13, N14 and N15. The numbers of virus-infected plants were, respectively, only 0, 0, 0 and 1, of 19, 48, 42 and 46, whereas about half of the susceptible control Yangmai 158 plants were infected (Table 1). To identify homozygousity of transgenic wheat plants, 100 independent plants were randomly sampled from T3 population of the four lines N12, N13, N14 and N15, to detect the NIb8 gene by PCR, and 99% plants were tested to be positive, which indicated that these lines were homozygous in Nib8 insertion locus after T3 generation (data not shown).

Table 1. Resistant testing of the T2 transgenic plants to the wheat yellow mosaic virus in field (2000)
LineTotal plants testedSusceptible plantsDI (%)
  1. A disease index (DI) was then calculated: DI (%) = ∑ (DS × Ni) × 100 / (3 × N), where DS is the disease scale (above), Ni the number of plants with this DS, and N the total number of plants observed.

N121900.00
N134800.00
N144200.00
N154612.17
Yangmai 158462247.83
Figure 2.

Molecular analysis of the transgenic wheat plants. (a) PCR testing for NIb8 gene in T1 generation plants. M, marker; 1, positive (plasmid) control; 2, negative control (Yangmai 158); 3, N12 line; 4, N13 line; 5, N14 line; 6, N15 line; 7, N17 line. (b) RT-PCR for NIb8 in T2 generation plants. M, marker; 1, positive (plasmid) control; 2 and 3, negative control (Yangmai 158); 4, N12 line; 5, N13 line; 6, N14 line; 7, N17 line; 8, N15 line. (c) Southern blotting for NIb8 gene in T6 generation lines using two restriction enzymes, EcoRI and SspI, to cut the genomic DNA. M, Marker; 2 and 8, N12 line; 3 and 9, N12-1 line; 1 and 10, N13 line; 4 and 11, N14 line; 5 and 12, N15 line; 6 and 13, N35 line; 7 and 14, Yangmai 158 (negative control); 15, positive control (plasmid).

Figure 3.

Identification of transgenic wheat for disease resistance in field. Transgenic wheat in disease nursery at Yangzhou, Jiangsu Province, in March 2001 (a), Yangzhou in March 2002 (b), Yangzhou in March 2005 (c), and Yizheng, Jiangsu Province in March 2005 (d). Y158, wild-type Yangmai 158; N12, N12-1, N13, N14 and N15, transgenic wheat lines.

To obtain marker-free transgenic wheat plants from the co-transformation experiments, the T3 plants derived from the positive N12 line were tested by PCR for the NIb8 and bar genes. Of the 268 plants tested, 28 had only NIb8 (without bar) and a marker-free line (N12-1) was developed from these plants (Figure 4). Later tests showed that the bar gene was also lost and the NIb8 retained in the T7 generations of N13, N14 and N15, whereas both genes were retained in N12 (Figure 4).

Figure 4.

Selecting marker-free lines using PCR and herbicide-resistant tests. (a) Herbicide-resistant testing of T7 transgenic lines using 150 ppm Liberty (herbicide); CK+, positive control (transgenic wheat lines) with herbicide-resistant gene; CK−, negative control (Yangmai 158). (b) Herbicide-resistant testing of T7 transgenic lines using 100 ppm Basta (herbicide); CK−, negative control (Yangmai 158). (c) PCR detection of NIb8 gene of T7 generation transgenic lines; M, DNA marker; 1, negative control (H2O), 2, negative control (Yangmai 158); 3, positive control (plasmid pUbi-NIb); 4, N12-1; 5, N13; 6, N14; 7, N15; 8, N12. (d) PCR detection of bar gene; M, DNA marker; 1, positive control (plasmid pUbi-NIb); 2, negative control (H2O); 3, N12; 4, N12-1; 5, N13; 6, N14; 7, N15; 8, negative control (Yangmai 158).

Southern blots of plant genomic DNA digested with EcoRI and SspI demonstrated single copy integration of the NIb8 gene in the five transgenic lines N12, N12-1, N13, N14 and N15 in the T6 generation (Figure 2c). To analyse the number of transgene locus and examine the heritability of NIb8 in N12-1 and N14, these two lines were crossed with WYMV-susceptible wheat variety Yangmai 15 (as female parent) and disease resistance of the F2 generation segregation population was tested in the field. In total, 693 resistant plants and 199 susceptible plants were found in the cross of Yangmai 15/N12-1, and 590 resistant plants and 183 susceptible plants in Yangmai 15/N14. The segregation ratios of 3.48 : 1 and 3.22 : 1, respectively, were consistent with Mendelian inheritance (Table 2), suggesting that NIb8 was integrated as a single locus in the N12-1 and N14 transgenic lines and that the resistance can be transferred to other wheat varieties by hybridization.

Table 2. Segregation ratio of resistant and sensitive disease plants in F2 population of transgenic lines and Yangmai 15
Cross combinationTotal plantsResistant plantsSensitive plantsRatio between resistant and sensitive plantsChi-square valuea
  1. a

    The critical value at 0.05 level is 3.84, and both of chi-square values are less than the critical value, which indicated that the segregation ratio between resistant and sensitive disease plants was 3 : 1 approximately.

Yangmai 15/N12-18926931993.48:13.44
Yangmai 15/N147735901833.22:10.72

WYMV resistance of the transgenic wheat in field trials

The lines N12, N13, N14 and N15 from 2001 (T3) to 2005 (T7), and N12-1 from 2002 (T4) to 2005 (T7) were identified for disease resistance in a disease nursery in Jiangsu Province, and seed from the disease-free plants were harvested and bulked each year. WT controls were always heavily infected with Disease Index (DI, see methods) >90% in most seasons, whereas the transgenic lines had a much lower DI (P < 0.01), often at about or below 5% (Figure 3b,c and d, Table 3). PCR tests showed that the NIb8 gene was present in the plants of each generation. These results indicate that transformation using antisense-NIb8 resulted in transgenic wheat plants with durable field resistance to WYMV.

Table 3. Field investigation of transgenic plants for wheat yellow mosaic virus resistance (DI %) from 2001 to 2005
Line2001 (T3)2002 (T4)2003 (T5)2004 (T6)2005 (T7)Mean value F value
  1. The percentage of sensitive plants in total tested plants was used as DI (%) to evaluate the resistance of the transgenic lines. The capital letter of A, B, C or AB meant the significance difference at 0.01 level.

  2. a

    Indicating significant difference at 0.01 level between wild-type and transgenic lines (F0.01 = 4.10). The difference between years were not significant (F0.05 = 2.87). The field testing was completed in disease nursery in Jiangsu Province by three times repeats.

N125.00 ± 4.58B5.50 ± 7.14C3.67 ± 5.19B4.60 ± 0.85B2.32 ± 0.44B4.22 ± 1.26BIntervarietal288.41a
N138.00 ± 8.00B5.50 ± 6.56C6.09 ± 4.72B5.10 ± 0.57B3.35 ± 0.66B5.61 ± 1.68BYear1.40
N146.33 ± 5.69B7.90 ± 3.43BC4.42 ± 6.45B4.30 ± 0.71B2.49 ± 0.74B5.09 ± 2.08B  
N156.00 ± 5.20B16.00 ± 4.16B2.08 ± 4.17B4.50 ± 0.57B1.20 ± 1.06B5.96 ± 5.93B  
N12-1/2.35 ± 2.73C0.00 ± 0.00B5.20 ± 0.99B1.58 ± 0.44B2.28 ± 2.18B  
Yangmai 15875.00 ± 18.52A95.00 ± 4.32A90.75 ± 5.38A99.60 ± 0.57A94.70 ± 3.82A91.01 ± 9.48A  

To test the resistance to different virus strains, plants of N12-1 were grown on three different sites: Yantai (Shandong Province), Xiping (Henan Province) and Yangzhou (Jiangsu Province). The site at Yantai was mostly infested with WYMV, but Chinese wheat mosaic virus (CWMV), an unrelated virus (genus Furovirus, family Virgaviridae) transmitted by the same vector (Diao et al., 1999), was occasionally detected, while only WYMV was detected in Xiping and Yangzhou. There are known to have differences between these WYMV populations (Sun et al., 2013). N12-1 was highly resistant to WYMV in Yantai, Xiping and Yangzhou; a small number of transgenic plants developed disease symptom in Yantai, which is due to the CWMV infection (detected by ELISA). In contrast, the WT was severely diseased at all locations (Table 4).

Table 4. Disease-resistant evaluation of transgenic line N12-1 in three different regions of China
RegionYantaia (WYMV + CWMV) DI (%)Xipinga (WYMV) DI (%)Yangzhoua (WYMV) DI (%)
  1. a

    Yantai is in Shandong Province, Xiping in Henna Province and Yangzhou in Jiangsu Province.

  2. b

    Indicating significant difference at 0.01 level between wild-type and the transgenic lines.

Lines
N12-15.00b0.00b0.00b
Yangmai 158100.00100.0052.30

Some major agronomic traits of the transgenic lines N12, N12-1, N13, N14 and N15 were compared to those of the WT Yangmai 158 in field disease nurseries in 2004 and 2005. In both years, the transgenic lines out yielded the WT, usually by at least 10% (P < 0.05) (Tables 5 and 6). Other apparent differences in agronomic traits were not statistically significant. In a field not infested with WYMV, there was no significant difference of grain yield or other agronomic traits between WT and transgenic wheat. It therefore appears that the increased yield of the transgenic wheat on the infested sites was a result of their excellent WYMV resistance.

Table 5. Main agronomic characters and yields of T6 transgenic lines in field experiments in 2004 in Yangzhou
LinesPH (cm)SL (cm)TNNGPEYT (t/h)YI (%)
  1. The yield comparing experiments were completed in disease nursery in Yangzhou by three times repeats; PH, plant height; SL, spike length; TN, the tiller number; NGPE, number of grains per ears; YT (tons per hectare), yield-test results calculated as yield of grains in plot (6.67 m2); YI, yield increased than that of Yangmai 158. The small letter of a, b or ab showed the significance difference at 0.05 level.

N1277.3210.7514.6334.797.49a12.03
N1381.0311.2614.2133.847.11ab6.31
N1477.2810.5815.4035.097.57a13.22
N1580.0710.9017.4836.417.54a12.72
N12-178.5511.1315.2036.897.82a16.85
Yangmai 15882.058.4813.7337.996.69b 
Table 6. Main agronomic characters and yields of T7 transgenic lines in field experiments in 2005 in Yangzhou
LinesMPH (cm)SL (cm)NSSNAPANGPETGW (g)YT (t/h)YI (%)
  1. The yield comparing experiments were completed in disease nursery in Yangzhou by three times repeats; M, maturity (Month/day); PH, plant height; SL, spike length; NSS, number of seeded spikelet; NAPA, number of ears per plant; NGPE, number of grains per ears; TGW, 1000-grains weight; YT (tons per hectare), yield-test results calculated by the yield in plot (6.67 m2); YI, yield increased than Yangmai 158. The small letter of a, b or ab showed the significance difference at 0.05 level.

N125/2888.08.918.86.452.340.67.7a16.6
N135/2889.09.419.16.654.140.37.3ab10.8
N145/2890.09.518.25.950.342.17.4a12.0
N155/2892.09.618.55.848.142.87.3ab11.1
N12-15/2889.09.318.56.251.740.57.5a14.0
Yangmai 1585/3085.08.718.75.848.939.46.6b 

Detection of WYMV accumulation in nontransgenic and transgenic plants

To investigate whether the absence of yellow mosaic symptoms in transgenic wheat plants is associated with the absence of virus accumulation, tissue printing and Western blot assays using an antibody specific for the WYMV CP were carried out. Leaves, stems and roots of WT Yangmai 158 (YM158) and transgenic line N12-1 were sampled 4 months after planting on WYMV-infested soil. In tissue printing assay, intense staining, indicating high levels of virus accumulation, was observed in upper leaves, lower leaves, stems and roots of WT plants, but not in those of N12-1 plants (Figure 5a). Similar results were also obtained from Western blot assays in which strong bands corresponding to WYMV CP were detected in leaves, stems and roots of Yangmai 158 plants, but not in those of N12-1 plants (Figure 5b). Together, these results indicate that WYMV is unable to accumulate in N12-1 plants.

Figure 5.

Wheat yellow mosaic virus (WYMV) accumulation in transgenic and nontransgenic wheat plants. (a) Tissue printing assay to detect WYMV accumulation in leaves stems and roots of susceptible nontransgenic control (Yangmai 158, YM158), resistant nontransgenic (Ning 9) control and transgenic (line N12-1) wheat plants grown on WYMV-infested soil. (b) Western blot analysis to detect WYMV accumulation in wheat plants described in panel A. Plants were sampled four months after planting. A polyclonal antibody specific for WYMV CP was used for the immunodetection.

The analysis of transgene- and WYMV-derived siRNAs in wheat plants

Next we analysed the accumulation of transgene-derived siRNAs in resistant transgenic wheat plant, because transgene RNA silencing is the most common mechanism in RNA-mediated virus resistance in transgenic plants (Simón-Mateo and García, 2011). First, we carried out Northern blot analysis using low molecular weight RNAs (30 μg) extracted from leaves of transgenic line N12-1. The blot was hybridized with a DIG-labelled DNA probe specific for WYMV RNA1 (position 5284-6495 nt) prepared by PCR. No transgene-derived siRNA was detected, whereas using the similar protocol with a DIG-labelled DNA probe, we were able to detect GFP transgene-derived siRNAs in an Agrobacterium co-infiltration assay where GFP was expressed in leaves of GFP-transgenic N. benthamiana line 16c (Voinnet et al., 1998). In an attempt to increase the sensitivity of detection, we gel-purified small RNAs with size ranging from 18 to 30 nt from a larger quantity of the low molecular weight RNA fraction (300 μg) and then used these in Northern blot as described above. However, we were still unable to detect transgene-derived siRNA in N12-1 plants.

Recently, deep sequencing technology has provided a novel method to analyse siRNA (Thomas and Ansel, 2010). We therefore generated cDNA libraries from the small RNA fraction extracted from leaves of N12-1 and WYMV-infected Yangmai 158 plants and analysed these by Illumina sequencing. About 3.8 and 2.6 million reads were obtained from N12-1 and Yangmai 158 libraries, respectively. Computational analyses were then carried out to identify the WYMV-derived siRNAs in each library, allowing for two mismatches with the virus genome reference sequence. A large number of siRNAs (36 703 reads; 1.4%) in the Yangmai 158 library were found to match the WYMV genome compared with only a few (43 reads; 0.001%) from the N12-1 library. Importantly, no siRNA from the N12-1 library was found to match nucleotide position 5284-6495 of WYMV RNA1, which is the sequence corresponding to the transgene. Taken together with the results of Northern blot analysis, these observations strongly suggest that transgene RNA silencing, the common mechanism of virus-derived disease resistance, was not involved in the WYMV resistance of transgenic wheat plants. In the Yangmai 158 library, WYMV siRNAs were almost equally sense and antisense (Figure 6a), and 21-nt was the most abundant size (Figure 6b). These siRNAs were distributed throughout the positive- and negative-strands of WYMV RNA1 and 2 with some siRNA hotspots (Figure 6c,d). This result provides the first experimental evidence that WYMV infection in wild-type wheat plants induces an RNA-silencing-mediated antiviral defence.

Figure 6.

The presence of wheat yellow mosaic virus WYMV-derived siRNAs in inoculated nontransgenic and transgenic wheat plants. (a) Frequencies of WYMV siRNAs in nontransgenic (Yangmai 158, YM158) and transgenic (line N12-1) wheat plants grown on WYMV-infested soil. (b) Size distribution of WYMV siRNAs derived from the YM158 library. (c) and (d), Distribution of WYMV siRNAs along the RNA1 (c) and RNA2 (d) segments of the WYMV genome.

Discussion

Novel transgenic wheat with durable field resistance to WYMV obtained by transformation of NIb8 gene from virus genome

There have been various reports that plant virus diseases can be controlled by transforming the plants with viral sequences and interfering with viral reproduction via the RNA interference pathway (Fahim et al., 2010, 2012; Kung et al., 2012; Pinto et al., 1999; Simón-Mateo and García, 2011). Resistance has been obtained by inserting viral genes in positive orientation (Pinto et al., 1999), in reverse (complementary) orientation (Sivamani et al., 2002), as long hairpin ds-RNA sequences (Fahim et al., 2010), as short hairpin ds-RNA sequences (Shimizu et al., 2011) and by introducing multiple artificial microRNAs (Fahim et al., 2012; Kung et al., 2012). However, the transgenic resistance obtained by these techniques has mostly been tested only in the greenhouse, and where field testing has been carried out, results have not always been consistent with those in the greenhouse. The only example of transgenic wheat in which resistance was maintained in both greenhouse and field is the expression of the viral origin antifungal protein KP4 to provide strong resistance to stinking smut (Tilletia tritici) (Schlaich et al., 2006; Zhou et al., 2000). However, transgenic wheat transformed with the replication enzyme gene or coat protein genes from wheat streak mosaic virus (WSMV; genus Tritimovirus, family Potyviridae) showed strong disease resistance in the greenhouse, but lost the resistance in field (Sharp et al., 2002). Stability of transgenic disease resistance has rarely been reported, and it is therefore significant that transgenic resistance to WYMV was identified continuously for 5 years at three different sites in our experiments (Figure 3, Table 3). We believe that this is the first reported transgenic wheat with durable disease resistance in field, and it provides novel germplasm for the breeding of wheat resistant to WYMV.

The resistance mechanism in the transgenic wheat plants

Transformation of plants with a transgene designed to express antisense viral RNA has been one of the most effective methods to generate plants resistant to viral infection. In some cases, the resistance in the transgenic lines is associated with high accumulation of antisense RNA from transcription of the transgene (Bendahmane and Gronenborn, 1997; Hammond and Kamo, 1995). However, resistance has also been achieved by an RNA-silencing-based mechanism through the accumulation of transgene-derived siRNA in transgenic lines (Zhang et al., 2005). Our analyses using Northern blot and deep sequencing of small RNAs showed that transgene-derived siRNAs did not accumulate in our resistant transgenic plants (Figure 6). These results suggest that transgene RNA silencing does not occur in this transgenic line and that resistance does not operate through the degradation of viral RNA by the homology-dependent mechanism that has been observed in many virus-resistant transgenic plants (Simón-Mateo and García, 2011). Southern blot analysis indicated that the four resistant transgenic lines each contain a single transgene copy and are apparently derived from independent transformations (Figure 2c). It is therefore unlikely that the resistance occurred because the gene transformation caused the disruption of a host gene required for WYMV infection. The transgene transcript was detected by RT-PCR (Figure 2b), and it is therefore possible that the accumulation of antisense RNA in the cells inhibits WYMV infection, for example by interference in viral genome replication. Another possibility is that antisense transcripts anneal viral genome RNA to form dsRNAs, which are then digested by cellular RNase. Given that WYMV infects wheat plant at low temperature (Kühne, 2009) and that RNA silencing is less active at low temperature (Szittya et al., 2003; Zhang et al., 2012), it could be suggested that noninvolvement of RNA silencing pathway in the resistance mechanism may be one of the reasons for the stable and potent resistance of our transgenic plants.

NIb8 confers resistance in various wheat cultivars

Transgenic disease resistance could be transferred from transgenic lines N12-1 and N14 into a susceptible variety Yangmai 15 (Table 2), indicating that NIb8 can confers resistance in a different wheat cultivar. We have also been able to transfer the NIb8 transgene from line N12-1 to other WYMV-susceptible commercial wheat varieties by multiple backcrossing and created some new lines with high resistance to WYMV and a single copy of the NIb8 gene from N12-1 line, some of which also have good agronomic traits. WYM disease has spread gradually from the middle and lower reaches of the Yangtze River to the north since the 1990s, and losses are becoming more serious year by year. Conventional resistance to WYMV is rare, but our results suggest that transgenic wheat has the potential to reduce the yield losses from this disease in China.

Experimental procedures

Construction of plasmids

The NIb gene of WYMV isolated from Yangzhou, Jiangsu Province, China (Chen et al., 2000), was amplified by RT-PCR. The NIb coding region of the RNA1 polyprotein was identified (nts 4912-6495), and sequencing analysis showed that it is highly homologous to the nucleotide sequences of the NIb gene of Japanese and other Chinese isolates (Figure S1). A part of the NIb gene (5284-6495) (Figure 1), designated NIb8, was amplified from the virus genome using BamHI linker primers NIb8F: 5′-CGGATCCATGATCAGAATGTTGGAAG-3′ and NIb8R: CGGATCCTTGGAG.

CTCAATGCTGCTATC-3′. The blunt-end fragment NIb8 and linear vector pUbi-35S were then linked to produce constructs with the NIb8 fragment inserted in either direction under the control of the ubiquitin promoter. The construct containing the reverse NIb8 insert fragment was then identified by sequencing and named pubi-NIb (Figure 1b). The assistant plasmid pAHC20 (Figure 1c) with selective marker gene bar (herbicide resistance gene) was prepared for co-transformation with plasmid pubi-NIb.

Transformation of wheat using the immature embryos

Fourteen-day-old immature embryos (IEs) of wheat cv. Yangmai 158 were cultured on SD2 medium to induce primary callus at 25 °C under dark conditions. The IEs precultured for 7 days were pretreated on high-osmotic medium (SD2 medium supplemented with 0.2 m mannitol and 0.2 m sorbitol) (Zheng and He, 1994) for 4 h and then bombarded by PDS-1000/He (Bio-Rad, Hercules, CA), with a 1100-psi split membrane and gold particles coated with the plasmid DNAs. After bombardment, the tissues were post-treated on high-osmotic medium for 16–18 h and then transferred to SD2 medium and maintained in the dark for 2 weeks without selection. The transformed calli were transferred onto 1/2 MS medium containing 1 mg/L KT + 3 mg/L Bialaphos + 1 mg/L NAA for differentiation at 24 °C under light for 4 weeks. Green shoots were transferred to 1/2 MS + 5 mg/L Bialaphos medium and continuously cultured under the same conditions for another 4 weeks for elongation. When they had grown to about 2 cm, the plantlets were moved onto 1/2 MS medium containing 0.5 mg/L IAA. The regenerated plants were transferred into pots, and when large enough, the leaves were sampled and genomic DNA was isolated.

Molecular screening of the putative transgenic wheat plants

For PCR testing, genomic DNA was extracted as described previously (Murray and Tompson, 1980). The full-length primers without BamHI sites were NIb8F-WB 5′-ATGATCAGAATGTTGGAAGACGCC-3′ and NIb8R-WB 5′-TTGGAGCTCAA. TGCTGCTATCACC-3′. The amplification conditions were 94 °C 5 min for denaturing; 94 °C 30 s, 53 °C 1 min 45 s, 72 °C 1 min 30 s for 35 cycles for amplification; 72 °C 7 min for prolongation and subsequent storage at 4 °C. For RT-PCR assays, total RNA was extracted from the transgenic wheat and wild-type plants according to the manual of the TRIZOL Kit (GIBCO BRL) and used to synthesize cDNA with random primers (TaKaRa cDNA Synthesis Kit, Dalian, Liaoning province, China). The primers and reaction condition of RT-PCR were the same as for PCR testing.

Resistant assessment of the transgenic wheat to WYMV in field

The transgenic wheat lines were evaluated for WYMV resistance from 2001 to 2005 in three field disease nurseries of Yizheng, Jiangyan and Lixiahe in Jiangsu Province. Seeds were sown in autumn (23 October) and harvested in summer (early June) of the following year. The trials received natural rainfall and were fertilized three times during the season. In 2000, when there were few seeds of the transgenic wheat lines, there was a single row (50 plants) of each test line. In 2001 and 2002 (T3 and T4 generation plants), the transgenic and wild-type wheat were sown in three replicate plots each of three rows of 1.4 m (40 plants). Disease incidence was evaluated on 100 plants per plot. From 2003 to 2005 (T5 to T7 generation plants), the plants were arranged in three replicate plots each of 6.7 m2 (1000 plants), and disease incidence was evaluated on 20 plants from each of five sampling points in each plot. In 2004 and 2005, the main agronomic characters of the transgenic wheat lines were also investigated. The plants were scored for disease class on a scale of 0 to 3 to evaluate severity of the disease (Hou et al., 1985): 0 = no visible symptoms, 1 = lightly streak mosaic leaf but plant not stunted, 2 = distinct mosaic streak covering one-half of the diseased leaf, and 3 = mosaic area covering three quarters of the diseased leaf and the plant obviously stunted. A disease index (DI) was then calculated: DI (%) = ∑ (DS × Ni) × 100/(3 × N), where DS is the disease scale (above), Ni the number of plants with this DS, and N the total number of plants observed. The DI and phenotypic data for the WYM responses were analysed using Statistical Analysis System, version 8.1 (SAS v8.1) (SAS Institute Inc, Raleigh, NC).

In 2003–2005, similar trials were also conducted on infested sites at Yantai (Shandong Province), Xiping (Henan Province) and Yangzhou (Jiangsu Province).

Selection of marker-free transgenic wheat by PCR and herbicide spraying

As NIb8 gene and bar gene were co-transformed into wheat in our study, it was possible to obtain marker-free plants in the offspring of T0 transgenic wheat. PCR tests were used to test T3 plants for the NIb8 and bar genes. Primers NIb8F-WB and NIb8R-WB were used for detecting NIb8, and primers Bar-1F 5′-CTGCACCATCGTCAACCACTACATC-3′ and Bar-1R 5′-AGCTGCCAGAAAC.

CCACGTCAT-3′ for the bar gene. The amplification conditions for bar gene were 94 °C 10 min for denaturing; 94 °C 30 s, 58 °C 1 min 30 s, 72 °C 1 min 15 s for 35 cycles for amplification; 72 °C 10 min for prolongation and storage at 4 °C. The transgenic offspring were also screened for marker-free plants by spraying herbicide (Liberty, 150 ppm or Basta, 100 ppm) at the seedling stage combined with PCR testing.

Copy number and segregation analysis of transgenic wheat plants

Genomic DNA was extracted from the leaves of the transgenic wheat plants. For Southern blotting, 35 μg genomic DNA was digested with EcoRI and SspI, separated by electrophoresis on 0.8% agarose gels and transferred to N+ Hybond nylon membranes. The probe was amplified using primers NIb8F-WB and NIb8R-WB from the vector pubi-NIb and was labelled with α-32P dCTP using TaKaRa Random Primer Labeling Kit. Southern blotting was carried out as described by Gao et al., (2005).

T4-generation plants of lines N12-1 and N14 were crossed with Yangmai 15 (a wheat variety susceptible to WYMV). 892 and 773 F2 plants from the respective crosses N12-1/Yangmai 15 and N14/Yangmai 15 were planted in WYMV nurseries in Yangzhou and evaluated for resistance.

Tissue printing and Western blot assays

Tissue printing assay was carried out as described previously (Andika et al., 2005). Preparation of protein samples, SDS-PAGE, electroblotting and immunodetection for Western blot analysis were performed as described previously (Sun and Suzuki, 2008). WYMV CP was detected using primary anti-CP (1 : 5000) polyclonal serum and secondary polyclonal AP-conjugated goat anti-rabbit IgG (1 : 10 000) (Sigma, St. Louis, MO).

Northern blot analysis, deep sequencing and bioinformatic analysis of small RNAs

Total RNAs were extracted from the leaves pooled of the ten wheat plants with Trizol (Invitrogen, San Diego, CA) according to the manufacturer's instructions. Gel electrophoresis, blotting and detection of siRNAs using the DIG system were carried out as described previously (Goto et al., 2003). A digoxigenin (DIG)-labelled DNA probe specific for WYMV RNA1 (nt 5284-6495) was prepared using the PCR DIG Probe Synthesis Kit (Roche, Mannheim, Germany).

The cDNAs of small RNA libraries were prepared using the Illumina TrueSeq Small RNA Sample Preparation Kit (Illumina, San Diego, CA) and then used for sequencing using the Illumina HiSeq2000 (Illumina). The adaptor sequences in the raw data sets derived from Illumina sequencing data were trimmed, and small RNA reads without an identifiable linker were removed. The reads of >30-nt or <18-nt were excluded for further analyses. Noncoding RNAs were then identified using the Rfam database and removed. Computational analyses were performed using Perl scripts. To search for viral siRNAs, reads from the two libraries were mapped to the WYMV RNA1 (AJ131981) and RNA2 (AJ131982) using BWA software (http://bio-bwa.sourceforge.net/) with two mismatches allowed.

Acknowledgements

This work was supported by the National Key Project for Research on Transgenic Biology (2013ZX08002-001) and the National 863 High-tech Project (2012AA10A309). We are grateful to Dr. Mike Adams for critically reading the manuscript.

Ancillary

Advertisement