Nitrogen is the most importance nutrition element for rice in plant growth, yield and quality (Dalling et al., 1985; Singh et al., 1998). In China, people massively use the nitrogen fertilizer in rice production to significantly increase the yield. However, the average utilization rate of nitrogen fertilizer in rice ranges from 28% to 37%, which was lower than the world average level about 15%~20% (Luo et al., 2003). Meanwhile, the excessive use of nitrogen fertilizer caused many problems, exacerbated the occurrence of disease and lodging of rice (Bohlool et al., 1992), as well as the environmental pollution. With the technologies of resource-saving and environment protection getting more and more attention, it would be of great theoretical and practical significance to explore and use the high-efficient nitrogen or low nitrogen-tolenrance rice germplasm resources.

There were significant genotype differences in rice under low nitrogen stress (Wu and Luo, 1996). In the low nitrogen conditions, the heritability of rice grain yield and nitrogen use efficiency was significantly higher than that in high nitrogen conditions. It was more effective to screem high efficient nitrogen and low nitrogen-tolerance rice varieties under the lower nitrogen levels (Piao et al., 2003; 2004). Yu Sibin (Yu et al., 2005) screened that Cisanggarung, Nippon-bare, Miyang23, Manawthukha, AT354, C71 and C70 were low nitrogen-resistant germplasms, and the donors DianTun502, Jiangxisimiao, Giza159, IR50, Jumlimarshi, Sadririce1, Molok, AP423, and Buy-dda were low- nitrogen-sensitive germplasms by employing more than 150 donor parents and backcross progenies, which were supported by the Global Rice Molecular Breeding Program, 948 Projects of Ministry of Agriculture, as the materials and taking biomass, tiller number, seed setting rate, spike size relative value as indices under the low nitrogen stress. Madhukar and UPR191-66 were selected as low-nitrogen-resistant genotypes germplasm based on chlorophyll SPAD values (Xu et al., 2005).

In this paper, we further evaluated the tolerance to low nitrogen of the introgression lines materials based on Xu (2005) in Global Molecular Breeding Programs.

1 Results and Analysis
1.1 Analysis of the basic agronomic characters
In the two Nitrogen levels, the SPAD value and the average value of some agronomic traits including the number of tillers, effective panicle, height, and grains per spike were similar between DJY1 introgression lines and CK (DJY1), there is no significant difference, and showed that the effect of backcross was obvious. The result of T-test showed that the SPAD value of introgression lines and the average value of agronomic traits of were significantly difference (Table 1), which illuminated that the changes of phenotypic value can be impacted on the Nitrogen level in the investigation. The SPAD value showed a normal distribution (Figure 1), moreover, the SPAD value of the introgression lines was significantly higher than the moderate nitrogen level (Yiliang nitrogen: 99.4 mg/kg) in relative lack of nitrogen in soil (Xundian nitrogen: 56.7 mg/kg). In addition, the average value of the agronomic traits investigated in the relative lack of nitrogen level was significantly lower than the moderate level of nitrogen, and the differences was showing in agronomic traits.

Table 1 Performance of agronomic traits of introgression lines at the two sites

Figure 1 Frequency distributing of SPAD Value in two sites

1.2 Correlation analyses of the SPAD values and some agronomic traits
In Xundian, the correlation analysis results showed that the SPAD value of introgression lines was highly significantly positive correlated with effective panicle, panicle length and grains per spike, and wassigni- ficantly positive correlated with height, and significantly negative correlated with tiller number, but was no correlation with neck length of spike and seed setting rate. In Yiliang, the SPAD value of introgression lines was very notablely positive correlated with height, effective panicle, neck length of spike, grains per spike, but was no correlation with tiller number, panicle length, seed setting rate (Table 2). Seen from table 2, we found that the SPAD value of the test introgression lines was quite significantly positive correlated with effective panicle, grains per spike in two test sites, which showed that the absorption and utilization of nitrogen were closely related to effective panicle and grains per spike. Therefore, we can take the effective panicle and grains per spike as indexes to evaluate the response of rice to nitrogen.

Table 2 Correlation analysis among the agronomic traits of introgression lines grown in the two sites

1.3 Screening of low nitrogen-tolerant germplasm
By two-tails test of the SPAD value of introgression lines, in 95% degree of confidence, the right-tailed critical value was 48.61 in Xundian, and that was 44.34 in Yiliang. In Xundian, SPAD values of 38 lines are higher than 48.61, and their donor parents were Zhong413, Heiheaihui, Zihui100, Gumei2, Huajingxian74, Yuexiangzhan, Q5, Y134, Cisanggarung, Khazar, ShweThweYinHy2, Bg300, Bg94-1, CR203, PSBRC28, PSBRC28, PSBRC66, IR58925B, IR50, X22, Porhveli, Chhomrong, Molok, VP15, GiZa14, Biniapan, Sapajira19-303, Ajaya, Amol3 (sona) and SML242, respectively. In Yiliang, have 31 lines were greater than 44.34, and their donor parents were Heiheaihui, Shengnong89366, Huajingxian74, Banaizhan, JP-5, Q5, B4122, Gang46B, Cisanggarung, Khazar, Nipponbare, Manawthukua, IR64, IR6855255-3-2, IR 58925B, M401, C71, Chhomrong, PaLung2, Molok, Ziri, Sapajira19-303 and Type3.
  
Heiheaihui, Q5, Cisanggarung, Khazar, IR58925B, Chhomrong, Molok and Sapajira19-303 as the donor of the introgression lines in both trail sites were in 95% right-tailed critical region, among them, the SPAD value of BC3F3 introgression line of the donor Khazar (No.1366, SPAD value was 49.54 in Xundian, SPAD value was 45.60 in Yiliang) and that of BC3F4 introgression line of the donor Chhom-rong (No.2155, SPAD value was 48.70 in Xundian, SPAD value was 45.40 in Yiliang) were greater than its critical value in two trail sites (Table 3), therefore, we infered that these two introgression lines were low nitrogen- resistant germplasms.

Table 3 Screened germplasms for tolerance to low-nitrogen

1.4 Screening of low-nitrogen-sensitive germplasm
By two-tails test, in 5% degree of confidence, the left-tail critical value was 42.03 in Xundian, and that was 34.72 in Yiliang. In Xundian, have 67 introgression lines of donor parents were lese than 42.03, and their donor parents were B4122, BR11, Bg90-2, Iksan438, Bg300, IR64, ASD16, Basmati370, Basmati, R50, Type3, Domsian, Bg304, Lemont, OM1723, KholeMavshi, ASD18, Ajaya, Budda, Doddi, Molok, Giza14, Chorofa, OM997, Jhona349, Milyang23, IR68552-55-3-2, Cs94, SML242, LA110, Madhukar, Chorofa, Jiangxisimiao, Chenghui448, Zhongyouzao8, Shennong89366, Yuqiugu and Gang16, respectively. In Yiliang, have 50 lines of donor parents were less than 34.72, their donor parents were PR106, IR64, ASD16, Basmati370, Tetep, KavanalLocal, Type3, Binam, AT354, OM1706, C70, KholeMavshi, UPR19166, Ajaya, Doddi, Rustylate, TekSiChut, SadriRice, Cs94, Serendahkunin, Chhomrong, Asomironi, Amol3 (sona), Madhukar, Malaysia, LA110, Diantun502, Fengaizhan, Yuqiugu, Zhongyouzao8, Yunhui72 and Gang16, respectively.

IR64, Basmati370, Madhukar, ASD16, Type3, KholeMavshi, Ajaya, Doddi, Cs94 and Yuqiugu as the donor of the introgression lines in both trail sites were in 5% degree of confidence left-tailed critical region, of which the SPAD values of the BC3F3 introgression lines of the donor Basmati370 (No.1626, SPAD value was 38.12 in Xundian, SPAD value was 30.62 in Yiliang), Type3 (No.1865, SPAD value was 41.72 in Xundian, SPAD value was 33.84 in Yiliang), KholeMavshi (No.2219, SPAD value was 40.80 in Xundian, SPAD value was 24.12 in Yiliang), Ajaya (No.2258, SPAD value was 39.36 in Xundian, SPAD value was 34.10 in Yiliang), Doddi (No.2275, SPAD value was 41.38 in Xundian, SPAD value was 32.60 in Yiliang) and Yuqiugu (No.2792, SPAD value was 41.26 in Xundian, SPAD value was 34.50 in Yiliang) were less than their critical value in these two sites (Table 4), consequently, we concluded that these six introgression lines were low-nitrogen-sensitive germplasms.

Table 4 Screened sensitive germplasms to low-nitrogen

2 Discussion
The SPAD values in different rice varieties were quite different under the same growth conditions. In addition, SPAD values were affected by different stages, leaves, specific leaf sites, and environment (Wu and Tao, 1999). In this study, we carried out the experiments under low nitrogen stress in two sites, and the SPAD values are different, which may be because of the planting time, survey time and weather, and other factors. However, we also planted the CK varieties and insisted in determining SPAD value of the flag leaf on rice heading time, fixed in the middle the part of the flag leaf, so we can effectively eliminate the experimental error by compared introgression lines with the CK, the SPAD values can better reflect the nitrogen status of plants. In the low nitrogen limitation conditions, the chlorophyll content of plant in late growth stage were higher, nutritional status of nitrogen in plant was relatively better, therefore, these germplasms were more resistant to low nitrogen, which was consistent with Xu (2005).

The results of the effective panicles and grains per spike of rice were highly significantly positive correlated with leaf SPAD value, which confirmed that nitrogen utilization was related to effective panicles and grains per spike, and was consistent with the study of Huang (2006) that the effective panicles could be used as the evaluation of rice nitrogen utilization.

In this test, we screened two low nitrogen tolerance introgression lines and six low nitrogen sensitive introgression lines, which can be used for plateau japonica rice breeding of nitrogen efficient. These eight germplasms donor parents didn't consist with resistance to low nitrogen and low nitrogen-sensitive parent identified by Yu (2005) and Xu (2005), which may be resulted from different evaluation indicators and methods. It is worth to noticing this issue in future research, and it is necessary to establish one simple efficient accurate and large-scale identification indicators and criteria of rice nitrogen response.

3 Materials and Methods
3.1 The testing materials and cultivation methods
We employed plateau japonica rice varieties the Dianjingyou 1 (referred to DJY1, hereinafter the same), which were wide adaptability, high yield, good integrated traits, as the recurrent parent in Yunnan province, and 151 rice varieties were donor parents for the Global Rice Molecular Breeding of 948 Project, the Ministry of Agriculture. They were successive backcrossing and self-cross from 2000 to 2005 to get different generations of the backcross populations, the total lines was 2064.

Field trials were in the Xundian County (elevation is 1980 m, soil nitrogen content was 56.7 mg/kg) and Yiliang County (elevation is 1600 m, soil nitrogen content was 99.4 mg/kg) in Kunming in 2006. The recurrent parent DJY1 and its BC3F3, BC3F4, BC4F3, BC2F4, BC2F5, BC2F6 generations cultivated in experimental field, sowed in late March and transplanted in late May, the recurrent parent DJY1 used as the control. Each material was planted into a line, 20 strains a line, single planting, spacing of 10 cm×20 cm. During the growth process, without any fertilizer, other measures were the same as the local management of conventional cultivation techniques.

3.2 Data investigation and processing
Using the SPAD-502 machine to determine the middle part of flag leaf SPAD values in rice heading stage, determinate single plan and 5 plants per line, we investigated tiller number, plant height, effective panicles, panicle length, neck length of spike and grains per spike in the whole growth period of rice and seed setting rate of post-harvest, and used Excel and SPSS software to deal with the mean, variance, standard error, variance coefficient, correlation coefficients and t-test analysis.

3.3 Evaluation indices and standards of nitrogen response
Using the SPAD value as evaluation indicators, the reaction to nitrogen with plant line as the unit of the tested materials was evaluated. The test materials were analyzed by the one-tailed test at 5% degree of confidence as screening criteria, of which the SPAD values of the same lines in the two sites in the right tail critical region were the resistant to low nitrogen germplasms, and the SPAD values of the same lines in the two sites in the left tail critical region were the low-nitrogen-sensitive germplasms.

Authors’ Contributions
CD and FRX conceived the overall study, performed the experiment designs, and drafted the manuscript. CFT, XXA, ELZ, YY Y, FFZ, TQY took part in some experiments and analyzed part data. LYD supervised the projects and revised the whole manuscript. All authors have read and approved the final manuscript.

Acknowledgment
This work was co-supported by a grant from the Major Programs of the Ministry of Agriculture (948) (No. 2006-G1), a grant from Green Super Rice Cultivation Program (863 Program), a grant from Key Projects in the National Science & Technology Pillar Program during the Eleventh Five-Year Plan Period (No.2006BAD13B01) and a grant from Intelligence Introducing Program of Yunnan Province (No. 2008PY049).

References
Bohlool B.B., Ladha J.K., Garrity D.P., and George T., 1992, Biological nitrogen fixation for sustainable agriculture: A perspective, Plant and Soil, 141(1-2): 1-11
 http://dx.doi.org/10.1007/BF00011307
 
Dalling M.J., 1985, The physiological basis of nitrogen redistribution during filling in cereals, American Society of Plant Biologists, 55-7l Huang N.R., Zhong X.H., and Zheng H.B., 2006, Selection of rice genotypes the high nitrogen utilization efficiency and its evaluation indices, Chinese Agricultural Science Bulletin, 22(6): 29-34
 
Luo Z.X., Su Z.S., Shi F.Z., Ruan X.M., and Piao Z.Z., 2003, The present situation and forecsting for high efficient nitrogen utilization of rice breeding, Zhongguo Nongxue Tongbao, Chinese Agricultural Science Bulletin, l9(1): 66-67
 
Piao Z.Z., Han L.Z., and Koh H.J., 2003, Variation of nitrogen use efficiency by rice genotype,Zhongguo Shuidao Kexue, Chinese Journal of Rice Science, 17(3): 233-238
 
Piao Z.Z., Han L.Z., Koh H.J., Lu J.A., and Zhang J.M., 2004, Selection effect of nitrogen use efficiency in rice, Zuowuo Xuebao, Acta agronomica sinica, 7(30):651-656
 
Singh U., LaJha J.K., Castillo E.G., Punzalan G., Timl.A., and Duqueza M., 1998, Genotypic variation in nitrogen use efficiency in medium and long-duration rice, Field Crops Res., 58(1): 35-53
 http://dx.doi.org/10.1016/S0378-4290(98)00084-7
 
Wu P., and Luo A.C., 1 996, Investigation on genetic background of leaf chlorophyll content variation in rice under nitrogen stress condition via molecular markers, Yichuan Xuebao, Acta Genetic Sinica, 23(6): 431-438
 
Wu L.H., and Tao Q.N., 1999, Nitrogen fertilizer application based on the diagnosis of nitrogen nutrition of rice plants (Oryza sativa L.) using chloroghyll meter, Zhejiang Nongye Daxue Xuebao, Journal of Zhejiang Agricultural University, 25(2): 135-138
 
Xu F.R., Tang C.F., Yu T.Q., Yan H.M., Zhou H., Li J., Jiang H.B., Ye C.R., and Dai L.Y., 2005, Screening of Rice Germplasms for Tolerance to Low-nitrogen Using SPAD-value by Chlorophyll Meter, Molecular Plant Breeding, 3(5): 695-700
 
Yu S.B., Zhao M.X., Zhao S.J., Zhou H.J., Tan Y.B., Xu G.C., Luo L.J., and Zhang Q.F., 2005, Development and selection of introgression lines with the identical genetic backguound of varieties Zhenshan 97B and 9311, Molecular Plant Breeding, 3(5): 629-626