Grain-filling directly relates to the productivity and quality of rice, and the degree and rate of rice grain-filling differ considerably according to the position to the grain's position on a panicle. In general, the earlier flowering and superior grains, which are located on the apical primary branches, are filled faster and formed larger and heavier grains, while the inferior grains are set on the proximal secondary branches with sterile or poor quality. To meet a high productivity requirement for the modern rice variety, the following issues have to be answered: (a) what is the molecular ecological process of grain-filling, (b) how can the grain-filling of inferior grains be enhanced, and (c) how can the yield of the large panicle rice, in particular, be increased. Previous studies have shown that nitrogen (N) fertilization has significantly positive effect on rice's grain-filling. However, whether N fertilization in the post-growth stage of rice is appropriate largely depending on the dynamic transport of N in rice leaves, which plays an important role in the physiological activity and hence affects the grain-filling of rice. Furthermore, the underlying molecular ecological effect of N fertilization on the grain-filling is less investigated particularly in large-panicle rice in the field. The present study investigated the characteristics of protein expression and their response to different N fertilizations during the grain-filling stage of an indica large-panicle rice cultivar (Jinhui 809), by the 2D technique. The field experiments were carried out at the Experimentation Station of Fujian Agriculture and Forestry University, Fuzhou (119.280E, 26.080 N), Fujian, China during the rice growing season (June-November) in 2006 and 2007. Field management was the same, except the N application ratios between the 2006 and 2007 experiments. In 2006, the 225 kg/hm² N were applied with the regular fertilization. In 2007, the constant of the total N supply, but the N fertilizer was given in two early to late growth stage application ratios. i.e., 7:3 and 6:4, as the treatments. The grain samples were collected in 2006 and the superior and inferior grains samples under the different N application ratios were collected separately in 2007 were used for the protein extraction. Then, the extracted grain proteins were separated with the two-dimensional gel electrophoresis(2DE) with the Isoelectric focusing (IEF) tube gels for the first dimension, and sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) for the second dimension (2D) separation. A total of 32 differentially expressed proteins at different grain-filling phases were detected by the 2D maps of grain proteome. Among them, 27 proteins were successfully identified by Mass Spectrometry (MS). And, these identified proteins in different grain-filling phases were grouped into five expression patterns according to their functions in starch synthesis, energy metabolism, signal transduction and gene expression, as well as their regulation in the defense response to stresses, etc.. The differential expression patterns of proteins in superior and inferior grains at different phases responding to different N application ratios were then analyzed. The results showed that 22 out of 26 differentially expressed proteins were identified in inferior grains, while 7 out of 8 proteins with differential expressions were identified in superior grains. Our results indicated that the grain-filling of the superior grains, especially in the large-panicle rice, had a higher environmental stability, while the reverse was true in the case of inferior grains. The present study also showed that the increases in N supply at grain-filling phases of rice favored in the up-regulation of the genes involved in physiological activities of inferior grain filling, including the signals transduction, gene expression, energy accumulation for the material metabolism, and hormone production etc. in inferior grains of rice, in turn, leading to increased yield. This study adds new significant insights to our current understanding of the regulation of rice grain filling by N application.