Introduction

Chlorophyll (Chl) is the most important photosynthetic pigment, and Chl content is a critical breeding trait because it is positively correlated with photosynthesis (Xu et al. 1997; Teng et al. 2004; Hu et al. 2009; Zhang et al. 2009; Wang et al. 2016). Previous reports have demonstrated that high-yielding varieties have a higher leaf Chl content and photosynthetic rate than the previously released low-yielding lines, particularly at heading stage (Peng et al. 2008). Increasing Chl content may be an effective way to increase dry matter production and grain yield in rice. Chl content is strongly affected by nitrogen (N) supply because N is a major component of Chl. In recent years, to obtain the maximum grain yield, rice varieties are selected to allow high chemical fertilizers input, including N input (Tong et al. 2006). However, excessive N supply will result in lower N use efficiency, which causes higher cost for agricultural production and environmental hazards (Tong et al. 2006; Kaushal et al. 2011; Hashim et al. 2015; Zhou et al. 2016). Insufficient N supply may lead to low Chl content, small leaves and low biomass production, which will consequently reduce grain yield and quality (Prasertsak and Fukai 1997; Adhikari et al. 1999). Increasing leaf Chl content under low N supply condition may be an effective way to improve N utilization efficiency.

Previous studies suggested that Chl content is controlled by multi-genes, usually called quantitative trait loci (QTL) (Wang et al. 2008). In recent years, with the rapid development of molecular markers and the application of high-resolution linkage map, some QTLs associated with Chl content have been identified. Wu and Luo (1996) detected three main-effect QTLs on chromosomes 2, 4, and 7 and one epistatic locus controlling Chl content using an F2 population of Palawan/IR42. Wang et al. (2003) discovered six QTLs associated with the contents of Chl a and Chl b on chromosomes 1, 4, and 11. Teng et al. (2004) detected three QTLs for Chl content on chromosomes 1, 3, and 8 using a double-haploid population of ZYQ8/JX17. Seven QTLs related to Chl content at tillering and heading stage were identified and mapped to chromosomes 1, 2, 3, 4, 6, and 8 using backcross-recombinant inbred lines (Dong et al. 2007). In other studies, Zuo et al. (2007) identified ten QTLs associated with Chl content at three different growth stages of rice on chromosomes 1, 3, 4, 5, 7, 11, and 12. Hu et al. (2009) detected 13 main QTLs controlling Chl content under well-watered and water stressed conditions on chromosomes 1, 2, 3, 4, 5, 6, and 10. Jiang et al. (2012) identified 22 QTLs related to Chl content using 94 recombinant inbred lines (RILs) derived from a cross between two japonica cultivars, Lijiangxintuanheigu (LTH) and Shennong 265 (SN265), and these QTLs distributed over all chromosomes except chromosome 5. Zhang et al. (2014) discovered five QTLs controlling Chl content on chromosomes 1, 4, 7, 12 and two QTLs on chromosomes 4 and 12 at heading and maturity stages, respectively. On the whole, these QTLs distributed on all chromosomes, mainly on chromosomes 1, 2, 3, 4 and 11. In these previous researches, some QTLs can be detected at several growth stages but the other QTLs can only be detected at a specific developmental stage (Wang et al. 2003; Zuo et al. 2007).

In a previous study, a high-density SNP-based genetic linkage map has been constructed using 132 RILs derived from a cross between 93-11 and Pei-Ai 64s (PA64s) (Gao et al. 2013). 93-11 is an indica restorer line, and PA64s is a photo-thermo-sensitive male sterile line with higher Chl content than 93-11. The pioneer two-line super hybrid rice variety, Liangyoupeijiu (LYP9), is derived from the cross between these two parents and has greatly increased yield.

The top three leaves of rice are the most important organs for photosynthesis and the main source of carbohydrates production (Foyer 1987). In the current study, using the genetic linkage map, we discovered 32 QTLs related to the Chl content of the top three leaves under two N conditions at two developmental stages. Among these QTLs, we identified a major-effect QTL, qFCC7 L controlling Chl content under low N condition. A chromosomal segment substitution line (CSSL7) was used to confirm this putative QTL. We performed fine mapping using a BC4F2 population. We also analyzed the candidate genes, which could provide a reference for cloning the target gene related to Chl content.

Materials and methods

Plant materials and genetic map

A total of 132 RILs from LYP9 and the high-density linkage map for this population were used in this study. The linkage map was constructed by using single nucleotide polymorphism (SNP) markers by Gao et al. (2013). For the development of CSSL7, an RIL containing the PA64s alleles at qFCC7 L region was selected by target markers and then successively backcrossed with 93-11 for four generations. Three markers M1, RM234, and RM21981 were used for marker-assisted selection of each generation. To avoid the interference from other QTLs, we developed some markers to identify the genotypes of the regions which also contained our target QTLs.

Field experiment and trait measurement

The 132 RILs and two parental lines were grown in Hangzhou, China (May, 2010) under two N conditions (LN, without N fertilizer; HN, with 150 kg/hm2 N fertilizer). The two parental lines and CSSL7 were grown in Hainan, China in 2013. The BC4F2 population with 3562 individuals was grown in Hangzhou (May, 2013) at LN condition. The plants of each line at the 25-day-old seedling age were transplanted into a four-row plot with six plants per row and a spacing of 20 cm × 20 cm.

Rice leaf Chl content exhibited dynamic change. Previous reports demonstrate that the growth rate around heading stage is significantly associated with the final yield in rice (Takai et al. 2006; Horie et al. 2006). Heading stage is an important period in the life cycle of rice because photosynthetic products are transported from source to the bank at this period. Recent studies discovered some QTLs controlling Chl content only at specific developmental stages (Zuo et al. 2007). Based on these research findings, we measured leaf Chl content at initial heading stage (a period between the end of booting stage and heading stage, when the flag leaf is fully expanded and flag leaf sheath is opened but the tip of inflorescence is not emerged from sheath) and heading stage. The Chl contents of the first leaf (FL), second leaf (SL) and third leaf (TL) from the top were evaluated by Soil and Plant Analysis Development (SPAD) meter (SPAD-502, Minolta Camera Co., Osaka, Japan), a portable device which provides a nondestructively method for estimating leaf Chl content (Chubachi et al. 1986). In some studies, leaf Chl content is presented as SPAD value which can also be used in estimating leaf photosynthesis (A-JIA et al. 2008; Takai et al. 2010). Three randomly selected plants from the middle of a row in each rice line were tested. For each observed leaf, SPAD values were measured in the upper, middle, and base sections of the leaf on main panicle, and the mean SPAD value of the leaf sections served as the phenotypical value of SPAD (Zhang et al. 2014). The Chl content (mg dm−2) of leaves was calculated as follows: Y = 0.0996 × −0.152 (Zhang et al. 2014), where X is the mean SPAD value measured by the Chl meter, and Y is the actual leaf Chl content. Photosynthetic rate of leaves was measured by a portable photosynthesis measuring system (LI-6499XT, http://www.ecotek.com.cn). The specific leaf area (SLA, cm2 g−1) of leaves was calculated as follows (Li et al. 2005a): SLA = S/G, S meant the leaf area measured by the leaf area meter, and G represented leaf dry weight.

Data analysis and QTL analysis

Phenotypic variations were analyzed using the SAS (Statistical Analysis System) V8.01 software. QTL analysis was performed using the MultiQTL1.6 package (http://www.multiqtl.com) with maximum likelihood interval mapping. The LOD threshold for each dataset was set based on a permutation test (1000 permutations, P = 0.05). Major-effect QTLs were accepted when their LOD scores were larger than 2.5. QTLs were named according to McCouch et al. (1997).

Mapping population

To fine map the major QTL qFCC7 L , a backcross population was constructed in our study. The RILs with PA64s genotype in the qFCC7 L region and most of the genetic background from 93 to 11 (based on the genotype of RILs) were selected to successively backcross with 93-11. Additionally, two markers M1 and RM234 (Table S1) were used for marker-assisted selection of each generation. As a result, a BC4F1 line with 93-11 genetic background exhibiting heterozygous (identified by the markers M1 and RM234) at the qFCC7 L region was constructed in Hainan. After self-crossing, we obtained a BC4F2 population for fine mapping of qFCC7 L .

Fine mapping of qFCC7 L

SSR markers were obtained from the Gramene database (http://www.gramene.org/). Additionally, we designed the insertions/deletions (InDels) markers M1, M2, and M3 within the target region on chromosome 7 using the InDels identified between 93-11 and PA64s (Gao et al. 2013) by two software, Primer Premier 5.0 and Oligo 7. All relevant primers are listed in Table S1. Total genomic DNA was extracted from young leaves using the CTAB method (Murray and Thompson 1980; Doyle and Doyle 1987).

RNA extraction and expression pattern analysis

Quantitative real-time polymerase chain reaction (qRT-PCR) was performed to analyze the expression pattern of the peptide transporter gene OsPTR4. Total RNA was isolated at heading stage using an RNA extraction kit (Axygen), and the first-strand cDNA was synthesized using the ReverTra Ace qRT-PCR RT Kit (TOYOBO CO. LTD.; http://www.toyobo.cn/) following the manufacturer’s instructions. The primers used in qRT-PCR analysis were designed using an online tool (https://www.genscript.com/ssl-bin/app/primer). The primers for amplification of the internal reference OsActin1 were 5′-CCATTGGTGCTGAGCGTTT-3′ (forward) and 5′-CGCAGCTTCCATTCCTATGAA-3′ (reverse). The primers for amplification of OsPTR4 were 5′-CAAGATTGGCTCACTCTCCA-3′ (forward) and 5′-AACCCAGAATGTCACGCATA-3′ (reverse). Each 10 μL reaction system consisted of (1) μL total cDNA, 5 μL of 2× SYBR Green PCR Master Mix (TaKaRa Co. LTD., Otsu, Japan; http://www.takara.com.cn), (2) μL of each 1 μM primer. The analysis of each sample was performed with four replicates. Reactions were run on the ABI PRISM®7900HT Sequence Detector (Applied Biosystems). The 2− ∆∆CT method (Livak and Schmittgen 2001) was adopted to analyze the relative expression level, and the two-tailed t test was used to analyze statistically significant difference.

Results

Performance of SPAD value in RILs and parents

The phenotypic differences between 93-11 and PA64s as well as the leaf SPAD value variations in the RIL population under two N conditions at two developmental stages are summarized in Fig. 1 and Table S2. The top three leaves of 93-11 showed lower SPAD values when compared with those of PA64s (Table S2). The SPAD values of the top three leaves in RILs all showed continuous distribution under LN and HN conditions at two different developmental stages. In addition, some lines showed apparent transgressive variations in the RIL population (Fig. 1). These results indicated that the Chl contents of the top three leaves were controlled by multiple genes, and the data satisfied the demands for QTL analysis.

Fig. 1
figure 1

Distribution of SPAD values of the top three leaves in the RIL population. a–c SPAD values of the top three leaves surveyed at initial heading stage. d–f SPAD values of the top three leaves surveyed at heading stage. White columns indicate the plants grown at low N condition. Grey columns indicate the plants grown at high N condition. Arrows with smooth end indicate 93-11, arrows with notch end indicate PA64s

Full size image

QTL analysis for Chl content

A total of 32 QTLs related to Chl content of the top three leaves under two N levels at two stages were detected (Fig. 2; Table S3). These QTLs were distributed on chromosomes 1, 4, 5, 6, 7, 8, and 12, with LOD scores between 2.51 and 9.46 and accounted for 6.0–20.8 % of the total phenotypic variation. Among these QTLs, 19 QTLs were clustered in three intervals on chromosomes 4, 6, and 7 (Fig. 2). Six QTLs were detected between the SNP markers SNP4-246 and SNP4-271, explaining 10.3–20.8 % of the phenotypic variations; seven QTLs were located in the region between SNP6-90 and SNP6-106, explaining 8.6–20.5 % of the phenotypic variations; six QTLs were detected between SNP7-160 and SNP7-192, explaining 6.0–13.9 % of the phenotypic variations. We considered that these 19 QTLs were actually consisted of three loci. The positive effect of most QTLs came from the alleles of PA64s, the parent with high Chl content (Table S3).

Fig. 2
figure 2

Locations of main-effect QTLs for chlorophyll content (SPAD value) in the RIL population of 93-11 and PA64s. QTLs detected in low N condition are designated with shapes used light color and QTLs identified in high N condition are marked with shapes used dark color. The number on the left of each chromosome is the marker’s genetic distance (cM)

Full size image

Fine mapping of qFCC7 L

Among these QTLs, we discovered that the QTL qFCC7 L controlled the Chl contents of the top three leaves under LN condition at initial heading stage and the Chl contents of FL and TL under LN condition at heading stage. This QTL explained 13.9 % of the total phenotypic variation of the FL Chl content, and the PA64s allele could increase the Chl content. We confirmed this QTL by the CSSL7 with 93-11 genetic background but containing a PA64s chromosome segment at qFCC7 L region (Fig. S1). The CSSL7 had a higher SPAD value and photosynthetic rate than 93-11 (Fig. 3a, b) and showed a lower SLA (Fig. 3c), suggesting that this locus may also influence photosynthesis and leaf thickness. To fine map this major-effect QTL, we constructed a BC4F2 population. Previous studies demonstrated that at least 50 % of the photosynthetic products for final yield are provided by flag leaf (Li et al. 1998). So we used the SPAD value of FL in BC4F2 population as the reference indicator for mapping. Three InDels and six SSR markers were used for fine mapping of qFCC7 L . After analyzing the genotypes and phenotypes of 3562 individuals in BC4F2 population, we finally positioned qFCC7 L between the markers M3 and RM21938 in a 124.5 kb interval on the long arm of chromosome 7 (Fig. 4). According to the Rice Genome Annotation Project Website (http://rice.plantbiology.msu.edu/), there were 16 open reading frames (ORFs) with different predicted biological functions in this region. LOC_Os07g41200 encodes a homolog of Arabidopsis thaliana LONGIFOLIA proteins; LOC_Os07g41240 encodes a putative cytochrome P450 protein; LOC_Os07g41250 encodes a putative peptide transporter protein; LOC_Os07g41260 encodes a putative PPR repeat domain-containing protein. The functions of these four genes were predicted to be related to yield, leaf color and N transport based on previous studies (Wang et al. 2015b; Yang et al. 2013; Fan et al. 2014; Tan et al. 2014).

Fig. 3
figure 3

Agronomic traits of leaves at heading stage. a Chl content of FL, SL, and TL of 93-11, CSSL7, and PA64s. Data was measured from six plants at Hainan. b Photosynthetic rate (PR) of 93-11, CSSL7, and PA64s. c Specific leaf area (SLA) of 93-11, CSSL7, and PA64s. Each bar indicates the mean ± SD of three independent replicates. Asterisks indicate significant difference (Student’s t test, *P < 0.05, **P < 0.01)

Full size image
Fig. 4
figure 4

Fine mapping for qFCC7 L . The black and white boxes indicate the chromosomal segments from PA64s or 93-11, respectively. ** means the SPAD values of lines L2, L3, L6, and PA64s which contained the qFCC7 L candidate region, have significant difference compared with 93-11 (P < 0.01). Student’s t test was used for statistical analysis, data are shown as mean ± SD, n = 3

Full size image

Candidate gene identification and expression analysis

The encoded peptide transporter belongs to the nitrate transporter 1/peptide transporter (NRT1/PTR) family which is involved in the uptake and transport of nitrate/peptide and has the functions related to N use efficiency (Fan et al. 2014; Li et al. 2015a). There was a predicted peptide transporter (PTR4) in the mapped interval, so we performed a further analysis of this gene. Deep resequencing showed that OsPTR4 contained five exons and four introns, and sequence variations existed between the parents but mainly in the introns (Fig. 5a). A total of 15 base variations were detected between the parents, including 4 base insertions and 11 base substitutions. Among these variations, four base substitutions caused the changes in three amino acids at the fourth and fifth exons (Fig. 5a). qRT-PCR detected obvious expression variations of OsPTR4 between 93-11 and PA64s during heading stage. The expression levels of OsPTR4 in 93-11 were lower than those in PA64s in all plant tissues (Fig. 5b). The OsPTR4 transcript level in the leaves was the highest, and that in the roots was the second, which were consistent with its uptake and transport capacities. Based on these results, we speculate that the differences in gene structure and expression level lead to the different functions between the parents.

Fig. 5
figure 5

Sequence differences and expression analysis of OsPTR4 between 93-11 and PA64s. a Sequence differences of OsPTR4 between 93-11 and PA64s. ATG and TAA denote start and stop codon. Black boxes indicate exons, discontinuous straight lines between black boxes represent introns. Sequence discrepancies are labeled as shown. Red alphabets represent changed amino acid residues. b Expression analysis of OsPTR4 between 93-11 and PA64s. Each bar indicates the mean ± SD of three independent replicates. Student’s t test was used for statistical analysis. ∗∗ represents significant difference at 0.01 level

Full size image

Discussion

Chl content is influenced by N supply because there is a certain relationship between N content and Chl content (Wang et al. 2002; Huang et al. 2008). Interestingly, there are very little differences between the Chl contents under zero N input and 150 kg/hm2 N input in the parental lines. The cross between 93-11 and PA64s produces a strong heterosis hybrid rice variety, LYP9, which can absorb more N fertilizer and has a slow reaction to N level. So the parental lines showed little differences in Chl content between two N levels. On the other hand, Chl content is controlled by multi-genes which have some interactions and can reduce the final effect.

To improve fertilizer use efficiency, we hope to dissect its genetic basis using the parents and the derived RILs. For genetic analysis and QTL mapping, we usually choose the parents with obvious phenotype differences. But in this study, the Chl content difference between parents was not very significant. Thus, the selection of parental lines seemed to not match well with the objective of the experiment. However, Li et al. (2005b) have confirmed that QTLs can be identified using the genetic populations without obvious parental differences, and some minor but useful QTLs are thus detected and can be mined via the construction of backcrossing lines (Li et al. 2015b). We have identified 32 such QTLs at two different growth stages. Although we did not observe much difference in Chl content for the parental lines, the presence of such a large number of QTLs indicates that Chl content, particularly in PA64s, is controlled by the interactions of multiple genes. Our study can supplement the discovery of Li et al. (2005b) that minor phenotypic variation may also result from multiple QTL identification. All these findings indicate that the parent PA64s is an elite sterile material for genetic dissection and breeding application.

Photosynthesis is the fundamental factor of dry matter production and grain yield, and high photosynthetic ability is one of the major breeding targets for further improving rice yield. Some studies have demonstrated a positive relationship between leaf photosynthesis and leaf Chl content (Kura-Hotta et al. 1987; Xu et al. 1997; Hu et al. 2009). Using different populations, many QTLs related to Chl content have been reported (Wang et al. 2003; Teng et al. 2004; Jiang et al. 2012). In this study, a mapping population developed from two parental lines, one with high Chl content but low yield potential (PA64s) and another with low Chl content but high yield (93-11), provided a unique opportunity to identify QTLs for the Chl content which contributed to the high heterosis of the pioneer hybrid rice LYP9 developed by crossing these two parents and would benefit the molecular breeding of new hybrid rice genotypes. A total of 32 main-effect QTLs for Chl content were detected (Fig. 2; Table S3), including 13 QTLs discovered under low N level and 19 QTLs identified under high N level. Previous reports have demonstrated that rice yield will remain stagnant with the increasing N utilization. The main reasons causing this phenomenon are unreasonable fertilization and the low use efficiency of N fertilizer (Borrell et al. 1998; Peng et al. 1996). Chl content is closely related to N supply. Under N-stress condition, the variation of Chl content can reflect the difference in N use efficiency of different genotypes. Therefore, the QTLs detected at low N level may contribute to the high tolerance to low N stress. They are useful for understanding the genetic mechanism underlying N use efficiency and reducing the excessive input of N fertilizer. Under N deprivation condition, nitrate reductase activity which affects N use efficiency is associated with the expression level of HvNar1 in barley (Xu et al. 2016). Some QTLs detected in this study were in or near the regions which have been reported previously. The QTLs between SNP1-10 and SNP1-27 are near to the region of qCHM1 (RM490–RM580) reported by Jiang et al. (2012) and HSPAD-1b (RM6902–RM8111) identified by Zhang et al. (2014). The QTLs between SNP4-246 and SNP4-271 are corresponding to qCH4, qCM4b and qCTH4 (RM255–RM559) found by Jiang et al. (2012) and qLSCHL4 (STS4-5–RM349) detected by Zhang et al. (2014). We also detected some new loci, the qSCC5 H between SNP5-203 and SNP5-226 controlling the Chl content of second leaf in high N level at heading stage, the qFCC8 L between SNP8-139 and SNP8-164 explaining 9.4 % of phenotypic variations of the FL Chl content at initial heading stage. We also discovered that the expressions of some QTLs had spatial-temporal specificity, and they could only be detected at a specific time. For example, the QTL between SNP6-106 and SNP6-137 only affected the Chl content at initial heading stage, whereas the QTL located between SNP5-203 and SNP5-226 on chromosome 5 was detected at heading stage. The Chl content of leaves might be affected by the variation of days-to-heading. Wang et al. (2015a) reported that Ghd7 is a major locus for the natural variation of Chl content. We discovered that some genes related to heading time were on the corresponding regions of QTLs for Chl content. In our study, qSCC6 IL /qTCC6 IL was near Se5/OsHY1/OsHO1 which also located on chromosome 6. Another major-effect QTL controlling heading time, DTH8/Ghd8/LHD1, is on the corresponding region of qFCC8 L . Actually, Gao et al. (2013) also detected this QTL for heading date on DTH8 using the same RILs in this study. Recently, Ding et al. (2016) discovered that the transcription levels of two photoperiod-sensitive regulators, Heading date 3a (Hd3a) and RICE FLOWERING LOCUS T 1 (RFT1), were down-regulated in leaves by nitrogen fertilizer. These findings suggested that Chl content and nitrogen fertilizer are closely related to heading time. Liu et al. (2015) identified 27 QTLs related to leaf shape (length and width) and the Chl content of top three leaves across five different environments using a set of CSSLs derived from the cross of 93-11 and PA64s, the same parental lines used in this study. Undoubtedly, some QTLs detected in Liu’s study are at the same regions identified in this research. For example, the QTLs between SNP1-10 and SNP1-27 are at the same region of qFLC1. The QTL between SNP8-38 and SNP8-62 is on the corresponding region of qFLC8-1. However, we also detected some QTLs different from the loci identified in Liu’s study. These results demonstrate that QTL detection is not only influenced by the parents but also limited by the construction of populations, developmental stages and environments.

Among these QTLs, only one QTL affected the Chl content of the top three leaves under low N level at initial heading stage, and it also controlled the Chl contents of FL and TL at heading stage (Fig. 2). Actually, Wu and Luo (1996) also discovered that this QTL can control leaf Chl content under nitrogen stressed condition. Therefore, qFCC7 L may be the same QTL as the previously identified QTLs. The CSSL7 which contained a PA64s chromosome segment in the qFCC7 L region with 93-11 genetic background showed a higher SPAD value than 93-11 (Fig. 3a). High Chl content per unit leaf area may reflect the presence of a larger number of chloroplasts per cell and/or high Chl content per chloroplast when assuming that leaf thicknesses do not change (Takai et al. 2010). However, previous studies have shown that SPAD value may sometimes reflect the variation in leaf thickness because the readings are based on the leaf Chl’s absorption of specific spectral bands which may be influenced by leaf thickness (Peng et al. 1993; Li et al. 2009). SLA is considered to be correlated with leaf thickness and was significantly lower in CSSL7 compared with that in 93-11 (Fig. 3c). These results suggest that the higher SPAD value of CSSL7 may result from its thicker leaves, and the relationship between leaf thickness and SPAD value needs further in-depth studies. Thicker leaves are considered to be important for increasing leaf photosynthesis because they can capture light energy efficiently by more Chl per unit leaf area, and they have sufficient mesophyll surfaces, allowing more efficient CO2 diffusion and transport (Terashima et al. 2006). Thus, the higher photosynthetic rate of CSSL7 may result from the increased Chl content or leaf thickness, or both (Fig. 3b). Although this QTL was only detected in one-year data, CSSL7 also showed a higher SPAD value than 93-11 at Hainan, which demonstrates that this QTL can be stably expressed at different conditions. Using a BC4F2 population, we fine mapped this QTL to a 124.5 Kb region between the markers M3 and RM21938 (Fig. 4), and this interval contained 16 ORFs. In fact, our team has detected a QTL controlling leaf width in this interval (Zhang et al. 2015). But whether this QTL is identical to qFCC7 L needs further study. We also performed a deep analysis of the candidate gene OsPTR4 and discovered the sequence variations and different expression levels between the parents which may lead to the different functions between parents (Fig. 5).

In modern cultivation conditions, the rice varieties with high yield always rely on the condition of high N input. Here, we have detected a major effect QTL qFCC7 L controlling Chl content at low N condition. Therefore, it may be useful for breeding the rice varieties capable of maintaining high leaf Chl content under low N condition. This is a good method to improve N use efficiency. Further researches should be carried out to confirm whether OsPTR4 contributes to Chl content. Identifying the locations of these QTLs can help further understanding the molecular genetic mechanisms of Chl content and photosynthesis. Therefore, mapping the QTLs controlling Chl content and leaf-related traits can accelerate the breeding new rice varieties with high photosynthetic capacity and yield via genome selection (Guo et al. 2014).