Quantitative trait loci analysis for the developmental behavior of tiller number in rice (Oryza sativa L.)

 A doubled-haploid rice population of 123 lines from Azucena/IR64 was used for analyzing the developmental behavior of tiller number by conditional and unconditional QTL mapping methods. It was indicated that the number of QTLs significantly affecting tiller number was different at different measuring stages. Many QTLs controlling tiller growth identified at the early stages were undetectable at the final stage. Only one QTL could be detected across the whole growth period. By conditional QTL mapping, more QTLs for tiller number could be detected than that by unconditional mapping. The temporal patterns of gene expression for tiller number could be different at different stages. Even an individual gene or genes at the same genomic region might have opposite genetic effects at various growth stages. Received: 7 July 1997 / Accepted: 10 February 1998     

Dynamic expression of nine QTLs for tiller number detected with single segment substitution lines in rice

Nine single segment substitution lines (SSSLs) in rice, which contain quantitative trait loci (QTLs) for tiller number on substituted segments detected in previous studies, were selected as materials to analyse dynamic expression of the QTLs in this study. These SSSLs and their recipient parent, Hua-jing-xian 74 (HJX74), were grown in four different environments and were measured for tiller number at nine different growth stages. An indirect methodology was applied in QTL mapping through analyzing multi-environment phenotypic data. Dynamics of three types of effects (including total effect, main effect, and QE interaction effect) of QTLs was released. It was shown that nine QTLs exhibited statistically significant effects only at certain stages. Effects of a QTL, although insignificant at certain stages, displayed dynamic change with the growth of rice plants. Two common features of nine QTLs were detected, one is no expression within 7 days after transplanting, and the other is opposite expression existed during the whole growth period. Nine QTLs largely focused on expression in certain stages, and accordingly were suggested to partition into three types, expression in prophase, both in prophase and in anaphase, and evenly during the whole stage. It may be reasonable explanation that dynamics of main effects of QTLs are likely due to gene expression selectly at certain times, while dynamics of QE interaction effects of QTLs might attribute to the subrogation of environmental factors. Examination of the association between QE interaction effect and specified environmental factors across stages may provide useful information on how an environmental factor regulates QTL expression. G. Liu and R. Zeng contributed equally to this work.     

Dynamic QTL Analysis of Rice Protein Content and Protein Index Using Recombinant Inbred Lines

Protein content (PC) and protein index (PI) play important roles in determining nutritional quality in rice (Oryza sativa L.). We used 71 lines derived from “Asominori/IR24” to analyze the developmental behavior of PC and PI through unconditional and conditional QTL mapping methods. In all, 10 unconditional QTLs and 6 conditional QTLs for PC, and 11 unconditional QTLs and 9 conditional QTLs for PI, were identified at four stages of grain filling. More were identified in the first three stages than at the final stage. Temporal patterns of gene expression for PC and PI differed over time, with several QTLs being expressed across two or three stages but many being expressed at only one stage. Some of these QTLs were closely linked with maturity QTLs reported previously. Many QTLs for PC and PI were co-localized, supporting the significant correlation found between PC and PI. Our results suggest that dynamic QTL mapping might be a valid means for revealing more genetic information about protein accumulations during seed development.     

Molecular dissection of developmental behavior of tiller number and plant height and their relationship in rice (Oryza sativa L.)

Plant height and tiller number are two important characters related to yield in rice (Oriza sativa L.). Zhenshan97 x Minghui63 recombinant inbred lines were employed to dissect the genetic basis of development of plant height and tiller number using conditional and unconditional composite interval mapping approaches. The traits were normally distributed with transgressive segregation in both directions. Increasingly negative correlations were observed between tiller number and plant height at five consecutive growth stages. A total of 23 and 24 QTL were identified for tiller number and plant height, respectively. More QTL were detected by conditional mapping than by conventional mapping. Different QTL/genes apparently controlled the traits at different developmental stages. Three genomic regions were identified as putative co-located QTL, which showed opposite additive effects on tiller number and plant height. Furthermore, in the period reaching maximum tiller number, the expression of QTL for tiller number was active, whereas that of QTL for plant height was inactive. These facts provided a possible genetic explanation for the negative correlations between the traits. The research demonstrates conditional mapping to be superior to conventional mapping for this type of research. Implications of the results for hybrid rice improvement are discussed.     

Functional mapping of quantitative trait loci associated with rice tillering

Several biologically significant parameters that are related to rice tillering are closely associated with rice grain yield. Although identification of the genes that control rice tillering and therefore influence crop yield would be valuable for rice production management and genetic improvement, these genes remain largely unidentified. In this study, we carried out functional mapping of quantitative trait loci (QTLs) for rice tillering in 129 doubled haploid lines, which were derived from a cross between IR64 and Azucena. We measured the average number of tillers in each plot at seven developmental stages and fit the growth trajectory of rice tillering with the Wang–Lan–Ding mathematical model. Four biologically meaningful parameters in this model––the potential maximum for tiller number (K), the optimum tiller time (t ₀), and the increased rate (r), or the reduced rate (c) at the time of deviation from t ₀––were our defined variables for multi-marker joint analysis under the framework of penalized maximum likelihood, as well as composite interval mapping. We detected a total of 27 QTLs that accounted for 2.49–8.54% of the total phenotypic variance. Nine common QTLs across multi-marker joint analysis and composite interval mapping showed high stability, while one QTL was environment-specific and three were epistatic. We also identified several genomic segments that are associated with multiple traits. Our results describe the genetic basis of rice tiller development, enable further marker-assisted selection in rice cultivar development, and provide useful information for rice production management.     

Dynamic QTL analysis of the Na+ content,K+ content,and Na+/K+ ratio in rice roots during the field growth under salt stress

Rice (Oryza sativa L.) is seriously impacted by global soil salinization. To determine the quantitative trait loci (QTLs) related to salt tolerance in rice roots, F_2:3 and BC₁F_2:3 populations derived from a cross between the cv. Dongnong 425 of high quality and yield and the salt-tolerant cv. Changbai 10, were studied at different development stages. Two genetic linkage maps of F_2:3 and BC1F_2:3 populations were constructed. A 66 mM NaCl solution was used to irrigate the field and to analyze the dynamic QTL of some rice root traits. Using unconditional and conditional QTL mapping methods, 30 unconditional QTLs and 16 conditional QTLs related to the 6 root traits were detected on the 9 rice chromosomes during different developmental stages. Fourteen pairs of unconditional and conditional QTLs were detected at the identical developmental stage in the identical population. A number of QTLs were detected at different developmental stages, however, many did not appear at the last stage. Remarkably, qRKC1 appeared continuously at multiple stages in both the populations suggesting its key role in regulating the salt tolerance of rice roots.     

Impact of epistasis and QTL×environment interaction on the developmental behavior of plant height in rice (Oryza sativa L.)

QTLs with epistatic effects and environmental interaction effects for the developmental behavior of plant height in rice were studied by conventional and conditional methods for quantitative trait loci (QTLs) by mapping with a doubled-haploid population of 123 lines from IR64/Azucena in three environments. The results showed that epistatic effects were important and most epistasis could be detected only by conditional QTL mapping, while most non–epistatic QTLs could be detected by both conventional and conditional methods. Many modificative QTLs showed only epistatic effects without their own additive effects at some stages. QTL×environment (QE) interaction effects were detected more often than QTL main effects for plant-height behavior, which might indicate that gene expression could be greatly affected by the environment. No QTLs had effects during the whole of ontogeny. Conditional QTL mapping might be a valid way to reveal dynamic gene expression for the development of quantitative traits, especially for epistatic effects. Received: 19 May 2000 / Accepted: 27 October 2000     

Quantitative trait loci analysis for the developmental behavior of Soybean (Glycine max L. Merr.)

Quantitative trait loci (QTLs) identified so far in soybean were mainly derived in the final stage of plant development, which did not apply to the exploitation of genetic effects that were expressed during a specific developmental stage. Thus, the aim of this study was to identify conditional QTLs associated with yield traits at a specific developmental interval of soybean plant. The 143 recombinant inbred lines developed from the cross of soybean cultivars ‘Charleston’ and ‘Dongnong 594’ were used for the developmental QTLs analysis of pod number in the main stem and plant height by composite interval mapping method combined with mixed genetic model. The results indicated that the number and type of QTLs and their genetic effects for the two agronomic traits were different in a series of measuring stages. A total of 10 unconditional QTLs in 6 linkage groups and 5 conditional QTLs in 3 linkage groups were identified for the pod number of the main stem, while 13 unconditional QTLs in 7 linkage groups and 12 conditional QTLs in 6 linkage groups were identified for plant height. Many QTLs that were detected in the early stages were different from those detected at the later stages. Some QTLs existed only at one stage and others existed across two or three stages. Five marker intervals (satt509-satt251, sat_099-sat_113, sat_113-OPAW19_4, satt457-OPC10_85, sat_095-OPBA08_5) were proven to be associated both with the development of pod number in the main stem and the development of plant height. The present study suggested that the development of pods and plant height in soybean were governed by time-dependent gene expression.     

Dynamic analysis of QTLs on tiller number in rice (Oryza sativa L.) with single segment substitution lines

Twelve single segment substitution lines (SSSLs) in rice, which contain quantitative trait loci (QTLs) for tiller number detected previously, were used to study dynamic expression of the QTLs in this study. These SSSLs and their recipient, Hua-Jing-Xian 74 (HJX74), were used to produce 78 crossing combinations first, and then these combinations and their parents were grown in two planting seasons with three cropping densities. Tiller number was measured at seven developmental stages. QTL effects including main effects (additive, dominance and epistasis), QTL?×?season and QTL?×?density interaction effects were analyzed at each measured stage. The additive, dominant and epistatic effects of the 12 QTLs as well as their interaction effects with the seasons and with the densities all display dynamic changes with the development. Eight QTLs are detected with significant additive effects and/or additive?×?season and/or additive?×?density interaction effects at least at one developmental stage, and all QTLs have significant dominant and epistatic effects and/or interaction effects involved in. For most of the QTLs dominant effects are much bigger than additive effects, showing overdominance. Each QTL interacts at least with eight other QTLs. Additive and dominant effects of these QTLs are mostly positive while epistatic effects are negative and minor. Most of the QTLs show significant interactions with planting seasons and cropping densities, but the additive effects of QTLs Tn3-1 and Tn3-2, the dominant effects of QTL Tn7 and Tn8, and the epistatic effects of 14 pairs of QTLs are stable across seasons and the dominant effect of QTL Tn3-3 and the epistatic effects of QTL pairs Tn2-1/Tn6-2, Tn2-1/Tn9 and Tn3-3/Tn6-3 are nearly consistent across cropping densities. This paper is the first report of dynamics on dominances and epistasis of QTLs for tiller number in rice and provides abundant information, which is useful to improve rice tiller number via heterosis and/or QTL pyramiding.     

Fine Mapping QTL for Drought Resistance Traits in Rice (<Emphasis Type="Italic">Oryza sativa</Emphasis> L.) Using Bulk Segregant Analysis

Drought stress is a major limitation to rice (Oryza sativa L.) yields and its stability, especially in rainfed conditions. Developing rice cultivars with inherent capacity to withstand drought stress would improve rainfed rice production. Mapping quantitative trait loci (QTLs) linked to drought resistance traits will help to develop rice cultivars suitable for water-limited environments through molecular marker-assisted selection (MAS) strategy. However, QTL mapping is usually carried out by genotyping large number of progenies, which is labour-intensive, time-consuming and cost-ineffective. Bulk segregant analysis (BSA) serves as an affordable strategy for mapping large effect QTLs by genotyping only the extreme phenotypes instead of the entire mapping population. We have previously mapped a QTL linked to leaf rolling and leaf drying in recombinant inbred (RI) lines derived from two locally adapted indica rice ecotypes viz., IR20/Nootripathu using BSA. Fine mapping the QTL will facilitate its application in MAS. BSA was done by bulking DNA of 10 drought-resistant and 12 drought-sensitive RI lines. Out of 343 rice microsatellites markers genotyped, RM8085 co-segregated among the RI lines constituting the respective bulks. RM8085 was mapped in the middle of the QTL region on chromosome 1 previously identified in these RI lines thus reducing the QTL interval from 7.9 to 3.8 cM. Further, the study showed that the region, RM212–RM302–RM8085–RM3825 on chromosome 1, harbours large effect QTLs for drought-resistance traits across several genetic backgrounds in rice. Thus, the QTL may be useful for drought resistance improvement in rice through MAS and map-based cloning.     

Genetic Relationships Among Panicle Characteristics of Rice (<Emphasis Type="Italic">Oryza sativa</Emphasis> L.) Using Unconditional and Conditional QTL Analyses

Using mixed-model-based composite interval mapping and conditional statistical methods, we studied quantitative trait loci (QTLs) with epistatic effects and QTLs by environment interaction effects for rice seed set percent (SSP), filled grain number per plant (FGP), and panicle length (PL). A population of 241 recombinant inbred lines was used which was derived from a cross between “Zhenshan 97” and “Minghui 63.” Its linkage map included 221 molecular markers. Our QTL analysis detected 28, 25, and 32 QTLs for SSP, FGP, and PL, respectively. Each QTL explained 1.37%∼13.19% of the mean phenotypic variation. A comparison of conventional and conditional mapping provided information about the genetic control system involved in the synthetic process of SSP, FGP, and PL at the level of individual QTLs. Conditional QTLs with reduced (or increased) effects were identified for SSP, which were significantly influenced by FGP or PL. Some QTLs could express independently for the given traits, thereby providing possibilities for simultaneous improvement of SSR and PL, and SSR and FGP. Epistasis was more sensitive to environmental conditions than were additive effects.     

QTL detected for grain-filling rate in maize using a RIL population

The grain-filling rate plays an important role in determining grain yield. To elucidate the genetic basis of the grain-filling rate, a set of 203 recombinant inbred lines was evaluated at two locations over 2 years. Quantitative trait loci (QTL) for grain-filling rate were detected using conditional and unconditional QTL analysis of genetic linkage maps comprising 217 SSR markers. The results showed that the grain-filling rate increased between 15 and 35 days after pollination, then decreased at the last two sampling times. Hybrids with high grain-filling rates determined the grain yield in those areas with a short growth season for maize. A total of 23 unconditional QTL for grain-filling rate were detected using the 100-kernel weight as the input data at different sampling stages. They were distributed on 10 chromosomes (except chromosome 9), and some QTL were detected at different sampling stages. In addition, nine conditional QTL were identified using the average increase in 100-kernel weight of per day between two sampling times, and six conditional QTL were detected simultaneously using the unconditional QTL mapping strategy. The QTL mapping results demonstrated that the grain-filling rate is controlled by a complex genetic mechanism, and the QTL detected at different sampling stages might be important contributors to grain yield in maize.     

Dynamic QTL analysis of linolenic acid content in different developmental stages of soybean seed

Linolenic acid (LN) in soybean (Glycine max L. Merr.) seed mainly contributes to the undesirable odors and flavors commonly associated with poor oil quality. LN deposition at various stages of soybean seed development had not been reported by 2010. The objects of this study were (1) to identify and measure quantitative trait loci (QTL) underlying LN content and (2) to estimate the QTL effects expressed from earlier seed developmental stages to drying seed of soybean. One hundred and twenty-five F_5:8 and F_5:9 recombinant inbred lines derived from the cross of soybean cultivars ‘Hefeng 25’ and ‘Dongnong L5’ were used for the identification of QTL underlying LN content from the 37 day (D) to 86D stages after flowering, at Harbin in 2008 and 2009. QTL × Environment interactions (QE) effects were evaluated using a mixed genetic model (Zhu in J Zhejiang Univ (Natural Science) 33:327–335, 1999). Twelve unconditional QTL and 12 conditional QTL associated with LN content were identified at different developmental stages. Most of the QTL explained <10% of phenotypic variation of LN content. Unconditional QTL QLNF-1, QLNC2-1, QLND1b-1, QLNA2-1 and QLNH-1 influenced LN content across different development stages and environments. Conditional QTL QLNF-1, QLNC2-1 and QLNH-1 were identified in multiple developmental stages and environments. Conditional and unconditional QTL clustered in neighboring intervals on linkage groups A2, C2 and D1b. Ten QTL with conditional additive main effects (a) and/or conditional additive × environment interaction effects (ae) at specific developmental stage were identified on nine linkage groups. Of them, six QTL only possessed additive main effects and seven QTL had significant ae effects in different developmental stages. A total of 13 epistatic pairwise QTL were identified by conditional mapping in different developmental stages. Two pairs of QTL only showed aa effects and five pairs of QTL only showed aae effects at different developmental stages. QTL with aa effects, as well as their environmental interaction effects, appeared to vary at different developmental stages.     

水稻叶绿素含量动态QTL分析

为剖析水稻不同生育时期叶绿素含量的变化动态及遗传机制,以‘Sasanishiki’（粳稻）、‘Habataki’（籼稻）及其杂交衍生的85个回交重组自交系（BILs）群体为材料,对控制水稻叶片叶绿素含量的数量性状基因位点（QTL）变化动态进行了分析。共检测到39个QTL,包括26个非条件QTL和13个条件QTL,分布在除第7和第11号染色体以外的10条染色体上,平均每个时期检测到3．25个非条件QTL。其中生育前期和生育中后期检测到的QTL位点较少,仅为1—3个;在生育中期（盛期）检测到的QTL位点相对较多,一般为4～5个。在生育期的中期和后期均能在第1和2号染色体上检测到控制叶绿素含量的QTL,并且这些QTL位点在1和2号染色体上呈现聚集现象。本研究同时发现了一些新的QTL位点,这些QTL将有助于我们更全面地了解叶绿素在不同发育时期的遗传基础。     

Identification of QTLs for Yield and Associated Traits in F₂ Population of Rice

Identification of quantitative trait loci (QTLs) controlling yield and yield-related traits in rice was performed in the F₂ mapping population derived from parental rice genotypes DHMAS and K343. A total of 30 QTLs governing nine different traits were identified using the composite interval mapping (CIM) method. Four QTLs were mapped for number of tillers per plant on chromosomes 1 (2 QTLs), 2 and 3; three QTLs for panicle number per plant on chromosomes 1 (2 QTLs) and 3; four QTLs for plant height on chromosomes 2, 4, 5 and 6; one QTL for spikelet density on chromosome 5; four QTLs for spikelet fertility percentage (SFP) on chromosomes 2, 3 and 5 (2 QTLs); two QTLs for grain length on chromosomes 1 and 8; three QTLs for grain width on chromosomes1, 3 and 8; three QTLs for 1000-grain weight (TGW) on chromosomes 1, 4 and 8 and six QTLs for yield per plant (YPP) on chromosomes 2 (3 QTLs), 4, 6 and 8. Most of the QTLs were detected on chromosome 2, so further studies on chromosome 2 could help unlock some new chapters of QTL for this cross of rice variety. Identified QTLs elucidating high phenotypic variance can be used for marker-assisted selection (MAS) breeding. Further, the exploitation of information regarding molecular markers tightly linked to QTLs governing these traits will facilitate future crop improvement strategies in rice.     

Unconditional and conditional QTL analysis of kernel weight related traits in wheat (Triticum aestivum L.) in multiple genetic backgrounds

Wheat thousand kernel weight (TKW) is a complex trait, and is largely controlled by several kernel traits, including kernel length (KL) and kernel width (KW). In order to reveal the genetic relationship between TKW and these kernel traits (KW and KL) as accurate as possible, we applied both unconditional and conditional mapping analyses to three distinct genetic populations, one DH population and two RIL populations. This report describes the identifications of 36 unconditional and conditional additive QTLs and 30 pairs of unconditional and conditional epistatic QTLs, all of which are closely associated with TKW. While the conditional additive locus Qtkw1B, detected in the RIL2 population, exhibited the largest contribution, explaining 14.12 % of TKW variance, the unconditional epistatic QTLs Qtkw3A-2/Qtkw5B.1, detected in the DH population, accounted for 11.95 % of phenotypic variance. This study also showed that, compared with unconditional mapping, conditional mapping resulted in very different numbers and different extent of effects of additive and epistatic QTLs that were associated with TKW when TKW was conditioned on kernel traits (KW and KL). These data strongly suggest that KW and KL indeed play a significant role in determining TKW. Furthermore, we demonstrated that the effects of the 25 additive QTLs for TKW were either entirely or largely determined by KW, while the effects of the other 25 additive QTLs for TKW were either entirely or largely affected by KL. We conclude that the conditional mapping can be useful for a better understanding of the interrelationship between the yield contributing traits at the QTL level.     

Kernel weight per spike: what contributes to it at the individual QTL level?

Spike length (SL), spikelet number (SPN) per spike, kernel number per spike (KNPS), and thousand-kernel weight (TKW) have strong genetic associations with kernel weight per spike (KWPS) in wheat. To investigate their genetic relationships at the individual quantitative trait locus (QTL) level, both unconditional and conditional QTL mapping for KWPS with respect to SL, SPN, KNPS, and TKW were conducted. Two related F_8:9 recombinant inbred line populations, comprising 485 and 229 lines, respectively, were used. The trait phenotypic performances of each population were evaluated in four different environments. Unconditional QTL mapping analysis identified 22 putative additive QTL for KWPS, five of which were stable QTL, and only QKwps-WJ-1B.2 showed significant additive-by-environment interaction effects. In comparison with unconditional QTL mapping analysis, conditional QTL mapping analysis indicated that, at the QTL level, KNPS and TKW contributed more to KWPS than did SL and SPN. Any unconditional QTL for KWPS detected in this study was associated with at least one of its four related traits. The present study will provide assistance in the understanding of the genetic relationships between KWPS and its related traits.     

Genomic regions affecting seed shattering and seed dormancy in rice

Non-shattering of the seeds and reduced seed dormancy were selected consciously and unconsciously during the domestication of rice, as in other cereals. Both traits are quantitative and their genetic bases are not fully elucidated, though several genes with relatively large effects have been identified. In the present study, we attempted to detect genomic regions associated with shattering and dormancy using 125 recombinant inbred lines obtained from a cross between cultivated and wild rice strains. A total of 147 markers were mapped on 12 rice chromosomes, and QTL analysis was performed by simple interval mapping and composite interval mapping. For seed shattering, two methods revealed the same four QTLs. On the other hand, for seed dormancy a number of QTLs were estimated by the two methods. Based on the results obtained with the intact and de-hulled seeds, QTLs affecting hull-imposed dormancy and kernel dormancy, respectively, were estimated. Some QTLs detected by simple interval mapping were not significant by composite interval mapping, which reduces the effects of residual variation due to the genetic background. Several chromosomal regions where shattering QTLs and dormancy QTLs are linked with each other were found. This redundancy of QTL associations was explained by ”multifactorial linkages” followed by natural selection favoring these two co-adapted traits. Received: 23 November 1998 / Accepted: 27 August 1999     

Molecular mapping and location of QTLs for drought-resistance traits in <Emphasis Type="Italic">indica</Emphasis> rice (<Emphasis Type="Italic">Oryza sativa</Emphasis> L.) lines adapted to target environments

Drought is a major limitation for rice production in rainfed ecosystems. Identifying quantitative trait loci (QTLs) linked to drought resistance provides opportunity to breed high yielding rice varieties suitable for drought-prone areas. Although considerable efforts were made in mapping QTLs associated with drought-resistance traits in rice, most of the studies involved indica × japonica crosses and hence, the drought-resistance alleles were contributed mostly by japonica ecotypes. It is desirable to look for genetic variation within indica ecotypes adapted to target environment (TE) as the alleles from japonica ecotype may not be expressed under lowland conditions. A subset of 250 recombinant inbred lines (RILs) of F₈ generation derived from two indica rice lines (IR20 and Nootripathu) with contrasting drought-resistance traits were used to map the QTLs for morpho-physiological and plant production traits under drought stress in the field in TE. A genetic linkage map was constructed using 101 polymorphic PCR-based markers distributed over the 12 chromosomes covering a total length of 1,529 cM in 17 linkage groups with an average distance of 15.1 cM. Composite interval mapping analysis identified 22 QTLs, which individually explained 4.8–32.2% of the phenotypic variation. Consistent QTLs for drought-resistance traits were detected using locally adapted indica ecotypes, which may be useful for rainfed rice improvement.     

Detection of QTLs with additive effects and additive-by-environment interaction effects on panicle number in rice (<Emphasis Type="Italic">Oryza sativa</Emphasis> L.) with single-segment substitution lines

A novel population consisting of 35 single-segment substitution lines (SSSLs) originating from crosses between the recipient parent, Hua-jing-xian 74 (HJX74), and 17 donor parents was evaluated in six cropping season environments to reveal the genetic basis of genetic main effect (G) and genotype-by-environment interaction effect (GE) for panicle number (PN) in rice. Subsets of lines were grown in up to six environments. An indirect analysis method was applied, in which the total genetic effect was first partitioned into G and GE by using the mixed linear-model approach, and then QTL (quantitative trait locus) analyses on these effects were conducted separately. At least 18 QTLs for PN in rice were detected and identified on 9 of 12 rice chromosomes. A single QTL effect (a + ae) ranging from −1.5 to 1.2 was divided into two components, additive effect (a) and additive × environment interaction effect (ae). A total number of 9 and 16 QTLs were identified with a ranging from −0.4 to 0.6 and ae ranging from −1.0 to 0.6, respectively, the former being stable but the latter unstable across environments. Three types of QTLs were suggested according to their effects expressed. Two QTLs (Pn-1b and Pn-6d) expressed stably across environments due to the association with only a, nine QTLs (Pn-1a, Pn-3c, Pn-3d, Pn-4, Pn-6a, Pn-6b, Pn-8, Pn-9 and Pn-12) with only ae were unstable, and the remaining seven of QTLs were identified with both a and ae, which also were unstable across environments. This is the first report on the detection of QE (QTL-by-environment interaction effect) of QTLs with SSSLs. Our results illustrate the efficiency of characterizing QTLs and analyzing action of QTLs through SSSLs, and further demonstrate that QE is an important property of many QTLs. Information provided in this paper could be used in the application of marker-assisted selection to manipulate PN in rice. Guifu Liu and Zemin Zhang contributed equally to this work.