Phenotypic variation among IR64 x TOG5681 rice (Oryza sativa x O. glaberrima) chromosome segment substitution lines (CSSL) in response to iron toxicity, and its associated QTLs

doi:10.21203/rs.3.rs-4842739/v1

Download PDF

Research Article

Phenotypic variation among IR64 x TOG5681 rice (Oryza sativa x O. glaberrima) chromosome segment substitution lines (CSSL) in response to iron toxicity, and its associated QTLs

https://doi.org/10.21203/rs.3.rs-4842739/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Iron (Fe) toxicity presents a significant challenge to rice production in lowland ecosystems globally. The identification of genetic factors responsible for Fe toxicity tolerance is crucial for the development of tolerant rice varieties. This study aimed to unravel the genetic bases of Fe toxicity tolerance using quantitative trait locus (QTL) mapping. We conducted phenotypic evaluations for Fe toxicity tolerance on 54 chromosome segment substitution lines (CSSLs) obtained from a cross between the moderately susceptible IR64 (O. sativa) and the tolerant donor TOG5681 (O. glaberrima) under Fe toxicity stress. QTL analysis was performed using agro-morphological traits and microsatellite genotypic data. High heritability estimates were observed for key traits such as leaf bronzing score (LBS) and grain yield. Several loci associated with agronomic traits, including plant height (qPH2.1), panicle number (qPN4.1), grain weight (qGW4.1), harvest index (qHI4.1), maturity (qMat6.1) and shoot weight (qSW6.1 and qSW11.1), were identified. Additionally, a major QTL (qLBS11.1) linked to leaf bronzing (LBS) was identified, housing the OsbHLHq11 gene involved in iron homeostasis regulation. A comparative analysis revealed colocation with previously reported QTLs, validating their significance. OsbHLHq11 was found to be conserved across diverse rice germplasm, including lowland NERICA (NEw RICe for Africa) varieties, which were developed through crossing O. sativaand O. glaberrima. The discovery of qLBS11.1 and the candidate gene OsbHLHq11 offers insights into the genetic mechanisms governing Fe toxicity tolerance, highlighting potential targets for breeding tolerant rice varieties using marker-assisted selection or genetic engineering strategies.

Iron toxicity

Rice

Quantitative trait locus

OsbHLHq11

Tolerance

Breeding

Iron toxicity is a complex disorder that severely limits lowland rice production in West Africa. After drought, it is the second most important abiotic stress in rice-cultivated areas in Africa (van Oort 2018). The occurrence and severity of iron toxicity depend on several environmental factors, including soil type, soil fertility, and water management. Leaf bronzing score (LBS), based on visual assessment of rice plants, has been widely used as a selection criterion for iron-toxicity tolerance. This is supported by the findings of Audebert and Fofana (2009) and Sikirou et al. (2016), who estimated that for every unit increase in LBS, rice grain yield is reduced by 390 kg ha^-1 and 20% in field and pot experiments, respectively.

Iron-toxicity tolerance is a polygenic trait, which improvement, so far, has not benefited from the application of molecular markers despite the report of several QTLs, including for leaf bronzing (reviewed by Sikirou et al. 2015; Onaga et al. 2016). Most of these QTLs were identified at the vegetative stage using hydroponic screening. To the best of our knowledge, only one study has reported QTLs for yield and yield components under natural iron toxicity conditions (Dufey et al. 2012). Inconsistencies between genotype ranking in hydroponics and natural field conditions have been reported (Becker and Asch 2005; Nozoe et al. 2008; Wang et al. 2008). Therefore, it is important to conduct QTL mapping studies in the target environment. However, field screening can sometimes be highly variable, even within the same plot (Gridley et al. 2006). A good alternative to field screening is pot screening using hotspot soil, which has been shown to be a reliable screening method (Sikirou et al. 2016).

Several studies have shown the high potential of Oryza glaberrima as a source of iron toxicity tolerance. Sahrawat and Sika (2002) demonstrated that CG14, the O. glaberrima parent of the upland NERICA (NEw Rice for Africa) varieties, has a strong resistance to iron toxicity. A study by De Dorlodot et al. (2005) and Majerus et al. (2007) found that an interspecific line (O. sativa x O. glaberrima) and TOG7105, an O. glaberrima accession, can store high concentrations of iron in their tissues without suffering major damage. Kang et al. (2011) identified three NERICA lines that use an avoidance strategy to take up less iron in conditions of iron toxicity. More recently, Sikirou et al. (2018) identified O. glaberrima accessions that outperformed O. sativa tolerant checks used in their study. Despite these findings, only one in-field QTL study for iron toxicity tolerance has been reported to date, using an interspecific O. sativa x O. glaberrima population (Dufey et al. 2015a). This study revealed new iron toxicity tolerance QTLs compared to earlier studies using O. sativa x O. sativa populations. However, the population used was made up of rice varieties adapted to upland conditions, where iron toxicity is very unlikely.

An interspecific population of chromosome segment substitution lines (CSSLs) derived from the lowland rice varieties IR64 and TOG5681 was created as part of the research project “Exploring Natural Genetic Variation: Developing Genomic Resources and Introgression Lines for Four AA Genome Rice Relatives”, which was funded by the Generation Challenge Program (GCP) and led by CIAT (Cali, Colombia) and IRD (Montpellier, France). Coincidentally, IR64 (O. sativa) and TOG5681 (O. glaberrima) are also the parents of the lowland NERICA varieties, among which, good sources of tolerance to iron toxicity have been reported (Dramé et al. 2010; Dufey et al. 2012). CSSLs have the particular advantage of carrying short genomic introgressions of the donor parent's genome, which facilitates quick identification of QTLs (Ali et al. 2010; Gutiérrez et al. 2010). These properties have triggered our interest in using the IR64 x TOG5681 CSSL population for genetic mapping of iron toxicity tolerance in rice.

In this study, we assessed the phenotypic variation in the IR64 x TOG5681 CSSL population for iron toxicity tolerance during the reproductive stage. We also mapped the QTLs that control agro-morphological traits under iron toxicity conditions. Stable QTLs were found, and promising lines were identified for further adaptability tests in the target environments of West Africa.

Plant Materials

A total of fifty-four chromosome segment substitution lines (CSSLs) derived from an interspecific cross between O. sativa (IR64) and O. glaberrima (TOG5681) were used in this study along with their parents. They were selected from a population of 353 lines (BC2F4/BC3F3 families), of which 61 of the lines covered together 95% of the TOG5681 genome (Lorieux and Tohme 2010). IR64 is an elite O. indica variety reported as moderately susceptible to iron toxicity (Dufey et al. 2009, 2012). TOG5681 showed consistent tolerance to iron toxicity across different environments (Sikirou et al. 2018).

Screening for Iron Toxicity Tolerance

CSSLs (BC3F7) and their parents were evaluated in pot experiments from August to December 2009 and March to July 2010 at the AfricaRice research station in Cotonou, Benin (06°25'N, 02°19'E, and 15m ASL). Pots were filled with 5 kg of soil collected from Niaouli, an iron toxicity hotspot site in Southern Benin (6°44'N, 2°08'25''E). The composite soil sample collected at the Niaouli site reported an extractable iron concentration of 944.6 ppm and a pH (water) of 5.14 (Hodomihou et al. 2012). Before pot filling, the collected soil was air-dried and sieved at 4 mm. The soil in the pots was then flooded for 3 weeks to allow the release of ferrous iron before transplanting. The experiments were set up as a randomized complete block design with three replications. Rice seeds were pre-germinated on humidified filter paper in Petri dishes, and two uniform seedlings were transplanted per pot. Fertilizer was applied as 1g of NPK 15-15-15 at transplanting (basal) and 40 days after transplanting (topdressing). The water level was maintained at 3 cm above the soil level throughout the growth period, and plants were grown outdoors.

Leaf bronzing at 65 and 76 days after sowing (DAS) was visually assessed following the standard evaluation system (SES IRRI, 1996). Plant height was measured at harvest and shoot biomass was determined as shoot dry weight after drying the straws at 74°C for 3 days. Grain yield was determined as filled grain weight per plant at 14% grain moisture content. Harvest index was calculated as the ratio between grain weight and shoot dry weight.

Phenotypic data were analyzed using Genstat software v4.2 (VSN International Ltd. 2003) and BMS v3.0.8 (VSN International Limited 2015). A one-way analysis of variance (ANOVA) was performed on all traits following a randomized block design model. Least squares means of rice lines within experiment were generated for each trait. Variance components for each trait were obtained using the REML option of the VARCOMP procedure. For this analysis, all factors were considered as random effects. Broad-sense heritability (H) for each trait was calculated as:

$$\:\text{H=}\frac{{\text{σ}}_{\text{G}}^{\text{2}}}{{\text{σ}}_{\text{G}}^{\text{2}}\text{+\:}\frac{{\text{σ}}_{\text{E}}^{\text{2}}}{\text{r}}}$$

where σ²_G is the genetic variance, σ²_E is the error variance, and r the number of replications.

Genetic correlations between two traits measured in the same environment within the same trial were calculated as follows (Bernardo 2010):

$$\:{\text{r}}_{\text{G12}}\text{=}\frac{{\text{Cov}}_{\text{12}}}{\sqrt{{\text{σ}}_{\text{G1}}^{\text{2}}\text{\:x\:}{\text{σ}}_{\text{G2}}^{\text{2}}}}$$

where $\:{r}_{G12},{Cov}_{12}$, $\:{\sigma\:}_{G1}^{2}\:\text{a}\text{n}\text{d}\:{\sigma\:}_{G2}^{2}$ are the genetic correlation coefficient between traits 1 and 2 within the same experiment, genetic covariance of traits 1 and 2, and the genotypic variances of traits 1 and 2, respectively.

QTL mapping

Genotypic data of the CSSL population at BC3F3 stage was obtained from the IRD-CIAT Rice Genetics & Genomics Laboratory, CIAT, Colombia. One hundred fifty-two (152) SSR markers from the universal rice core map (Orjuela et al. 2010), polymorphic between IR64 and TOG5681, were used to genotype the population. Following the genotyping, CSSLs' alleles like those of IR64 were encoded as A, alleles similar to those of TOG5681 were encoded as B, heterozygous sites were encoded as H and unknown and non-parental alleles were encoded as N.

Prior to the QTL analysis, a Kolmogorov-Smirnov normality test was conducted for each evaluated trait. Traits not corresponding to a normal distribution were transformed using logarithmic functions. QTL analysis was carried out using QGene v4.3.10 (Joehanes and Nelson 2008) through composite interval mapping (CIM) with automatic cofactor selection. Because calculating a genetic map using CSSLs produces unreliable recombination frequency estimates, the genetic map was rather extracted from the universal core map publication (Orjuela et al. 2010). The LOD threshold for identifying significant QTLs was determined after 1,000 iterations of permutations (∝ = 0.05). Additionally, parameters such as the additive effect and the phenotypic variance explained were estimated. Each identified QTL was named following the standard rice QTL nomenclature (McCouch 2008), beginning with "q" (short for QTL), followed by the trait name, rice chromosome number, a period ".", and a unique identifier (e.g., qGW4.1 for the first grain weight QTL on chromosome 4). The identified QTLs were compared with reported QTLs for iron toxicity tolerance. Marker positions (in Mbp) of the previously reported QTLs were obtained on Gramene (https://www.gramene.org/) using the 2009 annotated Nipponbare genome (IRGSP v1).

Gene ontology analysis

Genes present in the LBS QTL region were extracted and a gene set enrichment analysis was conducted with ClusterProfiler package (Yu et al. 2012) with a p-value and q-value cut-off of 0.05 and 0.1, respectively.

Phenotypic variation under iron toxicity

All recorded traits were assessed for variability across genotype, year, and their interaction. Analysis of variance indicated a significant effect of both genotype and year factors for all traits (Table 1). Furthermore, a significant interaction between genotype and year was observed for all traits except LBS at 76 DAS.

The calculated coefficients of variation (CV) for the traits showed a notable degree of genetic variation among the CSSLs. The highest CV was observed for grain weight (29.6 in experiment 2 and 26.9 in experiment 1), while the lowest CV was recorded for days to maturity (1.7 in experiment 1 and 7.2 in experiment 2) (Table 2). Heritability was relatively high in both experiments, ranging from 0.49 to 0.96 in experiment 1 and from 0.51 to 0.92 in experiment 2 (Table 2). Clear symptoms of iron toxicity were observed in plants, with LBS ranging from 2 to 5 in experiment 1 and from 3 to 6 in experiment 2 (Table 2).

Table 1

Effect of genotype, season (year) and their interaction, on the variation of the different traits measured in this study and the average value of the traits across all the genotypes tested.
Source of variation	Df	LBS^$ 65DAS^&	LBS 76DAS	Plant Height (cm)		Maturity (DAS^&)	Shoot weight (g/pot)	Grain weight (g/pot)	Harvest index
Genotype (G)	55	< 0.001	< 0.05	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
Year (Y)	1	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
G x Y	55	< 0.05	0.22ns	< 0.01	< 0.05	< 0.001	< 0.001	< 0.001	< 0.001
Grand Mean		4	4	73	14	129	20.5	7.3	0.36
SE		0.3	0.3	2.18	2.36	2.85	3.01	1.62	0.03
^$LBS: Leaf bronzing score; ^&DAS: Days after sowing; ns: not significant

Overall, the O. glaberrima parent, TOG5681, exhibited lower LBS (2 in experiment 1), greater plant height (82 cm in experiment 1), and earlier maturity compared with the O. sativa parent, IR64. Values for other traits fell within the same range for both parents, with a slight advantage for IR64 in terms of grain weight. Across all measured traits, transgressive segregation was evident within the CSSLs (Table 2). Some lines displayed more severe leaf bronzing than the least performing parent (up to a score of 6), while others yielded up to 20% and 31% more than the highest yielding parent in experiments 1 and 2, respectively. Comparing experiments revealed delayed maturity but higher grain weight in experiment 1.

Table 2

Heritability estimates of the different traits measured and their phenotypic variation among the CSSL lines and their parents.
Traits	Experiment	Heritability	CV (%)	Parents		CSSLs
Traits	Experiment	Heritability	CV (%)	IR 64	TOG5681	Average	Range
LBS 65DAS	1	0.70	17.5	3	2	3	2–5
LBS 65DAS	2	0.63	20.2	4	3	4	3–6
LBS 76DAS	1	0.71	21.7	3	2	3	2–5
LBS 76DAS	2	0.51	17.6	4	4	4	3–5
Plant height (cm)	1	0.88	10.7	71	82	71	59–99
Plant height (cm)	2	0.92	11.6	70	76	74	57–106
Panicle number	1	0.49	16.2	12.2	12.3	13	10–17
Panicle number	2	0.67	19.9	12.3	13	14	11–20
Maturity (days)	1	0.96	1.7	134	119	134	119–139
Maturity (days)	2	0.70	7.2	123	111	124	116–139
Shoot weight (g plant^-1)	1	0.55	14.8	21.6	21.8	21	19–24
Shoot weight (g plant^-1)	2	0.72	19.5	16.9	18.4	20	16–26
Grain weight (g plant^-1)	1	0.76	26.9	9.1	8.8	7.8	4.6–10.9
Grain weight (g plant^-1)	2	0.63	29.6	6.7	6.0	6.8	4.2–8.8
Harvest index	1	0.86	17.5	0.44	0.41	0.37	0.21–0.47
Harvest index	2	0.83	24.9	0.42	0.33	0.34	0.17–0.48
LBS: Leaf bronzing score; DAS: Days after sowing

In experiments 1 and 2, a significant and positive correlation was found between LBS at 65 DAS and maturity, shoot weight, panicle number, grain weight and harvest index (Table 3). The study revealed a negative genetic correlation between grain weight and both LBS at 65 and 76 DAS, between maturity and shoot weight, and also between maturity and grain weight (Table 3). In experiment 1, LBS at 76 DAS was positively correlated to maturity, while a negative correlation was found with shoot weight and harvest index.

Table 3

Genetic correlations between the different traits measured in this study.
Traits	LBS 65DAS	LBS 76DAS	Plant height	Panicle number	Maturity	Shoot weight	Grain weight	Harvest index
LBS 65DAS	1	0.45**	0.11ns	-0.15ns	0.26*	-0.1ns	-0.29*	-0.24ns
LBS 76DAS	0.69**	1	-0.00ns	-0.19ns	-0.03ns	-0.25ns	-0.28*	-0.12ns
Plant height	-0.08ns	-0.27*	1	-0.12ns	0.01ns	0.51**	0.07ns	-0.19ns
Panicle number	-0.19ns	-0.15ns	-0.26*	1	0.24ns	0.35**	-0.12ns	-0.34**
Maturity	0.41**	0.39**	-0.19ns	-0.02ns	1	0.33*	-0.48**	-0.63**
Shoot weight	-0.25ns	-0.51**	0.21ns	0.42**	-0.16ns	1	0.22ns	-0.34**
Grain weight	-0.43**	-0.61**	-0.03ns	0.34**	-0.26*	0.80**	1	0.83**
Harvest index	-0.44**	-0.53**	-0.17ns	0.17ns	-0.28*	0.47**	0.90**	1

Values below the 1:1 line correspond to the correlations obtained for experiment 1 and values above the 1:1 line correspond to correlations obtained for experiment 2.

ns: not significant; *: significant at P < 0.05; **: significant at P < 0.01

QTLs associated with trait variation in response to iron toxicity

Given the significant interaction between genotype and year, separate QTL analyses were conducted for experiments 1 and 2. A total of nine QTLs with LOD scores exceeding the 3.0 threshold value were identified, each accounting for a phenotypic variance ranging from 22.3–39.2% (Table 4).

In experiment 1, four QTLs were pinpointed on chromosomes 2 and 4, specifically for plant height, panicle number, grain weight and harvest index (Table 4). Notably, the same QTLs for plant height and grain weight were also detected in experiment 2. Additionally, five QTLs localized on chromosomes 6 and 11 were discovered in experiment 2, pertaining to harvest index, maturity, shoot weight, and LBS at 65 DAS.

Both parental lines made additive contributions to the identified QTLs (Table 4). IR64 contributed favorable alleles for QTLs on chromosome 4 associated with grain weight in both experiments 1 and 2, and for panicle number and harvest index in experiment 1. IR64 also contributed favorably to QTLs on chromosomes 6 and 11, related to harvest index and shoot weight in experiment 2. On the other hand, TOG5681 contributed favorable alleles for QTLs on chromosome 2 linked to plant height in both experiments 1 and 2, and for QTLs on chromosomes 6 and 11 associated with maturity, shoot weight, and LBS in experiment 2 (Table 4).

Table 4

QTLs mapped in the CSSL population derived from IR64 x TOG5681 in both experiments and their genetic parameters.
Trait	Locus name	Chr.	Position^a	Marker interval	LOD Score		PVE (%)		Additive effect^b
Trait	Locus name	Chr.	Position^a	Marker interval	Exp1	Exp2	Exp1	Exp2	Exp1	Exp2
Log(Plant height)	qPH2.1	2	24.1	RM262 - RM221	6.1	4.7	39.2	32.1	-0.04	-0.03
Panicle number	qPN4.1	4	8.1	RM261 - RM185	3.1	-	22.3		1.53	-
Grain weight	qGW4.1	4	28.1	RM252 - RM348	4.7*	3.9	32.3	27.4	1.29	0.85
Harvest index	qHI4.1	4	28.1	RM252 - RM348	3.4	-	24.5		0.05	-
Harvest index	qHI6.1	6	12.1	RM136 - RM20071	-	3.7		26.2	-	0.06
Maturity	qMat6.1	6	6.1	RM204 -RM3183	-	4.8*	-	32.8	-	-4.94
Shoot weight	qSW6.1	6	2.1	RM7158 - RM253	-	5.8*	-	37.7	-	-1.14
Shoot weight	qSW11.1	11	12.3	RM202 - RM21	-	5.2*	-	34.7	-	2.09
Log(LBS 65DAS)	qLBS11.1	11	20.3	RM21 - RM7443	-	4.2	-	28.9	-	-0.10
Chr: Chromosome number; Exp: Experiment; PVE: Percentage of variation explained
^aPosition on the chromosome corresponding to the peak LOD
^bA positive additive effect value indicates that the allele is contributed by IR64
^*Significant QTLs detected by the permutation test (α = 0.05)

Colocation of the QTLs identified with other QTLs and functional rice genes

Comparative analysis with reported QTLs showed that all the QTLs identified in this study are co-localized with QTLs that have already been found related to iron toxicity tolerance (Fig. 1).

The plant height QTL (qPH2.1) co-localized with three previously detected QTLs, including qSh2 (shoot height), qSwc2 (shoot water content), and qSfw2 (shoot fresh weight), found by Zhang et al. (2017). On chromosome 4, the panicle number QTL qPN4.1 found in this study was positioned between two iron toxicity QTLs (qFETOX-4-1 and qFETOX-4-2) reported by Wu et al. (2014) (Fig. 2). QTLs including qSW6.1, qMat6.1, and qHI6.1 from the present study were found clustered along with several agronomic traits QTLs related to shoot height (qSh6), shoot fresh weight (qSfw6), shoot dry weight (qSdw6), root dry weight (qRdw6), and shoot Fe concentration (qSFe6) (Zhang et al. 2017). Notably, the leaf bronzing score related QTL was found to be close to the gene Os11G0587600, an arsenic responsive gene (Singh et al. 2017), and to other agronomic traits QTLs, including shoot fresh weight (qSfw11), shoot dry weight (qSdw11), root dry weight (qRdw11), and shoot water content (qSwc11) identified by Zhang et al. (2017) (Fig. 1).

In-depth comparison of the marker positions with those used by Dufey et al. (2015b) and Fukuda et al. (2012) revealed a colocalization of loci previously detected for both agronomic and iron toxicity-related traits. Notably, detected QTLs for harvest index (qHI4.1), grain weight (qGW4.1), panicle number (qPN4.1), shoot weight (qSW11.1), and leaf bronzing score (qLBS11.1) share the same genomic regions with the genetic maps constructed by Dufey et al. (2015b) and Fukuda et al. (2012).

The iron toxicity QTL qLBS11.1 falls into a genomic region covering 438 genes listed in supplementary Table S1. Among them, 25 genes were predominantly enriched in disease-resistance genes (Supplementary Table S2 and Table S3). We detected the gene Os11g0601700 (also known as OsbHLHq11 or OsbHLH061), which belongs to the basic helix-loop-helix (bHLH) transcription factor family and plays a key role in Fe homeostasis regulation (Supplementary Table S1) (Wang et al. 2013, 2022). Taking advantage of the rice pan-genome database (https://cgm.sjtu.edu.cn/3kricedb/, accessed December 17, 2023), OsbHLHq11 appears to be a core gene of 3009 worldwide rice accessions, including NERICA genotypes developed by AfricaRice (NERICA 1, NERICA 2, NERICA 8, NERICA 9, NERICA-L-1, NERICA-L-27, and NERICA-L-20) (Supplementary Table S4).

Iron toxicity poses a significant challenge to rice production in various global regions, particularly in lowland rice ecosystems (van Oort 2018). The excessive accumulation of iron in plant tissues can lead to the formation of reactive oxygen species, triggering oxidative stress and ultimately causing reduced growth, yield losses, and even plant mortality (Sun et al. 2016). Understanding the genetic mechanisms that govern iron toxicity tolerance is crucial for developing rice varieties resilient to this stress. Quantitative trait locus (QTL) mapping offers a powerful method to unravel the complex genetic basis of iron toxicity tolerance in rice. Through molecular markers and segregating populations, QTL mapping identified genomic regions housing loci responsible for observed variations in iron toxicity tolerance (Dufey et al. 2009, 2012, 2015b; Wu et al. 2014; Zhang et al. 2017).

This study evaluated phenotypic variation for iron toxicity tolerance during the reproductive stage across 54 chromosome segment substitution lines (CSSLs) derived from a cross between the moderately susceptible cultivar IR 64 and the tolerant donor TOG 5681. The severity of iron stress was evident, with the susceptible lines exhibiting an average leaf bronzing score (LBS) of 6 on a 0–9 scale. Notably, LBS was higher in experiment 2 compared to experiment 1, paralleled by a decline in grain weight. This observation aligns with previous studies (Sikirou et al. 2016) that noted similar trends in pot screening studies using soil-based methods, estimating yield reductions of about 20% per unit of Fe-toxic soils.

Consistent with earlier findings, this study found a significant negative correlation between LBS and grain weight, supporting reports by Audebert and Fofana (2009), Audebert and Sahrawat (2000), Elec et al. (2013), Onaga et al. (2013), and Sikirou et al. (2016). Additionally, a negative correlation was observed between grain weight and maturity, indicating the impact of the plant's growth cycle on grain weight under iron toxicity conditions. Conversely, a strong positive correlation was found between grain weight and harvest index, along with a significant but not very strong correlation between LBS at 65 days after sowing (DAS) and maturity. These results emphasize the complex interplay among agro-morphological traits in response to iron toxicity stress.

The study demonstrated high heritability for all evaluated traits across both experiments, indicating consistent and reliable results. Particularly noteworthy were the heritability values for LBS at 65 DAS and grain yield, which were comparable to LBS at 76 DAS. These findings align well with previous reports and suggest that LBS is a reliable predictor of yield in rice breeding programs targeting iron toxicity resistance (Audebert and Fofana, 2009; Sikirou et al. 2016).

Comparative analyses with previously reported QTLs revealed co-localization of identified QTLs with genomic regions associated with iron toxicity tolerance, confirming their potential importance across diverse genetic backgrounds and populations. Notably, QTLs associated with key reproductive stage traits like panicle number, grain weight, and harvest index were identified in the CSSLs population, making them promising targets for further exploration in breeding programs. Although our study did not identify novel QTL regions beyond those reported in previous studies, the presence of O. glaberrima alleles within these regions introduces new allelic diversity. This novel allelic variation is valuable for breeding programs, potentially enhancing iron toxicity tolerance across various genetic backgrounds and populations.

The identification of a major QTL, qLBS11.1, linked to iron toxicity tolerance, led to the discovery of the OsbHLHq11 gene within this QTL’s region. OsbHLHq11, encoding a basic helix-loop-helix transcription factor involved in iron homeostasis regulation (Wang et al. 2013, 2022), emerged as a promising candidate gene underlying qLBS11.1, which may play a crucial role in modulating iron toxicity tolerance in rice. The widespread presence of OsbHLHq11 across diverse rice accessions, including lowland NERICA varieties, underscores its potential significance for rice cultivation in iron-toxic environments. Further investigations are required to confirm its functional role in regulating iron toxicity.

While the responses from IR64 and TOG5681 were not very contrasting under the experimental conditions of this study, additional experiments at higher stress intensity may be considered to ascertain the findings. Besides, Sikirou et al. (2018) identified highly tolerant O. glaberrima accessions that could be utilized in genetic mapping studies to yield even more robust and impactful QTLs. Overall, this study provides valuable insights into the genetic control of iron toxicity tolerance in rice, offering potential targets for marker-assisted breeding and crop improvement programs focused on developing tolerant rice varieties.

Funding: This research was funded by the Generation Challenge Program (GCP) through the project G3005.10 (Competitive grants 2005-2007 Project # 10) entitled “Exploring Natural Genetic Variation: Developing Genomic Resources and Introgression Lines for Four AA Genome Rice Relatives” and by the Bill and Melinda Gates Foundation through the "Stress Tolerant Rice for Africa and South Asia” (STRASA) project, grant number OPP1088843.

Data Availability Statement: Data are provided with this article and supplementary materials.

Supplementary Materials: Table S1: List of genes in the LBS QTL region found in this study (qLBS11.1); Table S2: GO enrichment results of genes in the LBS QTL region found in this study (qLBS11.1); Table S3: Resistance genes list located in the LBS QTL region found in this study (qLBS11.1); Table S4: Gene presence frequency of Os11g0601700 in rice pangenome; Table S5: List of SSR markers used in this study.

Author Contributions: KND and KS designed the study; ML provided the CSSL population and genotypic data;KND, MS, KS and RB performed the analyses; KND and MS wrote the original draft which was further improved by KS, RB and ML.

Acknowledgments: The authors would like to thank Marie Noelle Ndjiondjop for her guidance as well as the technical staff of Kazuki Saito who provided support in setting up the experiment and collecting data. The authors are also grateful to the donors who funded this study.

Conflicts of interest: The authors declare no conflict of interest.

Ali ML, Sanchez PL, Yu S bin, et al (2010) Chromosome segment substitution lines: A powerful tool for the introgression of valuable genes from Oryza wild species into cultivated rice (O. sativa). Rice 3:218–234. https://doi.org/10.1007/s12284-010-9058-3
Audebert A, Fofana M (2009) Rice yield gap due to iron toxicity in West Africa. J Agron Crop Sci 195:66–76. https://doi.org/10.1111/j.1439-037X.2008.00339.x
Becker M, Asch F (2005) Iron toxicity in rice—conditions and management concepts. J Plant Nutr Soil Sci 168:558–573. https://doi.org/10.1002/jpln.200520504
Bernardo R (2010) Variation in breeding populations. In: Woodburn, M.N. (Ed.) SP (ed) Breeding for Quantitative Traits in Plants. pp 115–143
De Dorlodot S, Lutts S, Bertin P (2005) Effects of ferrous iron toxicity on the growth and mineral composition of an interspecific rice. J Plant Nutr 28:1–20. https://doi.org/10.1081/PLN-200042144
Dramé KN, Saito K, Kone B, et al (2010) Coping with iron toxicity in the lowlands of sub-Saharan Africa: Experience from Africa Rice Center. In: Innovation and partnerships to realize Africa’s Rice potential. Proc 2nd Africa Rice Congress, Bamako, Mali, 22-26 March 2010. Africa Rice, Cotonou, Benin, pp 1.9.1-1.9.8
Dufey I, Draye X, Lutts S, et al (2015a) Novel QTLs in an interspecific backcross Oryza sativa × Oryza glaberrima for resistance to iron toxicity in rice. Euphytica 204:609–625. https://doi.org/10.1007/s10681-014-1342-7
Dufey I, Hakizimana P, Draye X, et al (2009) QTL mapping for biomass and physiological parameters linked to resistance mechanisms to ferrous iron toxicity in rice. Euphytica 167:143–160. https://doi.org/10.1007/s10681-008-9870-7
Dufey I, Hiel MP, Hakizimana P, et al (2012) Multienvironment quantitative trait loci mapping and consistency across environments of resistance mechanisms to ferrous iron toxicity in rice. Crop Sci 52:539–550. https://doi.org/10.2135/cropsci2009.09.0544
Dufey I, Mathieu AS, Draye X, et al (2015b) Construction of an integrated map through comparative studies allows the identification of candidate regions for resistance to ferrous iron toxicity in rice. Euphytica 203:59–69. https://doi.org/10.1007/s10681-014-1255-5
Fukuda A, Shiratsuchi H, Fukushima A, et al (2012) Detection of Chromosomal Regions Affecting Iron Concentration in Rice Shoots Subjected to Excess Ferrous Iron Using Chromosomal Segment Substitution Lines between Japonica and Indica. Plant Prod Sci 15:183–191. https://doi.org/10.1626/pps.15.183
Gridley HE, Efisue A, Tolou B, Bakayako T (2006) Breeding for tolerance to iron toxicity at WARDA. In: Iron toxicity in rice-based systems in West Africa. WARDA, Cotonou, pp 96–111
Gutiérrez AG, Carabalí SJ, Giraldo OX, et al (2010) Identification of a Rice stripe necrosis virus resistance locus and yield component QTLs using Oryza sativa × O. glaberrima introgression lines. BMC Plant Biol 10:. https://doi.org/10.1186/1471-2229-10-6
Kang D-J, Futakuchi K, Seo Y-J, et al (2011) Relationship of Fe-tolerance to morphological changes in roots in upland NERICA lines under Fe-treated hydroponic conditions. J Crop Sci Biotechnol 14:311–315. https://doi.org/10.1007/s12892-011-0021-7
Majerus V, Bertin P, Swenden V, et al (2007) Organ-dependent responses of the african rice to short-term iron toxicity: Ferritin regulation and antioxidative responses. Biol Plant 51:303–312. https://doi.org/10.1007/s10535-007-0060-6
Nozoe T, Agbisit R, Fukuta Y, et al (2008) Characteristics of iron tolerant rice lines developed at IRRI under field conditions. Japan Agric Res Q 42:187–192. https://doi.org/10.6090/jarq.42.187
Onaga G, Dramé KN, Ismail AM (2016) Understanding the regulation of iron nutrition: Can it contribute to improving iron toxicity tolerance in rice? Funct Plant Biol 43:709–726. https://doi.org/10.1071/FP15305
Orjuela J, Garavito A, Bouniol M, et al (2010) A universal core genetic map for rice. Theor Appl Genet 120:563–572. https://doi.org/10.1007/s00122-009-1176-1
Sahrawat KL, Sika M (2002) Comparative tolerance of Oryza sativa and O. glaberrima rice cultivars for iron toxicity in West Africa. Int Rice Res Notes 27:30–31
Sikirou M, Saito K, Achigan-Dako EG, et al (2015) Genetic improvement of iron toxicity tolerance in rice-progress, challenges and prospects in West Africa. Plant Prod Sci 18:423–434. https://doi.org/10.1626/pps.18.423
Sikirou M, Saito K, Dramé K, et al (2016) Soil-based screening for iron toxicity tolerance in rice using pots. Plant Prod Sci 19:489–496. https://doi.org/10.1080/1343943X.2016.1186496
Sikirou M, Shittu A, Konaté KA, et al (2018) Screening African rice (Oryza glaberrima) for tolerance to abiotic stresses: I. Fe toxicity. F Crop Res 220:3–9. https://doi.org/10.1016/j.fcr.2016.04.016
Singh PK, Indoliya Y, Chauhan AS, et al (2017) Nitric oxide mediated transcriptional modulation enhances plant adaptive responses to arsenic stress. Sci Rep 7:1–13. https://doi.org/10.1038/s41598-017-03923-2
van Oort PAJ (2018) Mapping abiotic stresses for rice in Africa: Drought, cold, iron toxicity, salinity and sodicity. F Crop Res 219:55–75. https://doi.org/10.1016/j.fcr.2018.01.016
VSN International Limited (2015) Breeding Management System v3.0.8. VSN International, Hemel Hempstead, UK
VSN International Ltd. (2003) Genstat for Windows 18th Edition. VSN International, Hemel Hempstead, UK. Web page: Genstat.co.uk
Wang L, Ying Y, Narsai R, et al (2013) Identification of OsbHLH133 as a regulator of iron distribution between roots and shoots in Oryza sativa. Plant Cell Environ 36:224–236. https://doi.org/10.1111/j.1365-3040.2012.02569.x
Wang W, Ye J, Xu H, et al (2022) OsbHLH061 links TOPLESS/TOPLESS‐RELATED repressor proteins with POSITIVE REGULATOR OF IRON HOMEOSTASIS 1 to maintain iron homeostasis in rice. New Phytol 234:1753–1769. https://doi.org/10.1111/nph.18096
Wang Y, Frei M, Wissuwa M (2008) An agar nutrient solution technique as a screening tool for tolerance to zinc deficiency and iron toxicity in rice. Soil Sci Plant Nutr 54:744–750. https://doi.org/10.1111/j.1747-0765.2008.00302.x
Yu G, Wang LG, Han Y, He QY (2012) ClusterProfiler: An R package for comparing biological themes among gene clusters. Omi A J Integr Biol 16:284–287. https://doi.org/10.1089/omi.2011.0118
Zhang J, Chen K, Pang Y, et al (2017) QTL mapping and candidate gene analysis of ferrous iron and zinc toxicity tolerance at seedling stage in rice by genome-wide association study. BMC Genomics 18:1–15. https://doi.org/10.1186/s12864-017-4221-5

No competing interests reported.

SupplementarySITableSI.xlsx

Download PDF

Reviews received at journal
03 Sep, 2024
Reviewers agreed at journal
14 Aug, 2024
Reviewers agreed at journal
14 Aug, 2024
Reviewers invited by journal
14 Aug, 2024
Editor assigned by journal
05 Aug, 2024
Submission checks completed at journal
02 Aug, 2024
First submitted to journal
01 Aug, 2024

You are reading this latest preprint version

Phenotypic variation among IR64 x TOG5681 rice (Oryza sativa x O. glaberrima) chromosome segment substitution lines (CSSL) in response to iron toxicity, and its associated QTLs

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Plant Materials

Screening for Iron Toxicity Tolerance

QTL mapping

Gene ontology analysis

Results

Phenotypic variation under iron toxicity

QTLs associated with trait variation in response to iron toxicity

Colocation of the QTLs identified with other QTLs and functional rice genes

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1