Genome-wide association studies identifies genetic loci related to fatty acid and branched-chain amino acid metabolism and histone modifications under varying nitrogen treatments in safflower (Carthamus tinctorius)
Fawad Ali A B , Mian A. R. Arif C , Arif Ali D , Muhammad A. Nadeem E , Emre Aksoy F , Allah Bakhsh G , Shahid U. Khan H I , Cemal Kurt J , Dilek Tekdal K , Muhammad K. Ilyas
L , Amjad Hameed C , Yong S. Chung M * and Faheem S. Baloch K *
A
B
C
D
E
F
G
H
I
J
K
L
M
Abstract
Effective identification and usage of genetic variation are prerequisites for developing nutrient-efficient cultivars. A collection of 94 safflower (Carthamus tinctorius) genotypes (G) was investigated for important morphological and photosynthetic traits at four nitrogen (N) treatments. We found significant variation for all the studied traits except chlorophyll b (chl b) among safflower genotypes, nitrogen treatments and G × N interaction. The examined traits showed a 2.82–50.00% increase in response to N application. Biological yield (BY) reflected a significantly positive correlation with fresh shoot weight (FSW), root length (RL), fresh root weight (FRW) and number of leaves (NOL), while a significantly positive correlation was also observed among carotenoids (C), chlorophyll a (chl a), chl b and total chlorophyll content (CT) under all treatments. Superior genotypes with respect to plant height (PH), FSW, NOL, RL, FRW and BY were clustered into Group 3, while genotypes with better mean performance regarding chl a, chl b C and CT were clustered into Group 2 as observed in principal component analysis. The identified eight best-performing genotypes could be useful to develop improved nitrogen efficient cultivars. Genome-wide association analysis resulted in 32 marker-trait associations (MTAs) under four treatments. Markers namely DArT-45481731, DArT-17812864, DArT-15670279 and DArT-45482737 were found consistent. Protein–protein interaction networks of loci associated with MTAs were related to fatty acid and branched-chain amino acid metabolism and histone modifications.
Keywords: biological yield, chlorophyll content, correlation, genome-wide association mapping, marker-trait association, morphological traits, photosynthetic traits, principal component analysis.
Introduction
Safflower (Carthamus tinctorius) is a self-pollinated plant belonging to the Compositae family, with a haploid genome size of 1.4 GB and 2n = 24 chromosomes (Kumari et al. 2017). Safflower is an under-utilised industrial crop cultivated in various geographic regions worldwide for various purposes, including dye manufacturing, edible oil extraction and various therapeutic uses (Weiss 2000). Archaeological findings from Syria dating back to approximately 7500 BC confirm safflower’s ancient history (Marinova and Riehl 2009). The cultivation of safflower then spread from this region to other parts of the world, such as Egypt, the Aegean and south-east Europe.
Nitrogen is a crucial element for crop production as it significantly influences leaf area development, photosynthetic efficiency and overall growth. Safflower has been observed to be particularly responsive to nitrogen compared to other nutrients (Weiss 2000). Inadequate nitrogen content in the soil inhibits growth (Sepaskhah and Barzegar 2010; Tafteh and Sepaskhah 2012), leading to a decline in the shoot-root ratio. Furthermore, nitrogen deficiency delays vegetative and reproductive phenological development and reduces leaf emergence rate (Gilbert and Tucker 1967; Jones and Tucker 1968; Steer and Harrigan 1986). Lower chlorophyll content under low nitrogen levels reduces photosynthesis (Evans and Terashima 1987; Fredeen et al. 1991). Conversely, excessive nitrogen usage can negatively impact ecosystems and contaminate subsurface water sources. Thus, determining suitable fertiliser nitrogen recommendations to achieve profitable crop yields while minimising environmental impacts is essential (Ryan et al. 2009).
The global demand for nitrogen is estimated at 117 million metric tonnes, with an expected annual growth of approximately 1.5% in the near future (FAO 2019). Farmers often rely on substantial amounts of nitrogenous fertilisers to achieve good harvests, contributing to the degradation of air, land and water quality (Hickman et al. 2014; Russo et al. 2017). Moreover, excessive nitrogen supply makes plants more susceptible to diseases and insect pests (Reddy 2017). Therefore, optimising and enhancing nitrogen intake in agricultural plants becomes essential to boost production without compromising the environment and natural resources (Sharma et al. 2023). Genetic approaches, such as identifying marker-trait associations (MTAs), can help create nitrogen-responsive safflower varieties with stable yields under nitrogen-limited conditions.
The ratio of the plant’s uptake of nitrogen to the total amount of nitrogen fertiliser applied is known as the nitrogen use efficiency. Reducing financial losses and environmental contamination is a critical component of sustainable agriculture (Ali et al. 2022). Many small grain crops have been found to have genotypic diversity in nitrogen uptake and utilisation efficiency, and research has been done on the prospect of increasing nitrogen utilisation efficiency through plant breeding. There is not much information available for safflower. Seasonal and genotypic variations in nitrogen transport and accumulation were observed (Koutroubas et al. 2004). Time of sowing was the primary factor influencing seasonal variations. During the filling period, there was a higher translocation to seed because autumn sowing outperformed spring sowing in terms of nitrogen accumulation up to anthesis. To successfully introduce safflower into a particular cropping system, it is necessary to identify the parameters influencing the crop’s nitrogen utilisation. Growers and breeders could utilise this knowledge to determine the most effective selection criteria that will maximise nitrogen exploitation, as well as to adopt the proper cultural methods (Koutroubas et al. 2008).
Genetic diversity for traits related to nitrogen deficiency tolerance is required to develop safflower cultivars that can thrive in low-input systems. Other agricultural plants, such as wheat (Triticum aestivum), have had quantitative trait loci (QTLs) regulating nitrogen uptake genetically mapped using bi-parental populations at various fertiliser application rates (An et al. 2006; Laperche et al. 2008; Xu et al. 2014; Mahjourimajd et al. 2016; Deng et al. 2017). However, no marker-trait associations using genome-wide association studies (GWAS) for response to nitrogen have been identified in safflower. Therefore, genotype and underlying QTL detection are essential to maintain high performance under low nitrogen conditions in safflower breeding programs aimed at improving nitrogen-deficiency tolerance.
Next-generation sequencing technologies, such as multiplex sequencing and genotyping by sequencing (GBS), provide vast amounts of genetic data for various applications (Raman et al. 2012). Due to high labour and consumable costs, current polymerase chain reaction (PCR)-based marker technologies are not practical for whole-genome analysis in association studies, genetic map development, extensive molecular analysis of collected germplasm, or genome-wide selection of preferred alleles (Raman et al. 2012).
Understanding the genetic basis of significant traits of interest is crucial for enhancing the performance of crop plants (Losos et al. 2013). GWAS and QTL mapping have been developed to determine the genetic basis of complex plant traits (Arif et al. 2012; Korte and Farlow 2013). While QTL mapping has some drawbacks, including population specificity and limited recombination (Arif et al. 2021, 2022), GWAS overcomes these limitations and allows markers to be used in any population (Korte and Farlow 2013). GWAS has been effective in identifying genomic regions affecting important attributes in various crop species (Wang et al. 2012; Arif et al. 2017; Monostori et al. 2017; Hazzouri et al. 2018; Akram et al. 2021; Nadeem et al. 2021). However, in safflower, many genomic regions/genes with various phenotypes of interest remain undiscovered despite the use of diverse germplasm and genome sequencing data. Therefore, this study aimed to investigate the best-performing safflower genotypes and identify the genetic basis of important morphological and photosynthetic traits under varying nitrogen treatments using DArTseq marker information.
Materials and methods
Plant materials and evaluation
In this study, 94 safflower (Carthamus tinctorius L.) genotypes from 26 different countries were used as plant materials. These lines were procured from the United States Department of Agriculture (USDA) (see Supplementary Table S1). The experiment was conducted in pots on 10 March 2023, at the Department of Botany, University of Baltistan, Skardu, Pakistan. The experiment followed a two-factorial design with three replications. Each experimental pot contained 1.8 dm3 of soil sieved through a 4-mm sieve. The soil texture consisted of a mixture of sand and soil in a ratio of 1:2, with different concentrations of micro- and macro-nutrients (Table 1). Three seeds from each safflower genotype were sown in the filled pots, and thinning was performed after germination, leaving only one seedling per pot for data recording. Tap water was added at regular intervals after every 3−4 days, according to the requirements of the genotypes.
| Property | ||
|---|---|---|
| pH | 8.6 | |
| Electrical conductivity (mS m−1) | 1.2 | |
| Organic matter (%) | 0.84 | |
| Available phosphorus (mg kg−1) | 2.7 | |
| Available potassium (mg kg−1) | 204 | |
| Saturation (%) | 32 | |
| Zn (mg kg−1) | 1.58 | |
| Fe (mg kg−1) | 4.8 | |
| B (mg kg−1) | 0.46 | |
| Mn (mg kg−1) | 7.3 | |
| Cu (mg kg−1) | 0.78 |
For each genotype, three plants were used to record data. Four different nitrogen (N) treatments were applied: T1, no N; T2, 60 mg dm−3 N; T3, 120 mg dm−3 N; and T4, 180 mg dm−3 N. The nitrogen concentrations were chosen following the study conducted by Anicésio et al. (2018). Urea fertiliser (Sona Urea, Fauji Fertiliser Company Limited, Skardu, Pakistan) was used as the nitrogen source. Different nitrogen doses were applied to treatments T2, T3 and T4 between 22 and 45 days after sowing. Safflower genotypes were harvested after 45 days to record important morphological and photosynthetic traits. Association analysis was performed using the mean value of each trait across all treatments.
Studied parameters
Plant height (PH) was measured in centimetres from the base of the stem up to the shoot apex. The number of leaves (NOL) was counted for each selected plant, and the same plant was weighed to record the biological yield (BY) in grams. The length of the roots (RL) was measured using a scale after washing the roots with tap water, and the fresh root weight (FRW) in grams was recorded using a digital balance. Likewise, the fresh shoot weight (FSW) in grams was also recorded for the same plants.
For the determination of total chlorophyll (CT) content, fresh leaves were immediately placed in a −70°C deep freezer. Chlorophyll a (chl a), chlorophyll b (chl b), carotenoids (C) and CT were measured using a UV spectrophotometer following the protocol proposed by Lichtenthaler and Buschmann (2001). The leaf surface was cleaned with double-distilled water, and a 200-mg leaf sample was weighed using an analytical balance. An 80% acetone solution was prepared and kept at 4°C. The leaf samples were ground with a mortar and pestle in 3 mL of 80% acetone, and after grinding, the volume was raised to 5 mL in a graduated test tube. The leaf samples were then incubated at 90°C for 5 min, followed by centrifugation for 10 min at 1585g. The resulting supernatant was transferred to a 1-cm pathlength cuvette, and its absorbance against a solvent blank was recorded in a UV-VIS spectrophotometer at wavelengths of A663 nm, A646 nm and A470 nm for chl a, chl b and C, respectively (Lichtenthaler and Buschmann 2001). Chl a, chl b, C and CT were calculated by using the formula described in Lichtenthaler and Buschmann (2001).
Genomic DNA isolation
Fresh, healthy and young leaves were collected and immediately frozen at −80°C in the laboratory to extract genomic DNA from each safflower genotype. DNA isolation for each genotype was performed using a pooled sample of leaves from ten different individual plants. The plants originated from the original seeds obtained from the gene bank. The recommended method by Diversity Arrays Technologies (https://www.diversityarrays.com) and the cetyltrimethylammonium bromide (CTAB) technique (Doyle and Doyle 1990) were used for DNA extraction. The DNA concentration was determined using NanoDrop (DeNovix DS-11 FX, USA) while its quality was determined by agarose gel electrophoresis (0.80%). DNA was then appropriately diluted, maintaining a concentration of 50 ng L−1 for genotyping by sequencing (GBS) analysis. Subsequently, the prepared DNA samples were sent to Diversity Array Technologies (Bruce, ACT, Australia; http://www.diversityarrays.com) for DArTseq analysis.
DArTseq-generated SilicoDArT marker analysis
The DArTseq technology, based on next-generation sequencing, effectively reduced complexity and facilitated the identification of genome regions containing active genes linked to plant traits of interest (Elshire et al. 2011). In this study, DArTseq was particularly useful in identifying genomic regions related to plant traits. The DNA samples were processed using digestion/ligation processes following the approach outlined by Kilian et al. (2012). Amplification of mixed fragments (PstI-MseI) was achieved using thirty rounds of PCR cycles. Detailed analysis of the silicoDArT markers can be found in the earlier research studies of Kilian et al. (2012) and Li et al. (2015).
Statistical analysis
Statistix 8.1 software was employed to generate descriptive statistics and conduct ANOVA. For data visualisation, boxplots were created using RStudio (ver. 1.0.153) and the ‘ggplot2’ package (Wickham 2016). Correlations were also visualised using the ‘qgraph’ package (Epskamp et al. 2012) in RStudio. Additionally, principal component analysis (PCA) was performed using the ‘factoextra’ package in R (ver. 4.1.3) to assess the relationship between genotypes and expressed phenotypes (Kassambara and Mundt 2017).
All images from the DArTseq platform were analysed using DArTsoft (ver. 7.4.7, Diversity Array Technologies). SilicoDArT markers, discovered using DArTseq and scored in a binary manner (Cömertpay et al. 2012; Baloch et al. 2017), were used for the analysis. A value of 1 or 0 indicated the presence or absence of the restriction fragment in the genomic representation of each sample, respectively. The markers were screened based on call rate, polymorphism information content (PIC) and repeatability. Markers with PIC, repeatability and call rates below 0.10, 100% and 0.80%, respectively, were discarded to prevent false inferences.
The STRUCTURE software (ver. 2.3.4) (Pritchard et al. 2000) was used to assess the population structure in the germplasm under investigation. To determine the optimal number of clusters (K subpopulations) ranging from 1 to 10, the method established by Evanno et al. (2005) was used with 10 separate runs for each K value. The initial burn-in period was set to 500 and 500,000 MCMC (Markov Chain Monte Carlo) iterations. Membership coefficients were used to assign each genotype to a particular population. The Neighbour Joining tree was created using the ‘hierfstat’ R tool in the R statistical software, while the PCoA was carried out using GenAlEx 6.5 (Peakall and Smouse 2006). Populations derived through neighbour-joining and PCoA were named and coloured using the same clustering pattern found with the model-based structure approach. Details about the population structure, neighbour-joining and PCoA followed the guidelines suggested by Ali et al. (2020b).
For marker-trait associations (MTAs) analysis, TASSEL 5.0.5 was used with a mixed linear model (MLM, Q + K) approach (Bradbury et al. 2007). Q-matrix (Q) and kinship (K) were used to correct for population and family structure during association analysis, following the recommendations of Nadeem et al. (2020) and Zia et al. (2020). The kinship matrix was computed using scaled identity and descent methods applied in TASSEL 5.0.5 (Bradbury et al. 2007). The P-value in the association study indicates the strength of the marker-trait relationship, and R2 represents the percentage of phenotypic variance explained by a significant marker (Jin et al. 2011). SilicoDArT markers significantly correlated with the trait of interest were defined based on Bonferroni and FDR thresholds of P = 0.01. Manhattan plots were created using the ‘CMplot’ R Package in R 4.0.0 statistical software for data visualisation.
Since full genome data of safflower is not publicly available yet, marker sequences were BLAST searched against the Arabidopsis thaliana genome (TAIR 10) via TAIR BLAST 2.9.0+ (Lamesch et al. 2012), and the loci with the highest hit score and E cut-off value of 0.001. Arabidopsis orthologues were then used to search for the protein–protein interaction networks through String (ver. 12.0) (Szklarczyk et al. 2019) with the minimum required interaction score of 0.9. To enhance the reliability and quality of the PPI network, we applied a filtering process using the ‘combined score’ with a recommended threshold of 400 (von Mering et al. 2005). Finally, the genes encoding for the proteins in each PPI network interacting with marker-associated loci were used as a set to determine the gene ontology (GO) enrichment for molecular function and biological process by using Fisher’s exact test with the Bonferroni correction for multiple testing (P < 0.01) (Thomas et al. 2022).
Results
Evaluation of morphological and photosynthetic traits
Analysis of variance (ANOVA) revealed significant differences among safflower genotypes, nitrogen treatments and between G × N interaction for all the studied traits, except for chl b (Table 2). Different nitrogen levels had a significant impact on the average performance of all the traits studied. PH, RL and CT exhibited notable differences between T1 and T3, whereas FSW, NOL, FRW, chl a, chl b and BY showed significant variations among T1, T2, T3 and T4 (Fig. 1).
| Trait | Source of variation | Mean squares | P-value | |
|---|---|---|---|---|
| BY | G | 2.1477 | 0.0000 | |
| N | 10.3932 | 0.0000 | ||
| G × N | 1.3602 | 0.0000 | ||
| C | G | 0.12221 | 0.0000 | |
| N | 1.62424 | 0.0000 | ||
| G × N | 0.04876 | 0.0000 | ||
| Chl a | G | 0.10357 | 0.0000 | |
| N | 0.61013 | 0.0000 | ||
| G × N | 0.0382 | 0.0000 | ||
| Chl b | G | 292.585 | 0.4978 | |
| N | 594.196 | 0.1097 | ||
| G × N | 291.856 | 0.5258 | ||
| FSW | G | 0.19306 | 0.0000 | |
| N | 2.37157 | 0.0000 | ||
| G × N | 0.08952 | 0.0000 | ||
| NOL | G | 9.349 | 0.0000 | |
| N | 204.298 | 0.0000 | ||
| G × N | 3.004 | 0.0000 | ||
| PH | G | 0.0288 | 0.0000 | |
| N | 0.07106 | 0.0000 | ||
| G × N | 0.01741 | 0.0000 | ||
| RL | G | 13.448 | 0.0000 | |
| N | 25.6083 | 0.0000 | ||
| G × N | 5.2892 | 0.0000 | ||
| FRW | G | 1.00771 | 0.0000 | |
| N | 5.82379 | 0.0000 | ||
| G × N | 0.39408 | 0.0000 | ||
| CT | G | 0.56742 | 0.0000 | |
| N | 6.82833 | 0.0000 | ||
| G × N | 0.2379 | 0.0000 |
Effect of N treatment on the phenotypic response of various plants traits in safflower. (a) PH, plant height; (b) NOL, number of leaves; (c) FRW, fresh root weight; (d) FSW, fresh shoot weight; (e) RL, root length; (f) C, carotenoids; (g) chl a, chlorophyll a; (h) chl b, chlorophyll b; (i) CT, total chlorophyll; and (j) BY, biological yield. Treatments are: T1, no N (blue); T2, 60 mg dm−3 N (yellow); T3, 120 mg dm−3 N (red); and T4, 180 mg dm−3 N (purple). Significant differences among treatments were based on t-test. *P < 0.1; ***P < 0.01; ****P < 0.0001; n.s., not significant.
Varying nitrogen levels significantly influenced all studied traits across treatments (Fig. 2). For example, PH increased from 1.77 cm in T1 to 1.83 cm, 1.86 cm and 1.82 cm in T2, T3 and T4, respectively. This implies an increase of 3.39%, 5.08% and 2.82% in T2, T3 and T4, respectively. FSW increased from 0.42 g in T1 to 0.51 g (21.43% increase), 0.63 g (50.00% increase) and 0.57 g (35.71% increase) in T2, T3 and T4, respectively. NOL increased from 7.10 in T1 to 8.12, 9.02 and 8.73 in T2, T3 and T4, respectively. RL increased from 15.18 cm in T1 to 16.13 cm, 16.90 cm and 16.66 cm in T2, T3 and T4, respectively. Similarly, FRW increased from 0.68 g in T1 to 0.88 g, 1.02 g and 0.91 g in T2, T3 and T4, respectively. This represents an increase of 29.41%, 50.00% and 33.82% in T2, T3 and T4, respectively. Additionally, chlorophylls a and b increased from 0.42 and 0.30 in T1 to 0.48 and 0.35, 0.51 and 0.38, 0.52 and 0.36 in T2, T3 and T4, respectively. C and CT increased from 0.37 and 1.08 in T1 to 0.47 and 1.30, 0.49 and 1.37, 0.55 and 1.44 in T2, T3 and T4, respectively. BY increased from 1.19 g in T1 to 1.39 g, 1.66 g and 1.48 g in T2, T3 and T4, respectively (Table S2).
Percentage increase in various traits under treatments T2 (60 mg dm−3 N, yellow), T3 (120 mg dm−3 N, red) and T4 (180 mg dm−3 N, purple) in relative to treatment T1 (no nitrogen). PH, plant height; FRW, fresh root weight; FSW, fresh shoot weight; NOL, number of leaves; RL, root length; chl a, chlorophyll a; chl b, chlorophyll b; C, carotenoids; CT, total chlorophyll; BY, biological yield.
Correlation, PCA and best-performing genotypes
The correlation analysis showed significantly positive associations between most of the studied traits in all treatments (Fig. 3, Table S2). In T1, BY demonstrated a significant positive correlation with FSW, FRW and RL. A significant positive correlation was also observed between C, chl a, chl b and CT in T1. In T2, BY exhibited a significantly positive correlation with FRW, FSW and NOL. Moreover, a significant positive correlation was recorded between C, chl a, chl b and CT in T2. In T3, all the studied traits displayed a significantly positive correlation with each other. Similarly, the studied traits revealed significantly positive correlations with one another, except for C, which showed a negative association with RL in T4.
Correlation among the studied traits under T1 (no N, blue), T2 (60 mg dm−3 N, yellow), T3 (120 mg dm−3 N, red) and T4 (180 mg dm−3 N, purple) treatments. Blue connecting lines indicate a positive correlation whose thickness corresponds to the strength of the correlation. Orange connecting lines indicate a negative correlation. For full details, see Table S3.
PCA showed that the first two principal components, having eigenvalues ≥1, accounted for a substantial portion of the variation (27.5% of the total variation), with PC1 contributing to 14.2% of the overall variance (Fig. 4). The second principal component (PC2) explained 13.3% of the total variation. This also indicates the complexity of the data involved. Based on the distribution of genotypes in the PCA biplot quadrants, all genotypes were classified into three distinct groups (designated as 1, 2 and 3). The genotypes in Cluster 3 shaded in blue performed better than the genotypes occupying Clusters 1 shaded in grey and 2 shaded in brown. Even in Cluster 3, the genotypes on the extreme right performed the best. These included Genotypes 36, 39, 44, 51, 66, 78, 82 and 94 among the 94 safflower panel. Even among them, Genotype 36 depicted the best phenotype in all treatments. Table 3 provides the means of the best genotype, population mean, and the percentage increase from the mean population for these selected genotypes. These top-performing genotypes can be valuable for future safflower breeding programs aimed at developing improved cultivars with enhanced NUE.
Biplot based on the principal component analyses showing the number of three clusters of safflower genotypes based on all the recorded traits in all treatments.
| Treatment | Trait | Genotypes | Mean of best lines from the biplot figure | % increase in the population mean | Population mean | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| G_36 | G_39 | G_44 | G_51 | G_66 | G_78 | G_82 | G_94 | ||||||
| Treatment 1 | PH | 1.86 | 1.86 | 1.52 | 1.69 | 1.95 | 2.12 | 1.86 | 1.95 | 1.85 | 6.15 | 1.76 | |
| FSW | 0.66 | 0.56 | 0.46 | 0.84 | 0.58 | 0.47 | 0.53 | 0.33 | 0.55 | 27.72 | 0.40 | ||
| NOL | 8.33 | 7.33 | 8.00 | 8.00 | 7.67 | 6.67 | 7.00 | 7.67 | 7.58 | 8.03 | 7.05 | ||
| RL | 16.00 | 29.80 | 16.26 | 19.05 | 31.33 | 21.34 | 14.22 | 20.91 | 21.11 | 31.73 | 14.62 | ||
| FRW | 0.78 | 2.30 | 0.87 | 0.81 | 1.68 | 0.86 | 0.94 | 0.68 | 1.12 | 43.87 | 0.64 | ||
| Chl a | 0.52 | 0.45 | 0.33 | 0.52 | 0.46 | 0.57 | 0.49 | 0.54 | 0.48 | 16.06 | 0.41 | ||
| Chl b | 0.38 | 0.35 | 0.37 | 0.34 | 0.34 | 0.35 | 0.33 | 0.38 | 0.35 | 19.69 | 0.29 | ||
| CT | 0.91 | 0.80 | 0.70 | 0.86 | 0.80 | 0.92 | 0.82 | 0.91 | 0.84 | 17.59 | 0.70 | ||
| C | 0.75 | 0.29 | 0.48 | 0.37 | 0.29 | 0.41 | 0.37 | 0.44 | 0.42 | 15.57 | 0.36 | ||
| BY | 1.44 | 2.87 | 1.33 | 1.65 | 2.26 | 1.33 | 1.46 | 1.01 | 1.67 | 38.52 | 1.04 | ||
| Treatment 2 | PH | 1.78 | 2.12 | 1.61 | 1.95 | 2.12 | 2.12 | 1.69 | 2.12 | 1.94 | 7.11 | 1.82 | |
| FSW | 0.67 | 0.67 | 0.65 | 1.44 | 0.69 | 0.56 | 0.77 | 0.87 | 0.79 | 39.32 | 0.49 | ||
| NOL | 8.00 | 8.33 | 10.00 | 10.33 | 8.67 | 9.67 | 10.00 | 11.00 | 9.50 | 16.91 | 7.99 | ||
| RL | 16.51 | 23.71 | 21.00 | 18.80 | 23.71 | 21.76 | 16.09 | 16.34 | 19.74 | 20.97 | 15.80 | ||
| FRW | 0.91 | 2.30 | 1.06 | 1.43 | 2.62 | 0.74 | 2.37 | 2.08 | 1.69 | 53.40 | 0.80 | ||
| Chl a | 0.53 | 0.55 | 0.49 | 0.59 | 0.54 | 0.59 | 0.84 | 0.67 | 0.60 | 22.40 | 0.47 | ||
| Chl b | 0.43 | 0.37 | 0.42 | 0.38 | 0.37 | 0.45 | 0.51 | 0.44 | 0.42 | 17.34 | 0.35 | ||
| CT | 0.96 | 0.92 | 0.91 | 0.96 | 0.91 | 1.04 | 1.35 | 1.11 | 1.02 | 20.32 | 0.82 | ||
| C | 0.75 | 0.35 | 0.52 | 0.47 | 0.35 | 0.59 | 0.57 | 0.47 | 0.51 | 9.57 | 0.46 | ||
| BY | 1.58 | 2.96 | 1.71 | 2.87 | 3.31 | 1.30 | 3.14 | 2.94 | 2.48 | 48.91 | 1.29 | ||
| Treatment 3 | PH | 1.86 | 1.69 | 1.78 | 2.12 | 1.86 | 2.12 | 1.52 | 1.86 | 1.85 | 0.43 | 1.86 | |
| FSW | 0.74 | 0.75 | 0.74 | 0.85 | 0.77 | 0.56 | 0.60 | 0.82 | 0.73 | 15.42 | 0.62 | ||
| NOL | 9.67 | 10.33 | 11.00 | 10.67 | 11.00 | 10.33 | 10.33 | 15.00 | 11.04 | 21.04 | 8.83 | ||
| RL | 17.02 | 21.17 | 21.59 | 16.43 | 21.08 | 14.05 | 18.71 | 18.37 | 18.55 | 10.73 | 16.75 | ||
| FRW | 0.96 | 1.29 | 1.46 | 0.67 | 1.40 | 1.11 | 0.62 | 2.58 | 1.26 | 21.81 | 1.00 | ||
| Chl a | 0.64 | 0.46 | 0.68 | 0.65 | 0.50 | 0.66 | 0.36 | 0.65 | 0.58 | 13.82 | 0.50 | ||
| Chl b | 0.47 | 0.42 | 0.37 | 0.47 | 0.41 | 0.45 | 0.32 | 0.38 | 0.41 | 8.97 | 0.38 | ||
| CT | 1.11 | 0.87 | 1.05 | 1.12 | 0.92 | 1.10 | 0.68 | 1.04 | 0.99 | 11.80 | 0.88 | ||
| C | 0.69 | 0.46 | 0.67 | 0.57 | 0.46 | 0.44 | 0.53 | 0.32 | 0.52 | 7.44 | 0.48 | ||
| BY | 1.71 | 2.04 | 2.21 | 1.52 | 2.17 | 1.67 | 1.21 | 3.41 | 1.99 | 19.47 | 1.62 | ||
| Treatment 4 | PH | 2.03 | 2.12 | 1.78 | 1.95 | 1.69 | 1.61 | 1.95 | 1.52 | 1.83 | 1.47 | 1.82 | |
| FSW | 0.84 | 0.53 | 0.84 | 0.74 | 0.50 | 0.77 | 0.59 | 0.42 | 0.65 | 14.97 | 0.56 | ||
| NOL | 10.33 | 9.00 | 11.00 | 11.00 | 9.00 | 10.33 | 9.00 | 11.00 | 10.08 | 15.63 | 8.61 | ||
| RL | 18.71 | 18.63 | 22.10 | 21.34 | 18.97 | 16.26 | 10.67 | 16.59 | 17.91 | 8.59 | 16.55 | ||
| FRW | 1.36 | 0.49 | 1.88 | 0.52 | 0.44 | 1.38 | 0.55 | 1.79 | 1.05 | 15.27 | 0.90 | ||
| Chl a | 0.52 | 0.36 | 0.76 | 0.77 | 0.31 | 0.74 | 0.52 | 0.66 | 0.58 | 11.79 | 0.52 | ||
| Chl b | 0.32 | 0.14 | 0.38 | 0.53 | 0.13 | 0.49 | 0.33 | 0.27 | 0.32 | −12.33 | 0.37 | ||
| CT | 0.84 | 0.50 | 1.15 | 1.30 | 0.44 | 1.23 | 0.85 | 0.93 | 0.90 | 3.18 | 0.88 | ||
| C | 0.76 | 0.41 | 0.71 | 0.78 | 0.46 | 0.61 | 0.45 | 0.46 | 0.58 | 6.05 | 0.55 | ||
| BY | 2.20 | 1.03 | 2.72 | 1.26 | 0.93 | 2.15 | 1.14 | 2.21 | 1.71 | 15.43 | 1.46 | ||
Marker-trait association analysis
A total of 12,232 highly informative DArTseq markers were utilised for genome-wide association mapping analysis with the MLM (Q + K) model across the four treatments. This analysis resulted in the identification of 32 significant marker-trait associations (MTAs) for the studied traits. Specifically, 10 MTAs were found for four traits under T1, while T2 showed 11 significant MTAs for five traits. T3 showed three MTAs for three traits and T4 showed eight MTAs for four traits (Table 4).
| Treatment | Trait | Marker | Chromosome # | P-value | R2 | Variance | |
|---|---|---|---|---|---|---|---|
| T1 | NOL | DArT-45481731 | 10 | 1.72E-04 | 0.17769 | 0.45531 | |
| NOL | DArT-45483063 | 8 | 1.45E-04 | 0.16989 | 0.45531 | ||
| NOL | DArT-38079559 | 2 | 1.79E-04 | 0.1665 | 0.45531 | ||
| NOL | DArT-45485734 | 2 | 1.94E-04 | 0.16358 | 0.45531 | ||
| FRW | DArT-17812864 | 8 | 1.81E-04 | 0.18955 | 0.09356 | ||
| FRW | DArT-22762716 | 2 | 1.41E-04 | 0.17631 | 0.09356 | ||
| CT | DArT-15670677 | 1 | 2.12E-04 | 0.17392 | 0.00426 | ||
| BY | DArT-38083567 | 12 | 1.58E-04 | 0.19237 | 0.1101 | ||
| BY | DArT-17812864 | 8 | 2.98E-04 | 0.1696 | 0.1101 | ||
| BY | DArT-22761898 | 2 | 4.28E-04 | 0.14891 | 0.1101 | ||
| T2 | FSW | DArT-100046395 | 12 | 3.16E-05 | 0.20615 | 0.00549 | |
| FSW | DArT-38083554 | 12 | 2.67E-04 | 0.16449 | 0.00549 | ||
| NOL | DArT-22764463 | 12 | 5.49E-05 | 0.21602 | 0.90253 | ||
| NOL | DArT-45478831 | 2 | 1.45E-04 | 0.18431 | 0.90253 | ||
| NOL | DArT-45481731 | 10 | 1.83E-04 | 0.16817 | 0.90253 | ||
| RL | DArT-15670186 | 6 | 1.93E-04 | 0.16663 | 0.24212 | ||
| FRW | DArT-45483148 | 8 | 6.90E-05 | 0.18796 | 0.06261 | ||
| FRW | DArT-45487501 | 7 | 1.97E-04 | 0.17352 | 0.06261 | ||
| FRW | DArT-15670279 | 4 | 1.78E-04 | 0.16505 | 0.06261 | ||
| BY | DArT-15670279 | 4 | 1.14E-04 | 0.17561 | 0.15657 | ||
| BY | DArT-45483151 | 5 | 3.06E-04 | 0.15853 | 0.15657 | ||
| T3 | FSW | DArT-100005164 | 12 | 2.56E-04 | 0.15559 | 0.0084 | |
| FRW | DArT-45482737 | 3 | 8.22E-05 | 0.18548 | 0.01315 | ||
| BY | DArT-45482737 | 3 | 2.52E-04 | 0.15878 | 0.04028 | ||
| T4 | FSW | DArT-100002034 | 8 | 1.56E-04 | 0.17007 | 0.01586 | |
| FSW | DArT-45485354 | 7 | 1.53E-04 | 0.16947 | 0.01586 | ||
| FSW | DArT-45487831 | 8 | 2.47E-04 | 0.15786 | 0.01586 | ||
| RL | DArT-45478819 | 2 | 1.44E-04 | 0.17828 | 1.18193 | ||
| FRW | DArT-15672551 | 1 | 1.53E-04 | 0.18146 | 0.0533 | ||
| FRW | DArT-15674217 | 1 | 1.94E-04 | 0.18142 | 0.0533 | ||
| FRW | DArT-45480200 | 5 | 2.30E-04 | 0.16453 | 0.0533 | ||
| BY | DArT-45487918 | 4 | 2.14E-04 | 0.16149 | 0.18141 |
In T1, we observed four significant markers in association with NOL, which were located on chromosomes 2 (SNPs DArT-38079559 and DArT-45485734), 8 (SNP DArT-45483063) and 10 (SNP DArT-45481731) (Fig. 5a). Likewise, there were two MTAs detected for FRW in T1 on chromosomes 2 (SNP DArT-22762716) and 8 (SNP DArT-17812864), were associated with FRW. Additionally, a single significant MTA was found to be associated with CT on chromosome 1 (SNP DArT-15670677), while three significant MTAs (SNPs DArT-22761898, DArT-17812864 and DArT-38083567) for BY were detected on chromosomes 2, 8 and 12, correspondingly, in T1.
A genome-wide scan (GWAS analysis) plot of SNP markers associated with various traits in 94 safflower accessions in different N treatments. Markers in (a) T1, (b) T2, (c) T3, and (d) T4. The plots show SNP-based Manhattan plots with the names, P-values and R2 of only significant SNPs. The chromosome numbers are shown on the x-axis, and the genome-wide scan −log10 (P-values) is shown on the y-axis. A box with a thick outline indicates that the SNP was associated with more than one trait.
In T2, we discovered two significant MTAs (SNPs DArT-100046395 and DArT-38083554) associated with FSW on chromosome 12 (Fig. 5b). Another three MTAs were detected for NOL on chromosomes 2 (SNP DArT-45478831), 10 (DArT-45481731) and 12 (SNP DArT-22764463). Moreover, one MTA for RL was detected on chromosome 6 (SNP DArT-15670186) and three MTAs (SNPs DArT-15670279, DArT-45487501 and DArT-45483148) were discovered for FRW on chromosomes 4, 7 and 8, correspondingly. Similarly, two MTAs (SNPs DArT-15670279 and DArT-45483151) were found for BY on chromosomes 4 and 5, respectively.
In T3, we identified single MTAs for each of the traits: FSW (SNP DArT-100005164 on chromosome 12) (Fig. 5c), FRW (SNP DArT-45482737 on chromosome 3) and BY (SNP DArT-45482737 on chromosome 3). Similarly, in T4, significant MTAs were found for FSW (SNPs DArT-45485354, DArT-100002034 and DArT-45487831 on chromosomes 7, 8 and 8, respectively), RL (SNP DArT-45478819 on chromosome 2), FRW (SNPs DArT-15672551, DArT-15674217 and DArT-45480200 on chromosomes 1, 1 and 5, correspondingly) and BY (SNP DArT-45487918 on chromosome 4) (Fig. 5d).
Marker-associated loci identification, PPI network and GO enrichment analysis
Since the complete genome data of safflower is not publicly available yet, we identified the Arabidopsis orthologues of marker-associated loci via a BLAST search (Table S4). We used a stringent threshold to select one Arabidopsis orthologue to represent each marker. Next, we draw the protein–protein interaction network of each treatment separately to get a deeper understanding of the potential interacting partners of each Arabidopsis orthologue (Fig. 6). Accordingly, the PPI network of the 10 loci associated with the SNP markers identified under T1 generated two clusters. The first cluster included 27 proteins while the second one was comprised of 12 proteins (Fig. 6a). Under T2, the PPI network of the 40 loci associated with the SNP markers generated two clusters. The first cluster included 28 proteins while the second one was comprised of 12 proteins (Fig. 6b). The PPI network of the 32 loci associated with the SNP markers identified under T3 generated one cluster (Fig. 6c). Similar to T1 and T2, the PPI network of the 35 loci associated with the SNP markers identified under T4 generated two clusters. The first cluster included 27 proteins while the second one is comprised of 8 proteins (Fig. 6d).
Protein–protein interaction network analysis of the loci associated with the SNP markers identified in different nitrogen treatments. (a) T1; (b) T2; (c) T3; (d) T4. PPI networks were generated in String.
To better understand the functions of the protein clusters obtained in PPI network analysis, we performed GO enrichment according to biological process and molecular function (Table 5). Cluster 1 in T1 is involved in chromatin remodelling and gene imprinting as well as maintenance of floral meristem identity while Cluster 2 includes proteins responsible for fatty acid derivative biosynthetic process and membrane lipid metabolic process. Cluster 1 proteins in T2 are related to the biosynthesis of lipoate, valine, isoleucine and xanthophyll whereas Cluster 2 proteins are involved in cutin-based cuticle development and fatty acid or sphingolipid biosynthetic processes. Proteins in T3 are significantly associated with ribosome assembly, leaf morphogenesis, and protein polyubiquitination and catabolism. Finally, Custer 1 in T4 is related to histone modifications such as acetylation and methylation while the proteins in Cluster 2 are involved in protein quality control and catabolism together with plastid organisation.
| Treatment | GO biological process | Fold enrichment | P-value | GO molecular function | Fold enrichment | P-value | |
|---|---|---|---|---|---|---|---|
| T1 | Cluster 1 | ||||||
| Transcription initiation-coupled chromatin remodelling | >100 | 4.73E-08 | Structural constituent of chromatin | >100 | 1.22E-16 | ||
| Maintenance of floral meristem identity | >100 | 5.09E-05 | Protein heterodimerisation activity | 62.52 | 6.79E-13 | ||
| Regulation of gene expression by genomic imprinting | >100 | 5.71E-07 | Chromatin binding | 20.63 | 4.29E-05 | ||
| Vernalisation response | >100 | 1.11E-06 | Structural molecule activity | 19.09 | 5.80E-10 | ||
| Maintenance of shoot apical meristem identity | 96.75 | 2.33E-04 | Protein dimerisation activity | 12.15 | 1.92E-07 | ||
| Cluster 2 | |||||||
| Very long-chain fatty acid biosynthetic process | >100 | 2.66E-22 | Ketoreductase activity | >100 | 1.05E-06 | ||
| Wax biosynthetic process | >100 | 9.65E-10 | 3-Hydroxyacyl-CoA dehydratase activity | >100 | 1.75E-06 | ||
| Fatty acid derivative biosynthetic process | >100 | 1.53E-09 | enoyl-CoA hydratase activity | >100 | 7.87E-06 | ||
| sphingolipid biosynthetic process | >100 | 1.29E-04 | Fatty acid synthase activity | >100 | 1.15E-05 | ||
| Membrane lipid metabolic process | 61.78 | 1.49E-05 | Acyltransferase activity, transferring groups other than amino-acyl groups | 49.39 | 2.62E-11 | ||
| T2 | Cluster 1 | ||||||
| Lipoate biosynthetic process | >100 | 6.01E-06 | L-isoleucine transaminase activity | >100 | 1.90E-13 | ||
| Valine biosynthetic process | >100 | 5.77E-18 | L-valine transaminase activity | >100 | 1.90E-13 | ||
| Isoleucine biosynthetic process | >100 | 2.91E-20 | L-leucine transaminase activity | >100 | 1.90E-13 | ||
| Xanthophyll biosynthetic process | >100 | 6.14E-10 | Lipoate synthase activity | >100 | 6.01E-06 | ||
| Branched-chain amino acid biosynthetic process | >100 | 1.79E-23 | Octanoyltransferase activity | >100 | 1.90E-08 | ||
| Cluster 2 | |||||||
| Very long-chain fatty acid biosynthetic process | >100 | 2.66E-22 | Ketoreductase activity | >100 | 1.05E-06 | ||
| Wax biosynthetic process | >100 | 9.65E-10 | 3-Hydroxyacyl-CoA dehydratase activity | >100 | 1.75E-06 | ||
| Cutin-based cuticle development | >100 | 9.60E-07 | Enoyl-CoA hydratase activity | >100 | 7.87E-06 | ||
| Fatty acid biosynthetic process | >100 | 4.43E-24 | Fatty acid synthase activity | >100 | 1.15E-05 | ||
| Sphingolipid biosynthetic process | >100 | 1.29E-04 | Acyltransferase activity, transferring groups other than amino-acyl groups | 49.39 | 2.62E-11 | ||
| T3 | Ubiquitin-dependent protein catabolic process via the N-end rule pathway | >100 | 6.91E-06 | Ubiquitin-activating enzyme activity | >100 | 1.15E-05 | |
| Adaxial/abaxial pattern specification | 58.36 | 2.23E-05 | Ubiquitin-like modifier activating enzyme activity | >100 | 1.95E-04 | ||
| Ribosome assembly | 42.86 | 5.38E-05 | Structural constituent of ribosome | 36.13 | 1.37E-17 | ||
| Leaf morphogenesis | 36.94 | 4.75E-06 | Structural molecule activity | 24.81 | 1.52E-15 | ||
| Protein polyubiquitination | 33.55 | 6.86E-06 | Ubiquitin-protein transferase activity | 17.62 | 1.36E-09 | ||
| T4 | Cluster 1 | ||||||
| Histone H3-K4 methylation | >100 | 2.60E-05 | Histone H4 acetyltransferase activity | >100 | 1.95E-05 | ||
| Histone acetylation | 98.32 | 4.97E-06 | Structural constituent of chromatin | >100 | 4.61E-24 | ||
| Histone lysine methylation | 92.36 | 2.54E-04 | Histone H3K4 methyltransferase activity | >100 | 6.11E-05 | ||
| Protein acetylation | 64.85 | 1.61E-05 | Histone acetyltransferase activity | >100 | 3.74E-06 | ||
| Histone modification | 55.21 | 3.76E-08 | Protein heterodimerisation activity | 85.96 | 5.35E-19 | ||
| Cluster 2 | |||||||
| Protein quality control for misfolded or incompletely synthesised proteins | >100 | 2.12E-16 | ATPase binding | >100 | 6.36E-18 | ||
| Chloroplast organisation | 42.86 | 3.72E-05 | ATP-dependent peptidase activity | >100 | 2.47E-16 | ||
| Proteolysis involved in protein catabolic process | 32.97 | 3.85E-09 | Serine-type endopeptidase activity | >100 | 2.31E-13 | ||
| Plastid organisation | 31.94 | 8.82E-05 | Serine hydrolase activity | >100 | 2.64E-12 | ||
| Protein catabolic process | 29.73 | 7.11E-09 | Endopeptidase activity | 64.09 | 7.51E-11 | ||
Discussion
The potential of crops to improve NUE has been harnessed to enhance global production by developing high-yielding cultivars (Pingali 2012). In recent years, numerous efforts have been made to create crop cultivars that can achieve equivalent or higher yields per input unit. However, the limited genetic variation in cultivated germplasm poses challenges, particularly for under-utilised crop plants like safflower. To effectively exploit natural diversity for safflower improvement, a global panel of 94 genotypes was evaluated in this study. We employed various morphological and photosynthetic traits as markers to characterise the phenotypic variation related to nitrogen uptake. A GWAS approach was used to identify the underlying genetic loci that could serve as DNA-based breeder-friendly markers.
Phenotypic variation
Evaluation of 94 safflower genotypes for various traits at different N treatments highlighted that all the measured traits, except chl b, were significantly different at the genotypic (G) and nitrogen (N) treatment levels. In addition, G × N interaction also imposed significant differences upon all traits. Overall, T3 (120 mg dm−3 N) was identified as the optimal nitrogen dosage for safflower cultivation, leading to a significant increase in mean performance for all studied traits. Some genotypes showed consistent phenotypes across various treatments, while others exhibited significant differences in response to different treatments (Bocianowski et al. 2019). The significant G × N interaction indicated that the genotypes exhibited altered performance for the studied traits under varying nitrogen treatments.
The overall performance of the germplasm in this study was notably affected by the different N treatments. For instance, PH increased by 3.39%, 5.08% and 2.82% in T2, T3 and T4, respectively, compared to the control treatment (T1), consistent with findings by Ferreira Santos et al. (2018). Semi-dwarf safflower PH was observed within the nitrogen range of 200 to 250 kg ha−1 (Ferreira Santos et al. 2018). FSW increased by 21.43%, 50.00% and 35.71% in T2, T3 and T4, respectively, compared to T1. Hasan et al. (2017) reported similar statistically significant changes in lettuce fresh weight/plant at varying N doses applied at different days after transplanting, with an increase in fresh weight/plant linked to optimal vegetative development. Higher leaf area resulting from nitrogen fertilisation promotes increased solar radiation incidence, carbon assimilation, growth and production (Cruz et al. 2007).
RL increased by 6.26%, 11.33% and 9.75% in T2, T3 and T4, respectively, compared to T1, while FRW increased by 29.41%, 50.00% and 33.82% in T2, T3 and T4, respectively. Moderate nitrogen fertilisation (240 kg ha−1) enhanced RL, root surface area and root biomass in cotton crop, resulting in a considerable increase in total root growth and biomass (Chen et al. 2020). Chl a increased by 14.29%, 21.43% and 23.81%, and chl b increased by 16.67%, 26.67% and 20.00% in T2, T3 and T4, respectively, compared to T1. Additionally, C increased by 27.03%, 32.43% and 48.65%, and CT increased by 20.37%, 26.85% and 33.33% in T2, T3 and T4, respectively, compared to T1. BY increased by 16.81%, 39.50% and 24.37% in T2, T3 and T4, respectively, compared to T1. These findings are consistent with previous research by Dordas and Sioulas (2008) and Bonfim-Silva et al. (2015) that demonstrated increased leaf nitrogen concentration, chlorophyll content, photosynthetic rate and stomatal conductance in response to elevated nitrogen levels. Nitrogen fertilisation also substantially increased safflower biomass (BY) (Dordas and Sioulas 2008). Furthermore, our results align with previous studies by Shahrokhnia and Sepaskhah (2016), which reported that nitrogen availability enhanced the productive components of safflower crops. Overall, T3 exhibited the highest increases in all studied traits compared to T1, T2 and T4, indicating that T3 is the optimum nitrogen level for safflower crop germination and growth.
Correlation analysis
Correlation analysis is a valuable tool for exploring the associations between various traits, enabling us to use the obtained information as a selection criterion for crop improvement (Ali et al. 2020a). In this study, we observed positive correlations of BY with FSW, RL, FRW and NOL. Additionally, all nitrogen treatments exhibited positive correlations among C, chl a, chl b and CT (Fig. 2). Adu et al. (2018) also reported significant correlations between FRW and leaf area as well as RL (r = 0.72 and 0.79, respectively). Furthermore, they found positive and significant correlations between FSW and FRW (r = 0.81*), RL and FSW (r = 0.84*) and total RL and FRW (r = 0.76*). The chlorophyll content of leaves indicates their nitrogen status, as nitrogen is crucial for chlorophyll formation (Linder 1980; Muñoz-Huerta et al. 2013). The findings of Yora et al. (2018), demonstrating a positive link between C and chl a, chl b and CT in okra (Abelmoschus esculentus) fruit, strongly support our results. Czyczyło-Mysza et al. (2013) also observed a positive correlation between CT and C.
Principal component analysis
Using PCA, the 94 safflower genotypes were grouped into three distinct clusters: Cluster 1 (29 genotypes); Cluster 2 (39 genotypes); and Cluster 3 (26 genotypes) (Fig. 3). PCA allowed us to break down extensive data collections into manageable and unrelated groups, potentially indicating underlying linkages. Selecting genotypes with optimum performance can be challenging due to multiple features to consider. Several multivariate techniques, such as cluster analysis, factor analysis and PCA, along with other indices, have been developed to address this issue (Akram et al. 2022). In our case, Group 3 exhibited the highest mean values for PH (1.86), FSW (0.66), NOL (8.97), RL (17.63 cm), FRW (1.16 g) and BY (1.82 g). Similarly, the safflower genotypes clustered in Group 2 displayed the highest mean values for chl a (0.53), chl b (0.38), C (0.51) and CT (1.42). Furthermore, the best-performing safflower genotypes were found in Group 3.
Selection of best-performing genotypes
Means of the genotype, population mean and percentage increase were employed to identify superior genotypes (Table 5). Genotype 36 exhibited the most favourable ideotype, showing positive genetic improvements for all traits. Genotypes 36, 39 44, 51, 66, 78, 82 and 94 were the top-performing genotypes across all treatments. Under all four treatments, these genotypes exhibited superior performance for all studied traits. Utilising safflower breeding efforts, we can enhance features related to nitrogen usage efficiency by utilising genotypes with more favourable alleles.
Genome-wide association mapping
The identification of loci influencing essential plant traits is crucial for marker-assisted breeding to enhance crop productivity. However, there is a dearth of studies related markers/loci associated with agronomic traits in safflower (Hamdan et al. 2008, 2012; Mayerhofer et al. 2010; García-Moreno et al. 2011; Pearl et al. 2014; Ebrahimi et al. 2017; Ambreen et al. 2018; Ali et al. 2020b, 2021). Moreover, no study exists that has reported associations with traits under different nitrogen treatments in safflower. In the current study, we successfully identified silicoDArT markers associated with important morphological and photosynthetic traits under different nitrogen treatments (Table 4).
Remarkably, most of the MTAs were treatment-specific, which aligns with Sharma et al. (2023), who also reported inconsistencies in MTAs obtained under different nitrogen treatments in wheat. However, we found one marker (DArT-45481731) that was significantly associated with NOL in both T1 and T2. Moreover, some markers were found to control more than one trait. For instance, DArT-17812864 was associated with FRW and BY in T1, while DArT-15670279 was significantly associated with FRW and BY in T2. Similarly, DArT-45482737 was significantly associated with FRW and BY in T3.
Genetic dissection of various traits using the GWAS is rare and started recently (Pushpa et al. 2023). For instance, Zhao et al. (2022) detected five loci associated with multiple traits (days to flowering, plant height, seed weight, seed protein and oil contents) on pseudo chromosomes 1, 7, 8 (two MTAs) and 12. We also detected six MTAs on the pseudo-chromosome for NOL (two MTAs), BY (in T1), FRW (in T2) and FSW (two MTAs in T4) indicating that this chromosome carries important loci that can be harnessed to accelerate safflower breeding. However, Ambreen et al. (2018) detected several MTAs related to oil contents, linoleic acid, oleic acid, 100-seed weight, plant height, capitula per plant, number of primary branches and flowering time on all the pseudo chromosomes of safflower except pseudo chromosomes 8 and 12. We detected our MTAs related to 10 traits in four different treatments on all the pseudo-chromosomes except chromosome 11 (Fig. 5). Hence, a wide variety of loci are present in safflower that control different traits in peculiar ways.
Marker-associated loci, PPI network and GO enrichment analyses
The PPI network and GO enrichment results revealed that the markers found in T1 and T4 of Cluster 1 are associated with the control of gene expression through chromatin remodelling by histone modifications including methylation and acetylation. Recent research has emphasised how chromatin regulation, which involves histone modification and chromatin structure, can influence the functioning of nitrogen transporters (Javed et al. 2022). These types of modifications regulate root system architecture, N uptake, translocation and accumulation, N metabolism, and tolerance against N deficiency (Zhang et al. 2023a). H3K27ac and H3K27me3 dynamics were recently been shown to affect tolerance under low N availability in wheat (Zhang et al. 2023b). Identified markers in Cluster 2 of T1 and T2 are related to long-chain fatty acid biosynthesis that is required for membrane lipid production, wax and cuticle biogenesis in plants. The same GOs were also induced under N deficiency in wheat with higher NUE (Wang et al. 2022). The genes involved in these GOs were downregulated in Brassica napus roots when treated with nitrate and ammonium (Zhou et al. 2022) while they were upregulated in a wheat cultivar that can inhibit nitrifying soil bacteria by secreting a fatty alcohol compound into the rhizosphere (Dongwei et al. 2024), suggesting fatty acid metabolism is globally essential in NUE in plants.
Identified markers in Cluster 1 of T2 are related to branched-chain amino acid (such as valine and isoleucine) biosynthetic process. GABA branched-chain amino acids were shown to accumulate less in roots and shoots of barley (Hordeum vulgare) under N limitation (Decouard et al. 2022). Degradation of GABA branched-chain amino acids increases under N deficiency to recycle the N backbone of these amino acids into the production of other amino acids, C backbone to TCA cycle, and mitochondrial energy production (Hildebrandt et al. 2015). There were two genes (BCAT5 and KCS6/CER6/CUT1/POP1) in Cluster 1 of T2 that are related to GABA branched-chain amino degradation and recycling under N deficiency. Interestingly, the marker associated with BCAT5 was significantly related to both FRW and BY in T2 while the marker associated with KCS6/CER6/CUT1/POP1 was significantly related to NOL in T1 and T2, indicating that these two markers are very important for NUE in safflower and their associated loci should be studied further under N deficiency. It was suggested that targeting branched-chain AA catabolism may be a promising strategy for enhancing NUE in plants (Dellero 2020).
Finally, the identified markers in T3 and Cluster 1 of T4 are related to protein degradation, chloroplast organisation, and leaf morphogenesis. Under N limitation, rubisco levels decrease which leads to reduced photosynthesis and altered chloroplast organisation (Raigar et al. 2022). Indeed, N limitation was shown to downregulate the genes involved in chloroplast organisation in a sensitive rice (Oryza sativa) cultivar (Tantray et al. 2022). Overall, this process is associated with alterations in leaf morphogenesis to adapt to N limitation. The PPI help to get more information about the target loci. These interactions themselves might serve a crucial role to enhance the functioning of nitrogen transporter, and dynamic model of organisations, and very long chain fatty acid synthesis in safflower. Furthermore, this study provide foundation for the future research work on the acetylation, enzymatic regulations and methylation in safflower.
Our study is the first that elucidated the genetics and response of safflower under variable nitrogen applications. The present study will provide new insights into uncovering the genetic basis of essential safflower plant traits related to nitrogen-efficient nutrient utilisation. It will also establish a foundation for future functional genomic investigations. The identified markers will serve as a valuable resource for combining favourable alleles, enabling the development of superior safflower cultivars with enhanced nitrogen utilisation capabilities. Furthermore, markers obtained across multiple treatments and significantly associated with more than one trait are of particular significance, as they present higher chances of validation. The loci associated with identified markers represent novel genes that can be considered in further studies in dissecting the NUE in plants.
Conclusion
The development of nitrogen-efficient cultivars depends on natural genetic variation and the use of simpler phenotypic approaches compatible with breeding pipelines. In this study, we characterised the phenotypic and genotypic variation in an international safflower panel of 94 genotypes. Using the GWAS technique, we identified and selected the best-performing safflower genotypes based on their responsiveness to nitrogen application and the underlying genetic loci. This approach offers a promising alternative for the development of nitrogen-efficient safflower cultivars.
Data availability
All data reported, used and cited in this research are present in the main text or in the supplementary materials of the manuscript.
Conflicts of interest
Muhammad Azhar Nadeem is a Guest Editor of Functional Plant Biology. To mitigate this potential conflict of interest they were blinded from the review process. The authors declare no other conflicts of interest.
Declaration of funding
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) of the Ministry of Education (2019R1A6A1A11052070), Republic of Korea.
Author contributions
Data curation, methodology, visualisation, writing – original draft: Fawad Ali. Data curation, formal analysis: Mian A. R. Arif, Allah Bakhsh and Muhammad K. Ilyas. Methodology, investigation, writing – original draft: Muhammad A. Nadeem, Amjad Hameed and Arif Ali. Formal analysis, visualisation, writing – review and editing: Emre Aksoy and Yong S. Chung. Supervision, conceptualisation: Shahid U. Khan, Cemal Kurt and Dilek Tekdal. Supervision, conceptualisation, writing – review and editing: Faheem S. Baloch. Funding: Yong S. Chung and Faheem S. Baloch.
Acknowledgements
The authors are grateful to the Scientific and Technological Research Council of Türkiye (TUBITAK) for providing a pre-doctoral fellowship under the 2216 program to Fawad Ali.
References
Adu MO, Asare PA, Asare-Bediako E, Amenorpe G, Ackah FK, Afutu E, Amoah MN, Yawson DO (2018) Characterising shoot and root system trait variability and contribution to genotypic variability in juvenile cassava (Manihot esculenta Crantz) plants. Heliyon 4, e00665.
| Crossref | Google Scholar |
Akram S, Arif MAR, Hameed A (2021) A GBS-based GWAS analysis of adaptability and yield traits in bread wheat (Triticum aestivum L.). Journal of Applied Genetics 62, 27-41.
| Crossref | Google Scholar | PubMed |
Akram S, Ghaffar M, Wadood A, Shokat S, Hameed A, Waheed MQ, Arif MAR (2022) A GBS-based genome-wide association study reveals the genetic basis of salinity tolerance at the seedling stage in bread wheat (Triticum aestivum L.). Frontiers in Genetics 13, 997901.
| Crossref | Google Scholar |
Ali F, Yilmaz A, Chaudhary HJ, Nadeem MA, Rabbani MA, Arslan Y, Nawaz MA, Habyarimana E, Baloch FS (2020a) Investigation of morphoagronomic performance and selection indices in the international safflower panel for breeding perspectives. Turkish Journal of Agriculture and Forestry 44, 103-120.
| Crossref | Google Scholar |
Ali F, Nadeem MA, Barut M, Habyarimana E, Chaudhary HJ, Khalil IH, Alsaleh A, Hatipoğlu R, Karaköy T, Kurt C, Aasim M, Sameeullah M, Ludidi N, Yang SH, Chung G, Baloch FS (2020b) Genetic diversity, population structure and marker-trait association for 100-seed weight in international safflower panel using silicoDArT marker information. Plants 9, 652.
| Crossref | Google Scholar |
Ali F, Nadeem MA, Habyarimana E, Altaf MT, Barut M, Kurt C, Chaudhary HJ, Khalil IH, Yildiz M, Cömertpay G, Shahid MQ (2021) Identification of genetic basis associated with agronomic traits in a global safflower panelusing genome-wide association study. Turkish Journal of Agriculture and Forestry 45, 834-849.
| Crossref | Google Scholar |
Ali MA, Ghazy AI, Alotaibi KD, Ibrahim OM, Al-Doss AA (2022) Nitrogen efficiency indexes association with nitrogen recovery, utilization, and use efficiency in spring barley at various nitrogen application rates. Agronomy Journal 114, 2290-2309.
| Crossref | Google Scholar |
Ambreen H, Kumar S, Kumar A, Agarwal M, Jagannath A, Goel S (2018) Association mapping for important agronomic traits in safflower (Carthamus tinctorius L.) core collection using microsatellite markers. Frontiers in Plant Science 9, 402.
| Crossref | Google Scholar |
An D, Su J, Liu Q, Zhu Y, Tong Y, Li J, Jing R, Li B, Li Z (2006) Mapping QTLs for nitrogen uptake in relation to the early growth of wheat (Triticum aestivum L.). Plant and Soil 284, 73-84.
| Crossref | Google Scholar |
Anicésio ECAD, Bonfim-Silva EDNA, Silva TJAD, Pacheco AB (2018) Nitrogen and potassium in safflower: chlorophyll index, biometric characteristics and water use efficiency. Revista Caatinga 31, 424-433.
| Crossref | Google Scholar |
Arif MAR, Nagel M, Neumann K, Kobiljski B, Lohwasser U, Börner A (2012) Genetic studies of seed longevity in hexaploid wheat using segregation and association mapping approaches. Euphytica 186, 1-13.
| Crossref | Google Scholar |
Arif MAR, Nagel M, Lohwasser U, Börner A (2017) Genetic architecture of seed longevity in bread wheat (Triticum aestivum L.). Journal of Biosciences 42, 81-89.
| Crossref | Google Scholar | PubMed |
Arif MAR, Shokat S, Plieske J, Ganal M, Lohwasser U, Chesnokov YV, Kocherina NV, Kulwal P, Kumar N, McGuire PE, Sorrells ME, Qualset CO, Börner A (2021) A SNP-based genetic dissection of versatile traits in bread wheat (Triticum aestivum L.). The Plant Journal 108, 960-976.
| Crossref | Google Scholar | PubMed |
Arif MAR, Agacka-Mołdoch M, Qualset CO, Börner A (2022) Mapping of additive and epistatic QTLs linked to seed longevity in bread wheat (Triticum aestivum L.). Cereal Research Communications 50, 709-715.
| Crossref | Google Scholar |
Baloch FS, Alsaleh A, Shahid MQ, Çiftçi VE, Sáenz de Miera LE, Aasim M, Nadeem MA, Aktaş H, Özkan H, Hatipoğlu R (2017) A whole genome DArTseq and SNP analysis for genetic diversity assessment in durum wheat from central fertile crescent. PLoS ONE 12, e0167821.
| Crossref | Google Scholar |
Bocianowski J, Niemann J, Nowosad K (2019) Genotype-by-environment interaction for seed quality traits in interspecific cross-derived Brassica lines using additive main effects and multiplicative interaction model. Euphytica 215, 7.
| Crossref | Google Scholar |
Bonfim-Silva EM, Paludo JTS, Sousa JVR, de Freitas Sousa HH, da Silva TJA (2015) Development of safflower subjected to nitrogen rates in cerrado soil. American Journal of Plant Sciences 6, 2136-2143.
| Crossref | Google Scholar |
Bradbury PJ, Zhang Z, Kroon DE, Casstevens TM, Ramdoss Y, Buckler ES (2007) TASSEL: software for association mapping of complex traits in diverse samples. Bioinformatics 23, 2633-2635.
| Crossref | Google Scholar | PubMed |
Chen J, Liu L, Wang Z, Zhang Y, Sun H, Song S, Bai Z, Lu Z, Li C (2020) Nitrogen fertilization increases root growth and coordinates the root–shoot relationship in cotton. Frontiers in Plant Science 11, 880.
| Crossref | Google Scholar | PubMed |
Cruz JL, Pelacani CR, Carvalho JEBD, Souza Filho LFDS, Queiroz DC (2007) Níveis de nitrogênio e a taxa fotossintética do mamoeiro “golden”. Ciência Rural 37, 64-71.
| Crossref | Google Scholar |
Czyczyło-Mysza I, Tyrka M, Marcińska I, Skrzypek E, Karbarz M, Dziurka M, Hura T, Dziurka K, Quarrie SA (2013) Quantitative trait loci for leaf chlorophyll fluorescence parameters, chlorophyll and carotenoid contents in relation to biomass and yield in bread wheat and their chromosome deletion bin assignments. Molecular Breeding 32, 189-210.
| Crossref | Google Scholar | PubMed |
Cömertpay G, Baloch FS, Kilian B, Ülger AC, Özkan H (2012) Diversity assessment of Turkish maize landraces based on fluorescent labelled SSR markers. Plant Molecular Biology Reporter 30, 261-274.
| Crossref | Google Scholar |
Decouard B, Bailly M, Rigault M, Marmagne A, Arkoun M, Soulay F, Caïus J, Paysant-Le Roux C, Louahlia S, Jacquard C, Esmaeel Q, Chardon F, Masclaux-Daubresse C, Dellagi A (2022) Genotypic variation of nitrogen use efficiency and amino acid metabolism in barley. Frontiers in Plant Science 12, 807798.
| Crossref | Google Scholar |
Dellero Y (2020) Manipulating amino acid metabolism to improve crop nitrogen use efficiency for a sustainable agriculture. Frontiers in Plant Science 11, 602548.
| Crossref | Google Scholar |
Deng Z, Cui Y, Han Q, Fang W, Li J, Tian J (2017) Discovery of consistent QTLs of wheat spike-related traits under nitrogen treatment at different development stages. Frontiers in Plant Science 8, 2120.
| Crossref | Google Scholar |
Dongwei D, Mingkun M, Xiaoyang Z, Yufang L, Kronzucker HJ, Weiming S (2024) Potential secretory transporters and biosynthetic precursors of biological nitrification inhibitor 1,9-decanediol in rice as revealed by transcriptome and metabolome analyses. Rice Science 31, 87-102.
| Crossref | Google Scholar |
Dordas CA, Sioulas C (2008) Safflower yield, chlorophyll content, photosynthesis, and water use efficiency response to nitrogen fertilization under rainfed conditions. Industrial Crops and Products 27, 75-85.
| Crossref | Google Scholar |
Doyle JJ, Doyle JL (1990) Isolation of plant DNA from fresh tissue. Focus 12, 39-40.
| Google Scholar |
Ebrahimi F, Majidi MM, Arzani A, Mohammadi-Nejad G (2017) Association analysis of molecular markers with traits under drought stress in safflower. Crop & Pasture Science 68, 167-175.
| Crossref | Google Scholar |
Elshire RJ, Glaubitz JC, Sun Q, Poland JA, Kawamoto K, Buckler ES, Mitchell SE (2011) A robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PLoS ONE 6, e19379.
| Crossref | Google Scholar |
Epskamp S, Cramer AOJ, Waldorp LJ, Schmittmann VD, Borsboom D (2012) qgraph: network visualizations of relationships in psychometric data. Journal of Statistical Software 48, 1-18.
| Crossref | Google Scholar |
Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14, 2611-2620.
| Crossref | Google Scholar |
Evans JR, Terashima I (1987) Effects of nitrogen nutrition on electron transport components and photosynthesis in spinach. Australian Journal of Plant Physiology 14, 59-68.
| Google Scholar |
FAO (2019) World fertilizer trends and outlook to 2020. Available at http://www.FAO.org/3/a-i6895e.pdf.
Ferreira Santos R, Bassegio D, Pereira Sartori MM, Zannoto MD, de Almeida Silva M (2018) Safflower (Carthamus tinctorius L.) yield as affected by nitrogen fertilization and different water regimes. Acta Agronómica 67, 264-269.
| Google Scholar |
Fredeen AL, Gamon JA, Field CB (1991) Responses of photosynthesis and carbohydrate-partitioning to limitations in nitrogen and water availability in field-grown sunflower. Plant Cell & Environment 14, 963-970.
| Crossref | Google Scholar |
García-Moreno MJ, Fernández-Martínez JM, Velasco L, Pérez-Vich B (2011) Molecular tagging and candidate gene analysis of the high gamma-tocopherol trait in safflower (Carthamus tinctorius L.). Molecular Breeding 28, 367-379.
| Crossref | Google Scholar |
Gilbert NW, Tucker TC (1967) Growth, yields, and yield components of safflower as affected by source, rate, and time of application of nitrogen. Agronomy Journal 59, 54-56.
| Crossref | Google Scholar |
Hamdan YAS, Pérez-Vich B, Fernández-Martínez JM, Velasco L (2008) Inheritance of very high linoleic acid content and its relationship with nuclear male sterility in safflower. Plant Breeding 127, 507-509.
| Crossref | Google Scholar |
Hamdan YAS, García-Moreno MJ, Fernández-Martínez JM, Velasco L, Pérez-Vich B (2012) Mapping of major and modifying genes for high oleic acid content in safflower. Molecular Breeding 30, 1279-1293.
| Crossref | Google Scholar |
Hasan MR, Tahsin AKMM, Islam MN, Ali MA, Uddain J (2017) Growth and yield of lettuce (Lactuca sativa L.) influenced as nitrogen fertilizer and plant spacing. Journal of Agriculture and Veterinary Science 10, 62-71.
| Google Scholar |
Hazzouri KM, Khraiwesh B, Amiri KMA, Pauli D, Blake T, Shahid M, Mullath SK, Nelson D, Mansour AL, Salehi-Ashtiani K, Purugganan M, Masmoudi K (2018) Mapping of HKT1;5 gene in barley using GWAS approach and its implication in salt tolerance mechanism. Frontiers in Plant Science 9, 156.
| Crossref | Google Scholar | PubMed |
Hickman JE, Palm CA, Mutuo P, Melillo JM, Tang J (2014) Nitrous oxide (N2O) emissions in response to increasing fertilizer addition in maize (Zea mays L.) agriculture in western Kenya. Nutrient Cycling in Agroecosystems 100, 177-187.
| Crossref | Google Scholar |
Hildebrandt TM, Nesi AN, Araújo WL, Braun HP (2015) Amino acid catabolism in plants. Molecular Plant 8, 1563-1579.
| Crossref | Google Scholar | PubMed |
Javed T, Singhal RK, Shabbir R, Shah AN, Kumar P, Jinger D, Dharmappa PM, Shad MA, Saha D, Anuragi H, Adamski R, Siuta D (2022) Recent advances in agronomic and physio-molecular approaches for improving nitrogen use efficiency in crop plants. Frontiers in Plant Science 13, 877544.
| Crossref | Google Scholar |
Jin FX, Ji SD, Xie XB, Kang JW, Ju HG, Ahn SN (2011) Detection of epistatic interaction of two QTLs, gw8.1 and gw9.1, underlying grain weight using nearly isogenic lines in rice. Breeding Science 61, 69-75.
| Crossref | Google Scholar |
Jones JP, Tucker TC (1968) Effect of nitrogen fertilizer on yield, nitrogen content, and yield components of safflower. Agronomy Journal 60, 633-634.
| Crossref | Google Scholar |
Kassambara A, Mundt F (2017) Package ‘factoextra. Extract and visualize the results of multivariate data analyses. Available at https://rpkgs.datanovia.com/factoextra/index.html
Kilian A, Wenzl P, Huttner E, Carling J, Xia L, Blois H, Caig V, Heller-Uszynska K, Jaccoud D, Hopper C, Aschenbrenner-Kilian M (2012) Diversity arrays technology: a generic genome profiling technology on open platforms. In ‘Data production and analysis in population genomics’. (Eds P François, B Aurélie) pp. 67–89. (Springer: Totowa, NJ, USA)
Korte A, Farlow A (2013) The advantages and limitations of trait analysis with GWAS: a review. Plant Methods 9, 29.
| Crossref | Google Scholar | PubMed |
Koutroubas SD, Papakosta DK, Doitsinis A (2004) Cultivar and seasonal effects on the contribution of pre-anthesis assimilates to safflower yield. Field Crops Research 90, 263-274.
| Crossref | Google Scholar |
Koutroubas SD, Papakosta DK, Doitsinis A (2008) Nitrogen utilization efficiency of safflower hybrids and open-pollinated varieties under Mediterranean conditions. Field Crops Research 107, 56-61.
| Crossref | Google Scholar |
Kumari S, Choudhary RC, Kumara Swamy RV, Saharan V, Joshi A, Munot J (2017) Assessment of genetic diversity in safflower (Carthamus tinctorius L.) genotypes through morphological and SSR marker. Journal of Pharmacognosy and Phytochemistry 6, 2723-2731.
| Google Scholar |
Lamesch P, Berardini TZ, Li D, Swarbreck D, Wilks C, Sasidharan R, Muller R, Dreher K, Alexander DL, Garcia-Hernandez M, Karthikeyan AS, Lee CH, Nelson WD, Ploetz L, Singh S, Wensel A, Huala E (2012) The Arabidopsis Information Resource (TAIR): improved gene annotation and new tools. Nucleic Acids Research 40(D1), D1202-D1210.
| Crossref | Google Scholar |
Laperche A, Le Gouis J, Hanocq E, Brancourt-Hulmel M (2008) Modelling nitrogen stress with probe genotypes to assess genetic parameters and genetic determinism of winter wheat tolerance to nitrogen constraint. Euphytica 161, 259-271.
| Crossref | Google Scholar |
Li H, Vikram P, Singh RP, Kilian A, Carling J, Song J, Burgueno-Ferreira JA, Bhavani S, Huerta-Espino J, Payne T, Sehgal D, Wenzl P, Singh S (2015) A high density GBS map of bread wheat and its application for dissecting complex disease resistance traits. BMC Genomics 16, 216.
| Crossref | Google Scholar | PubMed |
Lichtenthaler HK, Buschmann C (2001) Chlorophylls and carotenoids: measurement and characterization by UV-VIS spectroscopy. Current Protocols in Food Analytical Chemistry 1, F4.3.1-F4.3.8.
| Google Scholar |
Linder S (1980) Chlorophyll as an indicator of nitrogen status of conifer needles. New Zealand Journal of Forestry Science 1, 166-175.
| Google Scholar |
Losos JB, Arnold SJ, Bejerano G, Brodie ED, III, Hibbett D, Hoekstra HE, Mindell DP, Monteiro A, Moritz C, Orr HA, Petrov DA (2013) Evolutionary biology for the 21st century. PLoS Biology 11, e1001466.
| Crossref | Google Scholar |
Mahjourimajd S, Kuchel H, Langridge P, Okamoto M (2016) Evaluation of Australian wheat genotypes for response to variable nitrogen application. Plant and Soil 399, 247-255.
| Crossref | Google Scholar |
Marinova E, Riehl S (2009) Carthamus species in the ancient Near East and south-eastern Europe: Archaeobotanical evidence for their distribution and use as a source of oil. Vegetation History and Archaeobotany 18, 341-349.
| Crossref | Google Scholar |
Mayerhofer R, Archibald C, Bowles V, Good AG (2010) Development of molecular markers and linkage maps for the carthamus species C. tinctorius and C. oxyacanthus. Genome 53, 266-276.
| Crossref | Google Scholar |
Monostori I, Szira F, Tondelli A, Árendás T, Gierczik K, Cattivelli L, Galiba G, Vágújfalvi A (2017) Genome-wide association study and genetic diversity analysis on nitrogen use efficiency in a Central European winter wheat (Triticum aestivum L.) collection. PLoS ONE 12, e0189265.
| Crossref | Google Scholar |
Muñoz-Huerta RF, Guevara-Gonzalez RG, Contreras-Medina LM, Torres-Pacheco I, Prado-Olivarez J, Ocampo-Velazquez RV (2013) A review of methods for sensing the nitrogen status in plants: advantages, disadvantages and recent advances. Sensors 13, 10823-10843.
| Crossref | Google Scholar |
Nadeem MA, Gündoğdu M, Ercişli S, Karaköy T, Saracoğlu O, Habyarimana E, Lin X, Hatipoğlu R, Nawaz MA, Sameeullah M, Ahmad F (2020) Uncovering phenotypic diversity and DArTseq marker loci associated with antioxidant activity in common bean. Genes 11, 36.
| Crossref | Google Scholar |
Nadeem MA, Habyarimana E, Karaköy T, Baloch FS (2021) Genetic dissection of days to flowering via genome-wide association studies in Turkish common bean germplasm. Physiology and Molecular Biology of Plants 27, 1609-1622.
| Crossref | Google Scholar |
Peakall R, Smouse PE (2006) GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6, 288-295.
| Crossref | Google Scholar |
Pearl SA, Bowers JE, Reyes-Chin-Wo S, Michelmore RW, Burke JM (2014) Genetic analysis of safflower domestication. BMC Plant Biology 14, 43.
| Crossref | Google Scholar |
Pingali PL (2012) Green revolution: impacts, limits, and the path ahead. Proceedings of the National Academy of Sciences 109, 12302-12308.
| Google Scholar |
Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155, 945-959.
| Crossref | Google Scholar |
Pushpa HD, Kumaraswamy HH, Thomas HB, Ushakiran B, Sharma D, Anjani K, Sujatha M (2023) Innovative approaches for genetic improvement of safflower (Carthamus tinctorius L.): current status and prospectus. In ‘Smart plant breeding for field crops in post-genomics era’. (Eds D Sharma, S Singh, SK Sharma, R Singh) pp. 293–342. (Springer Nature Singapore: Singapore)
Raigar OP, Mondal K, Sethi M, Singh MP, Singh J, Kumari A, Sekhon BS (2022) Nitrogen use efficiency in wheat: genome to field. In ‘Wheat – Recent advances’. (IntechOpen) doi:10.5772/intechopen.103126
Raman H, Raman R, Nelson MN, Aslam MN, Rajasekaran R, Wratten N, Cowling WA, Kilian A, Sharpe AG, Schondelmaier J (2012) Diversity array technology markers: genetic diversity analyses and linkage map construction in rapeseed (Brassica napus L.). DNA Research 19, 51-65.
| Crossref | Google Scholar |
Russo TA, Tully K, Palm C, Neill C (2017) Leaching losses from Kenyan maize cropland receiving different rates of nitrogen fertilizer. Nutrient Cycling in Agroecosystems 108, 195-209.
| Crossref | Google Scholar |
Ryan J, Ibrikci H, Sommer R, McNeill A (2009) Nitrogen in rainfed and irrigated cropping systems in the Mediterranean region. Advances in Agronomy 104, 53-136.
| Google Scholar |
Sepaskhah AR, Barzegar M (2010) Yield, water and nitrogen-use response of rice to zeolite and nitrogen fertilization in a semi-arid environment. Agricultural Water Management 98, 38-44.
| Crossref | Google Scholar |
Shahrokhnia MH, Sepaskhah AR (2016) Effects of irrigation strategies, planting methods and nitrogen fertilization on yield, water and nitrogen efficiencies of safflower. Agricultural Water Management 172, 18-30.
| Crossref | Google Scholar |
Sharma A, Arif MA, Shamshad M, Rawale KS, Brar A, Burgueño J, Shokat S, Kaur R, Vikram P, Srivastava P, Sandhu N, Singh J, Kaur S, Chhuneja P, Singh S (2023) Preliminary dissection of grain yield and related traits at differential nitrogen levels in diverse pre-breeding wheat germplasm through association mapping. Molecular Biotechnology 65, 116-130.
| Crossref | Google Scholar |
Steer BT, Harrigan EKS (1986) Rates of nitrogen supply during different developmental stages affect yield components of safflower (Carthamus tinctorius L.). Field Crops Research 14, 221-231.
| Crossref | Google Scholar |
Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, Simonovic M, Doncheva NT, Morris JH, Bork P, Jensen LJ, von Mering C (2019) STRING v11: protein–protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Research 47(D1), D607-D613.
| Crossref | Google Scholar |
Tafteh A, Sepaskhah AR (2012) Yield and nitrogen leaching in maize field under different nitrogen rates and partial root drying irrigation. International Journal of Plant Production 6, 93-114.
| Google Scholar |
Tantray AY, Hazzazi Y, Ahmad A (2022) Physiological, agronomical, and proteomic studies reveal crucial players in rice nitrogen use efficiency under low nitrogen supply. International Journal of Molecular Sciences 23, 6410.
| Crossref | Google Scholar |
Thomas PD, Ebert D, Muruganujan A, Mushayahama T, Albou L-P, Mi H (2022) PANTHER: making genome-scale phylogenetics accessible to all. Protein Science 31, 8-22.
| Crossref | Google Scholar |
von Mering C, Jensen LJ, Snel B, Hooper SD, Krupp M, Foglierini M, Jouffre N, Huynen MA, Bork P (2005) STRING: known and predicted protein–protein associations, integrated and transferred across organisms. Nucleic Acids Research 33(suppl_1), D433-D437.
| Google Scholar |
Wang M, Jiang N, Jia T, Leach L, Cockram J, Waugh R, Ramsay L, Thomas B, Luo Z (2012) Genome-wide association mapping of agronomic and morphologic traits in highly structured populations of barley cultivars. Theoretical and Applied Genetics 124, 233-246.
| Crossref | Google Scholar |
Wang H, Ma Q, Shan F, Tian L, Gong J, Quan W, Yang W, Hou Q, Zhang F, Zhang S (2022) Transcriptional regulation mechanism of wheat varieties with different nitrogen use efficiencies in response to nitrogen deficiency stress. BMC Genomics 23, 727.
| Crossref | Google Scholar |
Xu Y, Wang R, Tong Y, Zhao H, Xie Q, Liu D, Zhang A, Li B, Xu H, An D (2014) Mapping QTLs for yield and nitrogen-related traits in wheat: influence of nitrogen and phosphorus fertilization on QTL expression. Theoretical and Applied Genetics 127, 59-72.
| Crossref | Google Scholar |
Yora M, Syukur M, Sobir S (2018) Characterization of phytochemicals and yield components in various okra (Abelmoschus esculentus) genotypes. Biodiversitas 19, 2323-2328.
| Crossref | Google Scholar |
Zhang H, Zhang X, Xiao J (2023a) Epigenetic regulation of nitrogen signaling and adaptation in plants. Plants 12, 2725.
| Crossref | Google Scholar |
Zhang H, Jin Z, Cui F, Zhao L, Zhang X, Chen J, Zhang J, Li Y, Li Y, Niu Y, Zhang W, Gao C, Fu X, Tong Y, Wang L, Ling HQ, Li J, Xiao J (2023b) Epigenetic modifications regulate cultivar-specific root development and metabolic adaptation to nitrogen availability in wheat. Nature Communications 14, 8238.
| Crossref | Google Scholar |
Zhao H, Savin KW, Li Y, Breen EJ, Maharjan P, Tibbits JF, Kant S, Hayden MJ, Daetwyler HD (2022) Genome-wide association studies dissect the G×E interaction for agronomic traits in a worldwide collection of safflowers (Carthamus tinctorius L.). Molecular Breeding 42, 24.
| Crossref | Google Scholar |
Zhou T, Wu P, Yue C, Huang J, Zhang Z, Hua Y (2022) Transcriptomic dissection of allotetraploid rapeseed (Brassica napus L.) in responses to nitrate and ammonium regimes and functional analysis of BnaA2.Gln1; 4 in Arabidopsis. Plant and Cell Physiology 63, 755-769.
| Crossref | Google Scholar |
Zia MAB, Demirel U, Nadeem MA, Çaliskan ME (2020) Genome-wide association study identifies various loci underlying agronomic and morphological traits in diversified potato panel. Physiology and Molecular Biology of Plants 26, 1003-1020.
| Crossref | Google Scholar |
