Abstract

Background

   Gamma ray induced mutation breeding has emerged as an excellent method
   for expedited development of improved varieties of rice, a staple food
   for more than half the world's population. However, the assessment of
   radiation induced variations are primarily phenotypic in nature. In
   this direction, evaluation of the metabolic signature of bio-active
   ingredients, which confer beneficial properties to rice, could be
   employed as a tool to select varieties which not only retain the health
   benefits of the parent variety but also exhibit better agronomic
   traits. The present study was, therefore, aimed at evaluating the
   metabolomic changes in the mutants of Gathuwan, an indigenous Indian
   rice with immunomodulatory properties. The mutant varieties were
   developed through gamma irradiation, and liquid chromatography-tandem
   mass spectrometry (LC–MS/MS)-based metabolic profiling was performed.

Results

   A total of 274 differentially expressed compounds were identified among
   Gathuwan and four of its mutants (mutant 6, mutant 7, mutant 8 and
   mutant 12), indicating that gamma irradiation induced stable metabolic
   alterations. Significant differences were observed in the phytochemical
   composition of mutants relative to the parent, emphasizing the
   importance of metabolic screening in functional rice breeding. Cluster
   analysis and phytochemical profiling revealed that mutant 6 was
   metabolically closest to the parent variety. Additionally, distinct
   phytohormonal variations among the mutants were observed which may
   account for the phenotypic differences in growth and development.

Conclusions

   Our study demonstrates that radiation-induced improvement in agronomic
   traits are accompanied by distinct alterations in phytochemical and
   phytohormonal profiles in stable rice mutants. These metabolic changes
   support the functional potential of the mutants and provide insights
   into the biological mechanisms underlying their traits. Among the
   mutants, mutant 6 emerges as a promising candidate due to its
   similarity to the parent in metabolite composition. Therefore,
   inclusion of metabolomic profiling as a selection criterion offers a
   powerful tool to identify robust and functionally superior rice
   varieties.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12870-025-07036-1.

   Keywords: Rice, Gamma rays, Mutation breeding, Metabolomics,
   Phytochemicals, Phytohormones

Background

   Rice is consumed by more than half of world’s population for calories
   and basic nutrition. In 2022–23, worldwide consumption of rice was
   520.4 million metric tons, of which a significantly large proportion
   was consumed by developing nations ([38]https://www.statista.com).
   Largescale dependency on rice makes it a crucial contributor to the
   global food security and captivates more emphasis toward sustainable
   rice production. However, two important aspects which need to be
   considered for ensuring food security in future are that rice is a crop
   with vast diversity in agro-morphological traits along with nutritional
   and sensory qualities. The most widely consumed form of rice is
   polished rice, wherein the processing stripes off most of the nutritive
   elements. Numerous studies have shown unpolished rice to be healthier
   than the polished rice due to retention of bran and germ, the
   anatomical fractions of rice containing many nutrients and
   phytochemicals [[39]1–[40]3]. Emerging evidence indicates that among
   the plethora of existing rice cultivars, the traditional ones which
   have adapted and stood the test of time over thousands of years,
   produce biochemicals with beneficial effects on human health and
   disease conditions [[41]4]. Many traditional rice varieties from
   countries like India, China, Thailand, Sri-Lanka etc. have been found
   to exhibit therapeutic effects as diverse as anti-oxidative,
   anti-allergic, anti-cancer, anti-diabetic, immune-stimulatory,
   cardioprotective, hepatoprotective etc., supporting activity based
   ethnic classifications of rice [[42]5–[43]9]. However, the lower yield
   of traditional varieties is a major drawback for their long-term
   commercial viability. Poor plant and panicle architecture along with
   long maturity duration are the factors responsible for the low yield.
   Hence there is a clear need to identify novel strategies that can be
   applied to overcome the undesirable traits associated with rice
   verities having health promoting properties.

   In this direction, inducing desired heritable changes in the phenotype
   of plants using physical mutagenic agents is an effective way to
   overcome the limitations of rice landraces within comparatively shorter
   span of time. Although radiation induced mutagenesis is a random
   process, it has the potential to generate wide combinations of traits
   providing opportunity to select and propagate the desirable one.
   Therefore, gamma ray induced mutagenesis has been extensively used for
   the improvement of yield, harvest time and efforts, nutrient use
   efficiency, stress tolerance and nutritional and organoleptic qualities
   in various crops [[44]10–[45]12], thus contributing significantly to
   food sustainability. Gamma rays induce broad spectrum mutations which
   manifests as diverse phenotypic, physiological and biochemical changes
   [[46]13–[47]15]. In crop improvement programs, major emphasis is laid
   on either phenotypic or physiological alterations. However, gamma rays
   have also been shown to alter contents of various biochemicals present
   in these varieties and also affect plant responses to various stresses
   [[48]16–[49]18]. These biochemical are often implicated in regulating
   the biological activities of different parts of plants and also known
   to be responsible for health promoting effects in humans
   [[50]19–[51]22]. In a study by Kim et al. (2004), metabolic changes led
   to balanced amino acid composition by enhancing the levels of limiting
   amino acids (lysine and tryptophan) in rice, along with increased
   protein content [[52]23]. Kongdin et al. (2023) also showed the effect
   of gamma rays on phenolic compound accumulation in rice under stress
   conditions [[53]24]. Taken together, these observations indicate that
   the mutant varieties may possibly acquire a different metabolic
   signature as compared to the parent variety due to altered biochemical
   pathways, particularly in case of plants of nutraceutical worth.

   It is well accepted that the metabolome of a plant is an aggregation of
   all the metabolites (primary and secondary) and is responsible for
   shaping the morphology and functional properties of the plant.
   Therefore, in recent years, there has been an increased focus on
   deciphering the metabolic signature of plant products associated with
   their biological activity in different organisms. A large number of
   metabolic investigations are being conducted in rice for diverse
   purposes such as metabolic profiling, discriminating different rice
   cultivars, plant interaction with biotic and abiotic factors, stress
   response and integrated studies to understand functional
   gene-metabolite relationships [[54]25–[55]30]. Hence, utilizing a
   metabolomics guided approach can prove to be a valuable tool for the
   improvement of indigenous rice having medicinal properties.

   In the present study, mutation breeding using gamma rays was employed
   to improve the agronomic traits of Gathuwan rice, a traditional rice
   variety originally collected from Abhanpur district of Chhattisgarh,
   India. It is being maintained as a part of India’s largest rice
   germplasm collection at Indira Gandhi Krishi Vishwavidyalaya (IGKV),
   Raipur, Chhattisgarh. We have recently showed that Gathuwan brown rice
   extract exhibits potent immune-suppressive functions, which may be
   responsible for its healing effects in the patients suffering from
   inflamed joints (arthritis) [[56]31]. The biological activity of
   Gathuwan rice was attributable to the enriched metabolites correlating
   with the functional response [[57]31]. This rice landrace, however,
   suffers marginalised cultivation mainly because of the tall height of
   the plant and longer maturity duration. Here, we have used gamma
   ray-induced mutation breeding to select the mutants with improved
   agronomic traits and employed metabolic profiling to identify the
   mutant which is closest to the parent. Our results highlight that
   metabolic profiling is particularly crucial for improving cultivars
   with biological properties resulting from interaction of specific
   components.

Methods

Mutant development

   For mutagenesis, pure seeds of Gathuwan rice were obtained from
   Department of Genetics and Plant Breeding, IGKV, Raipur in the year
   2019. The seeds (M0) were irradiated with 250 Gy of gamma rays at
   Nuclear Agriculture & Biotechnology Division (NA&BTD), Bhabha Atomic
   Research Centre (BARC), Mumbai. A population of 2500 plants was raised
   in M1 generation during Rabi season 2019–20 (Dry season 2019–20) and
   mother panicle of individual plant was harvested separately and kept in
   envelopes. M2 generation was raised using panicle-to-row method during
   Kharif season 2020 (Wet season 2020). Single panicle seeds were sown in
   single line at spacing of 15 cm, while inter-line distance was 20 cm. A
   thorough screening of approximately 18,000 plants in M2 generation for
   reduced plant height, early maturity duration and better yielding
   ability resulted in isolation of 9 mutants (plants with genetically
   altered phenotypes) with reduced plant height or flowering time or
   both. These mutants were tested for homozygosity in M3 generation
   during Rabi season 2020–21 (Dry season 2020–21), as the sowing was done
   using plant-to row method. They were further advanced to M4, M5 and M6
   generation for stabilization and agro-morphological evaluation during
   subsequent seasons. Four homozygous mutant lines in M6, along with the
   parent, were subjected to metabolomic analysis during Kharif season
   2022 (Wet season 2022). Comprehensive agronomic data was recorded, with
   plot replications, for mutants and parent in M6 generation.

LC–MS/MS profiling of Gathuwan and its mutants

Sample preparation

   Brown rice samples of Gathuwan and four mutants of Gathuwan, were
   ground to fine powder in liquid nitrogen. The extraction of metabolites
   from brown rice powder was carried out using method described by Lau et
   al. (2015), with few modifications [[58]32]. Briefly, 1 ml of
   extraction solvent mixture containing water, methanol and acetonitrile
   in 1:2:2 ratio, was added to the 50 mg of powdered brown rice samples.
   The samples were vortexed for 15 min at room temperature, followed by
   sonication (in water bath) for 15 min. The samples were centrifuged at
   10,000 rpm for 15 min at 4 °C. Supernatant was collected from each
   sample, transferred to vials for LC and then subjected to LC–MS/MS
   analysis.

Liquid chromatography with tandem mass spectrometry (LC–MS-MS) analysis

   The metabolite enriched extracts were injected into Agilent 1290
   infinity II liquid chromatography system with a C18 RRHD Zorbax column
   (20 × 150 mm, 1.8 μm particle size), coupled with QTRAP 6500 mass
   spectrometer (ABSciex). The separation of the metabolites was carried
   out using a 25-min LC (liquid chromatography) method. The mobile phase
   comprised of solvent A (0.1% formic acid in ultrapure water) and
   solvent B (0.1% formic acid in 90% acetonitrile) while the flow rate
   was kept constant at 0.25 ml/min. The gradient used in LC was: 2% B for
   1–10 min, 30% B at 10–14 mi, 60% B at 14–18 min, 95% B for 18–21 min
   and 2% B for 21–25 min. Analyst software version 1.6.3 was used for
   data acquisition and the Analyst Device Driver for setting the
   parameters for the analysis. Mass spectrometric data acquisition was
   carried out with IDA method (Information dependent acquisition) in low
   mass mode. The IDA method was built using the EMS (enhanced mass
   spectra) to EPI (enhanced product ion) modes. The top five spectra from
   the EMS mode were used for analysis in the EPI (MS/MS) mode, using high
   energy CID i.e. collision induced dissociation. The metabolite data was
   acquired in both positive and negative polarities at 4500 V and −4500 V
   respectively, with a probe temperature of 450 °C. The compound
   parameters were set at a declustering potential (DP) of 100 V,
   collision energy (CE) of 10 V. The data were acquired for all the
   samples, in triplicates, along with pooled Quality Control (QC). All
   the resultant files (wiff format) were analysed for metabolite hits.

Analysis for MS2 data

   The resultant files obtained from LC–MS/MS were directly uploaded for
   data processing using MZmine tool. The feature detection was carried
   out with an m/z tolerance of 0.05 Da. It was followed by chromatogram
   deconvolution using the noise amplitude algorithm with amplitude of
   noise as 1.5E2. Isotopes were detected using Isotopic peak grouper with
   an m/z tolerance of 0.25 Da and retention time tolerance of 0.2 min.
   Alignment of peaks was performed by Join Aligner with an m/z tolerance
   of 0.05 Da and a retention time tolerance of 0.5 min. After processing,
   the data was searched against the PlantCyc (BioCyc) metabolite database
   using MS2Compound software. The metabolite identification and
   annotation were carried out with a precursor mass tolerance of 0.05 Da
   and fragment mass tolerance of 0.5 Da. Adducts that were used for the
   positive mode were [M + H], [M + Na], [2 M + H] and [M + 2H] while, for
   the negative mode, [M-H], [2 M-H] and [M-2H] were used. The peak areas
   were mapped with their respective identifications. MS2 data matrix was
   subjected to final scrutiny for removal of any duplications.

Enrichment, pathway and statistical analyses

   Enrichment analysis of the annotated metabolites was carried out using
   Metaboanalyst 6.0 webserver. MS2 annotations were mapped against the
   server library. For unmatched compounds names, synonymous names were
   used from PubChem library. Kyoto Encyclopedia of Genes and Genomes
   (KEGG) based over-representation (enrichment) analysis of pathways and
   compound classes was performed for individual samples. Quantitative
   enrichment analysis was performed for pairwise comparison between each
   mutant and the parent. Sample data was normalized by median and log
   transformed for comparative weighted analyses. Statistical analyses,
   for pair-wise comparison of metabolite expression profiles of mutant
   and parent, included fold change analysis (FC threshold- 2.0), t-test,
   features correlation, Principal Component Analysis (PCA), Partial Least
   Squares-Discriminant Analysis (PLS-DA) and cluster analysis indicating
   the metabolite expression profiles of the samples. ANOVA was performed
   for simultaneous comparison of parent and all four mutants and
   dendrogram was built based on comparative results.

Results

Development of Gathuwan rice mutants

   In M2 generation, where the initial mutant screening was conducted,
   Gathuwan parent attained a height of 150 cm, reached 50% flowering in
   110 days and matured in 145 days. Following screening, 9 mutant plants
   exhibiting either reduced plant height (up to 30 cm shorter), earlier
   maturity (up to 15 days earlier), or both, were selected. In the M3
   generation, five segregating lines were discarded, and four homozygous
   mutant lines were advanced through M4 to M6 generations for
   stabilization and agro-morphological evaluation. Lines designated as
   mutant 6, mutant 7, mutant 8 and mutant 12 expressed stable expression
   of the traits in M3-M6 generations. In the M2 generation, plant height
   of these mutants ranged from 118 to 135 cm, with 50% flowering
   occurring between 95 and 98 days (Table [59]1). In the M6 generation,
   plant height ranged from 116 to 131 cm, and 50% flowering occurred
   between 91 and 102 days (Table [60]2). Among these, mutant 6 emerged as
   the most promising line, exhibiting consistent reduction in plant
   height and earlier flowering, along with significantly higher yield per
   plant compared to both the parent and other mutant lines. This increase
   in yield was primarily attributed to enhanced spikelet fertility. These
   findings highlight mutant 6 as a strong candidate for further
   evaluation and potential use in breeding programs with the aim of
   improving the yield and maturity in rice.

Table 1.

   Agronomic data of Gathuwan and its mutants isolated in M2 generation
   Name of genotype Days to 50% Flowering (days) Days to Maturity (days)
   Plant Height (cm) Panicle Length (cm) Flag Leaf Length (cm) Flag Leaf
   Width (cm) Effective Tillers/Plant Grain Yield/Plant (g)
   Gathuwan Parent 110 145 150 22 17.5 1.5 10 19
   Gathuwan Mutant-6 95 125 118 21 16.5 1.5 12 26
   Gathuwan Mutant-7 97 127 121 22 17.5 1.5 9 23
   Gathuwan Mutant-8 98 128 128 22 17.3 1.5 9 21
   Gathuwan Mutant-12 95 125 135 23 16.8 1.6 8 20
   [61]Open in a new tab

Table 2.

   Comprehensive agronomic data recorded for Gathuwan and its mutants in
   M6 generation. (n = 10)
   Name of genotype Days to 50% flowering Plant Height (cm) Panicle Length
   (cm) Total Tillers/plant Productive Tillers/Plant Flag Leaf Length (cm)
   Flag Leaf Width (cm) Sterile Spikelets/Panicle Fertile
   Spikelets/Panicle Total spikelets/Panicle Spikelet fertility % Grain
   Yield per plant (g)
   Gathuwan (Parent) 111 ± 1.41 152.25 ± 6.01 21.84 ± 1.18 8.50 ± 0.71
   8.50 ± 0.71 21.50 ± 0.71 1.51 ± 0.02 25.50 ± 3.54 88 ± 2.83
   113.50 ± 0.71 77.54 ± 2.98 18.50 ± 0.71
   Mutant 6 94 ± 1.41 126.75 ± 3.18 23.5 ± 0.71 10.5 ± 0.71 10 ± 0.00
   21.06 ± 0.86 1.49 ± 0.08 16 ± 2.83 83 ± 2.83 99 ± 0.00 83.84 ± 2.86
   23.5 ± 0.71
   Mutant 7 98 ± 1.41 116.5 ± 2.12 21.5 ± 0.71 8.5 ± 0.71 8 ±.00 20 ± 1.41
   1.44 ± 0.06 25.5 ± 2.12 74.5 ± 0.71 100 ± 1.41 74.51 ± 1.76 18.5 ± 0.71
   Mutant 8 102 ± 0.00 126 ± 1.41 20.5 ± 0.71 8.5 ± 0.71 8.5 ± 0.71
   20.5 ± 0.71 1.42 ± 0.03 24 ± 2.83 78.5 ± 0.71 102.5 ± 3.54 76.62 ± 1.95
   21.5 ± 0.71
   Mutant 12 91 ± 1.41 131 ± 1.41 21.5 ± 2.12 8 ± 1.41 8 ± 1.41
   21.5 ± 0.71 1.49 ± 0.03 26.5 ± 2.12 84 ± 2.83 110.5 ± 4.95 76.04 ± 0.85
   19.5 ± 0.71
   [62]Open in a new tab

   (The phenotypic data represents average values of plot replicates. In
   each replication, average values of measurements from five plants were
   considered)

Metabolite analysis

   A total of 274 compounds corresponding to well annotated MS2 peaks,
   were used for comparison between parent and mutants. These compounds
   belonged to various classes, viz., alkaloids, benzenoids, lignans,
   lipids, nucleosides, phenylpropanoids and polyketides, organic acids,
   organoheterocyclic, organic nitrogen, organic oxygen and organosulfur
   compounds. In the metabolic profiling, Gathuwan (parent) brown rice
   exhibited 141 compounds while brown rice of mutant 6, mutant 7, mutant
   8 and mutant 12 respectively showed 145, 113, 112 and 103 compounds,
   respectively. The number of compounds commonly shared among
   parent-mutant pair were 66, 57, 65 and 61 respectively. Out of these,
   28 metabolites were present in all of the groups whereas 44, 36, 16, 8
   and 16 compounds were unique to parent, mutant 6, mutant 7, mutant 8
   and mutant 12 respectively (Fig. [63]1A). mutant 6 had highest
   percentage (45.5%) of total parental metabolites. In order to study the
   functional relevance of metabolic changes, the metabolites were
   classified into primary and secondary metabolite classes as shown in
   Fig. [64]1B (S2.1 of Supplementary file S1).

Fig. 1.

   [65]Fig. 1
   [66]Open in a new tab

   Distribution of metabolites detected using LC–MS/MS among samples
   studied (A) Vein diagram showing common and unique metabolites among
   Gathuwan and its mutants and (B) Bar graph showing number of
   metabolites corresponding to different compound classes. Other
   phytochemicals include cyanohydrins, glucosinolates, phytocannabinoids
   etc. Other compounds include intermediary metabolites involved in the
   synthesis of final products

   To explore clustering patterns based on differential metabolic
   features, unsupervised principal component analysis (PCA) was performed
   to visualize group separation without prior classification. To further
   identify significant differences among groups, supervised partial least
   squares discriminant analysis (PLS-DA) was conducted, which maximizes
   group separation and highlights key discriminant metabolites. Pairwise
   comparison results from these analyses are discussed in the subsequent
   sections. Data normalization procedures and sample quality assessments
   are illustrated in Supplementary Figures [67]S1A–B, while the accuracy,
   goodness-of-fit (R^2), and predictive ability (Q^2) metrics that
   validate the quality of the multivariate models are presented in
   Supplementary Figures [68]S2A–D. These steps confirm the robustness and
   reliability of the statistical analyses performed.

Enrichment analysis

   Over-representation analysis of metabolic pathways in Gathuwan parent
   and mutants showed differences in relative significance and enrichment
   ratios (Fig. [69]2A-E). Enrichment of arginine biosynthesis, branched
   chain amino acid synthesis and Vitamin B6 metabolism pathways in
   Gathuwan are considered to be related with its immune-suppressive
   activity [[70]31]. We compared pathway enrichment among mutants with
   the parent variety and observed that all these three pathways were
   among top 25 highly enriched pathways in Gathuwan, wherein arginine
   biosynthesis had highest enrichment ratio and lowest p-value. Among
   mutants, mutant 6 and 12 had lower while mutant 7 had higher enrichment
   of arginine biosynthesis. mutant 8 and 12 had highest enrichment of
   branched chain fatty acid synthesis, followed sequentially by mutant 6,
   parent and mutant 7. However, enrichment ratios of metabolism of
   arginine and branched chain amino acids were similar among the samples
   undertaken. Vitamin B6 metabolism was highly perturbed in mutants as it
   was not among top 25 enriched pathways in mutants. Since phenylalanine
   is a substrate for synthesis of many secondary metabolites, synthesis
   and metabolism of phenylalanine was also compared among the samples. It
   was observed that phenylalanine metabolism was dominant over its
   synthesis in parent and mutant 6, with more enrichment in latter.
   mutant 7 also had higher prevalence of phenylalanine metabolism but it
   could be accounted to enriched synthesis. The two processes were
   mutually balancing in mutant 8 and mutant 12.

Fig. 2.

   [71]Fig. 2
   [72]Open in a new tab

   Metaboanalyst 6.0 based pathway enrichment analysis in (A) Gathuwan,
   (B) mutant 6, (C) mutant 7, (D) mutant 8 and (E) mutant 12. Top 25
   Pathways are arranged on the basis of their significance, which is also
   indicated by their significance values. Length of each bar represents
   the enrichment ratio while intensity of colour of the bar indicates
   their significance

Pathway analysis

   Pathway impact analysis is a measure to decipher the significant
   pathways based on abundance and topological importance of metabolites.
   parent variety had higher impact scores for secondary metabolite
   synthesis including betalain and isoquinoline alkaloids and tryptophan
   metabolism. Higher impact of both these secondary metabolite synthesis
   pathways was also observed among mutant 6, mutant 8 and mutant 12.
   mutant 8 had higher impact of tryptophan metabolism also among these
   three mutants. mutant 7 had comparatively higher impact scores but
   lower impact of pathways indicating a localized metabolic response than
   pathway alteration (Fig. [73]3A-E).

Fig. 3.

   [74]Fig. 3
   [75]Open in a new tab

   Metabolomic Pathway Analysis (MetPA) to highlight significant impact of
   pathways, based on topology and impact scores in (A) Gathuwan, (B)
   mutant 6, (C) mutant 7, (D) mutant 8 and (E) mutant 12. Pathways are
   displayed as circles on a plot between pathway impact score (x-axis)
   and log p-values (y-axis). The size of circle corresponds to impact
   score, while the intensity of color depends on the p-values. Highly
   impacted pathways with high statistical significance are mentioned in
   the plots

Metabolic diversification among rice mutants

   For testing the significantly altered features among mutants, fold
   change analysis (Threshold 2.0) and t-test (FDR adjusted p-value
   threshold 0.05) were employed, results of which are shown in Fig. [76]4
   A-D and Fig. S3 A-D (Supp. Figures) respectively. Fold change analysis
   of metabolites in mutants 6, 7, 8, and 12, with respect to the parent,
   showed the number of upregulated metabolites to be 92, 81, 68, and 68
   respectively, while downregulated metabolites numbered 110, 91, 93, and
   89 The number of metabolites showing statistically significant
   differences in mean abundance compared to the parent were 175, 142,
   139, and 109 for mutants 6, 7, 8, and 12, respectively (Table [77]S2.2
   in Supplementary File S1). The volcano plots shown in Fig. [78]4 E–H
   combine fold change and significance to identify biologically
   meaningful metabolic changes. Across all mutants, significantly
   downregulated metabolites outnumbered upregulated ones. Among these,
   mutant 6 had the highest number of upregulated and significant features
   compared to the parent.

Fig. 4.

   [79]Fig. 4
   [80]Open in a new tab

   Univariate analysis to identify significantly altered features between
   Gathuwan and mutants. A-D Fold change analysis (mutant/parent) to
   highlight biological significance, and (E–H) Volcano plot to identify
   statistically significant features respectively in parent-mutant 6,
   parent-mutant7, parent-mutant 8 and parent-mutant 8 pairs

   Principal Component Analysis was used to reduce dimensionality of the
   data. The cumulative score of first two principal components (PCs) was
   more than 80 percent in parent-mutant 6, parent-mutant 7 and
   parent-mutant 8 pairs, of which more than 70% variance was accountable
   to PC1 (S2.3, Supp. File S1). Even with supervised partial least square
   difference analysis (PLS-DA), the cumulative effects of two components
   were similar, with variance explained by component 1 remaining same
   (Fig. [81]5 A-D). In case of parent-mutant 12 pair, PC1 could explain
   67.2% of the variance. The distribution areas of replications of same
   rice sample agglomerated in the same group without overlapping with
   that of other sample. A distinct grouping of mutant sample triplicates
   from that of parent indicated that the former differed significantly
   from the parent. Loadings plot (Fig. S4 A-D, Supp. Figures) shows the
   metabolic interrelationships and contribution to the two principal
   components. Figure [82]6A-D shows top 40 important features identified
   using variable importance in the projection (VIP) scores of PC1.
   Heatmaps (Fig. [83]7A) showing expression profiles of metabolites
   across parent and mutants indicated 44 metabolites are downregulated in
   all the mutants, which were mainly concerned with amino acid
   metabolism. In cluster analysis, mutant 6 was found to be most similar
   to the parent than any other mutant (Fig. [84]7B). Triplicates of all
   samples showed good correlation amongst themselves and correlation
   among features led to distinct clusters of metabolites (Fig. S5 A-B,
   Supp. Figures).

Fig. 5.

   [85]Fig. 5
   [86]Open in a new tab

   Multivariate analysis for visualization of the difference between
   Gathuwan and its mutants, in reduced dimensions. Scores plot showing
   variant grouping of (A) parent and mutant 6, (B) parent and mutant 7,
   (C) parent and mutant 8 and (D) parent and mutant 12 in PLS-DA analysis

Fig. 6.

   [87]Fig. 6
   [88]Open in a new tab

   Top 40 important features identified using variable importance in the
   projection (VIP) scores of PC1 component, respectively in (A)
   parent-mutant 6, (B) parent-mutant7, (C) parent-mutant 8 and (D)
   parent-mutant 8 pairs

Fig. 7.

   [89]Fig. 7
   [90]Open in a new tab

   Cluster analysis to visualize the metabolic proximity between Gathuwan
   and its mutants. A Heatmaps showing expression profiles of all
   metabolites across the samples. The indicator bar on the left side of
   each heatmap shows the expression level marked by colour intensity.
   (For convenience, the heatmap has been fragmented into three parts) and
   (B) Dendrogram built on the basis on clustering of samples based on
   overall pattern of expression

Phytochemical composition analysis

   The trend in changes in the pools of secondary and primary metabolites
   in mutants were compared relative to the parent. More variability was
   observed in the number of metabolites belonging to different groups of
   secondary than primary metabolites. Number of terpenes and alkaloidal
   compounds were found to be increased in majority of mutants, while a
   decrease was observed in lipid compounds. Pools of amino acids
   (including proteinogenic and non-proteinogenic) were relatively more
   uniform in terms of number of metabolites, compared to any other group
   (Fig. [91]8A). However, the proportion and metabolism was different, as
   indicated by pathway analysis. The top 25 metabolic sets altered in
   mutants with respect to the parent and expression profiles of major
   groups of natural products have been shown in Fig. S6 A-D and Fig. S7
   A-D respectively. Among all mutants, mutant 6 had similar pools of
   majority of compound classes including amino acid, carbohydrate and
   vitamin contents and increased pools of terpenes, alkaloids and
   sterols, relative to the parent. Therefore, the expression pattern of
   parental bio-active metabolite sets was further explored in mutant 6
   (Fig. [92]8 B-D & S2.4 of Supp. file S1). With respect to the
   hydroxybenzaldehydes set of compounds, relative content of benzaldehyde
   and protocatechuic acid was higher in mutant 6 compared to the parent,
   while the content of hydroxybenzyl alcohol and dihydroxybenzaldehyde
   was lower. This suggested an increased reduction and reduced
   hydroxylation of phenolic compounds in the mutant. Compounds including
   Indole-methyl acetate, Indole-3-acetonitrile and N-acetylserotonin were
   enriched indoles among mutant 6. Their higher content may be a result
   of more conversion of tryptophan into indoles. Pyridoxamine (Vitamin B6
   forms) levels were downregulated in mutant 6, as was the pathway of its
   synthesis. However, 6,7-dimethyl-8-(1-D-ribityl)lumazine, a precursor
   of riboflavin, and niacinamide (Vitamin B3), were enriched instead.
   Thus, regarding the expression profiles of bioactive compound classes,
   mutant 6 is expected to retain its functional integrity.

Fig. 8.

   [93]Fig. 8
   [94]Open in a new tab

   Changes in metabolic profiles among mutants (relative to the parent) in
   relation to health effects. A Bar graph shows number of metabolites
   belonging to primary and secondary metabolite groups in different
   samples. B-D Box plots showing changes in the profiles of bio-active
   metabolic sets in mutant 6, relative to the parent. B
   Hydroxybenzaldehydes and Hydroxycinnamic acids, (C) Indoles and (D)
   Vitamin B complex components

Profiles of Plant growth regulators (PGRs) in relation to the plant height

   In the present study, metabolites belonging to different classes of
   PGRs, viz., Cytokinin (trans-zeatin-7-O-glucoside or Raphanatin),
   Gibberellin (Gibberellin A6), Brassinosteroid (26-hydroxycastasterone),
   Salicylic acid (SA)/Salicylate), indole precursors of auxin
   biosynthesis and precursor for ethylene biosynthesis
   (1-aminocyclopropane-1-carboxylate), were also found among the
   differential metabolites (Fig. [95]9A-G). Gibberellin, the major
   phytohormone regulating germination and plant height, was downregulated
   among all mutants, compared to the parent. Cytokinin, another classical
   plant hormone related to regulation of plant height, could not be
   observed among mutants, within the detection limits of the instrument.
   Indoleacetaldehyde and Indole-3-acetonitrile are immediate precursors
   of Indole acetic acid synthesis in parallel pathways [[96]33]. Both of
   these were detected in mutant 6, compared to presence of only the
   former one in the parent. SA was lower in all mutants compared to the
   parent, whereas brassinosteroid content was higher in majority of the
   mutants. Salicyclic acid levels governs plant resistance to pests but
   since the link between the hormonal level in seed and seedling is not
   distinctive, it cannot predict the mutant responses against pests. The
   immediate precursor of ethylene biosynthesis,
   1-aminocyclopropane-1-carboxylate, was detectable in mutant 6 and
   mutant 8 only.

Fig. 9.

   [97]Fig. 9
   [98]Open in a new tab

   Changes in phytohormonal profiles among mutants (relative to the
   parent). Violin plots showing changes in contents of phytohormones or
   their precursors. Violin plot for (A) Gibberellin A6, (B) Raphanatin,
   (C-D) two precursors of auxin, viz. C Indoleacetaldehyde and (D)
   Indoleacetonitrile, (E) Brassinosteroid, (F) Salicylate and (G)
   Precursor for Ethylene synthesis

Discussion

   Gamma ray mutagenesis has been widely used to generate genetic
   variation in crop plants, including rice. The present study demonstrate
   that gamma irradiation can be effectively utilized to generate rice
   genotypes with improved agronomic performance while retaining or
   enhancing the key metabolites associated with health benefits of the
   parent rice variety. In this study, four mutant lines were selected
   based on their stable expression of reduced plant height and maturity
   duration traits, from M3-M6 generations. Two of these mutants (mutant 6
   and mutant 7) exhibited higher yield than the parent variety. In view
   of the limited research on the metabolic effects of gamma radiation in
   genetically stabilized mutant population, we employed global
   metabolomics to evaluate stable gamma-induced rice mutants in
   comparison to the parent line, with the objective to identify lines
   that combine agronomic advantage with potential nutraceutical value. We
   observed a heterogenous pattern of metabolic changes among all four
   Gathuwan mutants, despite their relatively similar phenotype. Whether
   this is attributable to random and dispersed small genetic changes
   caused by the radiation or a consequence of interaction with matter led
   metabolic changes inherited further, is not clear.

   Global metabolomics revealed that, as compared to the parent, Vitamin
   B6 pathways was significantly affected among mutants and most of the
   downregulations were observed in metabolites pertaining to amino acid
   metabolism. Notably, the metabolism of arginine and branched-chain
   amino acids did not show significant alterations, unlike their
   synthesis. Therefore, these pathways do not provide substantial
   insights into the functional potency of the mutants compared to the
   parent. Further, phenylalanine metabolism was among the enriched
   pathway in all samples. The observed differences in content of
   phenolics such flavonoids, isoflavanoids, lignins, diarylhepatnoids
   implied changes in the relative importance of pathway nodes diverging
   from phenylalanine metabolism. Notably, mutant 6 expressed compounds
   from key functional phytochemical classes (e.g., alkaloids, sterols,
   and terpenes) that were absent in the parent line, while also
   minimizing the loss of existing metabolites. This makes mutant 6 a
   promising candidate for selection in medicinal applications. Compounds
   belonging to phenolics group were least affected in mutant 6. Although
   several metabolite groups such as indoles, hydroxybenzaldehydes,
   hydroxycinnamic acids and pyridoxamines, which have been previously
   linked to the functional roles of Gathuwan brown rice [[99]31], were
   altered in mutant 6, they appeared to be chemically modified to potent
   molecules with similar activity and mechanisms. This study, therefore,
   provides an integrated evaluation of the balance between biosynthetic
   and metabolic pathways, with a focus on the changes in overall
   metabolite pools rather than individual compounds. It also emphasizes
   on the functional relevance of these metabolic changes to identify the
   most promising candidate mutant.

   Further, we observed that classical phytohormones, GA, CK, IAA and
   ethylene were generally downregulated in mutants. Since rapid and
   uniform germination has been linked to variation in plant height
   differences [[100]34], it is plausible that plant height is influenced
   by the distribution of these phytohormones in different parts of the
   rice seed. They regulate the seed germination and have a decisive
   impact on plant height. Among them, GA and CK, are key regulators of
   germination [[101]35, [102]36]. In our study, reduced levels of
   gibberellins and zeatin in rice grain of mutants correlated with
   reduction in plant height. This hormone interaction is further
   supported by the known involvement of GA and CK crosstalk in regulating
   plant height through the OsNAC103 transcription factor [[103]37].
   Although brassinosteroids-GA interaction is also known to regulate
   plant height [[104]38], brassinosteroid levels in this case, may either
   not have a role in plant height regulation or the levels observed in
   mutant 6, 7 and 8 are high enough to exert suppressive effects on GA
   biosynthesis. Notably, mutant 6 contains immediate precursors of both
   IAA biosynthesis and ethylene biosynthesis, suggestion it may have
   distinct developmental pattern than the parent and other mutants.
   Higher spikelet fertility, a trait affecting the yield, has been
   associated with enclosed stigma under high heat conditions and GA
   application antagonizes the process by promoting exsertion of stigma
   [[105]39]. Further, Ethylene affects grain filling process by affecting
   amino acid biosynthesis [[106]40]. mutant 6, with reduced gibberellin
   and higher spikelet fertility, and higher ethylene content might be
   conforming to similar mechanism of increased yield under high heat
   conditions and thus, provides yet another prospect for which the mutant
   can be evaluated.

   Besides causing a wide range of mutations, Gamma rays may also trigger
   random epigenetic changes, particularly demethylation, which may
   significantly influence gene expression [[107]41]. Increasing evidence
   shows that epigenetic regulation, through DNA methylation, histone
   modifications, and small RNAs, plays a critical role in controlling
   plant metabolic networks, especially those governing stress responses
   and secondary metabolism [[108]42, [109]43]. These modifications
   influence gene expression both locally and distantly, shaping the
   plant’s developmental and physiological outcomes and may exhibit
   transgenerational inheritance [[110]43, [111]44]. Thus, the metabolic
   shifts observed among mutants may not be solely attributed to genetic
   mutations but could also reflect underlying epigenetic modifications
   induced by gamma irradiation. Future studies integrating epigenomic
   profiling will be essential to uncover the specific methylation
   dynamics associated with key metabolic traits. This will not only
   enhance our understanding of radiation-induced variation but also
   support the use of epigenetic diversity as a valuable tool in crop
   improvement, offering new avenues for trait enhancement without genetic
   erosion. The study, in addition to highlighting the potential changes
   in the biological activity of mutants and predicting their development
   and physiological responses to stress, also raises interesting
   possibilities to explore.

Conclusion

   In this study, metabolic profiles of agronomically superior mutants in
   the genetic background of Gathuwan, were analysed to assess their
   biochemical similarity to the parent. Among the mutants, mutant
   6—characterized by reduced plant height, early maturity, and higher
   yield—retained the highest percentage of parental metabolites and
   exhibited the most significant upregulation of features relative to the
   parent. The pool size of functional metabolite sets was also better
   preserved in mutant 6 than other mutants. Additionally, the bioactive
   metabolite sets and their functionality in the parent variety were
   largely preserved, despite a few modifications. Based on these
   findings, mutant 6 was identified as the closest to the parent in terms
   of phytochemical composition and as a promising candidate with respect
   to both agronomic and biochemical traits. The study demonstrates the
   potential of biochemical screening to aid in the selection of rice
   mutants with health-promoting properties. The integrative data provide
   a predictive basis for identifying the most promising candidate, in
   terms of bioactivity and agronomic performance. However, bioactivity
   testing and targeted quantification of metabolites are needed to
   validate these predictions.

Supplementary Information

   [112]Supplementary Material 1.^ (4.7MB, docx)
   [113]Supplementary Material 2.^ (17.5KB, xlsx)

Acknowledgements