Sowing date restructures carbon partitioning between defence and nutritional susceptibility to modulate stem borer resistance in sweet sorghum

Abstract

Sowing date acts as a deterministic regulator of plant-pest phenological alignment, however, its mechanistic role in restructuring host physiological and metabolic networks remains inadequately resolved. This study deciphers how sowing time restructures photosynthetic performance, structural composition, and biochemical metabolism to mediate stem borer (Chilo partellus) infestation in sweet sorghum. Four genotypes, categorised as susceptible (SUGARGRAZE, CSV 24SS) and tolerant (SPV 3078, SPV 3083), were evaluated under three different dates of sowing during the 2024 kharif season: 8^th June (D₁), 22^nd June (D₂), and 6^th July (D₃). Delayed sowing elevated mean infestation by 65.92%, from 41.67% (D₁) to 69.17% (D₃). Tolerant genotypes maintained photosynthetic stability, exhibiting higher net assimilation (23.45 µmol CO₂ m^-2 s^-1) and stomatal conductance (0.26 mol H₂O m^-2 s^-1) alongside reduced internal CO₂, concurrent with enhanced deposition of structural carbohydrates (cellulose, hemicellulose) and lignification. In contrast, susceptible genotypes displayed a metabolic shift toward nutritional enrichment, characterised by elevated accumulations of total soluble sugars, starch, protein, and free amino acids, traits strongly correlated with infestation severity (r = 0.97). Principal component analysis revealed two antagonistic trait assemblies; a defense module integrating photosynthetic performance, structural polymers, and secondary metabolites (phenolics, tannins, and saponins), and a susceptibility module comprising primary nutritional metabolites. The defense module was negatively correlated with infestation (r = -0.72 to -0.95). Early sowing promoted an integrated resistance phenotype through regulated carbon partitioning toward physical and biochemical defenses, whereas delayed sowing disrupted metabolic homeostasis, enhancing host nutritional quality and stem borer suitability. These results establish a physiological and metabolic framework in which sowing date directly modulates host plant resistance trajectories, providing a mechanistic basis for agronomic optimisation and trait-based selection in sweet sorghum.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-57668-y.

Introduction

Sweet sorghum (Sorghum bicolor L. Moench) is a versatile C₄ crop valued for its high biomass, substantial stalk sugar content, and adaptability to marginal and semi-arid environments, making it a crucial resource for food, fodder, and bioenergy production¹. As the fourth-largest global producer, India’s estimated output of 6.00 million metric tons constitutes approximately 9.5% of world production^2, underscoring the crop’s national agricultural significance. The crop accumulates high concentrations of readily fermentable soluble sugars in its stalks, predominantly sucrose (~75%), with minor proportions of fructose and glucose (~2.6%)³. However, this elevated sugar content concurrently heightens susceptibility to insect pests, most notably the spotted stem borer (Chilo partellus). This pest infests sorghum from early vegetative stages through harvest, resulting in characteristic damage such as dead hearts, stem tunnelling, disruption of assimilate translocation, and significant yield losses ranging from 15% to over 80% depending on the infestation severity and crop growth stage⁴. In this context, host plant resistance (HPR) forms a cornerstone of sustainable integrated pest management⁵, particularly in crops grown for livestock feed or bioenergy, where chemical insecticide use is constrained by environmental and residue concerns.

HPR is a complex phenotype arising from the coordinated interaction of structural, physiological, and biochemical traits⁶. Structural attributes such as cell wall components can physically impede insect penetration and reduce tissue digestibility, providing a first line of defence against larval establishment⁷. Physiological processes, particularly photosynthesis and carbon allocation regulate plant vigour and the availability of carbon substrates for defence biosynthesis⁸. Biochemical traits, especially phenolic compounds, condensed tannins, and saponins, act as feeding deterrents and antinutritional agents that reduce larval performance and development⁹. Genotypic variation in these resistance-associated traits has been documented across diverse sorghum germplasm, with superior genotypes demonstrating constitutively elevated structural carbohydrates and secondary metabolite profiles that deter larval establishment¹⁰. Despite this documented variation, the mechanisms through which agronomic practices regulate and amplify or suppress this intrinsic resistance traits remain poorly understood.

Sowing date is a fundamental agronomic factor that modifies crop phenology and environmental exposure, thereby altering the synchrony between vulnerable crop developmental stages and peak pest populations. Recent studies across diverse crop systems have demonstrated that optimal sowing schedules provide a cost-free agronomic tool for modulating biotic stress outcomes across agro-ecological zones¹¹. In sorghum and related cereal system, delayed sowing has been consistently associated with altered temperature and photoperiod regimes during critical developmental windows, potentially disrupting the constitutive defence machinery of the host plant and elevating its nutritional suitability for herbivore development. However, while phenological escape through temporal desynchronisation of crop vulnerability and pest peak populations is a recognised benefit of early sowing¹², it does not fully explain the differential susceptibility responses observed among genotypes under comparable infestation pressure, implying that sowing date also actively restructures the intrinsic biochemical and physiological resistance landscape of the host.

Despite the agronomic importance of this interaction, the mechanistic role of sowing date in restructuring the carbon partitioning trade-off between primary nutritional metabolism and secondary defence metabolism under natural Chilo partellus infestation in sweet sorghum remains uncharacterised. Specifically, no study has simultaneously resolved how sowing date shapes photosynthetic performance, structural cell wall composition, and the full spectrum of primary and secondary metabolite profiles across genotypes that differ fundamentally in their constitutive resistance levels. This mechanistic gap is critical because understanding it would move the management paradigm beyond phenological escape toward agronomic optimisation of inherent plant resistance, a distinction with significant implications for breeding strategy and sustainable IPM design. Understanding whether sowing date driven metabolic reprogramming is constitutively expressed or environmentally induced is particularly important for trait-based selection, since only constitutive traits offer breeding stability across variable agronomic environments¹³.

We hypothesised that early sowing promotes a defence-oriented metabolic phenotype in tolerant genotypes by enhancing carbon allocation toward structural carbohydrates and secondary metabolites, while delayed sowing disrupts metabolic homeostasis by elevating host primary metabolite pools (soluble sugars, reducing sugars, protein, and free amino acids), thereby increasing host nutritional suitability and stem borer susceptibility. This study the first simultaneous, mechanistically integrated characterisation of sowing-date-driven restructuring of photosynthetic performance, structural cell wall composition, carbon partitioning dynamics, and primary and secondary metabolite profiles in sweet sorghum genotypes contrasting in Chilo partellus tolerance under natural field infestation conditions. Therefore, this study was designed to identify the optimal sowing date for minimising stem borer infestation and to elucidate how sowing date modulates these key resistance traits, thereby providing a mechanistic framework for agronomic optimisation and trait-based selection in sweet sorghum.

Materials and methods

Experimental site and damage assessment

This field study evaluated the effect of sowing date on natural infestation by the stem borer (Chilo partellus) and associated HPR traits in sweet sorghum. The experiment was conducted at the research farm of Punjab Agricultural University (PAU), Ludhiana, Punjab, India (30°54ʹ20ʺ N, 75°47ʹ43ʺ E, 247 m above sea level) during the kharif 2024 cropping season. The experiment assessed four genotypes SUGARGRAZE, CSV 24SS, SPV 3078, and SPV 3083 under three sowing dates 08 June 2024 (D₁), 22 June 2024 (D₂), and 06 July 2024 (D₃) in factorial randomised block design with three replications. Seeds were sown by kera method at a seed rate of 50 kg ha^-1. Fertiliser was applied as per PAU recommendations at 100 kg N, 20 kg P₂O₅, and 25 kg K₂O ha^-1. Half the recommended dose of nitrogen along with the full dose of phosphorus and potassium were applied as a basal dose at sowing, and the remaining half dose of nitrogen was top-dressed after the first irrigation at 30 DAS. The plots measured 5.5 m x 1 m with 60 cm between rows and 15 cm between plants. The experimental soil was sandy loam in texture and the crop was raised following the standard agronomic practices of PAU. The crop was maintained under natural stem borer pressure without insecticide application to enable an unbiased assessment of resistance. Meteorological data including maximum and minimum temperature, relative humidity, and rainfall recorded throughout the crop season at the PAU meteorological observatory are presented in Supplementary Figure S1. Soil physicochemical properties of the experimental field are provided in Supplementary Table S2. Uninfested controls were not included as the objective was to compare genotypic responses under identical pest pressure. Stem borer infestation was assessed at 30 days after sowing. Observations were recorded from ten randomly selected plants per genotype per replication. Plants exhibiting characteristic stem borer damage symptoms, including pin holes, shot holes, and exit holes on the stem, were classified as infested. The % of plant infestation was calculated using the following formula:

Based on the plant infestation (%), the four genotypes were categorized into tolerant and susceptible groups (Table 1).

Table 1.

Plant infestation (%) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates.

Sr. No	Date of sowing	Plant infestation (%)	Genotypes	Plant infestation (%)
1	08.06.2024 (D1)	41.67c	SUGARGRAZE	75.56A
2	22.06.2024 (D2)	59.17b	CSV 24SS	73.33A
3	06.07.2024 (D3)	69.17a	SPV 3078	42.22B
			SPV 3083	35.56B

Data represent mean values; means followed by different lowercase letters indicate significant differences among dates of sowing, while means followed by different uppercase letters indicate significant differences among genotypes at p ≤ 0.05.

Photosynthetic efficiency

All physiological measurements were recorded at maximum sunshine hours from the uppermost fully expanded leaf at 30 DAS. Gas exchange parameters, including photosynthetic rate (A), stomatal conductance (gₛ), and internal CO₂ concentration (Cᵢ), were measured using a portable photosynthesis system (LCi, ADC Bioscientific Ltd., UK). Measurements were conducted under ambient CO₂ condition (~380 ppm) and natural irradiance, with photosynthetically active radiation incident on the leaf surface maintained at ≤ 1400 µmol m^-2 s^-1.

Structural cell wall components

Leaf and stem tissue were collected at 30 DAS, cut into small pieces and dried at a temperature of 50°-60°C, and processed by grinding in a mechanical grinder. Structural carbohydrates were analysed according to Van Soest et al.¹⁴. Acid detergent fibre (ADF) content was estimated from 1 g of dried plant sample transferred to a spoutless beaker, followed by addition of 100 ml of acid detergent solution. The contents were refluxed for 1 hr, after which the mixture was filtered through a previously weighed sintered glass crucible. The residue was washed with hot distilled water, followed by a single wash with acetone. The crucibles containing the residue were dried in a hot air oven at 80°C overnight, cooled, and weighed. ADF content was calculated and expressed as a percentage of dry weight using the following formula:

Neutral detergent fibre (NDF) content was estimated using 1 g of dried sample transferred to a spoutless beaker, to which 100 ml of neutral detergent solution was added. The mixture was refluxed for 1 hr after the onset of boiling. The contents were then filtered through a previously weighed sintered glass crucible and washed with hot distilled water, followed by a final wash with acetone. The crucibles containing the residue were dried in a hot air oven at 80°C overnight, cooled, and weighed. NDF was calculated and expressed as a percentage of dry weight using the following formula. Hemicellulose content was estimated indirectly by calculating the difference between NDF and ADF.

Acid detergent lignin (ADL) content was determined from the ADF residue by treatment with 72% sulphuric acid (w/w) and kept at room temperature for 3 hr to dissolve cellulose, leaving lignin as residue. After incubation, the acid was drained, and the residue was washed with distilled water. The crucibles were then dried in a hot air oven, cooled, and weighed. Subsequently, the crucibles were ignited in a muffle furnace at 500°C for 3 hr to ash the lignin. After ignition, the crucibles were cooled in a desiccator and weighed again. ADL content was calculated and expressed as a percentage of dry weight using the following formula. Cellulose content was estimated indirectly by calculating the difference between ADF and ADL.

Where, W = sample weight (g), W₁ = weight of crucible + ADL residue (g), and W₂ = weight of crucible + ash (g). ADL (%) is expressed as lignin content.

Biochemical parameters

Total soluble sugars were estimated spectrophotometrically according to the phenol-sulphuric acid method ¹⁵. Fresh leaf tissue (0.1 g) was homogenised and extracted with 80% ethanol. The extract was centrifuged, and an aliquot of the supernatant (1.0 ml) was reacted with 1.0 ml of 5% phenol and 5.0 ml of concentrated sulphuric acid. The developed orange-yellow colour was measured at 490 nm against a reagent blank. Sugar content was quantified using a standard calibration curve prepared with glucose (10-50 µg). Starch content was determined from the ethanol insoluble residue obtained after soluble sugar extraction, following the protocol of Clegg¹⁶. The dried residue was suspended in distilled water, followed by hydrolysis with perchloric acid under continuous stirring. After dilution and centrifugation, an aliquot of the supernatant (1.0 ml) was mixed with 1.0 ml of 5% phenol and 5.0 ml of concentrated sulphuric acid. The mixture was cooled, and absorbance was recorded at 490 nm. Starch content was calculated from glucose equivalents using a conversion factor of 0.9. Reducing sugars were assayed with the dinitrosalicylic acid (DNSA) method¹⁷. A plant extract aliquot (1.0 ml) was combined with 2.0 ml of distilled water and 3.0 ml of DNSA reagent. The mixture was heated in a boiling water bath, stabilised by adding 1.0 ml of potassium sodium tartrate solution, and cooled. Absorbance was measured at 510 nm, and reducing sugar concentration was determined using a glucose standard curve (20-100 µg).

Total soluble protein was quantified using the method of Lowry et al.¹⁸. Fresh tissue (0.2 g) was homogenised and extracted with 0.1 N sodium hydroxide. Proteins in the extract were precipitated with 20% trichloroacetic acid, and the resulting pellet was dissolved in 0.5 N sodium hydroxide. The protein solution was then reacted with Folin-ciocalteu reagent. After colour development, the absorbance was measured at 520 nm. Protein concentration was determined from a standard curve prepared with bovine serum albumin (20-100 µg). Free amino acids were estimated following the ninhydrin method¹⁹. An aliquot of the plant extract was combined with ninhydrin reagent and heated in a boiling water bath. The developed purple colour was measured spectrophotometrically at 570 nm. The concentration of free amino acids was calculated using a calibration curve prepared with L-glycine (20-100 µg).

Total phenolic content was measured using the Folin-Ciocalteu method according to Swain & Hills²⁰. Fresh tissue (0.5 g) was refluxed with 5.0 ml of 80% methanol and filtered. An aliquot of the methanolic extract was evaporated to dryness, and the residue was dissolved in 6.5 ml of distilled water. To this solution, 0.5 ml of Folin-ciocalteu reagent was added. After 5 minutes, 1.0 ml of sodium carbonate solution was added. Following incubation, the absorbance was recorded at 760 nm. Phenolic content was quantified using a calibration curve prepared with gallic acid standard (10-100 µg). Condensed tannins were estimated using the vanillin-HCl assay²¹. Fresh plant tissue (0.5 g) was extracted with 2.0 ml of 1% methanolic hydrochloric acid. An aliquot of the methanolic-HCl extract was reacted with vanillin reagent and incubated. The developed colour was measured spectrophotometrically at 500 nm. Tannin content was determined from a standard curve prepared with catechin (20-100 µg). Saponins were quantified following the method described by Shiau et al.²². Fresh plant tissue (0.5 g) was extracted with 5.0 ml of ethyl acetate and incubated for 24 hours. An aliquot of the ethyl acetate extract was reacted with 1.0 ml of anisaldehyde reagent and 2.0 ml of concentrated sulfuric acid. The mixture was incubated at 60°C, after which the absorbance was read at 435 nm. Saponin content was calculated based on a saponin-equivalent standard curve (50-300 µg). All biochemical results are expressed as mg per gram fresh weight (mg g^-1 FW).

Statistical analysis

Data are presented as mean ± SE (standard error) of three biological replicates. For all physiological, structural, and biochemical measurements, a single representative measurement was recorded per genotype per replication, yielding three observations per treatment combination corresponding to the three field replications. Stem borer infestation was assessed on ten randomly selected plants per genotype per replication, and the percentage of infested plants per replication was calculated and used as a single observational unit in the statistical analysis. Prior to analysis of variance (ANOVA), the assumptions of normality and homogeneity of variance were assessed using Shapiro-Wilk test and Levene’s test, respectively. All variables satisfied both assumptions (p > 0.05) and no data transformation was required. A two-way ANOVA under a factorial randomised block design was performed to evaluate the main effects of sowing date, genotype and their interaction (Table 4), where ANOVA indicated significant main effects or interactions. Post-hoc mean separation was conducted using Tukey’s HSD test at p ≤ 0.05. All ANOVA procedures were executed using SPSS software version 16.0, and significant differences among means are indicated by superscript letters in tables and figures. Pairwise linear correlations among all measured traits were assessed using Pearson’s correlation coefficient (r), with significance levels assigned at p ≤ 0.05 (^*) and p ≤ 0.01 (^**) (Fig. S3-S6). All bivariate correlation analyses were conducted in R version 4.3.2. To elucidate higher order structure in the dataset, Principal Component Analysis (PCA) was performed on the standardised trait matrix using SPSS. The percentage of total variance explained by each principal component is reported in the Results section. The PCA biplot graphically displays the covariation among genotypes, sowing dates, and the composite photosynthetic, biochemical, and structural profiles.

Table 4.

Factorial analysis of variance showing F-statistics and associated probability values (F, p) for the effects of genotype (G), date of sowing (DOS), and their interaction (G x DOS) on plant infestation percentage and structural, photosynthetic and biochemical traits in sweet sorghum genotypes evaluated across three sowing dates under natural stem borer infestation during the kharif season 2024.

Source	dF	PI (F,P)	Hemicellulose (F,P)	Cellulose (F,P)	Lignin (F,P)	PE (F,P)	gS (F,P)	iCO2 (F,P)	TSS (F,P)	Starch (F,P)	R. Sugar (F,P)	TPC (F,P)	FAA (F, P)	Phenol (F, P)	Tannins (F, P)	Saponin (F, P)
Genotypes (G)	3	41.16, 0.0000	18.75, 0.0000	13.67, 0.0000	5.96, 0.0039	8.45, 0.0006	7.60, 0.0000	9.72, 0.0003	182.76, 0.0000	189.91, 0.0000	26.27, 0.0000	33.07, 0.0000	11.71, 0.0001	119.19, 0.0000	142.72, 0.0000	84.93, 0.0000
Date of Sowing (DOS)	2	24.75, 0.0000	3.16, 0.0064	6.46, 0.0062	4.74, 0.0019	41.13, 0.0000	13.53, 0.0001	8.90, 0.0015	66.36, 0.0000	3.26, 0.0404	22.63, 0.0000	23.36, 0.0000	23.00, 0.0000	98.91, 0.0000	73.01, 0.0000	58.87, 0.0000
G x DOS	6	0.86, 0.5406	0.06, 0.9991	1.12, 0.3817	0.58, 0.7396	2.35, 0.0660	1.55, 0.2076	1.55, 0.2095	2.34, 0.0672	17.61, 0.0000	0.13, 0.9911	0.90, 0.5107	1.30, 0.2964	0.65, 0.6889	1.87, 0.1322	1.26, 0.3144

Traits assessed include plant infestation (PI; %); structural cell wall components- hemicellulose, cellulose, and lignin (%); gas exchange parameters- photosynthetic rate (PE; µmol CO₂ m^-2 s^-1), stomatal conductance (gS; mol H₂O m^-2 s^-1), and intercellular CO₂ concentration (iCO₂; µmol mol^-1); and biochemical parameters- total soluble sugars (TSS; mg g^-1 FW), starch content (mg g^-1 FW), reducing sugars (mg g^-1 FW), total protein content (TPC; mg g^-1 FW), free amino acids (FAA; mg g^-1 FW), total phenolics (mg g^-1 FW), condensed tannins (mg g^-1 FW), and saponins (mg g^-1 FW). Degrees of freedom for genotype (df = 3), date of sowing (df = 2), genotype x date of sowing interaction (df = 6), and error (df = 22) are presented. Bold values indicate statistical significance at p ≤ 0.05

Results

Stem borer infestation assessment

Natural field infestation by Chilo partellus was significantly influenced by both genotype and sowing date, with delayed sowing consistently amplifying pest pressure across all genotypes (Table 4). Delaying sowing from 08 June 2024 (D₁) to 06 July 2024 (D₃) progressively increased the mean plant infestation from 41.67% to 69.17%, representing a 65.92% increase in pest pressure (p ≤ 0.05; Table 1). Genotypes were clearly segregated into susceptible (SUGARGRAZE, CSV 24SS) and tolerant (SPV 3078, SPV 3083) phenotypes based on their infestation responses. SUGARGRAZE sustained the highest infestation at 75.56%, which was 112.47% higher than the most tolerant genotype SPV 3083 at 35.56% (p ≤ 0.05; Table 1). CSV 24SS exhibited 73.33% infestation while SPV 3078 recorded 42.22%, demonstrating greater tolerance than susceptible genotypes across all sowing dates (Table 1).

Structural cell wall components

Structural cell wall composition differed markedly between tolerant and susceptible genotypes, with sowing date further modulating these differences across the experimental period (Table 4). Tolerant genotypes SPV 3083 and SPV 3078 exhibited significantly higher concentrations of all structural cell wall components compared to susceptible genotypes across all sowing dates (p ≤ 0.05; Table 2). At D₁, SPV 3083 contained 33.60% hemicellulose, 46.68% cellulose, and 1.75% lignin, representing increases of 32.81%, 18.72%, and 52.17% respectively over SUGARGRAZE (p ≤ 0.05). Similarly, SPV 3078 showed 24.38% and 8.41% greater hemicellulose and cellulose than SUGARGRAZE at D₁ (p ≤ 0.05). Across all sowing dates, SPV 3083 consistently recorded the highest structural carbohydrate concentrations, followed by SPV 3078, with susceptible genotypes CSV 24SS and SUGARGRAZE occupying the lower end of this spectrum throughout the experimental period (p ≤ 0.05; Table 2).

Table 2.

Structural carbohydrate constituents (hemicellulose, cellulose, and lignin) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates

Genotypes	Date of Sowing	Hemicellulose (%)	Cellulose (%)	Lignin (%)
SUGARGRAZE	08.06.2024	25.30 ± 1.54aC	39.32 ± 1.03aB	1.15 ± 0.03aB
	22.06.2024	24.70 ± 1.42aC	38.67 ± 0.37aB	1.05 ± 0.09aB
	06.07.2024	22.33 ± 2.09aC	38.17 ± 0.85aB	1.01 ± 0.10aB
CSV 24SS	08.06.2024	28.70 ± 1.42aB	39.52 ± 0.92aB	1.20 ± 0.12aB
	22.06.2024	27.77 ± 0.90aB	38.80 ± 0.80aB	1.12 ± 0.10aB
	06.07.2024	26.33 ± 1.09aB	37.75 ± 0.78aB	1.08 ± 0.10aB
SPV 3078	08.06.2024	31.47 ± 0.43aAB	42.63 ± 0.82aA	1.45 ± 0.14aAB
	22.06.2024	30.13 ± 0.91abAB	41.16 ± 1.13aA	1.34 ± 0.20aAB
	06.07.2024	28.74 ± 0.55bAB	40.27 ± 0.52aA	1.13 ± 0.23aAB
SPV 3083	08.06.2024	33.60 ± 1.39aA	46.68 ± 2.66aA	1.75 ± 0.14aA
	22.06.2024	32.30 ± 1.37aA	44.43 ± 2.71abA	1.44 ± 0.17aA
	06.07.2024	31.53 ± 2.66aA	40.32 ± 1.27bA	1.22 ± 0.32aA

Each value represent mean ± SE of triplicates; means followed by different lowercase letters indicate significant differences among dates of sowing, while means followed by different uppercase letters indicate significant differences among genotypes at p ≤ 0.05.

Photosynthetic efficiency

Photosynthetic performance was differentially modulated by both genotype and sowing date, with tolerant genotypes sustaining significantly superior gas exchange characteristics across all three sowing dates (Table 4). Notably, sowing date emerged as the dominant driver of photosynthetic rate, exerting a stronger influence than genotype. At D₁, SPV 3083 recorded the highest photosynthetic rate of 23.45 µmol CO₂ m^-2 s^-1, which was 21.75% higher than SUGARGRAZE (19.26 µmol CO₂ m^-2 s^-1), coupled with the highest stomatal conductance of 0.26 mol H₂O m^-2 s^-1, which was 52.94% greater than SUGARGRAZE (p ≤ 0.05; Table 3). Concurrently, internal CO₂ concentration in SPV 3083 at D₁ was 96.00 µmol mol^-1, which was 34.09% lower than SUGARGRAZE (145.67 µmol mol^-1; p ≤ 0.05), indicating more efficient mesophyll carbon fixation in tolerant genotypes. This photosynthetic advantage was progressively compromised by delayed sowing, with photosynthetic rates declining by 23.83% in SPV 3083 and 14.43% in SUGARGRAZE from D₁ to D₃ (p ≤ 0.05), while internal CO₂ concentrations increased by 25.00% and 8.69% in the same genotypes respectively, suggesting progressive stomatal and non-stomatal limitation under delayed sowing conditions (Table 3).

Table 3.

Photosynthetic efficiency parameters (photosynthetic rate, stomatal conductance, and internal CO₂ concentration) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates.

Genotypes	Date of Sowing	Photosynthetic rate (µmol CO2 m-2 s-1)	Stomatal Conductance (mol H2O m-2 s-1)	iCO2 (µmol mol-1)
SUGARGRAZE	08.06.2024	19.26 ± 0.69aB	0.17 ± 0.01aB	145.67 ± 7.80aA
	22.06.2024	17.86 ± 0.86aB	0.15 ± 0.01aB	153.67 ± 11.29aA
	06.07.2024	16.48 ± 0.71aB	0.14 ± 0.01aB	158.33 ± 9.56aA
CSV 24SS	08.06.2024	20.74 ± 0.47aAB	0.19 ± 0.02aB	116.67 ± 5.36bAB
	22.06.2024	18.40 ± 0.55abAB	0.17 ± 0.01aB	122.00 ± 7.57bAB
	06.07.2024	17.43 ± 0.57bAB	0.16 ± 0.01aB	177.33 ± 8.11aAB
SPV 3078	08.06.2024	22.19 ± 0.86aA	0.20 ± 0.01aAB	108.33 ± 14.75aBC
	22.06.2024	21.21 ± 0.27aA	0.18 ± 0.02abAB	119.67 ± 14.17aBC
	06.07.2024	16.49 ± 1.03bA	0.17 ± 0.01bAB	134.33 ± 6.57aBC
SPV 3083	08.06.2024	23.45 ± 0.64aA	0.26 ± 0.01aA	96.00 ± 8.72aC
	22.06.2024	19.87 ± 0.38bA	0.18 ± 0.01bA	113.67 ± 13.59aC
	06.07.2024	17.86 ± 0.69cA	0.17 ± 0.01bA	120.00 ± 10.12aC

Each value represent mean ± SE of triplicates; means followed by different lowercase letters indicate significant differences among dates of sowing, while means followed by different uppercase letters indicate significant differences among genotypes at p ≤ 0.05.

Biochemical parameters

Primary metabolite accumulation revealed a clear nutritional enrichment pattern in susceptible genotypes that was progressively intensified by delayed sowing, while tolerant genotypes maintained comparatively restrained primary metabolite profiles across all sowing dates. Genotype explained the greatest proportion of variance in total soluble sugars and total protein, confirming that primary metabolite accumulation is predominantly under genetic control (Table 4). Starch content presented a unique response pattern, with a highly significant genotype x sowing date interaction (Table 4). At D₁, SUGARGRAZE contained the highest starch concentration of 60.70 mg g^-1, which was 80.54% higher than SPV 3083 (33.62 mg g^-1 FW; p ≤ 0.05; Fig. 2). Under delayed sowing, tolerant and susceptible genotypes responded in opposing directions, where starch in SUGARGRAZE declined by 8.00% from D₁ to D₃, and SPV 3083 accumulated starch reserves, increasing by 31.54% over the same period (p ≤ 0.05; Fig. 2). This divergent carbon storage response, depletion in susceptible genotypes versus accumulation in tolerant genotypes under identical delayed sowing conditions revealed that the capacity to maintain metabolic carbon reserves under phenological stress is a fundamental determinant of host plant resistance in sweet sorghum.

Fig. 2.

Starch content (mg g^-1 FW) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates. Data represent mean of triplicates; error bars represent SE of triplicate observations. Different lowercase letters denote significant differences among dates of sowing within a genotype, while different uppercase letters denote significant differences among genotypes at p ≤ 0.05.

Total soluble sugars increased progressively across sowing dates in all genotypes, with susceptible genotypes maintaining consistently higher concentrations throughout. At D₁, SUGARGRAZE recorded the highest total soluble sugar content of 18.73 mg g^-1 FW, which was 87.60% higher than SPV 3083 (9.98 mg g^-1 FW; p ≤ 0.05; Fig. 1). Concentrations rose further under delayed sowing; in SUGARGRAZE, levels increased by 23.17% from D₁ to D₃, reaching 23.07 mg g^-1 FW at D₃, a concentration 72.65% higher than SPV 3083 at the same sowing date (13.38 mg g^-1 FW; p ≤ 0.05; Fig. 1). Throughout the experimental period, SUGARGRAZE maintained the highest total soluble sugar concentrations, followed by CSV 24SS, SPV 3078, and SPV 3083 in descending order (p ≤ 0.05). Reducing sugars similarly increased across sowing dates in all genotypes, with susceptible genotypes consistently accumulating higher concentrations See Fig. 3 . At D₁, SUGARGRAZE exhibited the highest reducing sugar content of 6.16 mg g^-1 FW, which was 61.68% higher than SPV 3083 (3.81 mg g^-1 FW; p ≤ 0.05). All genotypes showed a progressive increase in reducing sugar content across sowing dates, with SUGARGRAZE consistently exhibiting the highest levels, reaching 8.01 mg g^-1 FW at D₃, which was 41.76% greater than SPV 3083 (5.65 mg g^-1 FW; p ≤ 0.05). Intermediate genotypes followed a consistent gradient, with CSV 24SS at 7.36 mg g^-1 FW and SPV 3078 at 6.18 mg g^-1 FW at D₃, reflecting the same susceptibility gradient observed across all primary metabolites (p ≤ 0.05; Fig. 3).

Fig. 1.

Total soluble sugar content (mg g^-1 FW) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates. Data represent mean of triplicates; error bars represent SE of triplicate observations. Different lowercase letters denote significant differences among dates of sowing within a genotype, while different uppercase letters denote significant differences among genotypes at p ≤ 0.05.

Fig. 3.

Reducing sugar content (mg g^-1 FW) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates. Data represent mean of triplicates; error bars represent SE of triplicate observations. Different lowercase letters denote significant differences among dates of sowing within a genotype, while different uppercase letters denote significant differences among genotypes at p ≤ 0.05.

The nutritionally favourable profile of susceptible genotypes was further characterised by significantly higher concentrations of total protein and free amino acids across all sowing dates. Total protein content reached a maximum of 16.09 mg g^-1 FW in SUGARGRAZE at D₃, which was 46.41% higher than SPV 3083 at the same sowing date (10.99 mg g^-1 FW; p ≤ 0.05; Fig. 4). The difference between the highest and lowest protein concentrations at D₁ was 69.12%, with SUGARGRAZE and CSV 24SS consistently accumulating greater protein reserves than their tolerant counterparts throughout the study period (p ≤ 0.05). Free amino acid concentrations followed a parallel trend, with SUGARGRAZE increasing by 90.87% from D₁ to D₃ to reach 0.71 mg g^-1 FW, which was 97.48% greater than the concurrent value in SPV 3083 (0.36 mg g^-1 FW; p ≤ 0.05; Fig. 5). CSV 24SS and SPV 3078 showed intermediate increases of 69.57% and 67.84% respectively from D₁ to D₃, while SPV 3083 showed the smallest increase of only 31.61%, reinforcing the divergent primary metabolic responses to delayed sowing across genotypic classes (p ≤ 0.05).

Fig. 4.

Total protein content (mg g^-1 FW) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates. Data represent mean of triplicates; error bars represent SE of triplicate observations. Different lowercase letters denote significant differences among dates of sowing within a genotype, while different uppercase letters denote significant differences among genotypes at p ≤ 0.05.

Fig. 5.

Free amino acid content (mg g^-1 FW) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates. Data represent mean of triplicates; error bars represent SE of triplicate observations. Different lowercase letters denote significant differences among dates of sowing within a genotype, while different uppercase letters denote significant differences among genotypes at p ≤ 0.05.

In direct contrast to primary metabolite accumulation, secondary metabolite profiles revealed a defence-oriented allocation strategy in tolerant genotypes that was consistently maintained across all sowing dates (Table 4). The absence of any significant genotype x sowing date interaction for secondary metabolites confirmed that the superior secondary metabolite profiles of tolerant genotypes represent constitutive genetic traits rather than environmentally induced responses, making them reliable targets for trait-based selection independent of agronomic management (Table 4). Total phenolic content was markedly elevated in tolerant genotypes across all sowing dates. SPV 3083 accumulated 5.98 mg g^-1 FW phenolics at D₃, which was 70.17% higher than SUGARGRAZE (3.52 mg g^-1 FW; p ≤ 0.05; Fig. 6). Although phenolic levels increased progressively in all genotypes from D₁ to D₃, the overall genotype ranking was consistently SPV 3083 ≥ SPV 3078 ≥ CSV 24SS ≥ SUGARGRAZE across all sowing dates (p ≤ 0.05). Condensed tannins followed a parallel pattern, with SPV 3083 reaching 3.52 mg g^-1 FW at D₃, which was 45.85% higher than SUGARGRAZE (2.41 mg g^-1 FW; p ≤ 0.05; Fig. 7), maintaining the same genotype ranking throughout. Saponin content showed the most pronounced intergenotypic differences among all secondary metabolites, with SPV 3083 accumulating 11.43 mg g^-1 FW at D₃, which was 37.38% higher than SUGARGRAZE (8.32 mg g^-1 FW; p ≤ 0.05; Fig. 8). The consistent genotype ranking SPV 3083 ≥ SPV 3078 ≥ CSV 24SS ≥ SUGARGRAZE was maintained across all sowing dates and all three secondary metabolite classes, confirming the stability and genetic robustness of the defence-oriented metabolic phenotype in tolerant genotypes.

Fig. 6.

Total phenolic content (mg g^-1 FW) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates. Data represent mean of triplicates; error bars represent SE of triplicate observations. Different lowercase letters denote significant differences among dates of sowing within a genotype, while different uppercase letters denote significant differences among genotypes at p ≤ 0.05.

Fig. 7.

Condensed tannins content (mg g^-1 FW) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates. Data represent mean of triplicates; error bars represent SE of triplicate observations. Different lowercase letters denote significant differences among dates of sowing within a genotype, while different uppercase letters denote significant differences among genotypes at p ≤ 0.05.

Fig. 8.

Saponin content (mg g^-1 FW) in sweet sorghum genotypes under natural stem borer infestation at different sowing dates. Data represent mean of triplicates; error bars represent SE of triplicate observations. Different lowercase letters denote significant differences among dates of sowing within a genotype, while different uppercase letters denote significant differences among genotypes at p ≤ 0.05.

Principal component biplot analysis

Multivariate analysis via PCA biplot (Fig. 9) confirmed a clear separation between tolerant and susceptible genotypes along a defence-nutrition axis. The first two principal components jointly accounted for 74.0% of total dataset variance (PC₁: 58.5%; PC₂: 15.5%), providing a robust multivariate framework for interpreting the observed trait associations. PC₁ distinctly separated genotypes along a defence versus nutrition gradient, with tolerant genotypes (SPV 3083 and SPV 3078) scoring positively and clustering with structural traits (cellulose, lignin, hemicellulose) and secondary metabolites (phenolics, tannins, saponins), alongside higher photosynthetic rate and stomatal conductance (Fig. 9C, D). Susceptible genotypes (SUGARGRAZE and CSV 24SS) occupied the negative region of PC₁, showing strong alignment with primary metabolites including total soluble sugars, starch, reducing sugars, total protein, and free amino acids, all of which correlated positively with plant infestation (Fig. 9A, B). PC₂ further resolved sowing date effects within each genotype, with earlier sowing (D₁) in tolerant genotypes associating more closely with photosynthetic efficiency and structural cell wall components, while later sowing (D₃) shifted these genotypes toward increased primary metabolite accumulation. Vector angles confirmed strong positive correlations within the defence trait cluster and within the nutritional trait cluster, while the two clusters were strongly negatively correlated with each other (r = -0.72 to -0.95), confirming the antagonistic nature of the defence-nutrition trade-off and the mechanistic basis through which sowing date modulates host plant resistance trajectories in sweet sorghum.

Fig. 9.

Principal Component Analysis (PCA) biplot showing the integrated effects of genotype and sowing date on photosynthetic, structural, and biochemical traits associated with stem borer infestation in sweet sorghum. Data points represent genotype: (A) SUGARGRAZE, (B) CSV 24SS, (C) SPV 3078, and (D) SPV 3083 across three sowing dates: D₁ (red), D₂ (green), and D₃ (blue). Trait vectors include free (AA), total protein content, total soluble sugars, reducing sugars (RS), internal CO₂ concentration (CO₂), photosynthetic rate (PE), hemicellulose (hemicell), stomatal conductance(gS), and cellulose (Cellul.). Angular relationships between vectors indicate correlation strength: acute angles (positive correlation), obtuse angles (negative correlation), and near-orthogonal angles (weak or no correlation). The biplot visually delineates the trade-off between growth-promoting primary metabolites and defense-related traits across genotypes and sowing regimes.

Discussion

The present study demonstrates that stem borer infestation in sweet sorghum elicited two fundamentally distinct physiological and metabolic responses, a regulated, defence-oriented strategy in tolerant genotypes SPV 3078 and SPV 3083, and a progressive breakdown of resource allocation in susceptible genotypes SUGARGRAZE and CSV 24SS. These contrasting phenotypes were governed by divergent patterns of carbon partitioning, secondary metabolite investment, structural integrity, and photosynthetic regulation under natural pest pressure across three sowing dates.

The progressive increase in stem borer infestation with delayed sowing observed in this study is consistent with the phenological synchrony hypothesis, wherein the timing of crop developmental stages relative to peak pest populations determines infestation intensity. In Punjab conditions, peak Chilo partellus infestation and egg laying activity coincides with the monsoon period of July to August, when temperature and humidity conditions are optimal for larval development. Early sowing on 08 June (D₁) allowed plants to advance beyond the most vulnerable early vegetative stage before peak borer pressure, thereby providing a degree of phenological escape. In contrast, delayed sowing on 06 July (D₃) placed the crop at the seedling and early tillering stages precisely during peak infestation pressure, maximising the overlap between vulnerable crop tissue and peak larval populations. This agronomic timing effect operated independently of genotypic background, as evidenced by the non-significant genotype x sowing date interaction for infestation, confirming that sowing date modulated absolute infestation levels uniformly across all genotypes. Recent studies have demonstrated that genotype by sowing date interactions govern key resistance associated traits across agro-ecological zones¹¹, and the present findings extend this understanding to the context of stem borer host plant resistance in sweet sorghum.

The superior structural integrity of tolerant genotypes SPV 3083 and SPV 3078 provides the primary mechanistic explanation for their resistance phenotype. The significantly higher hemicellulose, cellulose, and lignin concentrations in these genotypes constitute a formidable physical barrier against larval penetration and tissue damage. Lignin in particular plays a critical role in restricting borer establishment, as its deposition in cell walls increases mechanical resistance and reduces tissue palatability²³. The strong negative correlation between lignin and free amino acids (r = -0.83,p ≤ 0.01; Table S6) in tolerant genotypes indicates tightly controlled partitioning of phenylalanine, the common precursor for both lignin and phenolic biosynthesis via the phenylpropanoid pathway, ensuring that carbon flux was directed toward integrated structural and chemical defences²⁴. This coordinated lignin-phenolic investment reflects a constitutive resistance architecture that was expressed consistently across all sowing dates. In contrast, infestation in susceptible lines was associated with degradation of hemicellulose and cellulose without compensatory wound lignification⁷, facilitating larval establishment and tissue damage.

Photosynthetic performance emerged as a sensitive indicator of genotypic resistance status in this study, with tolerant genotypes maintaining significantly higher photosynthetic rates, stomatal conductance, and lower internal CO₂ concentrations than susceptible genotypes across all sowing dates. The strong negative correlation between total soluble sugars and photosynthetic efficiency in tolerant genotypes (r = -0.91 in SPV 3078; r = -0.84 in SPV 3083; p ≤ 0.01; Tables S5, S6) reflects a functional sugar signalling feedback mechanism, wherein elevated soluble sugars induced the downregulation of photosynthetic activity to prevent carbon overaccumulation and maintain metabolic homeostasis²⁵. This active regulation is analogous to the source-sink feedback observed in cereals under biotic stress, where photosynthetic efficiency is preserved through coordinated carbon management rather than passive response to damage. Comparable observations of biotic stress induced photosynthetic impairment have been reported in other host-pathogen systems, where pathogen attack disrupts photosystem II efficiency and reduces chlorophyll content progressively as infestation advances²⁶. In susceptible genotypes, the disruption of this sugar photosynthesis feedback loop, as evidenced by elevated internal CO₂ concentrations correlated with reduced photosynthetic efficiency (r = -0.60 in CSV 24SS,p ≤ 0.05,Table S4), suggests that photorespiratory stress exacerbated metabolic decline under infestation pressure.

The contrasting starch dynamics between tolerant and susceptible genotypes under delayed sowing, the only significant genotype x sowing date interaction across all measured traits represent the metabolic centrepiece of this study (Table 4). While susceptible SUGARGRAZE showed progressive starch depletion from D₁ to D₃, tolerant SPV 3083 accumulated starch reserves under the identical conditions. This divergent response reveals a fundamental difference in carbon allocation strategy between different genotypes. In tolerant genotypes, starch served as a sustained substrate pool for the biosynthesis of defensive secondary metabolites, as supported by the significant positive correlations between starch and phenolics, tannins, and saponins (SPV 3078: r = 0.72, 0.89, 0.83; SPV 3083: r = 0.87, 0.72, 0.62; p ≤ 0.01; Tables S5, S6). This strategic mobilisation of stored carbon toward defence biosynthesis without metabolic compromise reflects the bifunctional role of starch as both a carbon source and sink in plant stress responses, as documented by MacNeill et al.²⁷. In susceptible genotypes, starch reserves were rapidly hydrolysed under pest pressure, but the resulting carbon flux was inadequately channelled into defensive metabolite biosynthesis, as evidenced by the strong negative correlations between starch and secondary metabolites in SUGARGRAZE (r = -0.65, -0.52, -0.60,p ≤ 0.05; Table S3) and CSV 24SS (r = -0.79, -0.73, -0.82; p ≤ 0.01; Table S4). This metabolic misallocation mirrors observations in other cereal systems where genotypic differences in primary metabolite partitioning under stress determine tolerance outcomes^28,29.

The starch breakdown in susceptible genotypes fuelled a cascade of metabolic consequences that collectively enhanced host suitability for stem borer infestation. The resulting accumulation of total soluble sugars created a dual vulnerability, it provided a readily accessible nutrient source for larvae, increasing host attractiveness and pest suitability³⁰, while simultaneously disrupting the sugar-signalling feedback that maintains photosynthetic homeostasis. The conserved positive correlation between internal CO₂ and free amino acids across all genotypes (r = 0.40 to 0.88) points to a fundamental photorespiratory nitrogen recycling pathway that, in tolerant plants, likely supported defence-ready nitrogen pools without systemic stress, but in susceptible genotypes reflected protein degradation and a significant photorespiratory burden under severe infestation³¹. The elevated free amino acid concentrations in susceptible genotypes at later sowing dates further enriched the nutritional environment available to larvae, compounding the infestation advantage.

The accumulation of secondary metabolites, including phenolics, tannins, and saponin in tolerant genotypes occurred in a coordinated, resource-efficient manner that was constitutively maintained across all sowing dates. The absence of significant genotype x sowing date interactions for all three secondary metabolite classes confirms that these defence compounds are under strong genetic control rather than being environmentally induced responses. This constitutive nature is agronomically significant. The resistance advantage of SPV 3083 and SPV 3078 is reliably expressed regardless of sowing schedule, making these genotypes dependable for resistance-based pest management. Phenolics and condensed tannins deter larval feeding and development through multiple mechanisms including reduced digestibility, disruption of larval gut enzymes, and anti-feedant properties⁹. Saponins, which showed the most pronounced intergenotypic differences in this study, act as chemical deterrents through their amphiphilic properties and ability to disrupt insect midgut membranes. The positive lignin-starch correlation (r = 0.76,p ≤ 0.01) in susceptible SUGARGRAZE (Table S3) suggests a stress-induced, compensatory lignification response that was insufficient to prevent larval establishment, indicating that reactive rather than constitutive defence investment is inadequate under high pest pressure.

The PCA results integrated these findings into a coherent two-dimensional framework, confirming the antagonistic nature of the defence-nutrition trade-off along PC₁. The strong negative correlation between the defence module and the susceptibility module (r = -0.72 to -0.95; Fig. 9) demonstrates that carbon investment in primary nutritional metabolism inherently competes with carbon allocation toward structural and biochemical defences. This trade-off is mechanistically rooted in the shared dependence of primary and secondary metabolic pathways on common carbon precursors, whose allocation between competing pathways is regulated by the metabolic status of the plant. Early sowing promoted the defence-oriented arm of this trade-off in tolerant genotypes by ensuring adequate photosynthetic carbon supply before peak pest pressure, whereas delayed sowing disrupted this balance and amplified nutritional metabolite accumulation in susceptible lines, creating a progressively more favourable host environment for stem borer development. This interaction between agronomic timing and intrinsic metabolic architecture provides a mechanistic explanation for the differential pest suitability observed across genotypes and sowing dates.

In conclusion, this study establishes that sowing date restructures carbon partitioning between defence and nutritional susceptibility in sweet sorghum, thereby directly modulating host plant resistance against Chilo partellus. The physiological significance of these findings lies in the demonstration that tolerant genotypes SPV 3083 and SPV 3078 maintain a tightly regulated carbon economy under pest pressure, characterised by preserved photosynthetic integrity, constitutive structural fortification, and strategic starch mobilisation toward secondary metabolite biosynthesis. The biochemical significance is centred on the phenylpropanoid pathway-mediated defence architecture and the starch to defence carbon partitioning trade-off, which mechanistically distinguishes tolerant from susceptible genotypes at a fundamental metabolic level. The ecological significance resides in the phenological synchrony framework established here, wherein early sowing disrupts the temporal overlap between vulnerable crop stages and peak stem borer populations, amplifying the intrinsic resistance expression of tolerant lines while reducing the metabolic vulnerability of susceptible ones. Together, these converge on a unified conclusion that sustainable stem borer management with reduced dependence on the chemical inputs in sweet sorghum is achievable through the strategic combination of optimal sowing timing and genotype selection based on constitutive defence metabolite profiles, providing a physiologically grounded foundation for agronomic optimisation and trait-based resistance breeding.

Study limitations and future directions

The present study characterises host plant resistance mechanisms during a single cropping season (kharif 2024). While the mechanistic consistency observed across four genotypes and three sowing dates provides internal robustness to the conclusions, multi-year replication under varying agroclimatic conditions is recommended to assess the seasonal stability of these resistance trajectories. Year to year variation in monsoon onset timing, peak Chilo partellus flight activity and temperature regimes may modulate the absolute magnitude of sowing date effects, though the direction and mechanistic basis of the observed responses are expected to remain stable given their demonstrably genetic underpinning. Additionally, molecular validation through transcriptomic profiling of phenylpropanoid pathway genes and upstream carbon allocation regulators including sucrose synthase, would provide deeper mechanistic resolution and is identified as a priority for future investigation. Multi-location trials across Punjab’s diverse agroclimatic zones and economic analysis of sowing date-based pest management would further inform the scalability and cost effectiveness of the proposed agronomic strategy.

Practical implications for pest management

The findings of this study offer directly actionable guidance for sweet sorghum cultivation in Punjab and comparable agro-ecological contexts. Early sowing is recommended as the optimal strategy for minimising stem borer infestation, as it simultaneously promotes phenological escape from peak pest pressure and enhances the expression of integrated structural and metabolic defence traits in tolerant genotypes. Susceptible genotypes SUGARGRAZE and CSV 24SS, which are valued for their high sugar content and fermentable biomass, should be prioritised for early sowing to partially compensate for their inherent metabolic vulnerability through agronomic timing. Tolerant genotypes SPV 3083 and SPV 3078 demonstrated reliably superior resistance across all sowing dates and are recommended as primary choices where stem borer pressure is a consistent management challenge, reducing dependence on chemical insecticides in livestock feed and bioenergy production systems. The constitutive nature of secondary metabolite-based defence in these genotypes makes them suitable for trait-based selection in resistance breeding programmes targeting improved cell wall architecture, effective starch-to-defence partitioning, and stable photosynthetic regulation under pest pressure.

Author contributions

Harjeet Kaur: Performed experiments, collected data, writing - original draft; Harpreet Kaur Oberoi: Conceptualisation, methodology, supervision, writing - review & editing; Maninder Kaur: Data curation, supervision, writing - review & editing; A V Umakanth: Germplasm procurement, writing - review & editing; Ravinder Singh Chandi: Data validation, writing - review & editing; Shashi Pathania: Data analysis, statistical analysis, writing - review & editing. All authors discussed the results, contributed to the final manuscript, and approved the submitted version.

Declarations

Author contributions

Harjeet Kaur: Performed experiments, collected data, writing - original draft; Harpreet Kaur Oberoi: Conceptualisation, methodology, supervision, writing - review & editing; Maninder Kaur: Data curation, supervision, writing - review & editing; A V Umakanth: Germplasm procurement, writing - review & editing; Ravinder Singh Chandi: Data validation, writing - review & editing; Shashi Pathania: Data analysis, statistical analysis, writing - review & editing. All authors discussed the results, contributed to the final manuscript, and approved the submitted version.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The datasets generated and/or analysed during the current study are not publicly available due to their relation to ongoing, unpublished doctoral research conducted by the corresponding author. The datasets are, however, available from the corresponding author upon reasonable request.

Competing interest

The authors declare no competing interests.