Cytoplasm effects on rancidity related traits in pearl millet (Pennisetum glaucum (L.) R. Br.) hybrids Authors: P Sanjana Reddy, D Sravani, R Venkateswarlu, C Tara Satyavathi, M Shanti Priya Published: 2025 ABSTRACT AbstractBackgroundPearl millet is a nutritionally rich, climate-resilient cereal, but its large-scale utilization is constrained by rapid rancidity in flour, caused by lipid hydrolysis and oxidation. While nuclear variation for rancidity traits is studied, the contribution of cytoplasmic genomes remains less understood. This study assessed the influence of male-sterile (A1) and maintainer (B) cytoplasms on rancidity-associated traits in hybrids derived from iso-nuclear A/B pairs crossed with diverse restorers.ResultsA total of 64 hybrids were evaluated across two locations for alcoholic acidity, and activity of enzymes - lipase, lipoxygenase and peroxidase. Analysis of variance revealed significant effects of A-lines, explaining the largest share of variability (15.9–47.7%). Cytoplasmic effects were non-significant for alcoholic acidity but highly significant for enzymatic activities, contributing up to 27.8% of total variation in peroxidase. A-cytoplasm consistently enhanced lipase, lipoxygenase and peroxidase compared to B-cytoplasm, suggesting greater predisposition to rancidity. Strong cytoplasm × nuclear interactions were evident, with A-line × cytoplasm effects accounting for up to 24% of variation. Three hybrid nuclear backgrounds (IPMA 287 × 1138R, IPMA 267 × 1142R and IPMA 267 × 1233R) in A1-cytoplasm showed favourable rancidity profiles with reduced enzyme activities, highlighting scope for further exploitation.ConclusionsCytoplasmic background significantly influences rancidity-related enzymes in pearl millet, with A1-cytoplasm linked to faster lipid degradation. These findings emphasize the need to consider genetic background × cytoplasm interactions in breeding strategies to reduce rancidity in pearl millet.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12870-025-07926-4.Keywords: Pearl millet, Cytoplasmic effects, Rancidity, Lipase, Lipoxygenase, Peroxidase, Combining ability Background Pearl millet (Pennisetum glaucum (L.) R. Br.) is a climate-resilient cereal crop with remarkable adaptability to arid and semi-arid environments. It serves as a vital source of energy, proteins, and micronutrients, particularly for subsistence farming communities across Africa and South Asia [1]. Pearl millet is gluten free, contains high amounts of polyphenols, protein, minerals, vitamins, phytochemicals and antioxidants and is termed as “Smart food” [2]. Despite its nutritional superiority, widespread utilization of pearl millet is hindered by the rapid onset of rancidity in milled flour during storage, primarily due to high lipid content and LA, which compromise its shelf life and sensory quality. This rancidity significantly limits the crop’s industrial processing potential and consumer acceptance [3]. The rapid development of off flavors and taste within 5–7 days of milling causes drudgery to women in traditional pearl millet consuming areas [4]. The crop exhibits a high outcrossing rate (> 80%) due to its protogynous flowering nature, has two commercial cultivar options: open-pollinated varieties (OPVs) and hybrids, the latter possessing considerable commercial significance. The introduction of cytoplasmic male-sterility (CMS) systems has facilitated efficient and cost-effective hybrid seed production, thereby establishing hybrids as the predominant cultivar type in India. The CMS denotes a maternally inherited condition characterized by the inability to produce or release functional pollen. This sterility is governed by the interaction of cytoplasmic sterility-inducing factor(s) with homozygous recessive alleles (ms) at the nuclear locus. Cytoplasmic effects refer to the influence of non-nuclear genetic factors—primarily those housed in organelles such as mitochondria and chloroplasts—on the expression of traits. Unlike nuclear genes, which follow Mendelian inheritance, cytoplasmic genes are typically inherited maternally. The CMS arising from mitochondrial mutations, is widely exploited in hybrid seed production due to its ability to eliminate the need for manual emasculation. Similarly, chloroplast genes can affect photosynthetic efficiency, pigmentation, and adaptation to specific light or temperature conditions. Cytoplasmic inheritance, governed by mitochondrial and plastid genomes, can influence several metabolic pathways including lipid metabolism and oxidative processes. While previous studies have studied nuclear genetic variation for rancidity-associated traits, the role of cytoplasmic genomes remains unexplored. The interaction between cytoplasmic background and nuclear genome—termed cytoplasmic-nuclear interaction—may play a critical role in modulating rancidity development in pearl millet flour. The biochemical basis of rancidity is linked to the activity of enzymes such as lipases and the accumulation of free fatty acids, the genetic control of which may involve both nuclear and cytoplasmic components. The maternal inheritance of cytoplasmic factors means that different cytoplasmic backgrounds can significantly alter the expression of rancidity-related traits, including lipid composition, fatty acid profile, and enzyme activity levels. For breeding programs aiming to develop high-yielding, low-rancidity hybrids, careful selection of both male-sterile and restorer parents is essential. The CMS systems widely used in hybrid development must be evaluated for their potential impact on flour shelf life, as some CMS sources may carry undesirable rancidity-related effects. Molecular and quantitative genetic investigations have demonstrated that cytoplasmic factors can significantly contribute to the heritable variation observed in both qualitative and quantitative traits in plants [5]. In pearl millet, hybrids carrying either A1 (male-sterile) or B (maintainer) cytoplasm did not differ in their incidence of downy mildew (Sclerospora graminicola (Sacc.) Schroet.) [6]. Similarly, no significant influence of different CMS systems on the expression of resistance to downy mildew was reported [7]. The A1 cytoplasm, however, has been associated with moderate susceptibility to smut (Tolyposporium penicillariae Brefeld), downy mildew, and leaf blight (Xanthomonas spp.) in A1-based hybrids [8]. However, substantial differences exists between CMS and maintainer lines in biochemical parameters such as soluble proteins, free amino acids, peroxidase, and esterase activity, although chlorophyll content and phenolic levels remained unaffected [9]. Cytoplasmic effects are an unexplored but critical component of rancidity-related trait expression in pearl millet. Understanding and harnessing beneficial cytoplasm–nuclear combinations can lead to hybrids with improved grain quality, longer shelf life, and greater consumer acceptance, thereby enhancing the crop’s market potential and value in the food industry. This study investigates the effects of male-sterile (A1) and maintainer (B) cytoplasm, using a set of iso-nuclear A/B pairs on rancidity-related traits in pearl millet hybrids. Methods Selection of genetic material A total of about 100 pearl millet inbred lines, comprising 75 R-lines and 25 B-lines, were evaluated during 2023 rainy season at two locations using an alpha lattice design for agronomic traits and rancidity-related parameters [10]. Based on the analysis of rancidity and associated traits, parents exhibiting contrasting levels of rancidity (high and low) were identified. For the present study, four A-lines along with their corresponding four B-lines were crossed with 8 selected R-lines (Table 1), resulting in the development of 64 hybrids. Table 1List of parental lines involved in generating hybrids used for the current studyS. No.GenotypeCytoplasmRancidity level1267A1Male-sterileLow2267BMaintainerLow3287A1Male-sterileLow4287BMaintainerLow5211A1Male-sterileHigh6211BMaintainerHigh7264A1Male-sterileHigh8264BMaintainerHigh91071RRestorerLow101110RRestorerLow111138RRestorerLow121142RRestorerLow131095RRestorerHigh141101RRestorerHigh151198RRestorerHigh161233RRestorerHigh List of parental lines involved in generating hybrids used for the current study Thus a total of 64 hybrids derived from iso-nuclear A/B pairs were evaluated during the 2024 rainy season at two locations: ICAR-Indian Institute of Millets Research (ICAR-IIMR), Hyderabad, Telangana and Agricultural Research Station, Perumallapalle, Andhra Pradesh. The trials were laid out in a split–split-plot design with R-lines as main plots, A-lines as sub-plots, and cytoplasms (A₁ and B) as sub-sub-plots, in two replications. Each plot consisted of two rows of 4 m length spaced 60 × 12 cm. Recommended agronomic practices for raising a good crop. Grain samples from each replication were harvested at physiological maturity and used for rancidity-related biochemical assays that included Alcoholic acidity (AA), Lipase activity (LA), Peroxidase activity (POX) and Lipoxygenase activity (LOX). Flour of pearl millet grains was made in a UDY Cyclone Sample Mill (Model 3010-030, USA) using 1.0 mm screen. Chemical analysis Alcoholic acidity (AA) Method used to measure rancidity of pearl millet flour was as per standard methodology [11]. It measures the free acidity released from the hydrolysis of lipids into free fatty acids by lipases, breakdown of proteins into amino acids by proteolytic enzymes, as well as the presence of certain acids and salts. The 5 gm of sample flour was weighed in a stoppered conical flask and 50 mL of neutral ethyl alcohol (90% v/v) was added and allowed to stand for 24 h with occasional swirling. Then alcohol extract is filtered through a dry filter paper and 10 mL of the filtrate is titrated with 0.01 N standard sodium hydroxide solution using phenophtalein indicator. The development of pink colour is the end point. AA is calculated using the following formula and expressed as H2SO4 on dry basis. Lipase activity (LA) LA was estimated following the standard protocol [12] with minor modifications. Lipase hydrolyses the triglycerides into free fatty acids and glycerol and typically measures the amount of free fatty acids released during the hydrolysis process. The flour of samples was defatted in petriplates with hexane (for every 3 g of flour 5 mL of hexane was added) for 20 min at room temperature. Then, one gram of the defatted flour was taken into 10 mL screw capped tubes, added 1 mL of olive oil followed by 250 µL of distilled water, vortexed thoroughly and incubated at 40 ℃ for 3 h. The lipids from the blank were extracted immediately (no incubation). After incubation, 5 mL of hexane was added to the tubes, vortexed and then centrifuged at 1000 rpm for 2 min at 25 ℃. The supernatant was decanted into test tubes and the process was repeated three times. The hexane extracts were pooled and evaporated under nitrogen at 40 °C (Rapid Mini, Crescent Scientific, India). The residue obtained was dissolved in 4 mL of isooctane, transferred to screw capped tubes and added 1 mL of 5% cupric acetate. The tubes were centrifuged at 1000 rpm for 1 min at 25 ℃, pipetted out supernatant into cuvette and the absorbance was measured at 715 nm in a spectrophotometer. The absorbance of the sample was compared with the absorbance of oleic acid standards prepared in isooctane (1–10 mmolL− 1). Lipase activity was expressed as units/gram (U/g), where 1.0 U is defined as the micromoles of fatty acid liberated per hour. where, 1000 = conversion factor from mol/L to µEq/mL, 4 = volume of isooctane used to dissolve fat (mL), V = volume of olive oil added (mL), Af = absorbance of sample after incubation at 715 nm, Ai = absorbance of blank after incubation at 715 nm, € = molar absorptivity of oleic acid at 715 nm (M− 1cm− 1) – value, t = incubation time (h), l = path length (1 cm for a standard cuvette) and s = sample weight (g). Peroxidase activity (POX) POX was assayed by the standard method [13]. Peroxidase activity is an enzyme catalysed oxidation of specific substrates in the presence of hydrogen peroxide (H₂O₂) that results in a measurable product, often a colored compound, which can be quantified spectrophotometrically at 430 nm. The assay mixture in a total volume of 2.8 mL contained: 2.5 mL of 0.05 M potassium phosphate buffer (pH 6.8), 0.1 mL of 0.1% (w/v) o-dianisidine, 0.1 mL of 0.1 M H2O2 and 0.1 mL enzyme extract. The reaction was started by adding H2O2. Thus, omission of H2O2 from the reaction mixture served as blank. Increase in absorbance was monitored at 430 nm at an interval of 15 s for 2 min. One unit of enzyme activity was defined as the amount of enzyme required to increase 0.1-unit absorbance min− 1 g flour− 1. POX per gram flour is calculated using the following formula: Lipoxygenase activity (LOX) Lipoxygenase specifically catalyzes the oxygenation of double bonds in cis, cis-1,4-pentadiene moieties of unsaturated fatty acids leading to the formation of conjugated diene hydroperoxides which can be detected and measured spectrophotometrically at 234 nm. The reaction mixture in a final volume of 2.66 mL was prepared using 2.5 mL of 0.05 M acetate buffer (pH 4.2), 0.06 mL of 7.5 mM linoleic acid substrate solution and 0.1 mL of enzyme extract. The reaction was started by addition of the enzyme and progress of the reaction was monitored by recording increase in absorbance at 234 nm for 2 min against blank. One unit of enzyme activity was defined as the amount of enzyme required to increase 0.1-unit absorbance min− 1 g flour− 1. LOX per gram flour is calculated using the following formula: Statistical analysis Data from individual years were analyzed using analysis of variance (ANOVA) based on a split–split-plot design, with R-lines as main plots, A-lines as subplots, and cytoplasms as sub-subplots, using GenStat 12th Edition and further checked for homogeneity of error variances. In the combined ANOVA model, genotypes (combinations of cytoplasm, A-line, and R-line) were treated as fixed effects, while locations and replications were considered random effects. Combined ANOVA was used to assess the significance of hybrid × year interactions. Results The combined genotype × environment interaction ANOVA (table not presented here) indicated that the main effect of location was non-significant for all rancidity traits (AA, LA, LOX, POX). This indicates that environmental differences between the two testing sites did not substantially influence the expression of these traits, and the genotypes performed consistently for AA, LA, and LOX regardless of the location. Furthermore, genotype × location interactions were non-significant for all traits except LOX, suggesting that the majority of genotypes responded uniformly across environments for AA, LA, and POX. The significant interaction observed only for LOX implies that peroxidase activity exhibited differential genotype responses across locations, reflecting greater environmental sensitivity for this trait compared to the others. The analysis of variance of the split plot (Table 2) revealed significant contribution of nuclear genetic background of A-lines and R-lines in the expression of rancidity-related traits in pearl millet. The cytoplasm did not significantly influence overall rancidity as depicted by alcohol acidity while it significantly influenced the activity of enzymes especially POX. The interaction effects of A-line with R-line and between cytoplasm with A-line and with R-line were significant for all the rancidity related traits. The third order interaction between cytoplasm, A-line and R-line was non-significant for AA and significant for all the enzymatic effects (Table 2). Among the sources of variation, A-lines consistently accounted for the largest proportion of variability across all traits, contributing between 15.9% (AA) and 47.7% (POX). This highlights the predominant role of the female parent in governing these traits. In contrast, R-lines explained a relatively smaller proportion of the variation, ranging from 5.2% to 23.1%, suggesting that male parent effects were less pronounced (Fig. 1). The cytoplasm effect was trait-specific, being non-significant for AA but highly significant for enzyme activities, where it contributed substantially (4.2% for LA, 17.5% for LOX, and 27.8% for POX). Importantly, cytoplasmic–nuclear interactions were highly significant across most traits, particularly the A-line × cytoplasm interaction, which accounted for 24.1%, 12.3%, 17.2%, and 12.0% of variation in AA, LA, LOX, and POX, respectively. This was followed by R-line × cytoplasm interaction, which accounted for 22.6%, 4.0%, 4.1%, and 1.6% of variation in LA, LOX, and POX, respectively. The other interactions (A×R, A×R×Cytoplasm) contributed less towards rancidity related traits. These results indicate that the manifestation of rancidity-related traits is not solely dependent on cytoplasmic or nuclear factors in isolation but also arises from their interactions. Overall, the findings suggest that A-lines exert a major genetic influence on rancidity traits, while cytoplasm and cytoplasmic–nuclear interactions play crucial modifying roles, especially in enzyme activities associated with lipid degradation. Such insights are critical for guiding breeding strategies that aim to mitigate rancidity in pearl millet by carefully selecting maternal lines and cytoplasmic backgrounds. Table 2Analysis of variance for rancidity related traits in Pearl milletSource of variationd.f.Alcoholic acidityLipase activity (units/g tissue)Lipoxygenase (units/g tissue)Peroxidase (units/g tissue)R-line70.0071**43.7006**106,264**202,961**Residual70.00010.51153942129A-line30.0125**78.7368**407,782**1,856,138**R-line x A-line210.0034*14.0352**31,619**89,101*Residual240.00141.4354366942,188Cytoplasm10.0114 NS7.9069**172,275**1,083,601**R-line x Cytoplasm70.0177**7.5489**40,818*63,454**A-line x Cytoplasm30.0188**23.2489**169,234**465,334**R-line x A-line x Cytoplasm210.0028 NS11.8829**39,042**71,499**Residual320.00300.28515,35914,477*Significant at P < 0.05, **Significant at P < 0.05 Analysis of variance for rancidity related traits in Pearl millet *Significant at P < 0.05, **Significant at P < 0.05 Proportional contribution of genetic background of A-line, R-line, cytoplasm and their interaction effects on rancidity related traits in pearl millet Influence of A- vs. B-Cytoplasm (main effect) Cytoplasm main effects were non-significant for alcoholic acidity while it significantly influenced all enzymatic activities. The A1 cytoplasm had higher LA, indicating possibly faster lipid breakdown. It also had higher LOX activity, which could accelerate peroxidation of lipids and had higher POX, suggesting stronger oxidative enzyme system. Overall, A1 cytoplasm consistently shows higher LA, LOX and POX activity, which may predispose its hybrids to faster rancidity development compared to B-cytoplasm. Cytoplasm × genotype interaction The Table 3 depicts the performance of cytoplasm specific hybrids (R-lines crossed with A1or B cytoplasm). The differences of each pair of hybrids under both the groups involving common female genetic background and male genetic background are checked with the LSD at 5% for comparing at the same level of genetic background. Table 3Effect of cytoplasm on female parents and male parents for rancidity related traits in hybridsAlcoholic acidityLipase activity (units/g tissue)Lipoxygenase (units/g tissue)Peroxidase (units/g tissue)A1BA1BA1BA1BCytoplasm0.150.145.004.65*659.74607.86*703.71573.59*LSD 0.01 0.14 31.56 30.64 IPMA 2110.140.155.805.53539.26588.28535.41431.78*IPMA 2640.160.165.24*6.56738.54559.06*775.87484.01*IPMA 2670.160.10*4.433.18*734.77745.79829.04*926.61IPMA 2870.140.144.543.34*626.39538.30*674.53451.97*LSD (5%)# 0.03 0.27 63.11 61.27 1071R0.200.14*4.972.78*593.07548.57625.61539.141110R0.160.155.575.18*688.96625.17734.47629.48*1138R0.120.153.26*3.85642.08618.63646.75549.87*1142R0.130.103.233.39713.45588.90*795.59608.44*1095R0.130.166.666.32474.43*581.13467.33503.461101R0.170.12*4.50*5.05695.78605.53*766.90611.57*1198R0.180.11*5.844.48*752.84653.92*818.08634.01*1233R0.10*0.165.996.16717.32641.02774.96512.77*LSD (5%)# 0.04 0.38 89.25 86.65 # when comparing between same levels of nuclear background Effect of cytoplasm on female parents and male parents for rancidity related traits in hybrids # when comparing between same levels of nuclear background For AA, most A/B differences are not significant, except in a few cases (e.g., IPMA 267 among female nuclear background and 1071R, 1101R, 1198R among male parents), where B-cytoplasm shows lower acidity in the hybrids. The R-lines showed significantly higher LA in A1-cytoplasm in the nuclear background of IPMA1/B 267 and IPMA1/B 287 and B-cytoplasm in IPMA1/B 264 suggesting female nuclear background × female cytoplasm interaction. Similarly, the female lines showed significantly higher LA in A1-cytoplasm in the nuclear background of 1071R, 1110R and 1198R and in B-cytoplasm in the nuclear background of 1138R and 1101R suggesting male nuclear background × female cytoplasm interaction in the expression of LA in the hybrids. Consistently higher LOX is observed in A1-cytoplasm for most B and R-lines (nuclear backgrounds of IPMA1/B 264, IPMA1/B 287; 1142R, 1101R and 1198R) except for 1095R where B-cytoplasm showed higher LOX activity. Significant differences at individual nuclear backgrounds reinforce that A1-cytoplasm is linked to stronger lipid oxidation. Similar trend was observed for POX where A1-cytoplasm had significantly higher POX in the nuclear backgrounds of IPMA/B 211, IPMA/B 264 and IPMA 287 among female parents and 1110R, 1138R, 1142R, 1101R, 1198R and 1233R among male parents except in ICMA/B 267, where B-cytoplasm had higher POX activity indicating specific genotypic response. Thus, A1-cytoplasm tends to elevate enzyme activities (LA, LOX, POX), which are directly linked to rancidity development in pearl millet flour. However, some female lines and restorer lines perform better with A1-cytoplasm, highlighting that cytoplasm × nuclear interaction should be considered when selecting the parental lines for hybrid development. Mean performance The Table 4 presents mean data of the crosses between four A-lines (female nuclear backgrounds) in two cytoplasmic backgrounds (A1 & B) and eight R-lines (restorers). The AA values ranged from 0.09 to 0.24 in A-based hybrids and from 0.07 to 0.21 in B-based hybrids. Also, four hybrids in A1-cytoplasm background and 6 hybrids in B-cytoplasm background were significantly superior to counterpart hybrids suggesting that B-cytoplasm may confer lower rancidity potential. However, of the 32 hybrids in each cytoplasm, 25 A-cytoplasm based hybrids and 26 B-cytoplasm based hybrids were below the alcoholic acidity standard set by FSSAI (< 0.18). The LA ranged from 1.27 to 10.39 in A-based hybrids and from 1.21 to 11.13 in B-based hybrids. Similarly, for LA, nine hybrids in A1-cytoplasm background and 14 hybrids in B-cytoplasm background were significantly superior to counterpart hybrids suggesting that B-cytoplasm may confer lower LA leading to low rancidity. Lipoxygenase activity ranged from 253 to 996 in A1- cytoplasm based hybrids where three hybrids were significantly superior to B-cytoplasm based hybrids. The LOX ranged from 391 to 841 in B-cytoplasm background where six hybrids were significantly superior to counterpart hybrids. For POX, as compared to 4 hybrids in A-cytoplasm background, 14 hybrids in B-cytoplasm background were significantly superior to counterpart hybrids. POX ranged from 263 to 1199 units/g tissue in A1- cytoplasm based hybrids and from 276 to 1044 units/g tissue in B- cytoplasm based hybrids. Table 4Mean performance of hybrids for rancidity related traits across cytoplasms in isonuclear backgroundR-lineA-lineAlcoholic acidityLipase activity (units/g tissue)Lipoxygenase (units/g tissue)Peroxidase (units/g tissue)A1BA1BA1BA1B1071RIPMA 2110.160.143.514.10512.79486.91470.70296.24*IPMA 2640.200.193.833.64694.57496.51*703.71397.18*IPMA 2670.230.11*6.301.57*695.86732.90815.73*1043.70IPMA 2870.190.11*6.231.81*469.06477.95512.28419.441110RIPMA 2110.130.204.40*5.79590.00606.09582.14375.2*IPMA 2640.160.156.847.54706.17518.82*721.00548.52IPMA 2670.200.12*7.234.64*889.67804.341080.681013.32IPMA 2870.160.123.832.75*569.98571.43554.07580.861138RIPMA 2110.140.124.473.42*571.63629.15441.16383.46IPMA 2640.120.163.763.21662.77638.06604.94522.90IPMA 2670.130.123.54*5.17758.88673.141029.72860.72IPMA 2870.09*0.201.27*3.60575.04534.16511.16432.391142RIPMA 2110.130.125.414.40*665.24523.10677.11371.91*IPMA 2640.130.112.44*4.28776.95600.14795.48570.22*IPMA 2670.130.081.802.16811.56841.31817.09*1023.61IPMA 2870.140.113.282.72600.03391.05*892.66468.02*1095RIPMA 2110.120.1610.397.56*252.86*537.67262.85275.73IPMA 2640.13*0.217.708.26611.49461.26530.53330.4*IPMA 2670.150.114.284.35603.49666.89654.24*930.09IPMA 2870.130.174.26*5.13429.88*658.72421.70477.611101RIPMA 2110.150.114.663.68*518.23643.66534.17673.68IPMA 2640.240.16*4.81*11.13736.44594.18886.90517.86*IPMA 2670.170.104.502.18*686.00698.91679.44795.06IPMA 2870.120.114.033.20*842.45485.38*967.07459.69*1198RIPMA 2110.170.156.635.61*604.20667.53651.28641.69IPMA 2640.210.13*5.20*7.37996.32578.16*1198.54613.69*IPMA 2670.190.07*6.221.21*541.15*799.54552.95*889.70IPMA 2870.140.075.293.72*869.69570.43*869.56390.95*1233RIPMA 2110.10*0.216.95*9.72599.11612.16663.88436.31*IPMA 2640.100.177.317.04723.62585.34765.87371.28*IPMA 2670.090.071.56*4.13891.54749.301002.42856.66IPMA 2870.11*0.198.163.76*655.00617.26667.68386.82*Range0.09–0.240.07–0.211.27–10.391.21–11.13253–996391–841263–1199276–1044LSD (5%)#0.080.77178.50173.30# when comparing between same levels of nuclear background of the hybrid Mean performance of hybrids for rancidity related traits across cytoplasms in isonuclear background # when comparing between same levels of nuclear background of the hybrid Cytoplasm differences are not consistent across traits in individual nuclear backgrounds. In general, B-cytoplasm tended to lower AA and enzymatic activities in several hybrid combinations suggesting its potential in mitigating free fatty acid accumulation, limiting oxidative degradation of lipids. The observed differential responses again emphasize the importance of nuclear × cytoplasmic interactions, as the same cytoplasmic background could either suppress or enhance rancidity traits depending on the restorer line. Overall, hybrids such as 1071R × IPMA1 267 (B), 1071R × IPMA1 287 (B), 1138R × IPMA1 287 (A), 1142R × IPMA1 267 (A), 1198R × IPMA 267 (B) and 1233R × IPMA1 267 (A) exhibited a favourable profile with low AA and low LA, making them promising candidates for developing hybrids with improved storability and reduced rancidity. The cytoplasmic background notably influenced key rancidity-associated enzyme activities in pearl millet hybrids. The AA exhibited no significant difference between A1- and B-cytoplasm, indicating minimal cytoplasmic effect on this parameter. In contrast, LA, LOX, and POX were significantly higher in hybrids with A1-cytoplasm compared to those with B-cytoplasm, suggesting that A1-cytoplasm enhances both hydrolytic and oxidative rancidity potential. These enzyme-related differences align with broader biochemical understanding: lipase accelerates hydrolytic rancidity by releasing free fatty acids, serving as substrates for lipoxygenase, which in turn catalyzes oxidation to rancid products like hydroperoxides and aldehydes [14]. The roles of LOX and POX in lipid oxidation and development of off-flavors are well documented, including their exacerbating effect on shelf-life deterioration of pearl millet flour [14]. Moreover, genetic and metabolomic studies identify LA and associated free fatty acid accumulation as central to rancidity development in high-rancidity lines [15]. The presence of significant cytoplasm × genotype interactions suggests that nuclear–cytoplasmic interplay modulates these enzyme activities, a phenomenon observed in other cytoplasmic trait studies in pearl millet [16]. These interactions indicate that while A1-cytoplasm generally intensifies rancidity-linked enzymatic activity, specific nuclear backgrounds may counteract or exacerbate this effect. The data support the selection of nuclear background in A1-cytoplasm in breeding programs aimed at extending flour shelf-life by reducing lipolytic and oxidative enzyme activities. Correlation among rancidity-related traits Correlation analysis revealed that the magnitude and direction of associations between rancidity-related traits varied with cytoplasmic background. Under B-cytoplasm, AA exhibited a significant positive correlation with LA (r = 0.58, p < 0.01), indicating that lipase-mediated hydrolysis of triglycerides is a major contributor to free fatty acid accumulation in this background. Interestingly, AA showed a significant negative correlation with POX (r = − 0.58, p < 0.01), suggesting a possible protective role of peroxidase in mitigating rancidity by detoxifying lipid peroxides. In contrast, A1-cytoplasm displayed weak and non-significant correlations among these traits, implying a lesser cytoplasmic influence on trait expression. LOX and POX were strongly and positively correlated with each other in both cytoplasmic backgrounds (r = 0.91 and 0.81, p < 0.01), highlighting their coordinated role in lipid oxidation pathways. The negative association of LOX with AA under B-cytoplasm (r = − 0.40, p < 0.05) further suggests that increased LOX activity might shift lipid breakdown toward peroxidation rather than free fatty acid accumulation (Table 5). Overall, these findings emphasize that B-cytoplasm expresses stronger correlation. However, since A1-cytoplasm is used in hybrids, it is recommended to select nuclear backgrounds with reduced rancidity in A1-cytoplasm through simultaneous selection for low LA and moderate POX in hybrids. Table 5Correlation between rancidity related traits in Pearl milletTraitCytoplasmAlcoholic acidityLipase activity (units/g tissue)Lipoxygenase (units/g tissue)Lipase activity (units/g tissue)A10.16B0.58**Lipoxygenase (units/g tissue)A10.14-0.30B-0.40*-0.23Peroxidase (units/g tissue)A10.23-0.230.91**B-0.58**-0.350.81** Correlation between rancidity related traits in Pearl millet Discussion Most of the rancidity-associated biochemical traits are relatively stable and under strong genetic control, whereas LOX activity appears more influenced by micro-environmental variation. Such sensitivity of LOX may be linked to its role in oxidative stress pathways, making it more responsive to subtle environmental factors. Cytoplasmic and nuclear contributions to rancidity-related traits The present study provides comprehensive insights into the genetic and cytoplasmic determinants of rancidity in pearl millet hybrids by dissecting the effects of nuclear (A- and R-line) and cytoplasmic (A1 and B) backgrounds and their interactions. The significant variation observed among female and male parents for all rancidity-related traits demonstrates that nuclear genes exert strong control over both hydrolytic and oxidative pathways of lipid degradation. Among these, A-lines contributed the highest proportion of variance across all traits, particularly POX (47.7%) and LOX (40.7%), suggesting that the maternal nuclear genome plays a central role in modulating rancidity susceptibility. The differential contribution of R-lines, ranging from 5.2% to 23.1%, indicates that paternal effects, though significant, are secondary to maternal influence. This is consistent with the inheritance patterns of several metabolic traits that are predominantly maternally regulated, particularly when enzyme activities are linked to the endosperm or embryo tissues formed under maternal control. Earlier studies on pearl millet determined the genotypic variation among pearl millet hybrids, CMS lines and inbred lines for rancidity indicators/determinants [17]. The presence of significant A × R interaction further suggests that combining ability between female and male lines influences rancidity-related enzyme expression, as also observed in combining ability analyses for rancidity in pearl millet where non-additive gene action was predominant [3]. Cytoplasmic effects on lipid-degrading enzyme activities Although cytoplasm did not significantly affect alcoholic acidity, it exerted a profound effect on the activity of lipid-degrading enzymes such as LA, LOX and POX. The A1-cytoplasm consistently exhibited higher activity for these enzymes compared to the B-cytoplasm. This observation implies that A1-cytoplasm enhances the rate of both hydrolytic and oxidative rancidity processes in pearl millet flour. Biochemically, rancidity in pearl millet flour is triggered by hydrolytic breakdown of triglycerides catalyzed by lipase, resulting in free fatty acid release. These free fatty acids are then oxidized by lipoxygenase, producing hydroperoxides and volatile aldehydes responsible for off-flavors [15, 18]. Peroxidase, a terminal enzyme in the lipid oxidation cascade, amplifies oxidative degradation by converting peroxides into secondary oxidation products that deteriorate lipids and pigments. The elevated activities of these enzymes in A1-cytoplasm hybrids indicates their greater susceptibility to rancidity. The observed cytoplasmic influence on enzymatic activity may be attributed to differences in organellar genetic makeup, particularly mitochondrial or plastid genomes that regulate energy metabolism and reactive oxygen species (ROS) dynamics. Previous studies in maize have shown that S-type cytoplasm (CMS-S) compared to near-isogenic N-type cytoplasm lines display differences in redox status. The mitochondrial gene ORF355 (a CMS gene) is involved; its expression triggers changes in mitochondrial-to‐nucleus signalling (retrograde) that result in altered oxidative metabolism and enzyme activity [19]. The A1-cytoplasm of pearl millet may therefore promote a more oxidative cellular environment, enhancing lipid oxidation pathways. In contrast, the B-cytoplasm appears to maintain a more balanced redox state, potentially conferring stability to lipid components in the flour. Nuclear–cytoplasmic interactions on the breeding implications The study showed a high significance of cytoplasmic–nuclear interactions, particularly the A-line × cytoplasm interaction, which explained up to 24.1% of variation in AA and up to 17.2% in LOX. This clearly demonstrates that the manifestation of rancidity-related traits cannot be ascribed to cytoplasmic or nuclear effects alone but results from their complex interplay. Such nuclear–cytoplasmic interactions are well documented in cereals for traits such as fertility restoration, stress tolerance, and yield [8, 20, 21]. The interactions identified here reveal that specific nuclear backgrounds can either accentuate or mitigate the influence of A1- or B-cytoplasm on rancidity traits. For example, the hybrids involving IPMA 287 as female parent and 1138R and 1233R as male parents; IPMA 264 as female parent with 1095R as male parent exhibited favourable AA. Similarly, 14 hybrid combinations in A1 cytoplasm background and 25 hybrids in B-cytoplasm background exhibited favorable enzyme activity profiles. These results imply that while A1-cytoplasm may inherently favor higher lipolytic and oxidative activities, certain nuclear backgrounds can buffer this effect. This buffering could arise from nuclear-encoded antioxidant enzymes, regulatory proteins, or structural lipid components that limit peroxidation. The data also highlight that cytoplasm-specific hybrid combinations need to be evaluated rather than assuming a uniform cytoplasmic effect across genetic backgrounds. Previous studies also showed a differential behaviour of cytoplasms, both in combination with a common pollinator and across pollinators for several traits providing evidence for the distinctiveness of different cytoplasmic sources in pearl millet and for the influence of cytoplasmic factors on the phenotypic expression of nuclear genes [22]. Similar studies in pigeonpea involving A×R and B×R hybrids attributed the performance to cytoplasm × nuclear interactions. The differences varied in their direction and magnitude without showing a definite pattern in favour of a particular cytoplasm [23]. In the current study, the superior performance of several B-cytoplasm-based hybrids for reduced LOX and POX suggests that B-cytoplasm can be strategically utilized in breeding programs to mitigate rancidity risk. However, as most commercial CMS systems in pearl millet are A1-based, selection of nuclear backgrounds that minimize rancidity enzyme activity under A1-cytoplasm becomes essential for maintaining hybrid seed production efficiency while improving flour stability. Comparative trends and biochemical interpretation Mean performance analysis revealed that hybrids with A1-cytoplasm generally exhibited higher LA, LOX, and POX across nuclear backgrounds, confirming the oxidative tendency of this cytoplasm. Interestingly, AA—a proxy for overall rancidity—did not always follow the same trend, suggesting that accumulation of free fatty acids and their subsequent oxidation are not perfectly synchronized across genotypes. This could be due to inhibition of LA or differential efficiency of fatty acid oxidation among hybrids. The biochemical basis for A1-cytoplasm-associated rancidity may involve increased mitochondrial ROS production or altered lipid metabolism in developing grains. Mitochondria in cytoplasmic male sterile lines are known to exhibit modified electron transport chain efficiency, leading to increased oxidative stress [24]. The resultant oxidative environment could upregulate lipid oxidation enzymes, explaining the higher LOX and POX activities observed. Conversely, B-cytoplasm may represent a more “reductive” environment, reducing oxidative enzyme activation. The observed genotypic specificity of cytoplasmic effects indicates that both maternal and paternal nuclear genomes modulate organelle–nucleus signaling, influencing lipid homeostasis. Similar cytoplasm–nucleus specificity was observed in rice and wheat, where particular nuclear backgrounds altered mitochondrial gene expression and enzyme profiles linked to oxidative metabolism [25, 26]. Limited number of previous studies available [27, 28] propose that cytoplasmic influence on quality traits is not a major hindrance in hybrid rice breeding though there can be certain specific instances of significant impact of male sterile cytoplasm. This warrants the necessity of such studies in every cross combination under evaluation irrespective of the CMS line or cytosterile source and to select only those crosses where the cytoplasmic effects are absent or less predominant [29]. In general, the cytoplasmic influence for grain quality was found to be highly cross-specific and depended on the nuclear background of CMS line and fertility restorer [29]. Correlation structure and trait interdependence The correlation patterns among rancidity-related traits provide further evidence for cytoplasmic modulation of metabolic networks. Under B-cytoplasm, AA showed a strong positive correlation with LA (r = 0.58**), underscoring the direct contribution of lipase-mediated hydrolysis to free fatty acid accumulation. The significant negative correlation of AA with POX (r = − 0.58**) suggests that higher POX may play a detoxifying role by consuming lipid peroxides independent of free fatty acid accumulation. This indicates that peroxidase could serve as a compensatory antioxidant enzyme in B-cytoplasm, contributing to lower rancidity. Conversely, under A1-cytoplasm, the weaker and non-significant correlations among rancidity traits imply a more unregulated enzymatic response, possibly due to elevated ROS levels overriding metabolic control. The strong positive correlation between LOX and POX in both cytoplasms (r = 0.91** in A1 and r = 0.81** in B) confirms their coordinated involvement in the lipid oxidation cascade. The negative association of LOX with AA under B-cytoplasm (r = − 0.40*) further suggests that enhanced LOX activity may divert lipid degradation towards oxidation rather than free fatty acid accumulation, reducing the acidity index. Collectively, these correlation patterns reinforce that cytoplasmic type modulates not only enzyme activity magnitude but also their interrelationships within the rancidity pathway. Earlier studies have also shown that alcoholic acidity showed a significant positive association with LA and LOX [3]. Implications for hybrid breeding and grain quality improvement From a breeding standpoint, the results highlight that rancidity traits in pearl millet are under complex nuclear–cytoplasmic regulation involving both additive and interactive genetic effects. As A1-cytoplasm remains the standard in hybrid breeding due to its stable male sterility and availability of diverse restorers, the challenge lies in identifying nuclear genotypes that minimize rancidity while maintaining fertility restoration and yield potential. The identification of nuclear backgrounds that express lower LA or moderate POX under A1-cytoplasm—such as IPMA 264 and IPMA 267—offers a practical route for developing low-rancidity hybrids. Simultaneously, the demonstrated biochemical advantage of B-cytoplasm, which consistently reduced enzymatic activity and AA in many crosses, suggests its usage in the specific nuclear backgrounds in the development of new crosses and thereafter conversion on sterile-background. Previous work in pearl millet and sorghum has shown that substitution of cytoplasm can alter seed biochemical composition and stress resilience without adverse effects on fertility restoration [26, 30]. This further suggests studies on other cytoplasms in iso-nuclear background for developing hybrids combining both seed quality and agronomic superiority. These findings also emphasize the importance of incorporating biochemical screening for lipase and oxidative enzymes during hybrid development, alongside conventional yield and fertility parameters. Integrating these assays can provide early prediction of flour stability, reducing post-harvest rancidity losses that limit consumer acceptability of pearl millet-based products. Further elucidation of the molecular basis of cytoplasmic–nuclear interactions underlying rancidity will require integrating genomics and metabolomics approaches. Transcriptomic profiling of lipid degradation genes and ROS-scavenging enzymes in contrasting cytoplasmic backgrounds could identify candidate genes responsive to organelle–nucleus crosstalk. Metabolomic studies, focusing on free fatty acids, aldehydes, and antioxidant metabolites, could reveal specific biochemical markers associated with rancidity tolerance. Moreover, identification of mitochondrial or plastid sequence polymorphisms linked to low-rancidity phenotypes could pave the way for cytoplasmic substitution breeding. Ultimately, the findings of this study underscore that rancidity in pearl millet is a multifactorial phenomenon governed by the interplay between cytoplasmic, environment and nuclear genome. The cytoplasm acts as a modifier rather than a primary determinant, while nuclear–cytoplasmic harmony defines the biochemical fate of lipids post-harvest. Rational exploitation of this knowledge will enable breeders to design hybrids that combine high yield potential with improved grain storability and product quality, contributing to enhanced utilization of pearl millet in health and food industries. Conclusion The study showed that A-lines were the major contributors to rancidity-related traits in pearl millet, while R-lines had comparatively smaller effects. The cytoplasm significantly influenced enzyme activities (LA, LOX, POX) but not AA. Importantly, cytoplasmic–nuclear interactions, especially A-line × cytoplasm, played a substantial role in determining trait expression. These findings emphasize the need to consider both maternal lines and cytoplasmic backgrounds in breeding strategies to reduce rancidity in pearl millet. Supplementary Information Supplementary Material 1. Acknowledgments Clinical trial number Not applicable. Declaration of generative AI in the writing process During the preparation of the manuscript, the author(s) used ChatGPT in order to improve the readability and grammatical errors. After using this tool, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication. Abbreviations Alcoholic acidity Lipase Lipoxygenase Peroxidase (POX) Cytoplasmic male-sterility open-pollinated varieties Analysis of Variance Reactive Oxygen Species Authors’ contributions Conceptualization, Investigation, Methodology, Formal analysis, Data analysis, Review and editing: PSR, DS, RV ; Investigation : DS, RV, MSP; Data analysis: PSR, DS; Original draft writing: PSR, RV, CTS; Review and editing: PSR, DS, RV, CTS, MSP. Funding Authors are thankful for the funds provided by the ICAR-Indian Institute of Millets Research, Hyderabad. Data availability All data generated or analysed during this study are included in this published article and its supplementary information files. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement All data generated or analysed during this study are included in this published article and its supplementary information files.