**Background:** Micronutrient deficiency ("hidden hunger") affects approximately two billion people globally, particularly in low-income countries. Iron (Fe) and zinc (Zn) are essential for oxygen transport and enzyme function, and their deficiency is prevalent in sub-Saharan Africa (SSA) where maize is a staple crop. Biofortification through breeding is a cost-effective, sustainable intervention. However, low soil nitrogen (N) is a major constraint to maize production in SSA, causing 10–50% annual yield losses. Phytic acid (PA) in maize grain chelates Fe and Zn, reducing their bioavailability. This study aimed to determine combining ability, gene action, and heterosis for grain yield, Fe, Zn, and PA under low N and optimal conditions to identify superior parental lines and hybrids for target environments.
**Methods:** Six maize inbred lines (females) and three testers (males) were selected from 215 South African inbred lines based on low, medium, or high Fe and Zn concentration. Crosses were made using a line × tester (6 × 3) design, generating 18 F1 hybrids. The 18 hybrids and 9 parents were evaluated under low N and optimal conditions at three locations (Potchefstroom, Cedara, Vaalharts) over two seasons (2016–2017) in a randomized complete block design with two replicates. Low N plots were created by depleting soil N over several years without N fertilization. Fe and Zn were measured by atomic absorption spectrophotometry after dry ashing; PA was measured by a colorimetric method using Wade reagent. Molar ratios Fe:PA and Zn:PA were calculated as indicators of bioavailability. Combining ability was analyzed using line × tester analysis, and heterosis was calculated as mid-parent heterosis (MPH) and high-parent heterosis (HPH).
**Key Results:** Under optimal conditions, mean grain yield was 6.14 t/ha (range 2.46–9.33), mean Fe was 18.02 mg/kg (16.33–20.44), mean Zn was 20.39 mg/kg (18.36–23.79), and mean PA was 5.36 mg/g (4.63–6.06). Under low N conditions, mean grain yield was 2.50 t/ha (1.81–3.74), mean Fe was 16.23 mg/kg (11.80–19.44), mean Zn was 18.39 mg/kg (15.11–22.15), and mean PA was 5.89 mg/g (5.05–6.58). Low N reduced grain yield by 59%, Fe by 9%, and Zn by 9%, but increased PA by 10%, Fe:PA by 24%, and Zn:PA by 21%. Both additive and non-additive gene effects were important for all traits. Under low N, GCA/SCA variance ratios were >1 for Zn (4.29), PA (1.08), and Zn:PA (1.13), indicating predominance of additive gene action; ratios were <1 for grain yield (0.41), Fe (0.07), and Fe:PA (0.07), indicating non-additive gene action. Line CBY102 LM-1601 had positive GCA effects for grain yield and Fe under low N. Tester CBY358 LM-1857 was the best general combiner across both N conditions for most traits. Hybrid CBY102 LM-1601 × CBY358 LM-1857 showed significant positive SCA for grain yield under low N (0.21, P≤0.01). Hybrid CBY101 LM-1600 × CBY358 LM-1857 had significant positive SCA for Zn under low N (0.65, P≤0.01). Under low N, MPH for grain yield ranged from −31.82% to 28.08%, and HPH ranged from −35.04% to 24.25%. Narrow-sense heritability under low N was 0.53 for grain yield, 0.47 for Fe, 0.75 for Zn, 0.57 for PA, 0.47 for Fe:PA, and 0.43 for Zn:PA.
**Clinical Implications:** This study provides breeders with specific parental lines (CBY358 LM-1857 as tester, CBY102 LM-1601 as line) that combine high yield potential with elevated Fe and Zn content under both optimal and low N conditions. The finding that low N stress increased Fe and Zn bioavailability (higher Fe:PA and Zn:PA molar ratios) while reducing absolute mineral concentrations suggests that breeding for low-PA genotypes under low N could further improve mineral bioavailability. Hybrid CBY101 LM-1600 × CBY358 LM-1857 is a promising candidate for production in low N environments common in smallholder farming systems in SSA. The predominance of additive gene action for Zn and PA under low N indicates that selection progress is feasible for these traits. These results support the development of biofortified maize hybrids adapted to low-fertility soils, addressing both yield and nutritional security.