Anang Triwiratno1, Nirmala Friyanti Devy1*, R. Cinta Badia Ginting2, Joko Purnomo3, Agus Sugiyatno1, Farida Yulianti1, Baiq Nurul Hidayah1, Kurniawan Budiarto1, Khojin Supriadi4, Imro’ah Ikarini5, Emi Budiyati1, Hardiyanto1
1Research Center for Horticulture, Research Organization for Agriculture and Food, National Research and Innovation Agency of Indonesia, Cibinong Science Center, Cibinong, West Java, Indonesia
2Research Center for Applied Microbiology, Research Organization for Life Sciences and Environment, National Research and Innovation Agency, Cibinong Science Center, Cibinong, West Java, Indonesia
3Research Center for Estate Crops, Research Organization for Agriculture and Food, National Research and Innovation Agency of Indonesia, Cibinong Science Center, Cibinong, West Java, Indonesia
4Research Center for Food Crops, Research Organization for Agriculture and Food, National Research and Innovation Agency of Indonesia, Cibinong Science Center, Cibinong, West Java, Indonesia
5Research Center for Food Technology and Processing, Research Organization for Agriculture and Food, National Research and Innovation Agency of Indonesia, Yogyakarta, Gunung Kidul, Indonesia
*Corresponding author’s email: nfdevy@gmail.com
Received: 19 November 2025 / Revised: 25 February 2026 / Accepted: 28 March 2026 / Published Online: 10 April 2026
Abstract
Previous studies have demonstrated drought-induced physiological and molecular responses in citrus under controlled environments; however, whether these stress-response pathways operate predictably across contrasting field agroecosystems and under different propagation strategies remains unclear. This study aimed to determine how propagation strategy modulates citrus growth performance, leaf anatomy, stress-related gene expression, and untargeted metabolomic profiles across swamp and upland agroecosystems during early field establishment. Two different citrus species (Citrus nobilis L. cv. Pontianak Siam and C. reticulata Blanco cv. Keprok RGL) were established under open-field conditions in upland mineral soil and tidal-swamp acid-sulphate soil (pH 4.2–5.2) in East Kalimantan, Indonesia. The propagation techniques used were conventional chip budding and bud-assisted marcotting, locally known as ‘Okucang’. The environmental background was the primary driver of variation. In the soil PCA, the two systems (upland and swamp) were separated along PC1 (72.3% variance). Siam Pontianak, propagated through bud-assisted marcotting, produced significantly taller plants than other treatments, especially under upland conditions (P < 0.05). In contrast, Siam Pontianak propagated through chip budding showed a wider canopy in upland areas. Canopy development (P < 0.05) decreased under swamp conditions. Regardless of variety or propagation method, palisade mesophyll thickness was thicker in swamp areas than in upland areas. Secretory gland diameter was comparable between environments but tended to decrease under swamp conditions. Leaf Fe and Mn content were dominantly accumulated in the swamp area, whereas upland conditions favoured Ca–K enrichment and structural growth. Stress-responsive genes, Osmotin and aquaporins (PIP1/PIP2), were significantly upregulated under swamp conditions (P < 0.05), but their regulation varied with soil redox chemistry and propagation method. Metabolomic profiling additionally discriminated among agroecosystems (PLS-DA PC1 = 32.7%), with terpenoid- and antioxidant-related metabolites (VIP ≥ 1.2). Propagation effects were environment-dependent, reflecting genotype × environment × propagation interactions: bud-assisted marcotting enhanced stress-associated traits under swamp conditions, whereas chip budding promoted structural growth under upland conditions. These results extend previous controlled-drought experiments by demonstrating how established stress-response pathways operate within lowland agroecosystems shaped by soil mineral chemistry and propagation architecture.
Keywords: Citrus, Gene expression, Metabolomics, Propagation method, Swampland
