Response surface optimization and predictive modeling of shoot regeneration in Enicostemma littorale Blume using central composite design
摘要
Efficient and reproducible regeneration procedures are important for the establishment of reliable in vitro systems in medicinal plants. Here, a central composite design (CCD) consisting of 30 treatments was employed for modelling and optimisation of combined influences of kinetin (KI), 6-benzylaminopurine (BAP), indole-3-acetic acid (IAA) and maltose treatments on direct shoot regeneration from immature leaf explants of Enicostemma littorale Blume. Two responses were evaluated, which were the shoot regeneration frequency (Y1%) and the mean shoots per explant (Y2). Across the design space Y1 varied from 58 to 95% while Y2 varied from 6.0 to 30.0. Quadratic models were significant for both responses (Y1: F = 5.67, p = 0.0009; Y2: F = 5.91, p = 0.0008), with R2 (Y1) and R2 (Y2) were 0.84 and 0.85, respectively. Significant quadratic terms (p < 0.05) indicated the existence of pronounced curvature and mid-range optima and most linear and interaction terms were found not to be influential at the ranges tested. Box-Cox and residual diagnostics showed no need for transformation (λ ~ 1) and hence, both responses were analyzed on the original scale. Numerical optimisation showed an optimum value of 4.02 µM KI, 6.05 µM BAP, 4.01 µM IAA and 25.10 g L⁻1 maltose with predicted responses of 92.0% (Y1) and 8.40 (Y2). Experimental validation results were obtained as 93.20 ± 2.85% (Y1) and 8.70 ± 0.35 (Y2), which supports the practical predictability of the model. The optimized CCD-based protocol is a reproducible protocol for micropropagation and future in vitro applications, including secondary metabolites, in E. littorale.