Process optimization and transcriptomic profiling reveal an inverse relationship between gene expression and succinic acid production from sugarcane leaves

Abstract

Succinic acid is a valuable platform chemical widely used in the food, pharmaceutical, and bioplastics industries. This study combines process optimization with transcriptomic analysis to investigate succinic acid production from sugarcane leaf hydrolysate (SLH) using Actinobacillus succinogenes TISTR 1994 and to gain insight into the molecular mechanism underlying succinate biosynthesis from SLH-derived sugars. Systematic optimization identified MgCO<inf>3</inf> concentration, sugar concentration, and initial pH as key factors influencing succinic acid production. Under optimal conditions (52.84 g/L MgCO<inf>3</inf>, 49.68 g/L sugar, pH 7.94), succinic acid production reached 18.87 ± 0.42 g/L with a yield of 0.59 g/g, productivity of 0.39 g/L/h, and 95.3% sugar utilization efficiency, compared to 76.0% under non-optimized conditions. Genome-wide transcriptomic analysis revealed 1075 differentially expressed genes during mid-exponential phase. Notably, non-optimized conditions exhibited higher expression of key reductive tricarboxylic acid (TCA) cycle genes, including malate dehydrogenase (4.84-fold), fumarate reductase (2.00-fold), and phosphoenolpyruvate carboxykinase (1.55-fold), yet resulted in lower succinate titers. This inverse correlation between transcript levels and metabolic output suggests that environmental factors may influence fermentation performance potentially through mechanisms beyond transcription alone. Instead, optimized environmental conditions—particularly MgCO<inf>3</inf> concentration and pH—appeared to support succinate biosynthesis by providing favorable biochemical conditions, including pH stability for enzyme function and sufficient CO<inf>2</inf>/HCO<inf>3</inf>⁻ availability for carboxylation reactions. These findings suggest that process optimization influences succinic acid production from lignocellulosic biomass and that transcript levels alone may be insufficient predictors of fermentation performance, pointing to a potential contribution of post-transcriptional regulation.