γ‐aminobutyric acid (GABA) is a four‐carbon non‐protein amino acid presented in a wide range of organisms. In this study, a suppression subtractive hybridization (SSH) library was constructed using roots of a legume shrub, Caragana intermedia, with the combined treatment of 300 mm NaCl and 300 mm NaCl + 10 mm GABA. We obtained 224 GABA‐regulated unique expressed sequence tags (ESTs) including signal transduction, transcriptional regulation, hormone biosynthesis, reactive oxygen species (ROS) and polyamine metabolism, etc. The key H₂O₂‐generated genes, NADPH oxidase (CaGR60), peroxidase (CaGR61) and amine oxidase (CaGR62), were regulated at the mRNA level by 10 mm GABA, which clearly inhibited H₂O₂ accumulation brought about by NaCl stress in roots and leaves with the observation of 3,3′‐diaminobenzidine (DAB) staining. Similarly, 10 mm GABA also regulated the expression of 1‐aminocyclopropane‐1‐carboxylic acid (ACC) oxidase (ACO) genes (CaGR30 and CaGR31) and ethylene production in NaCl‐treated roots. Surprisingly, these H₂O₂‐generated genes were enhanced at the mRNA level by a lower concentration of GABA, at 0.25 mm, but not other alternative nitrogen sources, and endogenous GABA accumulated largely just by the application of GABA at either concentration. Our results further proved that GABA, as a signal molecule, participates in regulating the expression of genes in plants under salt stress.

To better understand the soil carbon dynamics and cycling in terrestrial ecosystems in response to environmental changes, we studied soil respiration, litter decomposition, and their relations to soil temperature and soil water content for 18-months (Aug. 2010–Jan. 2012) in three different-aged Pinus massoniana forests in the Three Gorges Reservoir Area, China. Across the experimental period, the mean total soil respiration and litter respiration were 1.94 and 0.81, 2.00 and 0.60, 2.19 and 0.71 µmol CO₂ m⁻² s⁻¹, and the litter dry mass remaining was 57.6%, 56.2% and 61.3% in the 20-, 30-, and 46-year-old forests, respectively. We found that the temporal variations of soil respiration and litter decomposition rates can be well explained by soil temperature at 5 cm depth. Both the total soil respiration and litter respiration were significantly positively correlated with the litter decomposition rates. The mean contribution of the litter respiration to the total soil respiration was 31.0%–45.9% for the three different-aged forests. The present study found that the total soil respiration was not significantly affected by forest age when P. masonniana stands exceed a certain age (e.g. >20 years old), but it increased significantly with increased soil temperature. Hence, forest management strategies need to protect the understory vegetation to limit soil warming, in order to reduce the CO₂ emission under the currently rapid global warming. The contribution of litter decomposition to the total soil respiration varies across spatial and temporal scales. This indicates the need for separate consideration of soil and litter respiration when assessing the climate impacts on forest carbon cycling.