Drought and salinity are two major limiting factors in crop productivity. One way to reduce crop loss caused by drought and salinity is to increase the solute concentration in the vacuoles of plant cells. The accumulation of sodium ions inside the vacuoles provides a 2-fold advantage: (i) reducing the toxic levels of sodium in cytosol; and (ii) increasing the vacuolar osmotic potential with the concomitant generation of a more negative water potential that favors water uptake by the cell and better tissue water retention under high soil salinity. The success of this approach was demonstrated in several plants, where the overexpression of the Arabidopsis gene AtNHX1 that encodes a vacuolar sodium/proton antiporter resulted in higher plant salt tolerance. Overexpression of AtNHX1 increases sodium uptake in vacuoles, which leads to increased vacuolar solute concentration and therefore higher salt tolerance in transgenic plants. In an effort to engineer cotton for higher drought and salt tolerance, we created transgenic cotton plants expressing AtNHX1. These AtNHX1-expressing cotton plants generated more biomass and produced more fibers when grown in the presence of 200 mM NaCl in greenhouse conditions. The increased fiber yield was probably due to better photosynthetic performance and higher nitrogen assimilation rates observed in the AtNHX1-expressing cotton plants as compared with wild-type cotton plants under saline conditions. Furthermore, the field-grown AtNHX1-expressing cotton plants produced more fibers with better quality, indicating that AtNHX1 can indeed be used for improving salt stress tolerance in cotton.

Industrial processes to produce ethanol from lignocellulosic materials are available, but improved efficiency is necessary to make them economically viable. One of the limitations for lignocellulosic conversion to ethanol is the inaccessibility of the cellulose and hemicelluloses within the tight cell wall matrix. Ferulates (FA) can cross-link different arabinoxylan molecules in the cell wall of grasses via diferulate and oligoferulate bridges. This complex cross-linking is thought to be a key factor in limiting the biodegradability of grass cell walls and, therefore, the reduction in FA is an attractive target to improve enzyme accessibility to cellulose and hemicelluloses. Unfortunately, our knowledge of the genes responsible for the incorporation of FA to the cell wall is limited. A bioinformatics prediction based on the gene similarities and higher transcript abundance in grasses relative to dicot species suggested that genes from the pfam family PF02458 may act as arabinoxylan feruloyl transferases. We show here that the FA content in the cell walls and the transcript levels of rice genes Os05g08640, Os06g39470, Os01g09010 and Os06g39390, are both higher in the stems than in the leaves. In addition, an RNA interference (RNAi) construct that simultaneously down-regulates transcript levels of these four genes is associated with a significant reduction in FA of the cell walls from the leaves of the transgenic plants relative to the control (19% reduction, P < 0.0001). Therefore, our experimental results in rice support the bioinformatics prediction that members of family PF02458 are involved in the incorporation of FA into the cell wall in grasses.Electronic supplementary materialThe online version of this article (doi:10.1007/s00425-009-1077-1) contains supplementary material, which is available to authorized users.