Changing concepts of dietary fiber: implications for carcinogenesis.

The dietary fiber (DF) hypothesis suggested that DF, as plant cell walls, protected against colorectal cancer. The implicit assumption in much historic literature was that the readily analyzed and quantified nonstarch polysaccharide (NSP) component was critical in cancer protection. However, the presence of polymeric phenolic components such as lignin or suberin has profound effects on the physicochemical properties of the cell walls and largely determines their physiological properties in humans. In certain groups of food plants, degradation of cell walls that contain neither lignin nor suberin releases ferulic acid and other hydroxycinnamic acids. These acids have antioxidant, antimutagenic, and other anticancer effects, including modulation of gene expression and immune response. Reexamination of literature on cancer protection suggests that plant cell walls containing significant amounts of phenolic components may be the most likely to protect against cancer. In the last 30 years, the definition of DF has been incrementally expanded to incorporate NSP extracted from plant cell walls and from sources other than plant cell walls, then resistant starch, and most recently nondigestible oligosaccharides. On the early definitions, increased consumption of DF could only be achieved by increasing the intake of whole grains, brans, or other food plant material that would increase the intake of plant cell walls containing significant amounts of phenolic components. However, the new definitions make it possible to increase "dietary fiber" without consuming any such materials. We suggest that this could have negative connotations for cancer risk in human populations.