and could have been used to establish an AI. However, establishing an AI also requires a clearly defined, reproducible indicator in humans sensitive to a range of intakes. Indicators that meet this criterion for establishing an AI are not currently available for any of these minerals, and therefore no AI was set.
Notwithstanding, observations of deficiency effects (e.g., on growth and development) in multiple animal species and data from limited human studies suggest beneficial roles for arsenic, boron, nickel, silicon, and vanadium in human health. These data clearly indicate a need for continued study of these elements to determine their metabolic role, identify sensitive indicators, and more fully characterize specific functions in human health.
Estimates of Tolerable Upper Intake Levels (UL) were set for boron, nickel, and vanadium. The ULs for boron and vanadium are based on animal data and have been set for adults at 20 mg/day and 1.8 mg/day, respectively. The UL for nickel is 1 mg/day. There were insufficient data using the model described in Chapter 3 to set a UL for arsenic and silicon.
There have been no studies to determine the nutritional importance of arsenic for humans. Although the metabolic function of arsenic is not well understood, one study in rats suggests that arsenic may have a role in the metabolism of methionine (Uthus and Poellot, 1992). Arsenic deprivation was associated with an increase in hepatic S-adenosyl-homocystine concentrations and a decrease in hepatic S-adenosyl-methionine concentrations. Arsenic deprivation has also been associated with impaired growth and abnormal reproduction in rats, hamsters, chicks, goats, and miniature pigs (Anke, 1986; Uthus, 1994). Arsenic has also been suggested to be involved with the regulation of gene expression (Meng and Meng, 1994). Arsenite is associated with changes in the methylation of core histones and therefore is active at the transcriptional level (Desrosiers and Tanguay, 1986).