The presence of serum in a culture medium makes it impossible to identify whether changed cellular functions are directly caused by a manipulation itself or mediated by a component in serum. Madin Darby canine kidney cells can survive in a serum-free medium for about 48 h. We took this advantage to examine whether low K(+)-induced up-regulation of Na,K-ATPase requires serum. We found that serum was essential for low K(+) to induce an increase in Na,K-ATPase binding sites as quantified by ouabain factor binding assays. In an attempt to identify which component was critical, we screened EGF, IGF1, PGE1 and transferrin to identify which one can replace serum. We discovered that transferrin was the single most important factor that mimicked about 80% to 90% of the effect of serum. Transferrin potentiated the effect of low K(+) on the Na,K-ATPase binding sites in a time- and dose-dependent manner. Furthermore, transferrin was also required for low K(+)-induced increase in alpha(1)-promoter activity, alpha(1)- and beta(1)-subunit protein abundance of the Na,K-ATPase. In the presence of transferrin, low K(+) enhanced cellular uptake of iron approximately by 70%. Inhibition of intracellular iron activity by deferoxamine (30 micromol/L) abrogated the effect of low K(+). We conclude that stimulation of the Na,K-ATPase by low K(+) is critically dependent on transferrin. The effect of transferrin is mediated by increased iron transport.