Large-conductance Ca(2+)-activated K+ (BK) channels are widespread and functionally heterogeneous. In other classes of K+ channels, functional heterogeneity derives from large gene families, alternative splicing, heterologous subunit composition, and functional modulation. The molecular basis of mammalian BK channel heterogeneity is unknown, since only a single gene (mSlo) has been identified. BK channels in native vascular smooth muscle have an apparent Ca2+ sensitivity approximately 10-fold greater than native brain or skeletal muscle channels, or cloned mSlo channels. Using mSlo as a low-stringency probe, we screened human arterial smooth muscle and genomic libraries extensively in search of genes or splice variants with novel properties. We isolated the human homologue of mSlo, including two novel splice variant forms, but found no other related genes. Electrophysiological characterization of the hSlo clones in Xenopus oocytes and Chinese hamster ovary cells gave BK currents that were not measurably different from mSlo currents. However, coexpression of hSlo with a recently cloned beta-subunit derived from smooth muscle dramatically increased apparent Ca2+ sensitivity. Thus alpha-subunits alone may not determine Ca2+ sensitivity of vascular smooth muscle BK channels. hSlo was mapped to human chromosome 10q23.1, and the genomic structure was analyzed. Immediately after the amino terminal, two unusual regions of trinucleotide repeating sequences are present. The first of these regions encodes polyglycine, and the second encodes polyserine. Both regions of repeated sequence are conserved between the mouse and human genome.