Human hyaluronidases have been considered to be the enzymes acting at the initial step in the catabolism of chondroitin sulfate (CS) in vivo. However, human hyaluronidase-1 digests CS more slowly than hyaluronan (HA), and its preferred substrate is HA rather than CS. We have identified a chondroitin hydrolase in Caenorhabditis elegans, which effectively degrades chondroitin but depolymerizes HA to a much lesser extent (Kaneiwa T, Yamada S, Mizumoto S, Montaño AM, Mitani S, Sugahara K. 2008. Identification of a novel chondroitin hydrolase in Caenorhabditis elegans. J Biol Chem. 283:14971-14979), suggesting the existence of CS-specific endoglycosidases in mammalian systems. In this study, human hyaluronidase-4 was demonstrated to be a CS-specific endo-beta-N-acetylgalactosaminidase. This is the first demonstration of a CS hydrolase in higher organisms. The specificity of a purified recombinant form of the enzyme was investigated in detail through the characterization of degradation products. The best substrate of the CS hydrolase was the galactosaminidic linkage in the sequence of a trisulfated tetrasaccharide GlcUA(2-O-sulfate)-GalNAc(6-O-sulfate)-GlcUA-GalNAc(4-O- or 6-O-sulfate), where GlcUA and GalNAc represent D-glucuronic acid and N-acetyl-D-galactosamine, respectively. The disaccharide unit on the nonreducing side, GlcUA(2-O-sulfate)-GalNAc(6-O-sulfate) (D unit), is rich in shark fin cartilage CS-D among various CS isoforms. CS hydrolase will be a useful tool for investigating CS-specific functions in tissues and cells. In addition, it may well be applicable to the treatment of acute spinal cord injuries as in the case of, or instead of, the bacterial CS lyase which has been used for recent clinical trials.