PURIFICATION AND CHARACTERIZATION OF A LIPID-MOBILIZING FACTOR ASSOCIATED WITH CACHEXIA-INDUCING TUMORS IN MICE AND HUMAN

A scheme is described for the purification of a lipid-mobilizing facto r from a cachexia-inducing murine tumor (MAC16) using a combination of ion exchange (Mono Q), exclusion (Superose), and hydrophobic (C-8) ch romatography, This process yields an active material with an apparent molecular weight of 24,000 with an overall purification of 3,500 from the tumor homogenate and representing 0.005% of the total protein pres ent, The material tends to aggregate to high molecular mass, is acidic (pI < 4), and displays heterogeneity of charge as evidenced by a broa d elution profile on ion exchange and exclusion chromatography and mul tiple peaks on hydrophobic columns. The purified material was heat and alkali (pH 10.4) labile and activity could be completely inhibited by sulfatase, suggesting that the negative charge could arise from sulfa te residues. There was no evidence that the material possessed triglgc eride lipase activity. Animals transplanted with the MAC16 tumor and w ith a delayed weight loss contained in their serum antibodies that rec ognized a M(r) 24,000 band on Western blots. This material copurified with the lipid-mobilizing factor. Such antibodies were not present in the serum of mice transplanted with the MAC13 tumor, which does not in duce cachexia, suggesting that the antibodies were directed to the ind uction of cachexia rather than the tumor itself. Urine from patients w ith cancer cachexia also contained a lipid-mobilizing factor which adh ered to DEAE-cellulose and gave an apparent M(r) of 24,000 by exclusio n chromatography. Western blotting using serum from MAC16 tumor-bearin g animals show ed the presence of a band of M(r) 24,000 in such fracti ons, which was not detected using serum from mice bearing the MAC13 tu mor. This band was not present in Western blots of urine from normal s ubjects. The fact that serum from mice bearing the MAC16 tumor can det ect the human lipid-mobilizing activity suggests a high degree of stru ctural similarity between the two and raises the possibility that cach exia in humans may be caused by the same species as in the mouse.