The constant level of various proteins including albumin and cellular components in intravascular pool in vivo is strictly controlled by an unknown homeostatic mechanism, although there are fluctuations seen in pathologic conditions. Because the majority of the IgG in the serum is regarded as self-reactive natural autoantibodies, IgG may have a role to react with all proteins in vivo. It is hypothesized that like an immune system, a homeostatic mechanism for the protein pool also has a sensitive role to identify and memorize the extent and repertoire of both normal and pathogenic proteins on an individual basis, and IgG may be one of the major players in performing these functions. This hypothesis may explain the unresolved clinical observations as followed: (1) the marked increased IgG levels observed in self-limiting diseases presumed to come from immunological insults such as acute poststreptococcal glomerulonephritis and Kikuchi-Fujimoto disease, (2) an immediate reduction of all protein levels except immunoglobulins after intravenous immunoglobulin (IVIG) treatment in Kawasaki disease, (3) a unified explanation for the variety of immunomodulating effects exerted by IVIG, (4) the IgG-enzyme complexes observed in benign conditions such as macroamylasemia and hyperphosphatasemia, and (5) the marked decreased IgG level, which is correlated with the albumin level in minimal change nephrotic syndrome. IgG may be a 'watch-dog' for the disturbances of protein homeostasis in vivo. IgG may control the pathogenic proteins that appeared in disordered states, and it may help prevent the loss of proteins in case of nephrotic syndrome.