Minireview both adipose and muscle lipoprotein lipase is small in vivo (Grunfeld and Feingold, 1992). Furthermore, TNFa does not appear to interfere with the clearance of triglyceriderich lipoproteins. Rather, TNFa appears to cause hyperlipidemia by stimulating hepatic lipogenesis and synthesis of very low density lipoproteins (VLDL)(Grunfeld and Feingold, 1991).

To study the effects of continuous exposure to TNFa, tumor cells engineered to express high levels of human TNFa have been injected into nude mice (Oliff et al., 1987). In the context of these expanding tumors, a profound wasting of the host animals occurs. It has been suggested that the relative efficiency of this protocol can be explained by the requirement for continual exposure to circulating TNFa for wasting to develop (Oliff, 1988). The ability of TNFasecreting tumor cells to cause wasting in nude mice differs greatly from animal to animal (Teng et al., 1991). These differences correlate closely with circulating TNFa levels. Mice containing 250 pglml or more TNFa in their blood show dramatic wasting, while those with lower levels show little or no significant weight loss despite an equal tumor burden. Thus, it appears that secreted TNFa can indeed cause a cachexia-like syndrome in animals, but primarily when continually secreted at relatively high levels and in the context of an expanding tumor mass. Utilizing various promoters and gene modifications, a number of transgenic mice expressing human TNFa have been generated (Keffer et al., 1991; Cheng et al., 1992). Weight loss has been observed in many of these transgenie animals. However, weight loss has always developed when accompanying other disease states, such as severe polyarthritis, skin disease, internal hemorrhaging, and necrosis of different tissues and organs. Therefore, it isdifficulttodetermine whetherweight loss isduedirectly to TNFa expression per se or indirectly to diseases caused by TNFa expression. In addition, it has not been possible to correlate weight loss with levels of serum TNFa protein in these animal models (Keffer et al., 1991). A straightforward interpretation of these experiments is difficult because of their complex design. The tumors secreting TNFa may express other factors that contribute to weight loss. TNFa itself induces a variety of other cytokines, and some of these, including interleukin-1, have the capacity to produce a wasting syndrome (Grunfeld and Feingold, 1992). It is not possible to conclude from these experiments that TNFa is the primary effector of the wasting as opposed to being a key agent that induces tumors to produce other cachectic substances. Attempts to define the role of TNFa independently of a rapidly expanding tumor or other serious illness have been provided by experiments that use a continuous infusion of recombinant TNFa into normal rats (Michieet al., 1989). These animals eat less and lose weight. However, when control animals were given food equivalent to that eaten by the infused animals (so-called pair feeding), the control animals lost the same or greater amounts of weight. This suggests that TNFa alone can cause anorexia but not