DNA fragmentation is a common feature of apoptotic cell death and we have previously suggested that the muscle wasting that accompanies cancer cachexia could be linked to an apoptotic phenomenon by which muscle cells lose not only protein but also DNA (Van Royen et al, 2000
). Apoptosis has already been described in human (Tews and Goebel, 1997
; Tews et al, 1997
) and rat (Dalla Libera et al, 1999
) atrophic muscle as well as in insect muscle (Schwartz et al, 1993
). In patients with malignant tumours, anorexia, weight loss, emaciation and progressive alterations of vital functions are common features associated with cancer cachexia (De Wys, 1985
). Although in some cases anorexia, gastrointestinal obstruction or malabsorption are responsible for the weight loss of cachectic patients (Balducci and Hardy, 1985
), it cannot be wholely attributed to these causes and therefore it has been postulated to be due to a decrease in the energetic efficiency of the cancer patient. Among the factors involved in decreasing the energetic efficiency, skeletal muscle protein turnover seems to have a very significant role as we have previously shown (see Argilés et al, 1997
, for review). In addition, apoptosis also seems to be present in cachectic muscle in different experimental tumour models (Van Royen et al, 2000
). The basic aim of the present investigation was to see if the changes that occur in DNA in skeletal muscle during experimental cancer cachexia are linked to TNF. To test this hypothesis we have used two different experimental approximations: chronic TNF administration to healthy rats and experimental cancer cachexia (induced by the Lewis lung carcinoma in mice) in gene-deficient mice for TNFRI.
Different mediators have been suggested to account for cancer-induced cachexia, but basically the presence of both tumoural and humoural (mainly cytokines, TNF in particular) compounds is associated with depletion of fat stores as well as of muscular tissues (Argilés and López-Soriano, 1999
). In fact, the balance between pro-inflammatory cytokines, their soluble receptors and the anti-inflammatory cytokines plays a key role in the development of the cachectic syndrome (Argilés and López-Soriano, 1998
). Our research group has demonstrated that TNF is involved in the activation of the ubiquitin-dependent proteolysis that takes place during tumour growth (García-Martínez et al, 1994
; Llovera et al, 1996
). We clearly show here that TNF is also involved in triggering DNA fragmentation in muscle during cancer cachexia, mainly through the TNFRI. Indeed, chronic administration of recombinant human TNF, which can only bind rat TNFRI, clearly induces DNA fragmentation, and the use of a tumour model (where the levels of circulating TNF are highly increased) confirm this fact.
Indeed, the Lewis lung carcinoma is a cachectic tumour that induces an important decrease in body weight without significant changes in food intake, at least in the two first weeks of tumour growth (Llovera et al, 1998
). Because TNFRI is absent from the cells of these animals, the data obtained here suggest that TNF can be involved in the muscle apoptotic mechanisms triggered by tumour growth through its binding with the TNFRI. However, TNFRI is not the sole receptor responsible for transduction of the death signal, even though it is the most important one. Under certain circumstances, TNFRII also either enhances the TNFRI death signal or, indeed, mediates death independently (Declercq et al, 1998
; Haridas et al, 1998
; Weiss et al, 1998
). The mechanism of the TNFRII death signal has not been characterised. The results clearly show that in the gene-deficient mice apoptosis is not induced by tumour growth to the same extent as in the wild-type animals. In fact, TNF has been shown to trigger apoptosis in many cell types (Obeid et al, 1993
; Ohta et al, 1994
; Sidoti-de-Fraisse et al, 1998
) including cardiac muscle (Krown et al, 1996
). In addition, a possible link between TNF and apoptosis has already been reported in ‘fast’ skeletal muscles in chronic heart failure (Dalla Libera et al, 1999
). Interestingly high circulating levels of TNF are detectable in both rat (Costelli et al, 1993
) and mouse (Llovera et al, 1998
) tumour models used in this study.
Furthermore, TNF-treatment induces Bcl-2 dephosphorylation, targeting this anti-apoptotic protein for degradation by the ubiquitin proteolytic system (Dimmeler et al, 1999
). Thus, TNF-induced apoptosis could be mediated by different cellular responses (Liu et al, 1996
), which include the activation of TNFRI death domain and, as a consequence, the caspase cascade and amplification of this apoptotic pathway by means of ubiquitin-dependent Bcl-2 degradation. Nevertheless, TNF binding to its receptors also induces cell proliferation and survival signals mediated by Bcl-2 activation of NF-κB (Liu et al, 1996
; Wang et al, 1996
). Therefore, cell survival depends on a delicate balance between the different TNF signalling pathways. For this reason, future investigations in our laboratory will concentrate on ascertaining the role of this and other cytokines in the activation of the apoptotic process associated with cancer cachexia in skeletal muscle.