Effect of a tumour-produced lipid-mobilizing factor on protein synthesis and degradation

Treatment of murine myoblasts, myotubes and tumour cells with a tumour-produced lipid mobilizing factor (LMF), caused a concentration-dependent stimulation of protein synthesis, within a 24 h period. There was no effect on cell number or [3H] thymidine incorporation, but a similar concentration-dependent stimulation of 2-deoxyglucose uptake. LMF produced an increase in intracellular cyclic AMP levels, which was linearly (r2= 0.973) related to the increase in protein synthesis. The effect of LMF was attenuated by the adenylate cyclase inhibitor MDL 12330A, and was additive with the stimulation produced by forskolin. Both propranolol (10 μM) and the specific β3-adrenergic receptor antagonist SR 59230A (10–5M), significantly reduced the stimulation of protein synthesis induced by LMF. Protein synthesis was also increased by 69% (P = 0.006) in soleus muscles of mice administered LMF, while there was a 26% decrease in protein degradation (P = 0.03). While LMF had no effect on the lysosomal enzymes, cathepsins B and L, there was a decrease in proteasome activity, as determined both by the ‘chymotrypsin-like’ enzyme activity, as well as expression of proteasome α-type subunits, determined by Western blotting. These results show that in addition to its lipid-mobilizing activity LMF also increases protein accumulation in skeletal muscle both by an increase in protein synthesis and a decrease in protein catabolism. © 2001 Cancer Research Campaign http://www.bjcancer.com

DEAE-cellulose and hydrophobic interaction chromatography (Todorov et al, 1996 ) Particulate material was removed from urine by centrifugation at 3000 g for 10 min followed by dilution with 4 vol 10 mM Tris. HCl, pH 8.0. DEAE cellulose (10 g l -1 of original urine) was added and the mixture was stirred for 2 h at 4˚C. The DEAE-cellulose was recovered by low speed centrifugation and LMF bioactivity, determined by glycerol release from epididymal adipocytes (Khan and Tisdale, 1999), was eluted with 0.5 M NaCl in 10 mM Tris. HCl, pH 8.0. The eluate was equilibrated against PBS and concentrated to 1 ml before further purification using a Resource-Iso HPLC column (Pharmacia Biotech, St Albans, Herts, UK) employing a decreasing (NH 4 ) 2 SO 4 concentration from 1.5 M. Active fractions containing LMF eluted at 0.6 M (NH 4 ) 2 SO 4 , and were desalted before use by washing 5 times against PBS using an Amicon filtration cell. LMF eluted as a single protein band of M r 43000 as determined by Coomassie blue staining of a 12% SDS polyacrylamide gel (Figure 1).

Precursor incorporation
C 2 C 12 myoblasts were seeded at 5 × 10 4 cells ml -1 and MAC16 at 10 5 cells ml -1 in 6-well multidishes containing 2 ml medium per well. After 24 h various concentrations of LMF was added and left for a further 24 h period. During the last 60 min of incubation the cells were incubated with 1.5 µmol L-phenylalanine containing 37 kBq of L-[2, 6-3 H] phenylalanine. The reaction was terminated by removal of medium and washing the cells 3 times with ice-cold PBS. The cells were then incubated at 40˚C for 20 min with 1 ml ice-cold 0.2 M perchloric acid, which was replaced with 1 ml 0.3 M NaOH and incubation continued at 4˚C for a further 30 min, followed by a further 20 min at 37˚C. Cellular protein was precipitated with 2 M perchloric acid (0.5 ml) for 20 min at 4˚C, followed by centrifugation at 3000 g for 10 min at 4˚C. The pellet, which comprised DNA and protein was dissolved in 1 ml of 0.3 M NaOH, and an aliquot (20 µl) was used to measure protein concentration using the Bio-Rad reagent (Sigma Chemical Co, Dorset, UK). The radioactivity in 0.5 ml was determined using a 2000CA Tri-Carb liquid scintillation analyser. The rate of protein synthesis was calculated as dpm µg protein -1 h -1 as described (Southorn and Palmer, 1990). The protocol for DNA synthesis was the same as above except that methyl [ 3 H] thymidine (0.5 µCi) was added at the same time as LMF and left for 48 h. Protein was precipitated using ice-cold 5% trichloroacetic acid for 1 h at 4˚C. Incorporation of 2deoxy-D [2, 6-3 H] glucose ([ 3 H] 2DG) was determined in C 2 C 12 myoblasts 24 h after addition of LMF. The media was removed, and the cells were rinsed once with Krebs Ringer bicarbonate buffer, followed by incubation for 30 min at 37˚C in a further 1 ml of Krebs Ringer bicarbonate, together with 0.1 mM [ 3 H] 2DG (sp. act 74 MBq mmol -1 ). Cells were washed 3 times with ice-cold PBS and incubated on ice for 1 h with lM NaOH (1 ml) and the amount of radioactivity incorporated was determined.
Cyclic AMP determination C 2 C 12 myoblasts were seeded at 4 × 10 4 cells ml -1 in 1 ml medium in a 24-well multi-dish and left for 48 h before addition of LMF for 30 min at 37˚C. The medium was removed and replaced with 0.5 ml 20 mM HEPES, pH 7.5, 5 mM EDTA and 0.1 mM isobutylmethylxanthine and then the cells were heated on a boiling water bath for 5 min, followed by cooling on ice for 10 min. The cell extracts were sonicated on ice followed by centrifugation at 5000 rpm for 15 min. To 50 µl of the cell extract was added 925 Bq of [8-3 H] cyclic AMP and 20 µg of cyclic AMP-dependent protein kinase and incubated for 2 h at 4˚C. Unbound cyclic AMP was removed by adsorption onto charcoal and the concentration of cyclic AMP in the sample determined by comparison with standard curves using known concentrations of cyclic AMP.

Determination of the activation of protein kinase A (PKA)
C 2 C 12 myoblasts were treated with LMF for 24 h and cytosolic fractions were produced by sonication in 0.25 M sucrose, 1 mM EDTA, 1 mM dithiothreitol, 10 µg ml -1 leupeptin and 10 µg ml -1 antipain followed by high speed centrifugation. The activation of PKA in the cellular supernatants was determined using the Pierce colourimetric PKA assay kit, Spinzyme™ Format (Rockford, IL, USA) using peptide substrate (Kemptide) labelled with fluorescent dye. The phosphorylated product was quantitated by measuring absorbance at 570 nm.

Measurement of proteasome activity
C 2 C 12 myoblasts were treated with different concentrations of LMF for 24 h and the activity of the 26S proteasome was determined according to the method of Orino et al (1991). Cellular supernatants were prepared in 20 mM Tris. HCl, pH 7.5, 2 mM ATP, 5 mM MgCl 2 and 1 mM dithiothreitol and incubated with the fluorescent substrate succinyl-LLVY-MCA (0.1 mM) in 100 mM Tris. HCl, pH 8.0. The reaction was terminated by the addition of 80 mM sodium acetate and the fluorescence was measured with an excitation of 360 nm and an emission of 460 nm. The protein concentration of the sample was determined using the Bradford assay (Sigma Chemical Co, Dorset, UK).

Determination of protein synthesis and protein degradation in soleus muscle after LMF administration
Ex-breeder male NMRI mice (average weight 44.50 g) were administered LMF (8 µg, b.d. iv) for 48 h. After termination soleus muscles were isolated and both protein synthesis and degradation were determined essentially as described (Smith and Tisdale, 1993). Protein synthesis was determined by the incorporation of L-[4-3 H] phenylalanine into protein during a 2 h incubation, and the rate of protein synthesis was calculated by dividing the amount of protein bound radioactivity by the amount of acid soluble radioactivity. Protein catabolism was determined by the tyrosine release assay as described (Lorite et al, 1997). Tyrosine release from soleus muscles was determined during a 2 h incubation in Krebs-Henseleit buffer containing 6 mM D-glucose, 1.2 mg ml -1 bovine serum albumin and 130 µg ml -1 cycloheximide. Tyrosine was quantitated by a fluorometric method (Waalkes and Undenfriend, 1957) at 570 nm on a Perkin-Elmer LS-5 luminescence spectrometer. The animal ethics meet the standards required by the UKCCCR Guidelines.

Assay of cathepsins L and B
C 2 C 12 myoblasts pre-incubated for 24 h with LMF were homogenized in 250 mM sucrose, 2 mM EGTA, 2 mM EDTA, 20 mM Tris. HCl, pH 7.4, containing 0.2% Triton X-100 followed by sonication. The supernatants formed after centrifugation at 18 000 g for 15 min were used to determine cathepsin activity as described (Lorite et al, 1998) using the fluorometric substrates N-CBZ-Phe-Arg-7-amids-4-methylcoumarin for cathepsin L and N-CBZ-Arg-Arg-7-amido-4-methylcoumarin for cathepsin B. The flourescence of the free aminomethylcoumarin was determined at an excitation wavelength of 370 nm and an emission wavelength of 430 nm.

RESULTS
A dose-response curve showing the effect of increasing concentrations of LMF on protein synthesis in C 2 C 12 myoblasts is shown in Figure 2A. Protein synthesis was increased in a concentrationdependent manner, with a maximal 40% stimulation above control values at a concentration of LMF of 580 nM (P < 0.001 from control). A similar dose-response relationship was obtained for LMF on protein synthesis in C 2 C 12 myotubes ( Figure 2B), and in MAC16 cells ( Figure 2C). LMF also produced a stimulation of 2-deoxyglucose uptake into C 2 C 12 myoblasts ( Figure 2D) and MAC16 tumour cells ( Figure 2E) with a dose-response curve similar to that for stimulation of protein synthesis. There was no effect of LMF on cell number or [ 3 H] thymidine incorporation into any cell line, showing the action of LMF to be specific for protein synthesis. LMF produced an early (within 30 min) increase in cyclic AMP levels in C 2 C 12 myoblasts, which was linearly (r 2 = 0.973, P = 0.004) related to the increase in protein synthesis after 24 h ( Figure 3). This suggests that the 2 effects may be related. There was an increase in protein kinase A (PKA) with increasing concentrations of LMF, which reached maximum stimulation at 58 nM LMF and was independent of LMF concentration up to 580 nM (P < 0.05 from control) (Figure 4). The stimulating effect of LMF on protein synthesis in C 2 C 12 myoblasts was attenuated by the adenylate cyclase inhibitor MDL 12330A , but not by the cyclic AMP-dependent protein kinase inhibitor H8 at 10 µM (Table 1). Both forskolin (25 µM) and dibutryl cyclic AMP (1 µM) stimulated protein synthesis, confirming a role for cyclic AMP in the process. Stimulation of protein synthesis by forskolin, but not by dibutryl cyclic AMP was additive with that of LMF (Table 1). The induction of protein synthesis in C 2 C 12 myoblasts by LMF was partially inhibited by a polyclonal antibody to zinc-α 2 -glycoprotein (ZAG) ( Table 2), and the non-specific β-adrenergic receptor antagonist, propranolol (10 µM) ( Figure 5). The specific β 3 -adrenergic receptor antagonist, SR59230A (Nisoli et al, 1996) at a concentration of 10 -5 M significantly reduced LMF stimulation of protein synthesis down to control levels ( Figure 6). This suggests that stimulation of protein synthesis by LMF may be mediated through a β 3 -adrenergic receptor. There was no effect on the LMF stimulation of protein synthesis by inhibitors of p70S6 kinase (rapamycin, 0.5 ng ml -1 ), mitogen-activated protein kinase (PD 98059, 0.63 µM), or of phosphatidylinositide-3-OH kinase (wortmannin, 0.02 µM, or LY 294002, 0.05 µM) ( Table 2). Administration of LMF (8 µg, i.v., b.d.) to ex-breeder male NMRI mice caused a progressive decrease in body weight (Figure 7), which became significantly different from control animals administered PBS within 24 h of the first injection. Despite this overall loss of body weight, which was exclusively fat, soleus muscles from LMF treated animals showed a 69% increase in protein synthesis (P = 0.006 from control) ( Figure 8A) and a 26% decrease in protein degradation (P = 0.03 from control) ( Figure 8B) 48 h after the first injection of LMF. Myosin levels were also increased in soleus muscle from mice receiving LMF ( Figure 8C). Densitometric analysis showed a 46 ± 9% (P < 0.05 from control) increase in myosin levels after LMF. In C 2 C 12 myoblasts LMF had no effect on the activity of the lysosomal enzymes, cathepsins L or B (data not shown), but produced a progressive decrease in the functional activity of the proteasome, as determined by the 'chymotrypsin-like' enzyme activity, using the fluorogenic substrate succinyl LLVY-MCA ( Figure 9A). Western blotting of cellular supernatants of LMF treated cells with MCP231 antibody, a murine monoclonal to the 20S proteasome, which reacts with the α-type subunits, showed a decrease in expression with increase in LMF concentration ( Figure 9B) parallelling the decrease in functional activity. Unlike LMF, PIF produced an increase in expression of the 20S proteasome α-subunits, which varied with the concentration, reaching a maximal 77% increase at 10.4 nm PIF ( Figure 10A). Addition of LMF (580 nM) completely attenuated the increased expression of the 20S α-subunits in the presence of PIF ( Figure 10B).

DISCUSSION
Loss of skeletal muscle protein is a characteristic feature of cancer cachexia leading to immobility of the patient and eventually to death (Tisdale, 1997). This process has been attributed to tumour production of PIF, which inhibits protein synthesis and increases protein catabolism in skeletal muscle (Todorov et al, 1996;Lorite et al, 1997). However, the present study shows that LMF, also produced by cachexia-inducing tumours, appears to oppose the action of PIF by increasing protein synthesis and decreasing protein degradation in muscle. This action of LMF has been demonstrated in the soleus muscles of mice administered LMF as well as in isolated myoblasts and myotubes. The effect of LMF on protein synthesis was attenuated, both by propranolol, a nonspecific β-adrenergic agonist as well as by SR59230A, which has been reported (Nisoli et al, 1996) to have a 10-fold selectivity for the β 3 -over the β 1 -AR, suggesting that the action of LMF may be mediated through a β-AR. β-Adrenergic agonists have been shown to lead to increased muscle protein synthesis, accompanied or followed by decreased protein degradation (Bell et al, 1998), through a cyclic AMP-dependent pathway. The increase in protein synthesis produced by LMF in C 2 C 12 myoblasts was also linearly related to increase in cyclic AMP levels and attenuated by the adenylate cyclase inhibitor, MDL 12330A . The mechanisms leading from an increase in cyclic AMP to increased protein synthesis have not been fully elucidated.
Gene regulation by polypeptide growth factors is thought to be mediated by transcription factors controlled either by the mitogenactivated protein (MAP) kinase pathway (Davis, 1993) or the p70S6 kinase (p70 s6k ) (Sturgill and Wu, 1991), which rapidly phosphorylates the S6 protein of the 40S ribosomal subunit. Stimulation of protein synthesis in L6 myoblasts was blocked by inhibitors of p70 s6k (rapamycin) and MAP kinase (PD-98059), as well as by inhibitors of phosphatidylinositide-3-OH kinase (wortmannin) (Kimball et al, 1998). However, neither rapamycin, PD-98059, wortmannin or the potent and specific phosphatidylinositide-3-OH kinase had any effect on LMF stimulation of protein synthesis in C 2 C 12 murine myoblasts, suggesting an alternative pathway was involved. While in COS-7 cells cyclic AMP activates the MAP kinase pathway (Faure et al, 1994) in rat phaeochromocytoma, PC12, it stimulates the MAP kinase isoenzyme extracellular signal-regulated kinase 1 (ERK1) (Frodin et al, 1994). In addition the ribosomal protein S6 is directly phosphorylated by cyclic AMP-dependent protein kinase (Wettenhall and Cohen, 1992). Transcriptional regulation following stimulation of adenylate cyclase can also be mediated by the family of cyclic AMPresponse element (CRE)-binding proteins (Habener, 1990). These factors are phosphorylated by PKA with increasing concentrations of cyclic AMP, leading to high stimulation in the transactivating potential (de Groot et al, 1993). These CRE-binding proteins may be involved in LMF stimulation of protein synthesis, although the  Figure 5 The effect of the non-specific β-AR antagonist propranolol (10 µM) on LMF-induced protein synthesis in C 2 C 12 myoblasts. Propranolol was added 1 h prior to LMF and protein synthesis was determined 24 h after addition of LMF (closed boxes) and compared with cells treated with LMF alone (open boxes). The values represent means ± SEM where n = 3 and the experiment was repeated 4 times. Statistical analysis was performed using one-way ANOVA with Student-Newman-Keuls test and differences are indicated as a, P < 0.05 from LMF alone and b, P < 0.01 from control 1 0 −5 10 −6 SR 59243A Figure 6 The effect of the specific β3-AR antagonist. SR59230A on LMFinduced protein synthesis in C 2 C 12 myoblasts. SR59230A was added 1 h prior to the addition of LMF (580 nM) and protein synthesis was determined after a further 24 h in the absence of LMF (open boxes) or in the presence of LMF (closed boxes). The values represent means ± SEM where n = 3 and the experiment was repeated 4 times. Statistical analysis was performed using one-way ANOVA with Student-Newman-Keuls test and differences are indicated as a, P < 0.001 from cultures not incubated with LMF and b, P < 0.01 and c, P < 0.001 for protein synthesis in the absence of SR59230A ) on body weight of exbreeder male NMRI mice (q) compared with animals administered PBS (q q). Body weight was measured prior to each injection and the weight loss is shown as means ± SEM where n = 5. The average body weight of the mice on initiation of the experiment was 44.50 ± 1.13 and was 41.34 ± 0.76 after 48 h treatment with LMF. Differences from control values were determined by Student-Newman-Keuls test and are indicated as a, P < 0.05; b, P < 0.01 and c, P < 0.001.
lack of inhibition of PKA by H8 at 10 µM would negate against the possibility. However, it is possible that higher concentrations of H8 are required for inhibition of PKA in this system. This requires further investigation.
Despite the stimulatory effect of LMF on protein synthesis there was no effect on DNA synthesis or cell number, suggesting a hypertrophic response to this tumour factor. Nevertheless protein synthesis was also enhanced in tumour cells suggesting that LMF is potentially a growth factor for the tumour. LMF also stimulated 2-deoxyglucose uptake into C 2 C 12 myoblasts which suggests that it facilitates glucose utilization. Administration of LMF to mice produces a decrease in blood glucose (Hirai et al, 1998) confirming the ability to stimulate glucose utilization. Li and Adrian (1999) have also reported that pancreatic cancer cells produce a bioactive factor which stimulates glucose uptake and utilization in murine myoblasts. Although the reported M r of this factor is much lower than that of LMF, in separate experiments (unpublished) we have shown LMF to undergo tryptic cleavage to yield a bioactive fragment of comparable molecular weight.
In addition to stimulation of protein synthesis LMF also attenuates protein catabolism in skeletal muscle. The ubiquitinproteasome system is considered to be the major pathway for selective protein breakdown in muscle, while lysosomal proteolysis plays only a minor role (Attaix and Taillander, 1998). While there is no effect of LMF on the lysosomal proteolytic enzymes cathepsins B and L, there is a significant inhibition of proteasome catalytic activity, through a decreased expression of the proteasome α-type subunits. Since protein catabolism in cachexia appears to be due to an up-regulation of proteasome expression (Lorite et al, 1998) then LMF appears to be antagonistic to tumour proteolytic factors, such as PIF, and as such may modulate the rate of loss of skeletal muscle mass. In the MAC16 murine cachexia model LMF bioactivity is maximally elevated at 15% weight loss and thereafter declines (Groundwater et al, 1990). In this model a decrease in protein synthesis and an increase in protein degradation is not seen until the weight loss exceeds 16% (Smith and Tisdale, 1993). This suggests that either PIF production is not apparent at low weight loss or that LMF attenuates the action of PIF. This can be determined by administration of the pure factors to mice.
Thus LMF increases muscle mass by an increase in protein synthesis and a decrease in protein catabolism. The effect on protein synthesis appears to arise from increases in intracellular cyclic AMP, possibly mediated through stimulation of a β 3adrenergic receptor. Since LMF was also found to stimulate