induction of muscle protein degradation and weight loss by ... · induction of muscle protein...

(CANCER RESEARCH 56. 1256-1261, March 15. 19961

Induction of Muscle Protein Degradation and Weight Loss by a Tumor Product1

Penio T. Todorov, Trudi M. McDevitt, Peter Cariuk, Brian Coles, Melanie Deacon, and Michael J. Tisdale2

Cancer Research Campaign Nutritional Biochemistry Research Group, Pharmaceutical Sciences institute, Aston University, Aston Triangle. Birmingham B4 7I-~TÂ¡P. T. T..

T. M. M., P. C.. M. D., M. J. T.], and Cancer Research Campaign Molecular Toxicology Group, Department of Biochemistry, University College antl Middlesex School ofMedicine, Windeyer Building, Cleveland Street. London WIP 6DS Â¡B.C.], United Kingdom

ABSTRACT

Spininoli", from mice bearing a cachexia-inducing tumor (MAC 16)

have been fused with mouse myeloma cells to produce hybridomas, whichhave been cloned to produce antibody reactive to a material which copu-rified with a lipid-mobilizing factor isolated from the same tumor. The

monoclonal antibody has been used to investigate factors potentiallyinvolved in the development of cachexia. The major protein detectable byimmunoprecipitation of a partially purified lipid-mobilizing factor was

Mr 69,000, whereas Western blotting showed two bands of Mr 69,000 andU, 24,000. Although the monoclonal antibody did not neutralize lipid-

mobilizing activity in an in vitro assay, it did neutralize a serum factorcapable of protein degradation in isolated gastrocnemius muscle. Affinitypurification of MAC 16 tumor homogenates using the monoclonal antibody yielded two immunoreactive bands of Mr 69,000 and Mr 24,000,which were further fractionated on a hydrophobic column (C8). Thismaterial was capable of inducing tyrosine release from isolated gastrocnemius muscle, and the effect could be blocked with the monoclonalantibody. The two immunoreactive bands from the hydrophobic columnwere capable of inducing weight loss in mice, whereas nonimmunoreactivefractions had no effect on body weight. The U, 24,000 species had aunique amino acid sequence, whereas the .I/, 69,000 species gave the samesequence as the .\/,. 24,000 material, together with that for albumin.The \l, 24,000 species contained carbohydrate, and lectin blotting showeda strong reaction with wheat germ and Erythrina crystagalli agglutinins.This suggests that the material is a glycoprotein or proteoglycanthat shows strong binding affinity for albumin, possibly through thecarbohydrate residues.

INTRODUCTION

Cancer cachexia is a complex metabolic problem characterized byweight loss with depletion of both skeletal muscle and adipose tissuemass ( I ). Although anorexia commonly accompanies cachexia,weight loss can occur even in patients with a normal energy intake (2).In addition, many patients either maintain body weight or lose weightwhile receiving calories that would be predicted to result in weightgain (3). These effects arise from an increased resting energy expenditure in some cancer patients (4). The absence of a direct nutritionalcause has led to numerous studies into the identification of mediatorsof the cachectic process.

Such mediators might be expected to have diverse biological effectsin view of the multifunctional host impairment observed in cachexia.A number of studies have suggested cytokines, such as tumor necrosisfactor-a and IL3-1 and IL-6, as mediators of the process of cachexia.

These molecules do not initiate direct lipid catabolism from adiposetissue but rather indirectly by inhibition of the enzyme lipoproteinlipase, preventing adipocytes from extracting fatty acids from plasmalipoproteins and resulting in a net flux of lipid into the circulation (5).Although capable of inducing protein degradation in vivo (6, 7), invitro incubation of skeletal muscle with the cytokines failed to show

Received 9/6/95; accepted 1/15/96.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

' This work was supported by Cancer Research Campaign Grant SPI518.2 To whom requests for reprints should be addressed.' The abbreviation used is: IL. interleukin.

a direct catabolic effect (7, 8), suggesting that the in vivo effects maybe triggered by an unknown intermediary factor. This view wassubstantiated with a study of cancer patients with weight loss thatshowed in vitro proteolysis-inducing factors in the serum not found in

healthy subjects (9). The bioactivity in 20% of the samples waspartially abrogated by antibodies to recombinant IL-I, suggesting thatthe accelerated breakdown of protein may be mediated by IL-I in

cooperation with other unidentified factors.Our own studies using the MAC 16 murine cachexia model have

also provided evidence for a proteolysis-inducing factor in the serum

of animals with weight loss (10). Protein degradation was specific tothe cachectic state since similar material was not present in the serumof mice bearing the MAC 13 tumor, which does not produce weightloss (11). Recently, we described a polyclonal antibody present in theserum of mice bearing the MAC 16 tumor but absent from micebearing the MAC 13 tumor, which showed immunoreactivity towardsa material of apparent molecular weight of Mr 24,000, which copu-rified with a lipid-mobilizing factor associated with cachexia (12).

This report describes the production of a monoclonal antibody to thismaterial using splenocytes from mice bearing the MAC 16 tumor andthe evaluation of this antibody in the purification of a potentialcachectic factor.

MATERIALS AND METHODS

Chemicals. RPMI 1640 tissue culture media and myoclone plus PCS werepurchased from GIBCO-BRL (Scotland. United Kingdom). Hypoxanthine-aminopterin-thymidine media supplement and BM-Condimed HI were pur

chased from Boehringer Mannheim UK (Lewes, East Sussex, United Kingdom), and the protein A-Sepharose was purchased from Sigma Chemical Co.

Ltd. (Poole, Dorset, United Kingdom).Preparation of Splenocytes for Fusion. Animals were transplanted with

the MAC 16 tumor and assessed for the development of antibodies to thelipid-mobilizing factor (12) using serum isolated from blood removed from the

tail vein. Suitable animals were killed, and the spleens were removed underaseptic conditions and placed in a 100-mm Petri dish containing 10 ml ofRPMI 1640 without serum and prewarmed to 37Â°C.The spleens were teased

apart until most of the cells had been released. The cells were then transferredinto a sterile centrifuge tube, leaving behind the larger pieces of tissue. Theresidue was washed with an additional 10 ml of RPMI 1640. and the washingswere combined. The cell suspension was allowed to stand for 2 min, afterwhich the supernatant was removed from the sediment and transferred to afresh centrifuge tube. These cells were then ready for fusion.

Preparation of Myeloma Cells for Fusion. Mouse BALB/c myelomaP3 X 63Ag 8.653 was obtained from the European Collection of Animal CellCultures (PortÃ³n Down, Wiltshire. United Kingdom). Cells were passaged inRPMI 1640 supplemented with 10% myoclone plus FCS. One day prior to thefusion, cells were diluted to a concentration of 5 x lOVml. and on the day offusion, cells were diluted with an equal volume of medium supplemented with20% myoclone plus FCS.

Fusion. The splenocytes and the myeloma cells were washed in separatecentrifuge tubes with RPMI 1640 without serum. After the wash. IO8 mousesplenocytes were mixed with 2 x 10* myeloma cells and centrifugea together

at 400 x g for 5 min. The supernatant was completely removed, and 1.5 ml ofpolyethylene glycol 1500 (w/v) prewarmed to 37Â°Cwas added to the pellet

drop by drop over a period of 1 min while the cells were continually stirred.This was followed by the addition of I ml of RPMI 1640 prewarmed to 37Â°C

over the next minute, followed by an additional 9 ml of RPMI 1640 over the

1256

Research. on September 20, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

PROTEIN-I)I:(Â¡RAI)ING FACTOR AND CACIIKX1A-INI31TING TUMORS

next 2 min. all with continuous stirring. The cells were centrifugea at 400 X g

for 5 min. the supernatant was removed, and the cells were resuspended in10 ml of RPMI 1640 supplemented with 20% myoclone plus PCS, 1X HATmedia supplement, and 10% BM-Condimed HI. The cells were transferred to

200 ml RPMI 1640 containing the above, and 100 ml of the cell suspensionwere dispensed into the wells of twenty 96-well microtiter plates and incubatedat 37Â°Cunder an atmosphere of 5% CO2 in air.

The clones were visible by microscopy at about day 5. and the fusionproduced about I(XX)hybridomas. Hybridomas were evaluated for productionof the requisite monoclonal antibody by an ELISA assay. Antigen obtainedfrom superÃ³se fractions of lipid-mobilizing activity purified from the M AC 16

tumor (12) was immobilized onto a PVC plate. Antibody was allowed to reactwith the antigen, and unbound antibodies were removed by washing. Thebound antibodies were detected by a second peroxidase-labeled antimouse

imniunoglohulin (Amershaml at 492 nm using an Anthos Leader 2001 platereader (Lab Tech Instruments). Using this assay, two positive hybridomaclones were identified. The antibody-producing cells were single-cell cloned

by limiting dilution, and this process was repeated four times.Propagation of Hybridomas and Isolation of Monoclonal Antibody.

The hybridomas were routinely passaged in RPMI 1640 containing 25 niMHEPES. 5% L-glutamine, and 10% fetal bovine serum (Myoclone Plus) under

an atmosphere of 5% CO, in air. Cells were routinely subcultured every 4 days.For maxima] production of monoclonal antibody, cells were left for 7 days, anda cell-free supernatant was obtained by centrifugation at 3000 rpm for 5 min

using a bench centrifuge. The supernatant was adjusted to pH 8 with IM Tris,and the antibody solution was slowly passed through a protein A-Sepharosecolumn (binding capacity. â€”¿�35mg human IgG/ml). The protein-A column waswashed with 60-column volumes of 100 mM Tris (pH 8), followed by 30-

column volumes of 10 mM Tris (pH 8). The antibody was eluted slowly with1(X)HIMglycine (pH 3). and fractions of 0.5 ml were collected. The fractionswere neutralized with 50 (Â¿Iof IM Tris (pH 8). The protein concentration of theindividual fractions was determined by the Bio-Rad method, and fractions

containing protein were pooled. The protein was concentrated by a diafloullrafiltration membrane (YMIO) in a stirred ultrafiltration cell (Amicon Inc..Beverly. MA). The concentration of monoclonal antibody was found to be 8.6mg per 1 of tissue culture supernatant. A control mouse IgG2a raised againsta rat Â¡mmunoglohulin was kindly supplied by M. Valeri (Institute of CancerResearch. Surrey. United Kingdom).

Purification of Lipid-mobilizing Factor from the MAC16 Tumor. This

was carried out as described previously ( 12).Purification of a Putative Cachectic Factor. Solid MAC16 tumors, ex

cised from mice with established cachexia. were homogeni/.ed in 10 mMTris-HCI (pH 8.0) containing 0.5 mM phenylmethylsulfonyl fluoride. 0.5 mM

EGTA. and I mM DTT. The homogenate was subjected to low speed centrifugation (4000 rpm for 15 min on a benchtop centrifuge), and ammoniumsulfate (40% w/v) was slowly added to the supernatant with stirring at 4Â°C.The

precipitate was removed by centrifugation (4500 rpm for 20 min), and thesupernatant was concentrated using a Microcon microconcentrator containinga membrane filter with a molecular weight cutoff of Mt K).(XX)(Amicon)against PBS. Any precipitate was removed by centrifugation (4500 rpm for 20min), and the supernatant was applied overnight to an affinity column con-

B

""//in' Â¿JLJ â€”¿�k

4 5

200

97

69

46

30

21

14

CkDa 1

97 -69 -

46

30

21 -14

kDa

97 â€”¿�

69 -

46 â€”¿�

30 â€”¿�

21 â€”¿�

14 â€”¿�

DkDa 1 2

66 -

45 -36 -

29 -24 -

Fig. 1. Immunoprecipitation of lipid-mobilizing factor with mouse monoclonal antibody. The details of the procedure are given in the legend Io Table I. A, Coomassie bluestaining of a 15% SDS-PAGE gel of immunoprecipitated material (Lanes 2 and 3)compared with the monoclonal antibody (Limes 4 and 5). Lime I contains [he molecularmarkers. B. immunoblol analysis of immunoprecipilated material using the MAC 16monoclonal antibody (Lime 2). Lane I contains the molecular markers. C Western blotof ELISA-positive fractions eluted from the affinity column (Table 3). lume I, molecularmarkers: Lanes 2-4. ELISA-positive fraction. D, silver staining of a 15% SDS-PAGE gel

of immunoreactive fraction isolated by affinity chromatography of a MAC 16 tumorhomogenate (Table 3).

Table 1 Immnnoprecipilalitm inni nentnili-iition of lipiil-mohili-m^ itctivity with a

mouse monoclonal imtihoilv

Treatment Â¿Â¿molglycerol/10" adipocytes/mg protein"

ControlImmunoprecipitation with MAC 16 lgG2'f

Immunoprecipitation with nonspecific lgG2''

Plus MAC 16 IgG2a'i

9.5 Â±0.51.4 Â±0.3'

11.2 Â±1.5

9.5 Â±0.7" Results are given for a typical experiment. The experiment was repeated on three

separate occasions, and the values did not differ by more than 59Ã•from those shown.''Protein A-Sepharose in 50 Â¡i\of 10 mvt Tris â€¢¿�HC1 (pH 8.0) was mixed with

lipid-mobilizing factor (7.5 ng) and monoclonal antibody (25 /ig) and incubated overnightwith continuous agitation. After centrifugation. the ability of the supernatant to liberateglycerol from freshly isolated murine adipocytes was determined as described previously(12).

cP < 0.001 by Student's 1 test.' Lipid-mobili/ing factor (7.5 /ig) was incubated with monoclonal antibody (25 fig)

overnight at 4Â°C.and glycerol release from adipocytes was determined.

taining monoclonal antibody coupled to Affi Gel Hz. according to the manufacturer's instructions (Bio-Rad. Hemel Hempstead. United Kingdom). The

coupling yielded 5 mg of antibody per ml of gel. The column was washed withPBS until no further protein eluted. and the retained proteins were eluled with100 HIMglycine HC1 (pH 2.5). After neutralization with 0.5 ml IM Tris â€¢¿�HC1

(pH 8), the Â¡mmunoreactivity of the individual fractions was determined by anELISA plate assay. Positive fractions were combined, and the volume wasreduced by ultrafiltration in an Amicon filtration cell against water until aprotein concentration of 2 mg/ml was obtained. Further tractionation wasachieved using a Brownlee Aquapore RP-300. Cs column (Applied BiosystemsLtd.) size 100 x 2.1 mm. Portions (60 /il. containing 70-90 /ng protein) wereapplied to the column, and the mobile phase was A. HPLC-grade water

(Fisons, Loughborough. United Kingdom) containing 0.069Ã• trifluoroaceticacid and B acelonitrile 190 (Romil Chemicals Ltd.. Loughborough. UnitedKingdom) containing 0.04% trifluoroacetic acid with a gradient of 2 to 65% Bin A over a 30-min period, followed by 65 to 100% B in A over 10 min and

100% B for 10 min. The flow rate was 0.2 ml/min. Absorbance was monitored

1257



PROTEIN-DEGRADING FACTOR AND CACHEXIA-INDUCING TUMORS

Table 2 Effect of serum on tyrosine release from isolateti gastrocnemiiis muscleDifferences were determined by two-way ANOVA.

TyrosineTreatment (nmol/mg/2 h)

Normal mouse serum 39 Â±2

Serum from cachectic MAC 16 mice 54 Â±6"

Scrum from cachectic mice + MAC16 monoclonal antibody 41 Â±3*

Cachectic mouse serum + nonspecific IgG2a 47 Â±2Normal mouse serum + antigen' 56 Â±5rf

Normal mouse serum -t- antigen + MAC16 monoclonal antibody 21 Â±2f

Normal mouse serum + antigen + nonspecific lgG2a 54 Â±3" P < 0.05 when compareti with normal mouse serum.

P < 0.01 when compared with serum from cachectic mice alone.Antigenic material was purified from the MAC16 tumor us described in Table 3.P < 0.01 when compared with normal mouse serum.

* P < 0.01 when compared with normal mouse serum + antigen.

at 214 nm, and protein peaks were collected into microfuge lubes containing0.1 ml of 0.1 M carbonate/bicarbonate buffer (pH 9.5). The acetonitrle wasremoved under a stream of nitrogen, and the volume was reduced to 0.1 ml.The immunoreactivity of the individual protein peaks was determined by anELISA plate assay.

ELISA Plate Assay. Samples were divided into two and immobili/.ed on a96-well Polyvinylchloride assay plate (Costar, Cambridge, MA) overnight at4Â°C.The liquid was removed by aspiration, and the wells were washed three

times with PBS + 0.1% Tween-20(200/j.1). Blocking solution (200 Â¿Â¿Iof PBS

containing 0.1% Tween 20 and 3% BSA) was added to the wells, and the platewas incubated for 2 h at 37Â°C.One-half of the sample was incubated with the

monoclonal antibody ( 10 /xg/ml) in blocking solution ( 100 /j.1) for I h at roomtemperature, while the other half was incubated in the same solution, but in theabsence of the antibody. After removal of the antibody solution, the wells werewashed six times before the addition of a protein A peroxidase conjugate(Sigma Chemical Co., Dorset, United Kingdom) and diluted 1 in 500 inblocking solution (100 /xl/well): then the plates were incubated for an additional 1 h at 37Â°C.The wells were washed six times and then the substrate

solution [Â«-phenylenediamine dihydrochloride (0.04%! and hydrogen peroxide

(0.012%) in 0.15 M phosphate citrate buffer. pH 5.0 (100 /xl/well)] was addedfor 30 min. The reaction was terminated by the addition of 0.2 M H,SO4 (50/il/well), and the absorbance was determined at 492 nm using a microplatereader (Anthos Labtec Instruments).

Western Blotting. For immunoblotting, gels were transferred to nitrocellulose membranes (Hoefer Scientific Instruments, San Francisco. CA) that hadbeen blocked with 5% Marvel in 0.15% Tween 20 in PBS at 4Â°Covernight.

The membranes were washed for 15 min and twice for 5 min in Tween/PBSand further incubated in the Tween/PBS containing 1.5% Marvel and 10 fig/mlof the monoclonal antibody for I h at room temperature. After being washedthree times as above, the filters were incubated for 1 h with protein Aperoxidase conjugate at a l:5000-fold dilution, followed by one 15-min washand four 5-min washes with 0.5% Tween 20 in PBS. Bands were detected with

an emission chemiluminescence (ECL) system (Amersham. Buckinghamshire.United Kingdom). For lectin blotting studies, gels were transferred to nitrocellulose membranes as above and blocked overnight. The washed membraneswere further incubated with Tween/PBS containing 1.5% Marvel, and thelectins were coupled to horseradish peroxidase (Sigma Chemical Co., Poole,Dorset. United Kingdom). The blots were visualized with the aid of a standard

color-detection scheme for horseradish peroxidase, which makes use of4-chloro-l-napthol and hydrogen peroxide.

Measurement of Protein Degradation. Female NMRI mice were killedby cervical dislocation, and their gastrocnemius muscles were quickly ligated.dissected out, and placed in ice-cold isotonic saline. To minimize diurnal

variation and to assure that animals were in the fed state, all animals weresacrificed between 9:00 and 10:00 am. The muscles were blotted, weighed, andcarefully tied via tendon ligatures to stainless steel incubation supports (13).This prevents contraction and improves protein balance and energy status (14).Protein degradation was measured by tyrosine release, since tyrosine rapidlyequilibrates between intracellular pools and the medium and is neither synthesized nor degraded. Muscles were preincubated in RPMI 1640 (3 ml)lacking phenol red in the presence of serum (280 jul) for 30 min at 37Â°Cin an

atmosphere saturated with O2:CO2 (19:1). The muscles were rinsed andincubated for an additional 2 h in Krebs-Henselit bicarbonate buffer containing6 mM D-glucose, 1.2 mg/ml BSA and 130 fig/ml cycloheximide with contin

uous gassing. At the end of the incubation, the butter was removed, depro-teinized with ice-cold 30% trichloroacetic acid (0.2 ml), and centrifuged at2800 x f>for 10 min: then the supernatant was used for the measurement oftyrosine by a fluorimetrie method (15) at 570 nm on a Perkin-Elmer LS-5

luminescence spectrometer.Gel Electrophoresis. Gels were prepared according to the method of

Laemmli (16) and consisted of a 5% stacking gel and a 15% resolving gel.Molecular weight standards used were rabbit muscle myosin (M, 205.000).

phosphorylase b (M, 97,000), BSA (M, 66.000). ovalbumin (M, 45,000),carbonic anhydrase (M, 29,000). trypsin inhibitor (M, 20,000), and a-lactal-

bumin (M, 14.200) and were purchased from Sigma Chemical Co. Ltd.Determination of Immunoglobulin and Subclass. The class and subclass

of the monoclonal antibody was determined using a mouse hybridoma sub-typing kit (Boehringer Mannheim I according to the manufacturer's instruc

tions.Body Composition Analysis. For the determination of the total carcass fat

and water content, each carcass was placed in an oven at 80Â°Cuntil constant

weight was reached. Carcasses were then reweighed. and the total fat contentwas determined by the method of Lundholm et al. ( 17). Lipids were extractedfrom the whole carcass with chloroform:methanol (1:1), ethanol-acetone (1:1),

and then pure ether, which was allowed to evaporate. Water content wascalculated from the wet and dry weights.

Statistical Analysis. Results are expressed as means Â±SEM. Comparisonof multiple groups has been made by ANOVA.

RESULTS

Using subclass-specific immunoglobulin conjugates from goat, the

monoclonal antibody isolated from mice bearing the MAC 16 tumorwas determined to be of the IgG2a subclass, and a monoclonalantibody of the same subclass was used as a control. To determine itthere was any interaction between the monoclonal antibody andlipid-mobilizing activity, material was purified from the solid MAC 16

tumor as described (12). mixed with monoclonal antibody bound toprotein A-Sepharose, and incubated with agitation overnight at 4Â°C.

After pelleting by centrifugation, no lipid-mobilizing activity could be

detected in the supernatant fluid (Table 1). In contrast, incubation witha control monoclonal antibody had no effect on lipid-mobili/jngactivity. Analysis of the precipitated protein using SDS-PAGE and

staining with Coomassie blue gave evidence for one additional band

Table 3 Purification of cachexia-inducing material from the MAC16 tumor

StageTumor

homogenate40%

(NH4)2S04AffinityC8

hplc 55%CH,CN81%CH3CNProtein

(mg)31601130.2561.2

X10~40.124Protein

recovery5.30.0123.8

X10"4

X 10~3A4yi

units438.63.8Recovery299Specificactivity

units/mg1687166731Purification-fold4270.18

1258




A

60 70Acetonilrile(%)

B

1 2 3kDa

200 -

97 -

69 -

46 -

30 -

21 -

14 -

Fig. 2. A, elution profile of affinity-purified antigen subjected to reverse phase hydro-

phobic chromatography on a Cs column with a gradient of acetonitrile in water. TheELISA reactivity of the individual fractions was determined by A4q2 (A), and the protein(O) was measured from the A2,4 in comparison with the absorption of a standard(albumin). B, Western blot of 15% SDS-PAGE gel of immunoreactive fractions elutingfrom the CK column. Lane /, molecular weight markers. Lane 2, material eluting at 81%acetonitrile. Lane 3, material eluting at 55% acetonitrile. C, silver stain of SDS-PAGE gelof material eluting at 55% acetonitriie from the Cs column.

97-69i-

46-

30-

21 -14-

of MT 69,000 (Fig. IA) in addition to the antibody heavy and lightchains. Western blotting of immunoprecipitated material showed twobands of Mr 69,000 and 24,000 (Fig. IÃŸ).Thus, although the antibodywas cloned to material of Mr 24,000 (12), it was also capable ofdetection of higher molecular weight material (Table 1). Preincuba-tion of MAC 16 tumor-derived lipid-mobilizing factor with monoclonal antibody at 4Â°Covernight failed to prevent the induction of

lipolysis in isolated murine adipocytes, suggesting that it is not aneutralizing antibody for this activity.

Since cachexia involves depletion of both adipose tissue and skeletal muscle mass, the effect of the MAC 16 monoclonal antibody ontyrosine release from isolated gastrocnemius muscle was also investigated. As reported previously (11), serum from mice bearing theMAC 16 tumor caused an increased protein degradation comparedwith serum from nontumor-bearing animals (Table 2), and this effect

was completely abolished by prior incubation of the serum with theMAC 16 monoclonal antibody while being unaffected by a nonspecificIgG2a. This suggests that this antibody is directed toward materialcapable of inducing muscle proteolysis and that this material must beclosely associated with material capable of causing lipid mobilization.

To determine the nature of the immunoreactive material, a purification scheme was devised that involved an initial ammonium sulfatefractionation of the MAC 16 tumor homogenate, followed by an affinity purification with monoclonal antibody coupled to Affi-gel Hz.

This matrix couples the antibody through aldehyde groups formed byoxidation of the carbohydrate residues to hydrazide groups on the gel,allowing the correct orientation of the antibody and resulting inincreased binding capacity per coupled IgG. Bound material waseluted from the column with glycine HC1 (pH 2.5), and fractionsreactive with the antibody were determined by an ELISA plate assay(Table 3). Western blotting of such fractions showed two immunoreactive bands of Mr 69,000 and Mr 24,000 (Fig. 1C), which were themain protein bands detected by silver stain after affinity chromatog-

raphy (Fig. ID). Bioactivity of the antigenic material was determinedby tyrosine release from isolated gastrocnemius muscle (Table 2).This showed that when antigenic material was added to the serumfrom nontumor-bearing animals, protein degradation was increased tothe level observed with serum from MAC 16 tumor-bearing mice with

cachexia. The elevation in tyrosine release could be inhibited bypreincubation with the MAC 16 monoclonal antibody prior to theaddition to the isolated gastrocnemius muscle, while being unaffectedby a nonspecific IgG2a monoclonal antibody.

Further fractionation of affinity-purified material was achieved

using reverse phase hydrophobic chromatography with a C8 columnand a gradient from 5 to 100% acetonitrile in water containing 0.06%trifluroacetic acid (Fig. 2A ). Immunologically reactive material eluted

Table 4 Effect of purified antigen on body weight of female NMRI mice, food and waler Â¡make,and plasma metabolic levels"

Acetonitrile(%)565878838688Weight loss(g)2.03Â±0.21*0.96

Â±0.291.84Â±0.38'0.80

Â±0.280.80

Â±0.170.39

Â±0.22Food

intakeg/mouse2.732.983.232.332.842.69Water(ml/mouse)1.31.51.41.21.72.2GlucosemM5.9

Â±0.26.4

Â±0.36.3

Â±0.26.0

Â±0.47.5

Â±0.76.0

Â±0.3TriglycÃ©ride

mM0.80

Â±0.10.79

Â±0.190.49

Â±0.080.59

Â±0. 130.97

Â±0.110.77

Â±0.18Fatty

acidmM0.97

Â±0.08'0.72

Â±0.021.16Â±0.12Â¿0.90

Â±0.130.66

Â±0.080.63

Â±0.05" Decrease in body weight of female NMRI mice after administration of fractions eluted from the C8 column. To each fraction was added PBS (3 ml), and the samples were placed

under a steady stream of nitrogen for 40 min to remove the acetonitrile. The fractions were concentrated by Ultrafiltration through an Amicon filtration cell containing a membranefilter with a molecular weight cutoff of 10.000 against PBS without calcium and magnesium, followed by two washes with the same buffer. Portions (150 fi\: 5 /ig protein) were injectedinto the tail vein of five female NMRI mice at 1.5-h intervals (10.30. 12.00, 13.30, and 15.00 h). Animals were weighed before each injection, with the final determination at 24 h.Results are for mice 24 h after the first injection and are the mean Â±SEM for four to six animals per group.

* P = 0.001 from 86 and 88% acetonitrile by one-way ANOVA.'' P = 0.03 from 86 and 88% acetonitrile by one-way ANOVA.

dP = 0.01 from 86 and 88% acetonitrile by one-way ANOVA.

1259




(S) (A)

Fig. 3. Amino acid sequence of p24 antigen.

as two fractions corresponding to 55 and 80% acetonitrile. Materialeluting at 55% acetonitrile showed a single band of Mr 24,000 onWestern blots (Fig. 2B), while material eluting at 80% acetonitrileshowed a single band of MT 69,000. Silver staining showed materialeluting at 55% acetonitrile to give a single band of Mr 24.000 onSDS-PAGE (Fig. 2C). The role of the antigen in cancer cachexia was

determined by administration of the individual fractions eluted fromthe Cs column to female NMRI mice and monitoring body weightchanges over a 24-h period. Changes in body weight followed changes

in elution of antigen, with fractions eluting at 55 and 80% acetonitrileproducing decreases of about 2 g in body weight over the 24-h period.

69 â€”¿�

30 â€”¿�

H-

I

4

3.5 -2.5 -

Fig. 4. A, detection of sugars in glycoconjugates by digoxigcmn glycan detection(Boehringer Mannheim) on affinity-purified MAC16 tumor run on a I5r/t SDS-PAGE gel.Lane I, molecular weight markers; Lane J. affinity-purified immunoreactive fraction:Lune }. non-glycosylated control protein (creatinase); IMIU'4, control glycoprotein (trans-ferrin). B, lectin blot of antigenic malerial eluting at 55r/r acetonitrile from the Cs column.

Lane l. Erylhrina cryslii^nlii agglutinin; Lane 2, wheat germ agglutinin: Lane 3,Telragontilobus purpureas agglutinin; Lane 4. ConA.

This confirms that the antigen is capable of producing cachexia. Therewas no significant change in food and water consumption between theindividual groups (Table 4), but there was an elevation of plasma freefatty acids in the groups administered material eluting at 55 and 80%acetonitrile.

Amino acid sequence analysis of the M, 24.000 material eluting fromthe Cx column with 55% acetonitrile showed a single species of 18(20)amino acids (Fig. 3). No further sequence information could be obtained.Material eluting at 80% acetonitrile gave the same amino acid sequence,together with the sequence for albumin. This suggests that the immunoreactive band of M,.69,(MX)is albumin bound to the AÃ•,24,(XX)material.Both bands stained for carbohydrate using the digoxigenin glycan detection kit (Fig. 4A ). Lectin blotting studies on the pure Mr 24,000 component showed a strong reaction with Triticum vulgaris wheat gemi agglutinin, which has specificity predominantly for W-acetylglucosamine (Fig.

4B). A strong reaction was also obtained with Erythrina crystagalliagglutinin, which has specificity for Gal ÃŸ(lâ€”*4W-acetylglucosamine.

Neither Con A. which has specificity for mannose, or Tetragoitotobuspurpureas agglutinin. which has specificity for fucose, gave a reaction.These results suggest that the malerial of Mr 24.Ã•XX)is a glycoprotein orproteoglycan, which shows strong binding affinity to albumin, possiblythrough the carbohydrate residues.

DISCUSSION

Previous studies with the cachexia-inducing murine MAC 16 tumor

model have provided evidence for circulatory tumor products capableof direct catubolism of both skeletal muscle and adipose tissue usingin vitro bioassays (18). Such catabolic factors were not present in thecirculation of a related tumor, MAC 13, which does not induce cachexia. Recently, we reported (12) that mice transplanted with theMAC 16 tumor and with delayed weight loss contained in their serumantibodies that interacted with a material of Mr 24.0IX). which copu-rified with a lipid-mobilizing factor. Such antibodies were not in the

serum of mice transplanted with the MAC 13 tumor, suggesting thatthe antibodies were directed to the induction of cachexia rather thanthe tumor itself. The present study describes the cloning of suchantibodies using splenocytes from mice transplanted with the MAC 16tumor and the use of such antibodies to characterize factors potentiallyresponsible for the cachexia.

The major protein produced by immunoprecipitation of a purifiedlipid mobilizing factor was of Mr 69,000. while Western blottingindicated immunoreactive bands at Mr 69.000 and Mr 24,000. Masunoet al. (19, 20) also reported the isolation of a lipolytic factor (toxo-hormone-L) from the ascites fluid of patients with hepatoma ( 19) and

mice with sarcoma 180 (20) of M, 70,000. Although not furthercharacterized in terms of structure, this material was similar to thelipid-mobilizing factor produced by the MAC 16 tumor in that bothwere acidic (pi < 4) and eluted from a DEAE-cellulose column

between 0.14 and 0.18 M salt (12).Although the monoclonal antibody was incapable of neutralizing

lipid-mobilizing activity in an in vitro assay, it was capable of complete inhibition of the enhanced protein degradation in isolated gas-

trocnemius muscle induced by serum from mice bearing the MAC 16tumor, thus providing some evidence for a relationship between thesetwo activities. Affinity purification of homogenates of the MAC 16

1260




tumor yielded immunologically reactive material of Mr 69,000 and Mr24,000, which could be further fractionated using Hydrophobie chro-

matography. This material was capable of stimulating protein degradation in the gastrocnemius muscle assay in a specific manner thatcould be inhibited by the MAC 16 monoclonal antibody.

Amino acid sequence analysis confirmed that the material of Mr24,000 had a unique sequence and showed no homology with any of therecognized cytokines. Material of Mr 69,000 contained albumin and thesame sequence as Mr 24,000. This suggests that binding to albumin isvery strong, since complete dissociation was not achieved, even afterfractionation using acetonitrile and trifluroacetic acid. Attempts weremade to obtain further sequence information by proteolytic digestion ofthe material of Mr 24,000. The predominant sequence after tryptic digestion was as shown in Table 5, with no evidence for another species. A lowlevel of non-peptide material was observed. A similar result was obtained

after overnight digestion with chymotrypsin, which again only gaveevidence for the NH-,-terminal sequence. Digestion with V8 protease

gave five prominent peaks, three of which corresponded to theNH2-terminal sequence and two to V8 autodigestion products. This

suggests that the material of Mr 24,000 is resistant to proteolyticdigestion or that no further structural information is present. Thisabnormal behavior may be explainable by the presence of carbohydrate. Although the material is resistant to digestion by proteolytic enzymes, degradation is achievable using chondroitinaseABC and AC, suggesting that it may be a sulfated heteropolysac-

charide. A recent study indicated that cell bound albumin boundstrongly to peptidoglycan and sulfated heparinoids to form a complex of Mr 70,000, which migrated as a single band on PAGE (21).

The role of this material in cancer cachexia is supported by an invivo study demonstrating that immunoreactive material eluting fromthe hydrophobic column was capable of inducing weight loss. Thiseffect occurred without a reduction in food and water intake, as isfound in animals bearing the MAC16 tumor (18). In this murinemodel, the energy expenditure has been shown to increase withincreasing weight loss and since food intake remains constant, anegative energy balance results (22). Thus, the material of M, 24,000possibly produces weight loss by increasing the energy expenditure.However, the mechanism by which this material produces cachexiaremains to be determined.

ACKNOWLEDGMENTS

We thank M. Wynter for assistance with the tumor transplantation.

REFERENCES

1. Tisdale. M. J. Cancer cachexia. Anticancer Drugs, 4: 115-125, 1993.2. Warnold, I.. Lundholm. K., and Scherstein, T. Energy balance and body composition

in cancer patients. Cancer Res.. 38: 1801-1807, 1978.

3. Heber, D., Byerley, L. O., Chi, J., Grosvenor. M.. Bergman, R. N., Colemann. M., andChlebowski, R. T. Pathophysiology of malnutrition in the adult cancer patient. Cancer(Phila.), 58: 1867-1873. 1986.

4. Fredrix, E. W. H. M., Soeters. P. B., Wouters, E. F. M., Deerenberg, I. M., vonMeyenfeldt, M. F., and Saris. W. H. M. Effect of different tumor types on restingenergy expenditure. Cancer Res., 51: 6138-6141, 1991.

5. Beulter, B., and Cerami, A. Cachectin and tumor necrosis factor as two sides of thesame biological coin. Nature (Lond.), 320: 584-588, 1986.

6. Flores, E. A., Bistrain. B. R.. Pomposelli. J. J.. Dinarello. C. A.. Blackbum, G. L., andIstfan, N. Infusion of tumor necrosis factor/cachectin promotes muscle catabolism inthe rat. J. Clin. Invest., 83: 1614-1622, 1989.

7. Garcia-Martinez, C., Lopez-Soriano, F. J., and ArgÃ¼es,J. M. Interleukin-6 does notactivate protein breakdown in rat skeletal muscle. Cancer Lett., 76: 1-4, 1994.

8. Goldberg. A.. Kettlehut, K.. Fagan. J., and Baracos, V. Activation of protein breakdown and prostaglandin E, production in rat skeletal muscle Â¡nfever is signalled bya macrophage product distinct from interleukin-1 or other known monokines. J. Clin.Invest., 81: 1378-1383, 1988.

9. Belizario, J. E., Katz, M.. Chenker. E., and Raw, I. Bioactivity of proteolysis-inducingfactors in the plasma proteins from cancer patients with weight loss. Br. J. Cancer. 63:705-710, 1991.

10. Smith. K. L.. and Tisdale. M. J. Increased protein degradation and decreased proteinsynthesis in skeletal muscle during cancer cachexia. Br. J. Cancer, 67: 680-685,

1993.11. Smith, K. L., and Tisdale. M. J. Mechanism of muscle protein degradation in cancer

cachexia. Br. J. Cancer, 68: 314-318, 1993.

12. McDevitt, T. M., Todorov, P. T., Beck, S. A., Khan. S. H.. and Tisdale. M. J.Purification and characterization of a lipid mobilizing factor associated with cachexia-inducing tumors in mice and man. Cancer Res., 55: 1458-1463, 1995.

13. Wu. G.. and Thompson, J. R. The effect of ketone bodies on alanine and glutaminemetabolism in isolated skeletal muscle from the fasted chick. Biochem. J., 255:139-143, 1988.

14. Baracos, V. E., and Goldberg. A. L. Maintenance of normal length improves proteinbalance and energy status in isolated rat skeletal muscle. Am. J. Physio]., 257:C588-C595, 1986.

15. Waalkes. T. P., and Undenfriend. S. A. A fluorometric method for the estimation oftyrosine in plasma and tissues. J. Lab. Clin. Med., 50: 733-736. 1957.

16. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head ofbacteriophage T4. Nature (Lond.). 277: 680-685, 1970.

17. Lundholm, K.. EdstrÃ¶m,S., Karlberg, L, Ekman, L., and Schersten, T. Relationshipof food intake, body composition and tumor growth to host metabolism in nongrow-ing mice with sarcoma. Cancer Res.. 40: 2515-2522, 1980.

18. Beck, S. A., and Tisdale. M. J. Production of lipolytic and proteolytic factors bya murine tumor-producing cachexia in the host. Cancer Res.. 47: 5919-5923,

1987.19. Masuno, H., Yoshimura, H., Ogawa, N., and Okuda. H. Isolation of a lipolytic factor

(toxohormone-L) from ascites fluid of patients with hepatoma and its effect onfeeding behaviour. Eur. J. Cancer Clin. Oncol., 20: 1177-1185. 1984.

20. Masuno, H., Yamasaki. N., and Okuda. H. Purification and characterization of alipolytic factor (toxohormone-L) from cell-free fluid of ascites sarcoma 180. CancerRes., 41: 284-288, 1981.

21. Dziarski, R. Cell-bound albumin is the 70-kDa peptidoglycan-, lipopolysaccharide-,and lipoteichoic acid-binding protein on lymphocytes and macrophages. J. Biol.Chem., 269: 20431-20436, 1994.

22. Plumb. J. A., Fearon, K. C. H.. Carter. K. B.. and Preston, T. Energy expenditure andprotein synthesis rates in an animal model of cancer cachexia. Clin. Nutr.. 10: 23-29,1991.

1261



1996;56:1256-1261. Cancer Res Penio T. Todorov, Trudi M. McDevitt, Peter Cariuk, et al. Tumor ProductInduction of Muscle Protein Degradation and Weight Loss by a

Updated version

http://cancerres.aacrjournals.org/content/56/6/1256

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/56/6/1256To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



induction of muscle protein degradation and weight loss by ... · induction of muscle protein...

Documents