interscapular RAT from obese rats was twice that of lean rats, but the amounts of DNA per BAT pad were similar. The greater weight per mg of DNA of BAT from ...
604th MEETING. CAMBRIDGE interscapular RAT from obese rats was twice that of lean rats, but the amounts of DNA per BAT pad were similar. The greater weight per mg of D N A of BAT from obese rats was probably due to greater triacylglycerol deposition per brown adipocyte. Mitochondria from obese-rat BAT bound less G D P per mg of mitochondria1 protein, but with a similar binding affinity: these findings agree with those of Holt & York (1982). When calculated per BAT pad. lean rats bind about 3 times as much G D P a s d o obcse rats. Despite this indirect evidence of decreased thermogenic capacity, BAT from obese rats did not show decreased basal or noradrenaline-stimulated oxygen consumption under these conditions in vitro. Therefore G D P binding may give an indication of maximal thermogenic capacity. but other factors may be of greater importance in controlling the oxidative function of BAT. Fatty acids or their acyl derivatives could be the physiological uncouplers of BAT mitochondria (Locke eta/.. 1982). and their availability within the brown adipocyte may be a more important determinant of thermogenic function than the maximum nucleotide-binding capacity. Obese Zucker rats have
727 a decreased BAT catecholamine turnover (Levin ef a/.. 1981) and are hypersulinaernic compared with lean rats (Zucker & Antoniades. 1972). Although the relevance of BAT to the greater energetic efficiency of the obese Zucker rat has yet to be demonstrated, these neuroendocrine factors, by enhancing fatty acid esterification and inhibiting lipolysis in BAT, may be responsible for decreased thermogenesis of this tissue in civo. Curtis-Prior. P. B.. Hanley. T. & Temple. N. J. (1975) Aria!llsl (I.ondon) 100. 105-1 I0 Holt. S. & York. D. A. (1982) Biochem. J. 208. 8 19-822 Jansky. L. (1973) B i d . Rev. Cornbridge Philos. Soc. 48.85-132 Levin. B. E.. Triscari. J. & Sullivan. A. C. (1981) Brain Res. 224. 3 53-3 66 Locke, R. M.. Rial, E., Scott, 1. D. & Nicholls, D. G. (1982) Eur. J . Biochern. 129. 373-380 Nicholls. D. G. (1976) Eur. J. Biochern. 62, 223-228 Rothwell. N . J. & Stock. M. J. (1979) Nulure (Loridon) 281. 31-35 Triandafillou. J. & Himms-Hagen. J. (1983) A m . J. Physiol. 244, E145-EI50 Zucker. L. M . & Antoniades. H. N . (1972) Endoerinologv 90. 1320-1330
Evidence from radioimmunoassay for a decreased concentration of mitochondria1 ’uncoupling’ protein from brown adipose tissue of genetically obese (oblob) mice M A R G A R E T ASHWELL. G R A H A M JENNINGS and PAUL TRAYHURN M R C Dunn Nurrition Unit, Downham’s Lane, Milton Road, Cambridge CB4 I X J . U . K . It has recently been shown that differences in the activity of brown adipose tissue can largely account for the difference in energy expenditure on non-shivering thermogenesis between genetically obese (oblob) mice and their lean counterparts. Not only is the noradrenaline-stimulated utilization of oxygen in uiiio by brown adipose tissue lower in obese mice than in lean siblings (Thurlby & Trayhurn. 1980). but purine-nucleotide-binding studies with isolated mitochondria (Himms-Hagen & Desautels, 1978: Goodbody & Trayhurn. 1982) have indicated that the activity of the proton conductance pathway across the mitochondrial inner membrane (Nicholls. 1979) is also decreased in the obese mouse. Purine nucleotides bind to a specific mitochondrial protein in brown adipose tissue, of mol.wt. 32000 (Nicholls. 1979). frequently referred to as ‘uncoupling’ protein. In non-hibernating species. the extent to which purine nucleotides bind to the uncoupling protein appears to reflect the ‘thermogenic’ activity of the tissue. However, changes in purine nucleotide binding have been associated with both long-term and short-term (Desautels et al.. 1978: Desautels & Himms-Hagen, 1979) increases in thermogenesis. so that the measurement of purine nucleotide binding alone cannot distinguish between a change in the actual content of uncoupling protein or a change in the number of available binding sites. Polyacrylamide-gel electrophoresis with sodium dodecyl
sulphate has shown that the proportion of polypeptides in the 32 000-mol.wt. region of sonicated mitochondria from brown adipose tissue of obese (oblob) mice is no different from that of lean mice (Hogan & Himms-Hagen, 1980), but this is an insensitive and less than specific method of determining the amount of uncoupling protein. Specific antisera have therefore been raised for the immunoassay of uncoupling protein (Cannon et a/.. 1982: Lean et al.. 1983: Ricquier ef al., 1983), and we have now measured the concentration of the protein in brown adipose tissue mitochondria from lean and obese (oblob) mice by radioimmunoassay (Lean ef al., 1983). The mice used were males. aged 6-1 1 weeks, of the ‘Aston’ variety. They were housed at 2 2 k 2 ” C in pairs of one obese (oblob) and one lean sibling (ob/+ or +/+). Mitochondria were isolated from interscapular brown adipose tissue (Goodbody & Trayhurn, 1982) and solubilized in 0.01% (w/v) Triton X-100 in phosphate-buffered saline. p H 7.6, at a concentration of 1 mg of rnitochondrial protein/ml. Uncoupling protein was measured by radioimmunoassay (Lean et al., 1983), using a mouse standard isolated by the method of Lin & Klingenberg (1980) from brown adipose tissue of cold-acclimated mice. The results in Table I show that the concentration of uncoupling protein is lower in the mitochondria from obese (oblob) mice than from lean siblings, the concentration in the obese being only 53% of the lean. Preliminary measurements of uncoupling protein were made on another obese mutant, the diabetic-obese (abldb) mouse, and the results (Table 1) also show a decrease in concentration of the protein, to 38% of that in lean controls (db/+). These decreases in uncoupling protein
Table I . Concenrrafion of ‘uncoupling’protein in brown adipose tissue mitochondria from lean and genetically obese (oblob)mice Uncoupling protein was measured by radioimmunoassay : for experimental details see the text. Data are also diabetic-obese (dbldb) mice and their lean (db/+) siblings. The results are expressed as mean values k s.E.M., of animals given in parentheses. * P < 0.0 1, **P < 0.001, compared with appropriate lean controls. Lean Obese Lean (obl+ or +I+) (oblob) (dbl+) Body wt. (g) 31.32 1.7 (16) 50.2k2.3 (16)** 22.9 & 1.5 (5) lnterscapular brown adipose tissue wt. (mg) 122k 17 (16) 4 I4 f 30 (16);; 156 22 (5) ‘Uncoupling‘protein (pglmg of mitochondrial protein) 25.2 2.8 (16) 13.4 2.5 (16)* 32.7 k 4.5 (5)
*
VOl. 1 1
included for adult with the numbers Diabetic-bese (dbldb) 33.4 f 2.0 (5)* 375 & I5 (5);; 12.3 & 3.3 (5).
728
BIOCHEMICAL SOCIETY TRANSACTIONS
parallel the decreased mitochondrial G D P binding reported previously for both mutants (Himms-Hagen & Desautels, 1978; Goodbody & Trayhurn. 198 1). It is concluded from the present study, by using a specific and sensitive radioimmunoassay for uncoupling protein, that the decreased thermogenesis of brown adipose tissue from adult obese mice, both oblob and dbldb, is due to a decreased concentration of the uncoupling protein and not simply t o a ‘masking’ of binding sites of purine nucleotides. Further studies are required to determine whether the decreased concentration of uncoupling protein is also evident early in the development of the obese syndrome. Cannon. B.. Hedin. A. & Nedergaard. J. (1982) FEBS Let/. 150, 129-132
Desautels. M. & Himms-Hagen. J. (1979) Can. J. Biochetn. 57. 968-976 Desautels. M.. Zaror-Behrens. G. & Himms-Hngen. J. ( 1978) Con. J . Biochern. 56.378-383 Goodbody. A. E. &Trayhurn. P. (1981) Biorhrrv.J. 194. 1019-1022 Goodbody. A . E. & Trayhurn. P. (1982) Biochirn. Biophvs. Arm 680. 119-126
Himms-Hagen. J. & Desautels. M. (1978) Biochem. Biophj,s. Rcs. Cornrnun. 83.628-634 Hogan. S. & Himms-Hagen. J. (1980) A m . J . PhJ7siol. 239, E301-E309 Lean, M . E. J.. Branch. W. J.. James. W. P. T.. Jennings. G. & Ashwell, M. (1983) Biosci. Rep. 3.61-71 Lin. C. S. & Klingenberg. M . (1980) FEBS Lerr. 113.299-303 Nicholls. D. G. (1979) Biochbn. BiophJs. Acta 549. 1-29 Ricquier. D.. Barlet. J.-P.. Garel. J.~M..Cornbes-George. M. & Dubois. M. P. (1983) Biochern. J. 210.859-866 Thurlbv. P. L. & Trayhurn. P. (1980) ffliigers Arch. 385. 193-201
Fatty acid oxidation in human skeletal-muscle mitochondria K I S H O R C H A N D R A GOHIL,* D A V I D A. J O N E S a n d R l C H A R D H. T. E D W A R D S
Department of Medicine, University College London School of Medicine, The Rayne Institute, Universitv Strcci. London 655. U.K. Defects in oxidative pathways have been demonstrated a s one of the causes of muscle dysfunction (Land & Clark, 1979). With the use of conventional polarographic techniques studies of mitochondria1 function cannot be made on small samples of human muscle that can be routinely obtained by the needlebiopsy technique (Edwards et a/.. 1980). We have. however, shown that the activities of components of the electron-transport chain and the rate-limiting enzymes of the tricarboxylic acid cycle can be routinely determined in needle biopsies of muscle by measuring the reduction or oxidation of added cytochrome c (Gohil et a/.. 1981: Gohil & Jones, 1983). This technique is now extended to measure rates of palmitoylcarnitine and palmitoylC o A oxidations in mitochondria prepared from small samples of tissue. Nine normal subjects and 23 patients with some mitochondrial abnormality were studied. Mitochondria were prepared from 50-100mg samples of muscle as previously described (Gohil et a/.. I98 I). Oxidation of palmitoylcarnitine was determined in samples of mitochondria by measuring the reduction of added cytochrome c at 5 5 0 n m . The assay medium in a volume of 560-570pl contained: 50mMpotassium phosphate buffer. pH 7.3. 5 mMMgCI,. 0.5% bovine serum albumin. 0.2 mMADP. 0 . 2 m ~ c y t o c h r o m ec. 0 . 2 m ~ N a C Nand 10-2Opl portions of mitochondria. The reaction was started with I mM-malate and 50p~-palmitoylcarnitine.Pyruvate oxidation was measured under the same conditions except that pyruvate ( 5 mM) replaced palmitoylcarnitine. Oxidation of palmitoyl-CoA was measured in lysed mitochondria prepared by three cycles of freezing and thawing. Samples of volume 20-4Opl were used for assay under the conditions used for the assay of the oxidation of palmitoylcarnitine except that 1 10pM-pahnitoyl-CoA replaced palmitoylcarnitine. Activity of cytochrome c oxidase was determined by measuring the oxidation ferrocytochrome c as previously described (Gohil et a/.. I98 1). Carnitine palmitoyltransferase activity was measured essentially a s described by Bieber et a/. ( 1972). The 2 3 patients could be divided into two groups o n the hacis
* Present address: Membrane Bioenergetics Group, Lawrence Berkeley Laboratory. University of California. Berkeley. CA 94720. U.S.A.
of the activities of the various pathways. In most (group I. 18 patients) all the activities were much lower than those of the normal subjects: oxidation of palmitoylcarnitine. oxidation of palmitoyl-CoA. cytochrome c oxidase activity and carnitine palmitoyltransferase activity were 23%. 43%. 33% and 23% of normal respectively. This suggests a decrease in the numbers of normal mitochondria. In the remaining five patients (group 2) the oxidation of palmitoylcarnitine was below the limit of detectibility ( < 0 . 0 3 p n o l / m i n per g). In the nine normal subjects and the patients in group 1 this activity was (means & s.E.M.) 0.73 + 0. I 7 rn = 9) and 0. I7 ? 0.04 ( n = 9) respectively. In all the patients in
0 Normal subjects
Patients
Patients (group 21
(group 1 I
Fig. 1. Oxidation of palnrito~~lcarnitine by mitochondria from normal hirman siibjects and .from paiienis with sonw mitochondrinl ahnormalitj~ Oxidation of palmitoylcarnitine is expressed per unit of cytochrome c oxidase in nine normal subjects (mean -t s.E.M.). in the nine patients in group I ( m e a n ? s.E.M.) and in the five patients in group 2 (individual values given). Some of these later patients were biopsied on more than one occasion. 1983
