biochemistry

Biochem. J. (1988) 249, 897-902 (Printed in Great Britain) 897 Induction of dihydrolipoamide dehydrogenase in 3T3-L1 cells during differentiation Donna J. CAROTHERS, Gabriel PONS and Mulchand S. PATEL* Department of Biochemistry, Case Western Reserve University School of Medicine, Cleveland, OH 44106, U.S.A. The activity and turnover of dihydrolipoamide dehydrogenase (E3), the common component of the three 2-oxoacid dehydrogenase complexes, were measured during the differentiation of 3T3-L1 preadipocytes into 3T3-LI adipocytes. The specific activity of E3 increased approx. 3-4-fold in 3T3-L1 adipocytes differentiated under a regimen of insulin, dexamethasone and 3-isobutyl-1-methylxanthine for 48 h, followed by insulin alone thereafter. A rabbit antibody to pig heart E3 quantitatively precipitated the enzyme from 3T3-L1 adipocytes. By using immunoprecipitation and gel electrophoresis, a 3.3-fold increase was observed in E3 protein in 3T3-Ll adipocytes as compared with 3T3-Ll preadipocytes, on a DNA basis. Pulse-labelling experiments with L-[35S]methionine revealed a 3.5-fold increase in the rate of synthesis of E3 in 3T3-LI adipocytes compared with that observed in 3T3-LI preadipocytes. In contrast, the apparent half-lives of the E3 in 3T3-LI preadipocytes (43 h) and 3T3-LI adipocytes (33 h) were not significantly different. Therefore, the 3-4-fold increase in the specific activity of E3 in 3T3-L1 adipocytes resulted from an increased rate of synthesis of the enzyme. INTRODUCTION Dihydrolipoamide dehydrogenase (E3) (EC 1.8.1.4) is the common component of the three 2-oxoacid dehydrogenase complexes, pyruvate dehydrogenase complex (PDC), 2-oxoglutarate dehydrogenase complex (2OGDC), and branched-chain oxoacid dehydrogenase complex (BCOADC). Each complex also contains two other catalytic components specific for each complex: 2-oxoacid dehydrogenase (E1) and dihydrolipoamide acyltransferase (E2) [1]. PDC and BCOADC are regulated by covalent modification by their specific kinases and phosphatases acting on the a subunit of E1 [2,3]. These complex activities are also regulated by metabolites [4-8]. All of these modulations alter active PDC without concomitant alteration in total PDC activity. Changes in total PDC activity have been found in developing rat brain [9], liver [10], lactating mammary gland [11] and differentiating 3T3-L1 cells [12-14]. The differentiation process converting the 3T3-Ll preadipocytes into 3T3-LI adipocytes is associated with the acquisition of many morphological and biochemical characteristics of adipose tissue [15-17]. Differentiation is spontaneous in preadipocytes, occurring over several weeks; confluent cells treated with a variety of hormones or drugs (insulin, dexamethasone, 3-isobutyl-1-methylxanthine etc.) differentiate more rapidly [14-19]. Differentiation of the 3T3-L1 preadipocytes into 3T3-L1 adipocytes is accompanied by increases in the activities of enzymes responsible for synthesis of triacylglycerol [20-22] and long-chain fatty acids [23-26]. Since PDC plays a central role in the incorporation of glucose carbon into long-chain fatty acids, changes in PDC activity have also been monitored in the differentiating 3T3-L1 cells [12]. The activity of the complex itself, as well as the pyruvate dehydrogenase component El, increase severalfold with differentiation. The increased PDC-E1 activity coincides with increased PDCE1 protein, resulting from increased relative rates of synthesis for the two non-identical subunits of PDC-E1 [12]. BCOADC activity also increases severalfold during differentiation of 3T3-LI cells, but the 2-OGDC activity is increased only 3-fold [13]. The component enzymes of PDC and BCOADC are each increased in specific activity during this differentiation [13]. In this paper we report that the increased activity of E3 is correlated with increased E3 protein in the differentiated 3T3-L1 cells, and that this increase in protein content results from an increase in the rate of synthesis of the enzyme. EXPERIMENTAL Materials Bovine insulin, 3-isobutyl- 1 -methylxanthine, dexamethasone, DL-6,8-thioctic acid amide (lipoamide), NAD+, biotin and dihydrolipoamide dehydrogenase were obtained from Sigma Chemical Co. (St. Louis, MO, U.S.A.) L-[35S]Methionine was purchased from New England Nuclear (Boston, MA, U.S.A.). Highly purified bovine kidney dihydrolipoamide dehydrogenase, isolated from PDC, was very generously provided by Dr. Lester J. Reed of the University of Texas, Austin. Cell culture and differentiation 3T3-LI preadipocytes obtained from the American Type Culture Collection (A.T.C.C. CCI 92.1) were grown as described previously [12,13]. At confluence (day 0) the cells were treated in one of two ways. In Treatment I, cells were maintained in supplemented Eagle's minimal essential medium (complete medium) Abbreviations used: E3, dihydrolipoamide dehydrogenase; PDC, pyruvate dehydrogenase complex; 2-OGDC, 2-oxoglutarate dehydrogenase complex; BCOADC, branched-chain oxoacid dehydrogenase complex; E1, 2-oxoacid dehydrogenase; E2, dihydrolipoamide acyltransferase. * To whom reprint requests should be addressed. Vol. 249 898 [12]. Cells thus treated remained as preadipocytes. In Treatment II [12], cells were maintained in given complete medium supplemented with insulin (10 jg/ml), dexamethasone (0.25 AM), 3-isobutyl- 1 -methylxanthine (0.5 mM) and biotin (8 gg/ml) for 48 h, and maintained thereafter in complete medium supplemented with only insulin and biotin. This treatment caused differentiation of 3T3-L1 preadipocytes into adipocytes. Enzyme assay Cell extracts for E3 assay were prepared from cell monolayers scraped into phosphate-buffered saline (Dulbecco's 'A'); cells were washed thrice in phosphatebuffered saline and resuspended in a hypo-osmotic extraction buffer (20 mM-potassium phosphate, pH 7.5, 1 % Triton X-100) containing proteinase inhibitors (2 mM-EDTA, 2 mM-EGTA, 0.2 mM-phenylmethanesulphonyl fluoride, 0.5 mg of leupeptin/l, 0.7 mg of pepstatin/l) [27]. Cell suspensions were maintained on ice (30 min), frozen and thawed thrice, and kept again on ice 30 min before E3 was assayed. DL-Lipoamide was converted into the reduced form (dihydrolipoamide) chemically [28]. E3 activity was assayed spectrophotometrically in the direction of dihydrolipoamide oxidation (modified from [29]); the reaction mixture contained 50 mM-potassium phosphate, pH 8, 1.5 mmEDTA, pH 7, 1.5 mM-NAD', 2 mM-dihydrolipoamide and 0.5 mg of Triton X-100/ml. A milliunit (munit) of activity is defined as 1 nmol of substrate oxidized/min at 37 'C. Protein was determined by the method of Lowry et al. [30], with bovine serum albumin as standard. Immunological procedures Cell extracts, prepared as described above, were centrifuged at 20000 g for 30 min at 4 'C. E3 (approx. 300 munits) was immunoprecipitated from the supernatant fraction with 50,1u of rabbit anti-(pig heart E3) antiserum [31]. The immunoprecipitates were collected by centrifugation ofthe pellet through a 'sucrose cushion' (1 M-sucrose/ 10 mM-L-methionine/0.5 % Triton X-100/ 0.5 % Triton X-405) and washed as described previously [12]. The washed precipitates were dissociated by 10 min incubation in 2 % (w/v) SDS and 20 mM-dithiothreitol at 100 'C and subjected to electrophoresis [32]. The gels were stained with 0.25 % Coomassie Blue, destained [32], and densitometry was performed to determine staining intensity [12]. Immuno-cross-reactivity of the anti-E3 serum with E3 in 3T3-L1 adipocytes was measured by immunoprecipitating L-[35S]methionine-labelled E3 from cells grown in a medium containing L-[35S]methionine for 48 h. The immunoprecipitates were subjected to SDS/ polyacrylamide-gel electrophoresis, stained for protein, destained, and cut into 2 mm-thick slices. Gel slices were dissolved in 30 % (v/v) H202 containing 1 % (v/v) NH3, and the radioactivity was determined by liquidscintillation counting. One major peak of radioactivity was detected, co-migrating with Coomassie-Blue-stained E3. Furthermore, to demonstrate that all the radiolabelled E3had been immunoprecipitated, a sample of cell extract from unlabelled 3T3-L1 adipocytes, containing a known amount of E3 activity, was added to the supernatant fraction of the first immunoprecipitate. More antiserum was added, and the second immunoprecipitate was analysed as described above. The E3 band was visible on the Coomassie-Blue-stained gel, but no significant D. J. Carothers, G. Pons and M. S. Patel amount of radioactivity co-migrated in this position, indicating that all the labelled E3 had been precipitated with the first addition of antiserum. Relative rate of synthesis of E3 Cell monolayers, at confluence in 60 mm-diam. dishes, were incubated in 2 ml of medium containing 20 /LM L-[35S]methionine (100 #uCi) [12]. In a preliminary experiment, the incorporation of label into total protein and into E3 protein in both preadipocytes and adipocytes was observed to be linear over a 2 h period (results not shown). At the end of the labelling period, the medium containing L-[35S]methionine was removed and the cells were washed twice in phosphate-buffered saline containing 2 mm unlabelled L-methionine. The cells were harvested, solubilized, and E3 was immunoprecipitated. Cellular extract from unlabelled 3T3-LI adipocytes was added as carrier, if necessary, to give a total E3 activity of 300 munits. The immnoprecipitates were subjected to SDS/polyacrylamide-gel electrophoresis, and the radioactivity in E3 was determined. Relative rates of synthesis were calculated as the ratio of the radioactivity in E3 to the radioactivity in total trichloroacetate-precipitable protein [33]. Degradation of E3 Cell monolayers of both 3T3-L1 preadipocytes and adipocytes maintained for 6 days after initiation of Treatment I or II were incubated in 2 ml of L-methioninemodified medium containing 100 ,M-L-[35S]methionine (100 1tCi for preadipocytes, 50 ,uCi for adipocytes), and the appropriate supplements to maintain Treatment I or II. After 24 h of incubation the cells were washed with sterile phosphate-buffered saline containing 2 mm unlabelled L-methionine, and then incubated in complete medium (400 #,M-L-methionine) [12]. At the end of the 5 h chase period, and every 12 h thereafter, cells were harvested, solubilized, and processed for the determination of radioactivity in E3 and total protein. RESULTS Changes in the specific activity of the E3 in 3T3-L1 cells We examined the effects of insulin, dexamethasone and 3-isobutyl- l-methylxanthine on the increase in specific activity of 3T3-L1 cells during differentiation (Fig. 1). Confluent 3T3-L1 preadipocytes (Treatment I) showed no increase in E3 specific activity over the 17-day span of the experiment. Morphologically, the cells remained fibroblast-like for this period. The specific activity of E3 in the 3T3-L1 adipocyte (Treatment II) increased markedly, reaching a 3.5-fold increase on day 8 of the treatment, and then declined over the next 4 days, to a value twice that seen in the preadipocytes. In a separate experiment, a 9.9-fold increase in E3 activity per dish and a 2.6-fold increase in the DNA content per dish were observed on day 8 of the treatment. A 3.8-fold induction of the E3 activity per cell, based on DNA content, was similar to a 3.5-fold increase in the specific activity of E3 reported in Fig. 1, suggesting no change in the protein/DNA ratio during differentiation of 3T3-L1 cells. This finding is consistent with the observed 4-fold increase in the DNA content and no significant change in the protein/DNA ratio of 3T3-L1 cells during differentiation, found previously by an identical regimen [34]. 1988 Dihydrolipoamide dehydrogenase in 3T3-LI cells 899 400 CN 0 ._; E Cur wt 0 4 12 8 Time past confluence (days) 16 Fig. 1. Differentiation-induced changes in the specific activity of E3 in 3T3-Ll adipocytes Monolayer cultures of 3T3-Ll preadipocytes were grown to confluence (day 0) and subjected to either Treatment I (0) or Treatment II (@) as described in the Experimental section. E3 activity was measured in cell extracts at the times indicated. The results are expressed as means+ S.D. for six separate dishes. c 100 E :C 80 60- 40 - 20 0 25 75 50 Antibody added 100 (pl) Fig. 2. Immunotitration of E3 activity in 3T3-Ll preadipocytes and adipocytes Cell extracts were prepared from 3T3-L1 preadipocytes (0) or adipocytes (@) differentiated for 6 days, and a constant amount of E3 activity was precipitated with the indicated volumes of anti-E3 serum as described in the Experimental section. Residual E3 activity was measured in the supernatant fractions. To distinguish between an increase in E3 specific activity being due to increased catalytic activity or to increased E3 protein, we compared immunotitrations of a constant amount of E3 activity from both 3T3-LI adipocytes and preadipocytes with increasing amounts of anti-E3 serum. Fig. 2 shows that the imrnunotitration curves were similar, indicating that the specific activity of the enzyme was unchanged during differentiation of the adipocytes. To determine the amount of E3 in the 3T3-L1 preadipocytes and adipocytes, E3 was immunoVol. 249 1 3 2 Fig. 3. Determination of immunoprecipitable E3 in 3T3-L1 preadipocytes and adipocytes Extracts of 3T3-Ll preadipocytes (Treatment I) or adipocytes (Treatment II for 8 days) were prepared, and E3 was immunoprecipitated. Immunoprecipitates were subjected to SDS/polyacrylamide-gel electrophoresis, and the gel was stained for protein. Lane 1: purified bovine kidney E3. Lane 2: 3T3-L1 preadipocyte extract containing 72 ,tg of DNA. Lane 3: 3T3-L1 adipocyte extract containing 72 ,ug of DNA. precipitated from equal numbers of cells (based on DNA) from both cell types (Fig. 3). In this experiment the specific activity of E3in the adipocytes was increased 3.9-fold compared with preadipocytes, whereas a 3.3-fold increase in the amount of precipitable E3 protein was quantified by densitometry of the stained gel. The heavily stained band migrating below the E3 band is IgG heavy chain. Turnover rate of E3 To determine the mechanisms responsible for the increase in E3 protein, the relative rate of synthesis and apparent rate of degradation of E, were determined in differentiating 3T3-LI cells. As found in other experiments, incubation with insulin, dexamethasone and 3-isobutyl- 1-methylxanthine caused a marked increase in the E3 specific activity in adipocytes; preadipocyte E3 activity was unchanged over this period (Fig. 4a). The 900 D. J. Carothers, G. Pons and M. S. Patel 1200 .L - E - Q 6. 40 2 C 800 0 Q E a -o .3 E .4_ 400 u cr .0 x .2 -c E U, w ._ ._A C c - .0 C: 0 w a 03 c 03 4-0 .4_ .(U cc x 0) x m 0 6 12 0 6 Time past confluence (days) 12 Fig. 4. Relative rate of synthesis of E3 in 3T3-L1 preadipocytes (0) and adipocytes (0) (a) Changes in the specific activity of E3. (b) Incorporation of L-[35S]methionine into trichloroacetic acid-precipitable cellular protein. (c) Incorporation of L-[35S]methionine into E3. (d) Relative rate of synthesis of E3. Preadipocytes (Treatment I) or adipocytes (Treatment II) were labelled with medium containing L-[35S]methionine, and E3 was precipitated from the cell extracts as detailed in the Experimental section. The relative rate of synthesis for E3 is expressed as the ratio of radioactivity in E3 protein to radioactivity in total cellular protein. Results are shown as means+ S.D. for three separate dishes. rate of incorporation of L-[35S]methionine into total protein increased approx. 5-fold in differentiating cells on day 9, and declined thereafter (Fig. 4b). The radioactivity in total protein in preadipocytes increased only marginally during the 15-day period. Incorporation of L-[35S]methionine into E3 increased about 16-fold in adipocytes as compared with preadipocytes on day 9 (Fig. 4c), and declined thereafter. Incorporation of L-[35S]methionine into E3 in preadipocytes was essentially unchanged. Incorporation of L-[36S]methionine into E3 relative to incorporation into total protein (the relative rate of synthesis of E3) increased approx. 3.5-fold in differentiated 3T3-LI cells by day 9. The relative rate of synthesis then declined to a value approaching that of the preadipocytes on day 15 (Fig. 4d). To determine the apparent rate of degradation of E3, cells were labelled with L-[35S]methionine, and the decrease in radioactivity in total cellular protein and in E3 was monitored over a 60 h period (Fig. 5). The halflife of labelled total protein in adipocytes (68 h) was somewhat longer than that in preadipocytes (48 h) (Fig. 5a), in keeping with previous studies [12]. The half-life of E3 in 3T3-L1 adipocytes (33 h) was not markedly altered compared with that in preadipocytes (43 h) (Fig. 5b). DISCUSSION Previous work in this laboratory [12-14] has been directed towards an understanding of the mechanisms underlying the induction of the PDC (and the two other 2-oxoacid dehydrogenase complexes) in differentiating 3T3-LI cells. The activities of PDC, BCOADC and 2OGDC are all increased during the conversion of 3T3-Ll preadipocytes into 3T3-LI adipocytes, although the magnitudes of the increases differ greatly: approx. 10fold for BCOADC, 7-fold for PDC and 3-fold for 2OGDC [13]. In another experiment, assays of the component enzyme activities for PDC and BCOADC showed a 4-5-fold increase in specific activity for each of the three catalytic components [13]. Studies of the E1 component of PDC showed that the increase in the specific activity of PDC in differentiated 3T3-L1 cells was associated with increased synthesis of both subunits of E1 [12]. 1988 Dihydrolipoamide dehydrogenase in 3T3-LI cells & > 5 901 the 3T3-L1 cells might be similar. In Escherichia coli the E1 and E2 genes (aceE and aceF respectively) for PDC and the E1 and E2 genes for 2-OGDC are regulated by their respective operons. E3 is encoded by a gene (lpd) linked to the ace operon, but it also has its own promoter, which may be regulated by uncomplexed E3 acting as a repressor [35]. Although eukaryotic genes do not have an operon structure, it is possible that the E3 gene may be co-ordinately regulated with the genes coding for the other components of PDC. The differentiating 3T3-L1 adipocyte may provide a system in which modulation of component enzymes may be studied at the genetic level. (a) Co co°- IEA= - CC v V 0. .2_2 "a 2 0*0 (b) cc 2 .° 5 12 24 36 Time (h) 48 60 Fig. 5. Apparent rate of degradation of E3 in 3T3-L1 preadipocytes (0) and adipocytes (0) (a) Total cellular protein. (b) E3. Monolayer cultures of 3T3-LI preadipocytes or adipocytes were maintained in culture for 6 days, pulse-labelled for 24 h with L-[35S]methionine-containing medium, and chased for 5 h with unlabelled L-methionine (2 mM) as detailed in the Experimental section. Zero time is designated at the end of the 5 h chase period. At that time, and for every 12 h thereafter, E3 was immunoprecipitated and the radioactivity in E3 was measured. Each point is the mean of three separate dishes. Linear-regression analysis was used to fit the line to the data points. The E3 component is common to all three 2-oxoacid dehydrogenase complexes. Previous work [13] has demonstrated that E3 activity increases concordantly with the increased activities of the three complexes. The present study demonstrates that the increase in E3 activity is due to an increased synthesis of enzyme protein. This is consistent with the induction process observed for the E1 component of PDC [12]. The apparent variations of increase in the separate components of the multienzyme 2-oxoacid dehydrogenase complexes requires some explanation. In experiments where PDC activity increased approx. 7-fold with differentiation, a 6-fold increase in the E1 component was measured [12]. However, in a separate set of experiments, when increases in all three component activities were measured during differentiation of 3T3-L1 adipocytes [13], a 4-S-fold increase was observed for all three components ofPDC. In contrast, BCOADC-E1 increased in activity 17-fold and BCOADC-E2 and -E3 activities were increased by 4-5-fold, whereas whole-complex activity increased 10-fold [13]. Assuming that the three complexes have similar catalytic efficiencies towards their respective substrates and contain similar amounts of E3 per molecule of complex, it can be calculated that E3 in PDC, 2-OGDC and BCOADC represents approx. 90, 8 and 20 respectively of E3 activity in differentiated 3T3-L1 cells [13]. For this reason, the overall increase in E3 activity that we observed is consistent with the magnitude of increase in 'total' PDC activity [13]. The measured relative rate of E1 synthesis (0.02-0.025%) [12] is in good agreement with that reported here for E3 (0.04%). The parallel kinetics of induction of PDC and BCOADC suggest that the signals which regulate production of these complexes during differentiation of Vol. 249 This research was supported by National Institutes of Health grant AM-20478. G. P. is a Fulbright-MEC Fellow. We are grateful to Dr. Lester J. Reed for a generous gift of purified PDC-E3. We thank Dr. Douglas Kerr and Dr. Richard Miller for their helpful comments while the manuscript was in preparation. REFERENCES 1. Reed, L. J., Pettit, F. H., Yeaman, S. J., Teague, W. M. & Bleile, D. M. (1980) Enzyme Regul. Mech. Action, Proc. FEBS Meet. 60, 47-56 2. Linn, T. C., Pettit, F. H. & Reed, L. J. (1969) Proc. Natl. Acad. Sci. U.S.A 62, 234-241 3. Paxton, R. & Harris, R. A. (1982) J. Biol. Chem. 257, 14433-14439 4. Garland, P. B. & Randle, P. J. (1964) Biochem. J. 91, 6c-7c 5. Bremer, J. (1969) Eur. J. Biochem. 7, 535-540 6. Tsai, C. S., Burgett, M. W. & Reed, L. J. (1973) J. Biol. Chem. 248, 8348-8352 7. Pettit, F. H., Yeaman, S. J. & Reed, L. J. (1978) Proc. Natl. Acad. Sci. U.S.A. 75, 4881-4885 8. Parker, P. J. & Randle, P. J. (1978) Biochem. J. 171, 751-757 9. Cremer, J. E. & Teal, H. M. (1974) FEBS Lett. 39, 17-20 10. Knowles, S. E. & Ballard, F. J. (1974) Biol. Neonate 24, 41-48 11. Coore, H. G. & Field, B. (1974) Biochem. J. 142, 87-95 12. Hu, C.-W. C., Utter, M. F. & Patel, M. S. (1983) J. Biol. Chem. 258, 2315-2320 13. Chuang, D. T., Hu, C.-W. C. & Patel, M. S. (1983) Biochem. J. 214, 177-181 14. Patel, M. S., Raefsky, C., Hu, C.-W. C. & Ho, L. (1985) Biochem. J. 226, 607-611 15. Green, H. & Kehinde, 0. (1974) Cell 1, 113-116 16. Green, H. & Meuth, M. (1974) Cell 3, 127-133 17. Green, H. & Kehinde, 0. (1974) Cell 5, 19-27 18. Mackall, J. C., Student, A. K., Polakis, S. E. & Lane, M. D. (1976) J. Biol. Chem. 251, 6462-6464 19. Russel, T. R. & Ho, R. (1976) Proc. Natl. Acad. Sci. U.S.A. 73, 4516-4520 20. Kuri-Harcuch, W. & Green, H. (1977) J. Biol. Chem. 252, 2158-2160 21. Coleman, R. A., Reed, B. C., Mackall, J. C., Student, A. K., Lane, M. D. & Bell, R. M. (1978) J. Biol. Chem. 253, 7256-7261 22. Spooner, P. M., Chernick, S. S., Garrison, M. M. & Scow, R. 0. (1979) J. Biol. Chem. 254, 1305-1311 23. Student, A. K., Hus, R. Y. & Lane, M. D. (1980) J. Biol. Chem. 255, 4745-4750 902 24. Weiss, G. H., Rosen, 0. M. & Rubin, C. S. (1980) J. Biol. Chem. 255, 4752-4757 25. Goodridge, A. G., Fisch, J. E. & Glynias, M. J. (1984) Arch. Biochem. Biophys. 228, 54-63 26. Freytag, S. 0. & Utter, M. F. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 1321-1325 27. Ho, L., Hu, C.-W. C., Packman, S. & Patel, M. S. (1986) J. Clin. Invest. 78, 844-847 28. Reed, L. J., Koike, M., Levitch, M. E. & Leach, F. R. (1958) J. Biol. Chem. 232, 143-158 29. Ide, S., Hayakawa, T., Okabe, K. & Koike, M. (1967) J. Biol. Chem. 242, 54-60 D. J. Carothers, G. Pons and M. S. Patel 30. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, 265-275 31. Carothers, D. J., Raefsky-Estrin, C., Pons, G. & Patel, M. S. (1987) Arch. Biochem. Biophys. 256, 597-605 32. Haney, P. M., Bolinger, L., Raefsky, C. & Patel, M. S. (1984) Biochem. J. 224, 67-74 33. Mans, R. J. & Novelli, G. D. (1961) Arch. Biochem. Biophys. 94, 48-53 34. Freytag, S. 0. (1981) Ph.D. Dissertation, Case Western Reserve University, Cleveland 35. Guest, J. R., Darlison, M. G., Spencer, M. E. & Stephens, P. E. (1984) Biochem. Soc. Trans. 12, 220-223 Received 1 May 1987/16 September 1987; accepted 5 October 1987 1988