Cholesterol Metabolism in Diabetes: the Effect of Insulin on the Kinetics of Plasma Squalene*

Cholesterol Metabolism in Diabetes: The Effect of Insulin on the Kinetics of Plasma Squalene*

BRIAN M. FRIER AND CHRISTOPHER D. SAUDEK Department of Medicine, Cornell University Medical College, New York, New York 10021 Downloaded from https://academic.oup.com/jcem/article/49/6/824/2679269 by guest on 29 September 2021

ABSTRACT. The turnover of isotopically labeled squalene curve yield an estimated rate of chloesterol synthesis based on formed in plasma from [14C]mevalonate has been used to meas- the likely assumption that mevalonate pool size did not decrease. ure cholesterol synthesis in diabetics over a 7-h period. Five Insulinization increased this calculated mean rate of cholesterol patients were studied while in poor diabetic control (mean synthesis from 961 ± 151 to 1206 ± 223 mg/day (P < 0.025). The daytime glycemia, 349 mg/dl) and at a later date once improved use of squalene kinetics to evaluate changes in cholesterol syn- control was established by multiple daily insulin injections (mean thesis deserves further study, particularly in metabolically un- glycemia, 175 mg/dl). This degree of diabetic control resulted in stable states such as diabetes in which conventional methods for an increase in the fractional conversion of [14C]mevalonic acid measuring cholesterol synthesis are difficult to apply and to to [14C]squalene from 55.2 ± 1% to 67.5 ± 4% (P < 0.025). These interpret. (J Clin Endocrinol Metab 49: 824, 1979) data together with the area under the squalene specific activity

NUMBER of studies have shown that cholesterol has serious limitations in the measurement of cholesterol A metabolism is affected by diabetes. Plasma choles- synthesis in vivo when applied to an unstable condition terol concentration, the simplest indication of cholesterol such as diabetes. homeostasis, is mildly but significantly elevated in dia- Kinetic analysis of plasma squalene has recently been betic populations (1-4). Experimental diabetes in rats developed as a method for measuring cholesterol synthe- causes changes in lipoprotein composition (5, 6), and in sis directly (14, 15). Isotopically labeled mevalonic acid subhuman primates, diabetes causes increased suscepti- (MVA) is infused iv, and the kinetics of endogenously bility to cholesterol-induced atherosclerosis (7, 8). At the labeled plasma squalene (appearance and die away) are cellular level, diabetes suppresses hepatic 3-hydroxy-3- measured over a 7-h period. By quantitating the utiliza- methylglutaryl-coenzyme (HMG-CoA) reductase (9-11), tion of squalene into cholesterol, this method avoids the although the details of this effect have not been well assumption of no net input of cholesterol into the plasma studied. from slowly turning over cholesterol pools. By virtue of Two reports have addressed the effect of diabetic plasma squalene's relatively rapid turnover rate, it mea- control on cholesterol balance in humans (12, 13). Each sures synthesis in a single day. The short duration of the found that fecal bile acid excretion is diminished by study is both a theoretical and a practical advantage. insulinization, but one study noted a compensatory in- Results of the squalene kinetic method agree closely with crease in fecal neutral steroids (12) while the other did cholesterol synthetic rates established in stable subjects not (13). The importance of this discrepancy is that, by cholesterol balance (15, 16). assuming subjects were in metabolic balance, a dimin- In the present report we have applied the method of ished net sterol excretion would suggest diminished cho- squalene kinetic analysis to diabetes. By studying sub- lesterol synthesis under the influence of insulin (13). The jects in poor diabetic control and later in improved study from our laboratory (12) did not document a diabetic control, we have determined the effect of this change in net sterol excretion, but cholesterol balance clinical manipulation on the disposition of circulating mevalonate, and we have considered the interpretation of these results as they bear on cholesterol synthesis. Received April 24, 1979. Address requests for reprints to: Dr. Christopher D. Saudek, De- partment of Medicine, Clinical Research Center, Adult Unit, New York Materials and Methods Hospital-Cornell Medical Center, 525 East 68th Street, New York, New York 10021. Patients * This work was supported in part by NIH Grant HL-20488; the General Clinical Research Centers Branch, Division of Research Re- Five diabetics were studied at the Cornell Clinical Research sources (DRR-0047); and the New York Diabetes Association. Center. Clinical characteristics are noted in Table 1; all were 824 CHOLESTEROL METABOLISM IN DIABETES 825 maturity-onset diabetics within 120% of ideal body weight. The plasma, which was then saponified under mild conditions in mean duration of diabetes was 12.0 yr, and while four of the KOH (10%, wt/vol) and methanol (70%) at 70 C for 2 h. The five had some evidence of diabetic complications, none had nonsaponifiable fraction was recovered with three extractions clinically evident renal, hepatic, or intestinal disease. of petroleum ether, and this was concentrated and transfered All subjects volunteered for the study and gave fully informed to an 8 X 1-cm column of alumina oxide (Alumina, AG 7, 200- consent, in accordance with procedures established by the mesh, Bio-Rad Laboratory, Richmond, CA). Squalene and institutional review board. squalane were recovered using 50 ml petrol ether; cholesterol remained in the column. Four fifths of this hydrocarbon fraction 14 Study design was counted for C on a Tri-Carb counter (model 3255, Packard Co., Downers Grove, IL). One fifth was applied to a gas-liquid Patients were studied initially while in chronically poor dia- chromatograph (model 4710A, Hewlett Packard Co., Palo Alto, betic control and then 3-4 months later while in improved CA) using 1% Dexsil 300 on 100- to 200-mesh Supelcoport Downloaded from https://academic.oup.com/jcem/article/49/6/824/2679269 by guest on 29 September 2021 diabetic control. After the first infusion, insulin treatment was (Supelco, Inc., Bellefonte, PA) for squalene concentration. The initiated (in 4 patients) or insulin dosage was increased (in 1 14C to 3H cholesterol ratio was measured 21 or more days after patient). For at least a month before the second study, subjects the infusion, at a time when all [14C]MVA was converted to were maintained on 2 or more injections of insulin daily, in [14C]cholesterol (14, 15). This ratio was used to calculate the combinations of short and intermediate acting preparations. dose of administered squalene (see below). Plasma was sapon- Glycemia was monitored by routine urine glucose testing at ified in KOH-methanol as described above, cholesterol being home (data not presented) and by measurement of plasma extracted with petroleum ether, and the extract was counted glucose 13 times throughout the study day. Thus, while diabetic for 14C and 3H. control on the infusion day correlated with the degree of glu- For analysis of the 14C to 3H cholesterol ratio after the second cosuria in preceding weeks, the measured glycemia during the infusion, the isotope die-away curves from the first infusion infusion was considered most relevant to the study of insulin's (given 15-19 weeks earlier) were extrapolated and appropriate effect during that particular day. corrections were made. Maximum error in this technique was judged to be 3%, since residual radioactivity from the first Infusion procedures infusion was only 5-15% of that due to the second infusion, and the die-away curve of the first infusion was relatively linear at On the morning of infusion, patients remained fasting. Intra- 15-19 weeks. venous lines were established in each arm, one being used for The calculation of cholesterol synthesis depends upon the isotope infusion and the other for blood sampling. RS-[5- I4 14 fact that squalene is obligatorily metabolized to cholesterol in C]MVA ([ C]MVA; Schwarz/Mann, Orangeburg, NY; 240- the active cholesterol-synthesizing organs, so that cholesterol 260 juCi; SA, 15 mCi/mM) was filtered through a Millipore filter synthesis equals squalene synthesis X 0.943 (to account for the (Swinnex-24, 0.45 /im, Swinnex Millipore Co., Bedford, MA) 3 cleaved carbon atoms in demethylation steps between la- into 150 cc sterile saline. [l,2- H]Cholesterol ([3H]cholesterol; nosterol and cholesterol). Squalene synthesis is calculated as New England Nuclear, Boston, MA; SA, 53 mCi/mM) was follows: squalene synthesis (milligrams per day) = dose of purified on thin layer chromatography, suspended in ethanol, squalene (disintegrations per min) •*• area under squalene spe- 14 and then added to the same saline bottle as [ C]MVA. Each cific activity curve (disintegrations per min/mg-day). 3 subject received [ H]cholesterol in the first study from the same The area under the specific activity curve of plasms squalene lot as in the second. After thorough mixing, the isotopes were was calculated by computer program based on trapezoidal administered to the subject within 30 min of preparation in measurements of the upswing in specific activity and calculation saline. Two baseline plasma samples were obtained and the of the least square best fit for the die-away portion. Not all of isotopes were infused iv over a 5-min period. The infusion line the [14C]MVA was converted to squalene. First, the S-MVA of and all glassware and filters were rinsed with petroleum ether and chloroform for determination of residual isotopes. TABLE 1. Clinical characteristics After infusion of the [14C]MVA and [3H]cholesterol, 12 cc blood were withdrawn in 1 mg/ml EDTA for analysis of plasma Total daily insulin squalene specific activity and plasma triglyceride, cholesterol, Pa- Duration dose" Age % Compli- and glucose concentrations at 15, 30, 60, 90, 120, 180, 240, 300, tient Sex of diabe- (yr) IBW cations* 360, and 420 min (total blood drawn, 144 cc). Blood was cooled no. tes (yr) 1st in- 2nd infu- immediately and plasma was separated within 1 h. During the fusion sion day, the subject ingested a standard breakfast and lunch, ex- 1 72 M 100 15 0 60N, 20R BR, CHD, cluding foods which contain significant amounts of squalene PVD (17). Subjects received their customary insulin dose after iso- 2 41 F 115 14 40N 90N, 40R BR tope infusion. 3 59 F 110 6 0 25N PVD, HC 4 65 F 100 6 0 25N, 5R 5 64 F 120 19 0 20N BR, N Analyses IBW, Ideal body weight. Determination of plasma squalene specific activity was per- " N, NPH insulin; R, regular insulin. formed according to the method described in greater detail * BR, Background retinopathy; CHD, coronary heart disease; PVD, elsewhere (18). Briefly, 2 /xg squalene were added to 3 ml peripheral vascular disease; N, neuropathy; HC, hypercholesterolemia. 826 FRIER AND SAUDEK JCE&M • 1979 VoU9 No 6 the isomeric mixture was biologically inactive. Of the biologi- concentrations were significantly affected in this small cally active, JR-[14C]MVA, a variable portion was converted to group of patients (Table 2). There was no consistent or nonsterol metabolites, as demonstrated by Popjak and col- significant weight gain between infusions despite the leagues (19, 20). Accepting that all squalene formed was metab- improvement in glycemia. olized to cholesterol, however, the use of [3H]cholesterol in the 14 3 A typical squalene specific activity curve is pictured in infusion and the measurement of C to H cholesterol ratio Fig. 2. It will be noted that [14C]squalene appears in the at a later date permitted calculation of the actual dose of 14 14 14 3 plasma within minutes of [ C]MVA infusion and rises to [ C]squalene: squalene dose (disintegrations per min) = C: H cholesterol in plasma X [3H]cholesterol administered (disinte- a peak specific activity at about 90 min, dying away in grations per min). log-linear fashion over the subsequent 6 h. Knowing the dose of i?-[14C]MVA administered, the dose of Data from the squalene kinetic analyses are listed in [3H]cholesterol, and the plasma 14C to 3H ratio at a later date, 14 Meal Downloaded from https://academic.oup.com/jcem/article/49/6/824/2679269 by guest on 29 September 2021 the percent conversion of i?-[ C]MVA to cholesterol could also Poor control 3 be calculated: percent conversion = dose of [ H]cholesterol Improved control (disintegrations per min) -s- dose of #[14C]MVA x 14C:3H. 400 n Plasma cholesterol and triglycerides were analyzed using the Auto Analyzer, methods N-74a and N78a (Technicon, Tarry- town, NY). Plasma glucose was measured by the Beckman glucose analyzer (Beckman Co., Mountainside, NJ), which uses a glucose oxidase method. Student's paired t test (21) was used for statistical analysis, with each subject studied before and after achieving improved diabetic control.

Results Table 2 lists changes in plasma glucose, cholesterol, 9 a.m. Noon 4 p.m. and triglyceride concentrations. The mean (±SE) daytime FIG. 1. Mean plasma glucose concentrations (±SEM) over the 7-h plasma glucose while ingesting the standardized meals study period. Subjects were given two meals, at 0930 and 1200 h, and was 349 ± 20 mg/dl in the poor control period and 175 received their customary morning dose of insulin (for the improved ± 15 mg/dl during the improved control period (P < control studies) at 0900 h. Improved control plasma glucoses are all 0.001, paired control), representing an approximate halv- significantly less than their corresponding poor control points (P < 0.01). ing of mean glycemia. The area under the curve of daytime glycemia decreased from 2386 to 1256 mg/dl-h. Figure 1 illustrates the mean plasma glucose curves 00 throughout the day. At all 12 time points, glucose was 80 significantly less during the improved control period (P 60 x < 0.01). Neither total plasma cholesterol nor triglyceride 40 TABLE 2. Plasma glucose, cholesterol, and triglyceride concentration (milligrams per dl) e 20 Plasma glucose Plasma Plasma e Patient Infusion" choles- triglyc- a. x no. 6 Mean Range terol" eride 10 - 422 356-516 226 286 CO 8 235 173-306 269 298 c 6 - 317 219-378 179 165 O 4 D - 158 90-226 205 147 cr X, CO 320 223-381 254 272 2 - 170 132-213 223 294

322 261-362 197 66 i i i i 1 1 1 158 92-218 172 64 100 200 300 400 500 600 Minutes 366 312-415 187 255 FIG. 2. A typical squalene specific activity curve vs. time is depicted. 153 99-198 215 134 14 [ C]Squalene is measured in plasma after iv administration of " Infusion I, Poor control; II, improved control. [14C]MVA. The line drawn is the least squares best fit to this log-linear h Mean of three or more fasting samples. decay. CHOLESTEROL METABOLISM IN DIABETES 827

Table 3. Plasma squalene concentration changed variably which is assumed to eliminate any effect of the treatment from the poor control to the improved control periods, as on net flux of cholesterol. did the area under the squalene specific activity curve. The injection of [14C]MVA into the circulating pool of The fraction of [14C]MVA converted to [14C]squalene, MVA and the measurement of its appearance as [I4C]- however, increased in the improved control period in all squalene has obvious technical advantages to the balance five subjects (mean of 55.2 ± 2% in poor control period method of estimating cholesterol synthesis. However, by vs. 67.5 ± 4.2% in improved control; P < 0.025). This entering the cholesterol synthetic pathway beyond the increased fractional conversion was associated with an presumed rate-limiting step (the reduction of HMG-CoA increase in cholesterol synthesis in all five subjects, from to MVA), the results may be affected by changes in 964 ± 75 mg/day in the poor control to 1206 ±111 mg/ mevalonate pool size as well as by changes in cholesterol day in the improved control (P < 0.025; Fig. 3). 1500 Downloaded from https://academic.oup.com/jcem/article/49/6/824/2679269 by guest on 29 September 2021 The fractional conversion of MVA to squalene corre- I 1 Poor control lated with the derived figure for cholesterol synthesis (r !£223 Improved control = 0.82; P < 0.001) more closely than did the area under the squalene specific activity curve (r = 0.19; P = NS). 1000 T Discussion There is no single or universally applicable method available to measure cholesterol synthesis in man. Sterol balance requires lengthy hospitalization to establish met-

Fractional Cholesterol Plasma squalene Area (dpm/mg- Patient no. Infusion synthesis a day)6 (Mg/dl) (mg/day) 1 I 17.9 180,000 54.0 828 II 19.4 188,734 61.5 911

2 I 25.2 154,757 52.1 856 II 23.4 134,730 61.1 1237

3 I 33.9 152,063 57.0 954 II 16.3 125,638 68.0 1331

4 I 14.5 150,879 55.4 933 II 10.3 171,127 63.8 1016

5 I 26.1 125,261 57.6 1251 II 16.1 147,719 83.6 1535

Mean ± SEM I 23.5 ± 3.4 152,600 55.2 ± 1.0 964 ± 75.4 II 17.1 ± 2.1 153,600 67.5 ± 4.2 1206 ± 111

P (I vs. II) NS NS <0.025 <0.025 ° Mean of 13 plasma squalene determinations on day of infusion. 6 Calculations described in text. 828 FRIER AND SAUDEK JCE & M • 1979 VoU9 • No 6 synthesis. In our studies, if insulin caused a smaller MVA insulin administered in the usual clinical manner (sc, as pool size, this could yield the apparent increase in cho- a mixture of short and long acting forms). Beyond the lesterol synthesis noted. direct effect of insulin, this treatment clearly affects a While MVA pool size has not been directly measured variety of factors (for example, lipolysis, protein synthe- in man, data are available on plasma MVA concentra- sis, and dietary intake) which could indirectly affect tions which suggest that a decrease of MVA pool size is cholesterol synthesis. Furthermore, this mode of treat- not a credible explanation for our results. First, the ment is far from the finely tuned pattern of insulin release amount of i?-[14C]MVA administered (8.3 jumol, distrib- seen in normal insulin-glucose homeostasis. Thus, a great uted into an extracellular fluid volume of approximately deal more must be learned about the effects of normal, 14 liters) would yield a mevalonate concentration of relatively short-lived insulin secretion on cholesterol syn- about 0.59 JUM, or 9 times the basal blood MVA concen- thesis before making recommendations relevant to the Downloaded from https://academic.oup.com/jcem/article/49/6/824/2679269 by guest on 29 September 2021 tration of 67 nM estimated by Hagenfeldt and Hellstrom clinical treatment of diabetes. (22). This increased concentration would probably over- ride any change in basal mevalonate concentration. Sec- Acknowledgments ond, MVA concentration in the rat is inversely propor- The authors wish to thank Demosthenes Tambakos for excellent tional to manipulations of liver HMG-CoA reductase technical assistance and Ms. Maria Lopez and Mary Lynne Smith for preparation of the manuscript. We also thank Dr. Donald McNamara induced by dietary cholesterol. Diabetes is known to for reviewing the manuscript and offering helpful suggestions. reduce this enzyme in the rat liver (9-11) and insulin restores its activity. Therefore, our insulin treatment References would be expected to increase the MVA concentration, not decrease it. Third, two studies using similarly large 1. Medalie, J. H., C. Papier, J. B. Herman, U. Goldbourt, S. Tamir, H. N. Neufeld, and E. Riss, Diabetes mellitus among 10,000 adult men. and variable doses of isotopically labeled MVA in man I. Five-year incidence of associated variables, Isr J Med Sci 10: have yielded consistent results. McNamara et al. (15) 681, 1974. found that squalene kinetic data compare well with sterol 2. Albrink, M. J., P. H. Lavietes, and E. B. Man, Vascular disease and serum lipids in diabetes mellitus. Observations over 30 years (1931- balance data in measuring cholesterol synthesis using 1961), Ann Intern Med 58: 305, 1963. doses of [14C]MVA identical to ours. Schwartz et al. (23) 3. Kaufmann, R. L., J. P. H. Assal, J. S. Soeldner, E. G. Wilmshurst, derived comparable kinetic data of cholesterol and bile J. R. Lemaire, R. E. Gleason, and P. White, Plasma lipid levels in diabetic children. Effect of diet restricted in cholesterol and satu- acid turnover over a range of MVA administration vary- rated fats, Diabetes 24: 672,1975. ing from 70-285 juCi MVA (5-3H or 2-14C, specific activity 4. Sharma, D., B. C. Bansal, and C. Prakash, Serum lipid studies in not given). The above considerations suggest that a de- insulin-dependent diabetics below the age of 30 years, J Indian MedAssoc 54: 416, 1970. crease in MVA pool size under the influence of insulin is 5. Bar-On, H., P. S. Roheim, and H. A. Eder, Hyperlipoproteinemia not a likely explanation of our results. in streptozotocin-treated rats, Diabetes 25: 509,1976. Because calculations used in deriving cholesterol syn- 6. Schonfeld, G., C. Birge, J. P. Miller, G. Kessler, and J. Santiago, Apolipoprotein B levels and altered lipoprotein composition in thesis include the measurement of the area under the diabetes, Diabetes 23: 827, 1974. squalene specific activity curve (as denominator) and the 7. Lehner, N. D., T. B. Clarkson, and H. B. Lofland, The effect of measurement of administered dose of squalene (as nu- insulin deficiency, hypothyroidism and hypertension on atherosclerosis in the squirrel monkey, Exp Mol Pathol 15: 230, 1971. merator), we were able to determine which factor was 8. Lehner, N. D., T. B. Clarkson, F. P. Bells, R. W. St. Clair, and H. consistently changed by insulinization. The calculated B. Lofland, Effects of insulin deficiency, hypothyroidism and hy- dose of administered squalene increased in all five sub- pertension on atherosclerosis in the squirrel moneky, Exp Mol Pathol 16: 109, 1972. jects upon establishing improved diabetic control. This 9. Lakshmanan, M. R., C. M. Nepokroeff, G. C. Ness, R. E. Dugan, indicates that the fractional conversion of [14C]MVA to and J. W. Porter, Stimulation of insulin of rat liver hydroxy- [14C]squalene is increased by insulin. methylglutaryl CoA reductase and cholesterol-synthesizing activi- ties, Biochem Biophys Res Commun 50: 704, 1973. The possibility that uncontrolled diabetes causes de- 10. Bhathena, S. J., J. Avigan, and M. E. Schreiner, Effect of insulin pressed cholesterol synthesis and that this state is re- on sterol and fatty acid synthesis and hydroxymethylglutaryl CoA versible by insulin is consistent with several other pieces reductase activity in mammalian cells grown in culture, Proc Natl AcadSci USA 71: 2174, 1974. of evidence. Fasting has many metabolic similarities to 11. Nakayama, H., and S. Nakagawa, The influence of streptozotocin uncontrolled diabetes (low plasma insulin, increased diabetes on intestinal 3-hydroxy-3-methylglutaryl coenzyme A re- FFA, hyperketonemia, and utilization of lipids rather ductase activity in the rat, Diabetes 26: 439, 1977. 12. Saudek, C. D., and E. Brach, Cholesterol metabolism in diabetes. than carbohydrates as the primary energy substrate), I. The effect of diabetic control on sterol balance, Diabetes 27: and fasting is characterized by diminished cholesterol 1059, 1978. synthesis (24). As noted, diabetes depresses HMG-CoA 13. Bennion, L. J., and S. M. Grundy, Effects of diabetes mellitus on cholesterol metabolism in man, N Engl J Med 296: 1365, 1977. reductase in hepatic microsomes (9, 10), and insulin 14. Liu, G. C. K., E. H. Ahrens, Jr., P. H. Schreibman, P., Samuel, D. reverses this effect (9). J. McNamara, and J. R. Crouse, Measurement of cholesterol syn- The present data tested the effect of large doses of thesis in man by isotope kinetics of squalene, Proc Natl Acad Sci CHOLESTEROL METABOLISM IN DIABETES 829

USA 72: 4612, 1975. alonate in rats and man not leading to sterols, J Biol Chem 250: 15. McNamara, D. J., E. H. Ahrens, Jr., P. Samuel, and J. R. Crouse, 1771, 1975. Measurement of daily cholesterol synthesis rates in man by assay 20. Edmond, J., and G. Popjak, Transfer of carbon atoms from meva- of the fractional conversion of mevalonic acid to cholesterol, Proc lonate to n-fatty acids, J Biol Chem 249: 66, 1974. Natl Acad Sci USA 74: 3043, 1977. 21. Colton, T., Statistics in Medicine, Little, chapt. 4, Brown, and Co., 16. Goodman, DeW. S., R. P. Noble, and R. B. Dell, Three-pool model Boston, 1974, p. 99. of longterm turnover of plasma cholesterol in man, J Lipid Res 14: 22. Hagenfeldt, L., and K. Hellstrom, Blood concentration and turnover 178, 1973. of circulating mevalonate in the rat, Life Sci 11: 669, 1972. 17. Liu, G. C. K., E. H. Ahrens, Jr., P. H. Schreibman, and J. R. Crouse, 23. Schwartz, C. C, M. Berman, Z. R. Vlahcevic, L. G. Halloran, D. H. Measurement of squalene in human tissue and plasma: validation Gregory, and L. Swell, Multicompartmental analysis of cholesterol and application, J Lipid Res 17: 38, 1976. synthesis in man: characterization of the hepatic bile acid and 18. Saudek, C. D., and G. C. K. Liu, Plasma squalene: lipoprotein biliary cholesterol precursor sites, J Clin Invest 61: 408, 1978. distribution and kinetic analysis, J Lipid Res 19: 827, 1978. 24. Miettinen, T. A., Fecal steroid excretion during weight reduction in 19. Fogelman, A. M., J. Edmond, and G. Popjak, Metabolism of mev- obese patients with hyperlipidemia, Clin Chim Ada 19: 341, 1968. Downloaded from https://academic.oup.com/jcem/article/49/6/824/2679269 by guest on 29 September 2021

Male Reproductive Function

Concept Accommondation will be at the Iluka, a modern and comfortable motel. The Gold Coast is an infor- The Satellite Symposium on Male Reproductive mal holiday playground and dress is causual. Function will be held following the Sixth Interna- tional Congress of Endocrinology and will combine discussions of the basic scientific and clinical as- Participants pects of this important subject in a relaxed setting. Invited participants include: David de Kretser The programme is designed to be flexible and will (Monash University, Melbourne), Fernand Labrie include addresses by invited speakers and short (L'Universite Laval, Quebec), Richard Santen presentations of contributed material for which (Pennsylvania State University, Hershey), and time will be limited to 10 minutes. Ample time will Emil Steinberger (University of Texas, Houston). be allowed for discussion. Closing Dates Venue Queensland's Gold Coast is renowned for its cli- Abstracts/Registration: December 1, 1979. mate, beaches, surf and fine restaurants. Surfers Abstracts should be submitted on a similar form to paradise is in the heart of the Gold Coast and is a that required by the Sixth International Congress short drive from the rain forests of the hinterland of Endocrinology. and Brisbane, the capital of Queensland. The cli- mate is excellent with warm summer days averag- All correspondence and requests for further infor- ing 28°C (86°F) during the day and dropping to a mation should be addressed to: Dr. R. Mortimer, pleasant 22°C (72°F) at night. Access from Mel- Endocrinologist, Royal Brisbane Hospital, Herston, bourne is by air to the Coolangatta Airport. Qld. 4006