Prenatal Androgenization of Lambs: 11. Metabolism in Adipose Tissue and Liver'

L. R. Hansen, J. K. Drackley2, L. L. Berger, D. E. Grum, J. D. Cremin, Jr.3, X. Lin, and J. Odle

Department of Animal Sciences, University of Illinois, Urbana 61801

ABSTRACT: In vitro measurements of metabolism were made in subcutaneous and perirenal adipose tissue (AT) and liver from prenatally androgenized ewe lambs (TE), control ewe lambs (CE ), and control wether lambs ( C W). In adipose tissue slices, release of glycerol or fatty acids into the medium was not different among treatments, but glycerol release was greater ( P < .O 1) from subcutaneous AT than from perirenal AT. Basal fatty acid release and the free fatty acid pool were greater ( P < .05) for perirenal AT than for subcutaneous AT; fatty acid release and the fatty acid response (increased NEFA in media and tissue) were increased more by lipolytic stimuli in subcutaneous AT than in perirenal AT. Adipose tissue from CW had the greatest ( P < .05) fatty acid response under conditions of near-maximal stimula- tion; rates from TE were intermediate to those from CW and CE. Incorporation of glucose into fatty acids and glycerol in subcutaneous AT was lowest ( P < .05) for TE. Oxidation of glucose and acetate to C 0 2 and incorporation of acetate into fatty acids or glycerol in subcutaneous AT, glucose and acetate metabolism in

perirenal AT, and cellularity measurements for both AT did not differ among treatments. In liver slices, oxidation of [l-14C]propionate to C 0 2 was greater ( P < .05) for CE than for TE or CW, and gluconeogenic capacity from [l-14C]propionate tended to be greater ( P < . l o ) for CE than for TE. Glucose and C 0 2 production from [2-14Clpropionate, [U-l4C1alanine, or [U-14Clglycerol and total and peroxisomal first cycle of @-oxidation of [l-14C]palmitate were not altered by prenatal androgenization or sex. There were no effects ( P > . l ) of prenatal exposure t o testosterone on mitochondrial protein content of liver, rates of mitochondrial state 3 or state 4 respiration, the ratio of ADP:oxygen in the presence of respiratory sub- strates, or hepatic contents of lipid, triglyceride, or glycogen. Protein content of liver was greater ( P < .05) for CW than for CE; TE were intermediate. Collectively, there were minimal modifications of in vitro metabolism in AT or liver attributable to prenatal androgenization or sex that would directly influence ADG and carcass composition.

Key Words: Pregnancy, Testosterone, Lambs, Adipose Tissue, Liver, Metabolism

Introduction

Fat deposition occurs earlier and faster in female lambs than in ram or wether lambs (Shelton and Carpenter, 1972). Excessive fat deposits in ewe lambs decrease efficiency of meat production and retail carcass value (Carpenter et al., 1964). Prenatal androgenization decreased subcutaneous and internal

lThe authors extend their gratitude to Karen Williamson and Laurel Brown for their assistance.

2To whom correspondence should be addressed: 260 Animal Sciences Lab, 1207 W. Gregory Drive.

3Present address: Dept. of Nutr. Sci., Univ. of California- Berkeley, 212 Morgan Hall.

Received August 11, 1994. Accepted March 8, 1995.

J . Anim. Sci. 1995. 73:1701-1712

fat of ewe lambs (DeHaan et al., 19871, resulting in carcasses more similar to those of wether or ram lambs, but the mechanisms for this response remain unclear. Fat accretion depends on the balance between rates of lipolysis and lipogenesis in adipose tissue (AT); alterations of either or both of these rates by prenatal androgenization could produce lambs with leaner carcasses.

Prenatally androgenized ewe lambs had enlarged livers (DeHaan et al., 1987; Jenkins et al., 1988), which might contribute t o the altered growth patterns of these animals because of the extensive metabolic activity in liver. The ruminant liver is a primary site of gluconeogenesis from propionate, amino acids, and glycerol (Leng, 1970) and a major site of mitochon- drial and peroxisomal 0-oxidation of long-chain fatty acids (Grum et al., 1994). Rats treated with de- hydroepiandrosterone ( DHEA; a steroid precursor)

1701

1702 HANSEN

showed hepatocyte hypertrophy, peroxisomal prolifer- ation, and altered mitochondrial ultrastructure (Bellei et al., 1992). Female goats injected with testosterone had mitochondrial respiration rates similar to those of male goats (Brown et al., 1988). Whether prenatal exposure to testosterone might result in similar effects in sheep is not known.

To identify mechanisms for increased ADG (De- Haan et al., 1987; Hansen et al., 1995) and decreased fat deposition (DeHaan et al., 1987) in prenatally androgenized lambs, our objectives were to character- ize the effects of prenatal androgen exposure on basal and catecholamine-stimulated rates of lipolysis, incor- poration of glucose and acetate into lipids, and size and number of adipocytes in subcutaneous and perirenal AT and rates of gluconeogenesis, mitochon- drial and peroxisomal first cycle of P-oxidation, and mitochondrial respiration in liver.

Materials and Methods

Lambs

All procedures were conducted under protocols approved by the University of Illinois Laboratory Animal Care Advisory Committee. Lambs from the growth trial described by Hansen et al. (1995) were used for the study of metabolism in AT and liver. Pregnant Suffolk ewes were implanted with testoster- one propionate between d 40 and 70 of gestation. Prenatally androgenized ewe lambs ( TE), control ewe lambs (CE) , and control wether lambs (CW) were blocked according to weight at the beginning of the feeding trial. Lambs ( n = 7 per group) were slaugh- tered by block when the heaviest lamb in each block reached approximately 55 kg. Lambs were allowed ad libitum access to feed before slaughter to minimize effects of lipolytic stimuli that might occur after feed withdrawal. Because of the large number of in vitro measurements made, only one of the three lambs of each block was slaughtered per day. Prenatal an- drogenization increased ADG by 13% without affecting carcass composition, but liver weights tended t o be greater than those from CE (Hansen et al., 1995).

Adipose Tissue Sampling and Assays

Sample Collection and Preparation. Subcutaneous AT was collected from the area over the shoulder and internal AT was obtained from the perirenal depot. Tissues were obtained within 5 min postmortem and placed immediately into bottles containing Krebs- Henseleit bicarbonate buffer ( K B ; 118 mM NaC1, 4.7 mM KCl, 1.25 mM CaC12, 1.19 mA4 KH2PO4, 1.16 mM MgS04, and 25 mA4 NaHC03) supplemented with 10 mM HEPES (KB/HEPES) and maintained at 38°C and pH 7.4. Bottles were placed into an insulated bottle containing water at 38°C for transportation to the laboratory.

ET AL.

Slices of AT weighing approximately 35 mg and approximately 250 pm thick were prepared with a hand microtome constructed to the specifications of Faulhaber et al. ( 1972). Slices were prepared, blotted, and weighed under a heat lamp positioned to maintain the ambient temperature of the work area at approxi- mately 38°C. Slices were prepared and loaded into flasks within 1 h after slaughter.

Lipolysis. The method of McNamara and Hillers (1986) was adapted for measuring lipolysis in AT. Approximately 70 mg of tissue slices was added to flasks containing 2.0 mL of KBBEPES buffer with 3% (wt/vol) BSA ( KBLHEPESBSA); flasks were prein- cubated for 20 min in a shaking water bath at 38°C to stabilize the rate of lipolysis. After the preincubation period, the medium was aspirated and replaced with fresh medium, which contained lipolytic stimulators (i.e., effectors) when appropriate. Tissue treatments included lipolysis under basal ( B ) conditions (i.e., in KBBEPESBSA only), and lipolysis stimulated with

M epinephrine bitartrate ( E ) , M epinephrine bitartrate plus M theophylline plus 1 unit adenosine deaminase ( ETAD), or lo-* M norepinephrine ( NE). Effectors were purchased from Sigma Chemical (St. Louis, MO). Treatment with ETAD was designed to provide an estimate of near- maximal lipolytic capacity (Iliou and Demarne, 1987). Flasks were gassed with 02:COz (95:5) and incubated an additional 2 h, then placed on ice to stop metabolic activity. Tissue from triplicate flasks was pooled and frozen ( -2O"C), and the medium from each flask was frozen individually (-20°C) until it was assayed for contents of glycerol and NEFA. Tissue and media NEFA were extracted and quantified by microtitration according to the method of Kelley ( 1965). Concentrations of glycerol in media were determined enzymatically as described by Eggstein and Kuhlmann (1974).

Lipogenesis. For measurement of lipogenic capacity, approximately 100 mg of tissue slices from subcutane- ous or perirenal AT was placed into 25-mL Erlen- meyer flasks containing 3 mL of incubation buffer (McNamara and Hillers, 1986) with 3% (wt/vol) BSA at pH 7.4. The medium also contained 5 mM glucose, 10 mM sodium acetate, .3 IU bovine insulin (Sigma Chemical), and either 3 pCi of D-[U-14Clglucose (New England Nuclear, Boston, MA) or 1.5 pCi of Na-[l- l4C1acetate (Research Products International, Mount Prospect, IL). These concentrations of substrates and insulin yielded maximal lipogenic rates. Each tissue type and radiotracer combination was incubated in triplicate.

Radiolabeled C02 was collected and quantified by liquid scintillation spectrometry (Beckman Model LS 6000 IC, Beckman Instruments, Fullerton, CA) as described previously (Cremin et al., 1994). Lipids in the tissue and medium were extracted according to Bligh and Dyer (1959) and washed as described by

PRENATALLY ANDROGENIZED LAMBS: METABOLISM 1703

Folch et al. (1957). The resulting extracts were dried under moving air and dissolved in 10 mL of chloro- form, and radioactivity in glycerol and fatty acids was determined as described by Baldner et al. (1985).

Adipose Tissue Cellularity. On the day of slaughter, 10-g samples of AT from subcutaneous and perirenal sites were blotted and frozen until examined for cellularity characteristics. Tissue was examined microscopically according to the frozen-cut method described by Sjostrom et al. (1971). Mean cell diameter and number of adipocytes per gram of tissue were calculated from 300 adipocytes per sample.

Liver Sampling and Assays

Gluconeogenesis. The liver of each lamb was ob- tained and weighed within 5 min postmortem. A portion of the liver was placed immediately into ice- cold phosphate-buffered (.015 W .9% (wthol ) NaCl at pH 7.4. Rates of gluconeogenesis from propionate, alanine, and glycerol were determined as described previously (Cremin et al., 1994). Liver slices approxi- mately 100 pm thick and weighing 20 to 40 mg were prepared with a Krumdieck tissue slicer ( K and F Research, Birmingham, AL) within 30 min postmor- tem. Approximately 100 mg of tissue was blotted, weighed, and placed into 25-mL Erlenmeyer flasks containing 3 mL of KB at pH 7.4. Substrates were 5 mM L-alanine with 1 pCi per flask of L-[U-I4C]alanine (Amersham International, Amersham, U.K.) or 5 mM glycerol with 1 pCi of [2-14Clglycerol (ICN Radiochem- icals, Irvine, CA). Gluconeogenesis from propionate was measured in flasks containing 3 mL of KB and 10 mM sodium propionate. For the first four complete blocks (i.e., 12 sheep), 1 pCi of Na-[2-14Clpropionate (ICN Radiochemicals) per flask was used as the tracer, whereas blocks five through seven (nine sheep) used 1 pCi of Na-[l-14Clpropionate (Amer- sham International) per flask as the tracer.

Radiolabeled C02 and glucose were quantified by liquid scintillation as described previously (Cremin et al., 1994). Glucose derived from propionate was isolated by anion-cation exchange chromatography using the three-column procedure described by Mills et al. (1981). The five-column procedure (Mills et al., 1981) was used for isolating glucose originating from alanine. Glucose originating from glycerol was isolated by pouring the sample through a hydrogen-charged resin in tandem with a borate-charged resin (Mills et al., 1981). The borate column was washed twice with 5-mL aliquots of 20 mM sodium tetraborate and the eluent was discarded. The glucose bound to the borate- charged resin was eluted with one 3-mL and one 5-mL aliquot of 1 N acetic acid (Terrettaz and Jeanrenaud, 1990). The resulting eluent containing the isolated glucose was treated as described by Mills et al. (198 1). All flask treatments were incubated in triplicate and metabolic rates were calculated on a tissue wet-weight basis.

P-Oxidation of Fatty Acids. Hepatic total and peroxisomal fatty acid oxidation were measured in liver homogenates as described previously (Grum et al., 19941, based on the methodology of Veerkamp and van Moerkerk ( 1986). Aliquots of homogenate were incubated in medium containing 1 mM Na-palmitate plus 1.35 pCi of Na-[l-14Clpalmitate (ICN Radiochem- icals) complexed with BSA in a molar ratio of 5:l. To measure peroxisomal @-oxidation, flasks also con- tained 50 p M antimycin A and 10 p M rotenone to inhibit mitochondrial @-oxidation. Accumulation of radiolabel from [l-14C]palmitate in acid-soluble metabolites and 14C02 was measured by liquid scintillation as described by Grum et al. (1994).

Mitochondrial Respiration. At slaughter, a portion of the liver was placed in ice-cold isolation buffer (Mersmann et al., 1972) at pH 7.2. Liver tissue was homogenized in isolation buffer with a hand-held tissue grinder and a mitochondrial fraction was prepared according to procedures described by Mers- mann et al. ( 1972). All procedures were conducted a t 4°C. The protein concentration of the resultant mitochondrial suspension was determined by the method of Lowry et al. (1951).

Mitochondrial respiration was measured in respira- tory medium (Aprille and Asimakis, 1980) with a YSI Model 5300 Biological Oxygen Monitor and 5331 Standard Oxygen Probe (Yellow Springs Instrument, Yellow Springs, OH). Oxygen uptake was monitored with a 2-channel chart recorder (LKB Produkter, Bromma, Sweden). Final concentrations of respiratory substrates included 5 mM pyruvate plus 5 mM malate, 5 mM glutamate plus 5 mM malate, or 10 mM succinate. State 4 rate of 0 2 uptake by mitochondria was measured in 2.6 mL of respiratory medium, 100 pL of substrate, 100 pL of 27% (wt/vol) BSA, and 100 pL of mitochondrial suspension ( 4 to 5 mg of mitochondrial proteirdml of buffer). Twenty-five microliters of 90 p M ADP was added to the reaction mixture to quantify state 3 rate of 0 2 consumption. Mitochondrial respiratory measurements were per- formed in triplicate for each substrate. Respiratory control ratio ( RCR) and ADP:O ratio were calculated as described by Estabrook ( 1967 j .

Liver Composition. On the day of slaughter, 10 g of liver was frozen ( -20°C) until analyses for contents of total lipid, triglyceride, glycogen, and total protein. Liver total lipid content was determined gravimetri- cally after ch1oroform:methanol (2: 1, vol/vol j extrac- tion (Folch et al., 1957). Triglyceride concentration was assayed according to the method of Foster and Dunn ( 1973 1. Hepatic glycogen was determined as described by Lo et al. (1970), and the method of Lowry et al. ( 195 1) was used to assay total protein concentration.

Statistical Analyses

Data for lipogenesis, AT cellularity, liver composi- tion, and rates of liver metabolism were analyzed by

1704 HANSEN ET AL.

ANOVA for a randomized complete block design. Lipolytic data were analyzed as a split (i.e., tissue site) - split (i.e., effector type) - plot design; the main effect of treatment was tested using the treatment x block mean square as the error term. Tissue site and the site x treatment interaction were tested by using the site x treatment x block mean square as the error term. The residual mean square was used as the error term t o test flask treatment (effector) effects and appropriate interactions. The design for glucose and C02 derived from Na-[l-14Clpropionate contained three complete blocks and the analysis for Na42- 14C]propionate as a tracer contained four complete blocks. All statistical calculations were performed using the GLM procedures of SAS ( 1985 1. Means were separated by using the protected least significant difference. Significance was P < .05; comparisons with P < . l0 were discussed as tendencies.

Results and Discussion

Adipose Tissue Metabolism

Lipolysis. Glycerol release from subcutaneous and perirenal AT into the media, representing complete hydrolysis of triglycerides, is presented in Table 1. Rates of glycerol release were not affected ( P = .78) by treatment but were greater in subcutaneous AT than in perirenal AT. A tendency ( P = .l41 for a treatment x site interaction was detected; for CE, rates of glycerol release were similar for subcutaneous and perirenal AT (163.4 and 138.6 nmol.2 h-I/100 mg of

Table 1. Glycerol release from subcutaneous and perirenal adipose tissue slices from lambsa

tissue), whereas for TE and CW, rates for subcutane- ous AT were more than twofold greater ( P < .001) than those for perirenal AT (196.8 vs 97.2 and 180.0 vs 74.8, for TE and CW, respectively). Rates of glycerol release were increased by ETAD in sub- cutaneous AT, but not in perirenal AT (effector x site interaction, P < .001). The enhanced lipolytic stimula- tion of subcutaneous AT in the presence of ETAD presumably occurred from the addition of theophylline and adenosine deaminase, which inhibited phos- phodiesterase and accumulation of adenosine, respec- tively, and thereby allowed a greater glycerol release.

Incubation of subcutaneous AT with E or NE did not affect glycerol release, but these catecholamines decreased glycerol release in perirenal AT (effector x site interaction, P < .001). Etherton et al. (1977) reported that in vitro glycerol release from subcutane- ous AT doubled in the presence of a catecholamine; such an effect of E or NE was not observed in our study. Previous studies also have reported catechola- mine-stimulated increases of glycerol release from perirenal AT of lambs (Etherton et al., 1977; Du- quette et al., 1984). Partial hydrolysis of triglyceride may have been increased by E or NE stimulation of AT because the concentration of NEFA in media tended to increase (Table 2). Subcutaneous AT glycerol release seemed t o be more sensitive to catecholamine stimulation than that from perirenal AT.

Fatty acid release into the media (Table 2) did not differ among treatments ( P = .91) or between tissue sites ( P = .95). The interaction of effector and tissue site was significant. Basal fatty acid release was

Table 2. Fatty acid release from subcutaneous and perirenal adipose tissue slices from lambsa

Treatmentb

Site and effectorC TE CE cu7 Treatmentb

Site and effectorC TE CE cw Subcutaneous - nmol.2 h-'/100 mg of tissue -

Be 162 135 132 Ee 121 116 124 ET AD^ 319 294 368 NEe 185 108 97

Be 132 206 116

ETADe 156 172 124

SEM 30 30 30

Perirenal

Ef 47 82 28

NE^ 54 94 32

aProbabilities of greater F for effects in model: treatment, P = .78; site, P < ,001; site x treatment, P = .14; effector, P c ,001; effector x treatment, P = .67; effector x site, P < .OOl; effector x site x treatment, P = .73.

bTE = treatment ewe lambs; CE = control ewe lambs; CW = control wether lambs.

'B = basal; E = epinephrine; ETAD = epinephrine plus theophyl- line plus adenosine deaminase; NE = norepinephrine.

d,e,fFor effector x site interaction, means (not shown) lacking a common superscript letter differ ( P < .05).

Subcutaneous - nEq.2 h-'/100 mg of tissue - Bi 96 96 83 Egh 252 277 211 ET AD^ 539 416 598 ~ ~ e f 394 318 321

Bh 167 220 235 Perirenal

Efgh 271 306 276 ETADe 3 18 370 457

SEM 48 48 48 ~ ~ e f g 294 33 1 338

aProbabilities of greater F for effects in model: treatment, P = .91; site, P = .95; site x treatment, P = .38; effector, P c .001; effector x treatment, P = .20; effector x site, P < .001; effector x site x treatment, P = .91.

bTE = treatment ewe lambs; CE = control ewe lambs; CW = control wether lambs.

CB = basal; E = epinephrine; ETAD = epinephrine plus theophyl- line plus adenosine deaminase; NE = norepinephrine.

d,e,f,g,hJFor effector x site interaction, means (not shown) lack- ing a common superscript letter differ ( P < .05).

PRENATALLY ANDROGENIZED LAMBS: METABOLISM 1705

greater ( P < .05) from perirenal AT than from subcutaneous AT. In subcutaneous AT, ETAD caused the greatest fatty acid release, followed by NE and E. In perirenal AT, fatty acid release was increased by ETAD and NE, but not E; in agreement with results for glycerol release, the catecholamines caused less stimulation of fatty acid release in perirenal AT than in subcutaneous AT. These results are in agreement with the increased fatty acid release observed when subcutaneous and perirenal AT from market ewe lambs were stimulated with E or E plus theophylline (Etherton et al., 1977).

The content of NEFA in subcutaneous and perirenal AT, referred to as the fatty acid pool, is presented in Table 3. The fatty acid pool was greater ( P . O 1) in perirenal AT than in subcutaneous AT but did not differ ( P = . l71 among treatments. The fatty acid pool was increased ( P < .05) by ETAD or E, but not NE. Increases of the fatty acid pool in subcutaneous or perirenal AT indicate an activated lipolytic response in the presence of E.

The fatty acid response, defined as the sum of the fatty acid pool plus fatty acid release when stimulated by an effector minus the sum of the fatty acid pool plus fatty acid release under basal conditions, is presented in Table 4. The fatty acid response was not different ( P = .62) when averaged across treatments but was greater ( P < .O 1) for subcutaneous than for perirenal AT. The interaction of effector x site was significant; the fatty acid response was greatest in both tissues in the presence of ETAD, but the relative increase was less in perirenal AT. Incubation with NE caused a greater fatty acid response in subcutaneous than in perirenal AT.

Table 3. Fatty acid pool in subcutaneous and perirenal adipose tissue slices from lambsa

Treatmentb

Site and effectorCd TE CE cw Subcutaneous - nEqil00 mg of tissue -

B 156 186 182 E 166 161 226 ETAD 207 207 285 NE 152 159 181

B 203 240 277 E 206 319 337 ETAD 213 308 362 NE 230 262 296

SEM 21 21 21

Perirenal

aProbabilities of greater F for effects in model: treatment, P = .17; site, P = ,002; site x treatment, P = .47; effector, P < .001; effector x treatment, P = .24; effector x site, P .18; effector x site x treatment, P = .49.

bTE = treatment ewe lambs; CE = control ewe lambs; CW = control wether lambs.

‘B = basal; E = epinephrine; ETAD = epinephrine plus theophyl- line plus adenosine deaminase; NE = norepinephrine.

dMeans by effector type: B = 207f; E = 236e; ETAD = 264d; NE = 213ef (means lacking a common superscript letter differ, P < .05).

Table 4. Fatty acid responsea in subcutaneous and perirenal adipose tissue slices from lambsb

TreatmentC

Site and effectord TE CE cw Subcutaneous - nEq2 h-I/100 mg of tissue - Efg 166 156 172

NE^ 294 196 236

Eg 106 165 101 ET AD^ 160 219 307 NEg 154 133 122

SEM 56 56 56

ETADe 495 341 617

Perirenal

aFatty acid response = [(fatty acid pool + fatty acid release),t;,,- latq - (fatty acid pool + fatty acid release)b,,,l].

Probabilities of greater F for effects in model: treatment, P = .62; site, P = .004; site x treatment, P = .43; effector, P < .001; effector x treatment, P = ,052; effector x site, P = .004; effector x site x treatment, P = .86.

CTE = treatment ewe lambs; CE = control ewe lambs; CW = control wether lambs.

dE = epinephrine; ETAD = epinephrine plus theophylline plus adenosine deaminase; NE = norepinephrine.

e,f,gFor effector x site interaction, means (not shown) lacking a common superscript letter differ ( P < .05).

For the fatty acid response, the interaction of effector x treatment approached significance ( P = .052). The fatty acid response to ETAD was greater ( P < .05) for CW than for TE and CE (461 vs 328 and 280 nEq.2 h-l/lOO mg of tissue, respectively); means for TE and CE did not differ ( P = .40). The greater response to ETAD for CW resulted from the combina- tion of nonsignificantly increased fatty acid release from adipocytes and nonsignificantly increased fatty acid pool remaining in the tissue. Whether the greater near-maximal fatty acid response for CW could contribute to decreased carcass fat deposition is unknown.

Coupled with the 45% greater (although nonsignifi- cant) fatty acid response t o ETAD for TE than CE, the greater fatty acid response observed for CW suggests that the presence of testosterone during fetal develop- ment of AT may affect maximal lipolytic capacity expressed at maturity. Testosterone and growth hor- mone increased the number of 0-adrenergic receptors in cultured rat (Xu et al., 1990) and ovine (Watt et al., 1991) adipocytes, respectively. The effects of prenatal exposure to testosterone in lambs on the number of &adrenergic receptors in sheep adipose tissue are unknown. The hypothesis that maximal lipolytic response could be determined during fetal sexual differentiation remains to be tested directly.

The ratio of the sum of the fatty acid pool and fatty acid re1ease:glycerol release was not different ( P = .92) among treatment groups but was greater ( P < .01) in perirenal than in subcutaneous AT (data not tabulated). Across treatments and tissue sites, incu- bation with E caused an increase ( P < . 0 5 ) in the ratio, indicating stimulation of partial triglyceride

1706 HANSEN ET AL.

Table 5. Incorporation of D-[U-14Clglucose and Na-[l-l4C1acetate into CO2, fatty acids, and glycerol in subcutaneous and perirenal adipose tissue slices from lambs

Treatmenta

Measurement TE CE cw SEM

Subcutaneous nmol.2 h-'/100 mg of tissue Glucose to C02 62 80 70 10 Glucose to fatty acids -2b 24' gbc 6 Glucose to glycerol sb 22c 12bc 6 Acetate to C02 32 34 36 8 Acetate to fatty acids 144 132 206 36 Acetate to glycerol 0 2 -4 2

Glucose to C02 44 42 42 4 Glucose to fatty acids 4 4 0 6 Glucose to glycerol 12 6 -2 6 Acetate to C02 14 18 6 6 Acetate to fatty acids 18 4 14 6 Acetate to glycerol 2 2 -4 2

Perirenal

aTE = treatment ewe lambs; CE = control ewe lambs; CW = control wether lambs. blcMeans within a row lacking a common superscript letter differ ( P < .05).

hydrolysis; this increase was particularly large in perirenal AT from TE and CW (11.3 and 14.5, respectively, vs 5.8 for CE; SEM = 3.3). Incubation with NE also tended to increase the ratio of total fatty acid to glycerol release, but incubation with ETAD had no effect; these findings indicate that lipolysis under conditions of near-maximal stimulation was essen- tially complete. Etherton et al. ( 197 7) observed an average ratio of fatty acid re1ease:glycerol release of less than 2:l for subcutaneous and perirenal AT, but they made no measurement of fatty acid pool, which could have increased the ratio.

Overall, sex or prenatal androgenization caused few differences of in vitro lipolytic capacity, although responses of TE generally were more similar to those of CW than CE. The small differences of in vitro lipolytic capacity may be consistent with the lack of differences among groups for subcutaneous or internal fat measurements in these lambs (Hansen et al., 1995). Lipolysis measurements also fail to explain the decreased concentrations of NEFA and glycerol in plasma after feeding in TE and CW, but not in CE (Hansen et al., 1995).

Lipogenesis. Oxidation of glucose to C 0 2 in sub- cutaneous AT was not different among treatment groups (Table 51, although mean glucose oxidation was 22.5% lower for TE than for CE. Rates of conversion of glucose t o fatty acid and glycerol were lowest for TE and highest for CE ( P < .OS>. Acetate metabolism was not affected by gender or prenatal testosterone treatment. Typically, acetate is the major carbon source for fatty acid synthesis in lambs (Hanson and Ballard, 1967; Ingle et al., 1972b), but glucose was incorporated into fatty acids a t a substan- tial rate in AT from CE. Fatty acid synthesis from glucose in sheep is limited by minimal activity of the citrate cleavage pathway; however, increasing glucose

availability increases the conversion of glucose to fatty acid (Ballard et al., 1972). The concentration of glucose in plasma was more sensitive to withholding of feed during the finishing phase for CE than for TE or CW (Hansen et al., 19951, suggesting a sex-related difference in regulation of glucose concentration in blood. In addition, Sidhu et al. (1973) reported that glyceride synthesis was greater in ewes than in rams; wethers were intermediate. The enhancement of capacities for utilization of glucose as a precursor of fatty acids and glycerol in CE could contribute to their propensity to deposit more lipid in AT than CW (DeHaan et al., 1987). Conversely, the decreased capacity for incorporation of glucose into lipids in TE may contribute t o decreased subcutaneous fat and leaner carcasses.

Decreased capacity for glucose utilization in sub- cutaneous AT of TE occurred despite the doubling of insulin concentration in plasma of TE compared with CE (Hansen et al., 1995). In ovine AT, insulin is known to increase uptake of glucose (Vernon and Finley, 1988), oxidation of glucose to C 0 2 and incorporation of glucose into lipid (Mears and Mendel, 19741, and fatty acid synthesis from acetate (Vernon and Finley, 1988). McCann and Reimers (1985) observed greater concentrations of insulin in serum of obese heifers than in serum of lean heifers and a diminished response of glucose in serum to the administration of exogenous insulin in the obese animals. Whether prolonged exposure of AT in TE to high concentrations of insulin (DeHaan et al., 1990; Hansen et al., 1995) might down-regulate the AT capacity for lipid synthesis from glucose in a similar manner is not known.

Average plasma glucose concentrations were simi- lar among treatments despite greatly increased insu- lin in TE and CW compared with CE (Hansen et al.,

PRENATALLY ANDROGENIZED LAMBS: METABOLISM

Table 6. Incorporation of alanine, glycerol, and propionate into CO2 and glucose by liver slices from lambs

1707

Treatmenta

Substrate and productsb TE CE cw SEM

nmol.2 h-l1100 mg of tissue ~ - [ ~ - l ~ ~ ] a l a n i n e c02 456 520 490 36 Glucose 116 104 88 14

[2-14~~g~ycerol c02 160 154 140 12 Glucose 284 206 282 60

Na-[l-l*Clpropionate c02 2,946‘ 3,664d 2,626‘ 174 Glucose 684e 1,018f 762ef 96 Glucose:C02 .22 .28 .30 .04

Na-[2-14C]propionate c02 46 66 24 22 Glucose 272 322 228 34 G1ucose:COn 6.84 7.58 10.43 2.24

aTE = treatment ewe lambs; CE = control ewe lambs; CW = control wether lambs. bConversion of 5 mM L-alanine, 5 mM glycerol, or 10 mM Na-propionate to C02 and glucose. ‘PdMeans within a row lacking a common superscript letter differ ( P < ,051. e,fMeans within a row lacking a common superscript letter differ ( P ,101.

1995). Coupled with the observations of decreased glucose utilization by subcutaneous AT in the present study, these results suggest that insulin insensitivity in AT may be a sexually differentiated response to prenatal exposure to androgens. The mechanism of this effect may be similar to the antagonistic action that growth hormone exerts on insulin-stimulated lipogenesis in AT of sheep (Vernon, 1982; Vernon and Finley, 1988) and cattle (Etherton et al., 1987). Although the growth hormone secretory profile was not altered by prenatal androgenization of sheep, Klindt et al. ( 1987) speculated that the increases of lean muscle growth associated with prenatal testoster- one exposure are related to a change in sensitivity to growth hormone at the tissue level.

Incorporation of glucose or acetate into COz, fatty acid, or glycerol in perirenal AT was not altered by sex or prenatal androgenization (Table 5 ) . With the exception of the oxidation of glucose t o C 0 2 , metabolic activity of perirenal AT was minimal for all treatment groups. Lipogenic activity in perirenal AT of sheep decreases with maturity (Ingle et al., 1972a); thus, the low lipogenic activity of perirenal AT could be attributable to the stage of maturity of the sheep. Alternatively, difficulties in handling the tissue be- cause of temperature sensitivity and fragility may have resulted in lower lipogenic rates; however, the substantial rates of glucose oxidation indicate that the tissue was metabolically viable. Because perirenal AT development in the fetal lamb precedes subcutaneous AT development by approximately 2 wk (Alexander, 19781, the timing of testosterone exposure may have differential effects on subsequent metabolic activities in subcutaneous vs perirenal AT.

Cellularity. Mean cell diameter and number of adipocytes per gram of wet tissue were not different among treatment groups. The mean cell diameters for subcutaneous AT were 98.1, 103.1, and 101.9 pm for TE, CE, and CW, respectively. The observations for the number of adipocytes per gram of tissue (2.11, 1.70, and 1.88 milliodg of tissue for TE, CE, and CW) are similar to those reported by Haugebak et al. ( 1974) for finishing market lambs fed for ad libitum intake. Mean cell diameters for perirenal AT were 93.5, 88.4, and 94.9 pm for TE, CE, and CW, respectively; the number of adipocytes was 2.67, 2.06, and 2.31 milliodg of tissue. The apparent insulin insensitivity of CE and CW evidently was not as- sociated with a larger adipocyte size.

Adipocyte hyperplasia is a determinant of the potential size of a fat depot, and adipocyte hyper- trophy seems to be the major factor affecting amount of fat deposition (Broad and Davies, 1980; Broad et al., 1980). Prenatal exposure to testosterone did not alter AT cellularity of the finished ewe lambs, but it is possible that less adipocyte hyperplasia occurred, which would decrease the potential size of AT depots.

Liver Metabolism

Gluconeogenesis. Rates of conversion of alanine, glycerol, and propionate to glucose and C02 are presented in Table 6. The conversions of alanine or glycerol to C 0 2 and glucose were not affected by sex or prenatal androgenization. Amino acid utilization by hepatic tissue depends on the availability to and extraction by the liver. The ADG was greater for TE than for CE, but feed efficiency was unchanged

1708 HANSEN

(Hansen et al., 1995); changes in amino acid supply and utilization likely were small and did not result in changes in hepatic capacity for metabolism of alanine. The substantial rates of oxidation of alanine to CO2 suggest that this amino acid could be an important energy source for hepatic tissue, although measure- ments of alanine oxidation would need to be made in the presence of other hepatic fuels to make this conclusion.

Glycerol oxidation to C02 and conversion to glucose increases during fasting in sheep (Bergman et al., 1968). In our study, average concentrations of glycerol in plasma did not differ among treatment groups, but the increased glycerol concentrations for TE and CW during the pre-feeding state (Hansen et al., 1995) were not accompanied by changes of capacity for the conversion of glycerol to glucose or C02.

The use of Na-[l-14C]propionate demonstrates the capacity of the liver to oxidize a considerable amount of propionate to CO2 in the absence of alternative fuels. Prenatal androgenization “masculinized” liver metabolism of propionate, in that CE had a greater ( P < .05) rate of Na-[l-14C]propionate oxidation to C02 and conversion to glucose than either TE or CW. No significant differences were observed in the conversion of Na-[2-l4C1propionate to C02 or glucose, although a pattern similar to that for Na-[l-14C]propionate me- tabolism was noted. Lyle et al. (1984) observed a similar increase of propionate metabolism in fasted steers and suggested that decreasing the supply of glucogenic substrates increased hepatic enzyme activi- ties; therefore, the efficiency of the utilization of glucogenic substrates increased. For lambs in our study, the gain to feed ratios were similar among treatment groups, but ADG was greater for TE than for CE (Hansen et al., 1995); therefore, TE consumed more feed to support the greater ADG. Because propionate production is related to DE intake (Her- bein et al., 1978), less propionate probably was available to CE; decreased propionate availability may have increased activity of hepatic enzymes involved in the metabolism of propionate.

An altered profile of growth hormone secretion, a modified sensitivity of tissues to growth hormone, or alterations in tissue sensitivity to steroid hormones have been proposed as mechanisms responsible for the growth performance advantage of prenatally an- drogenized ewe lambs (Klindt et al., 1987). Adminis- tration of exogenous growth hormone to lactating dairy cows increased the conversion of Na-[l- 14Clpropionate to glucose and C 0 2 (Pocius and Herbein, 19861, in contrast to our data for TE. The greater capacity for propionate metabolism in CE might be a physiological mechanism to compensate for the smaller livers of CE compared with TE or CW (DeHaan et al., 1987; Jenkins et al., 1988; Hansen et al., 1995); however, in the present study, liver weights as a percentage of carcass weight did not differ

ET AL.

significantly among treatments (3.9, 3.7, and 3.7% for TE, CE, and CW).

The greater amount of 14C02 derived from [l- 14C]propionate vs [2-14C]propionate indicates that the conversion of [l-14C]propionate t o 14C02 occurs during gluconeogenesis or in less than one complete turn of the Krebs cycle. During gluconeogenesis, this C02 is evolved either at the decarboxylation of oxaloacetate or pyruvate, or during the Krebs cycle at the decarboxylation of isocitrate or a-ketoglutarate. In contrast, [2-14C]propionate is not converted to 14C02 unless the propionate carbon remains in the Krebs cycle for more than one complete cycle. Calculated from means in Table 6, the ratio of [l- 14Clpropionate-derived C02:[2-14Clpropionate-derived C 0 2 was greatest for CW ( 1091, possibly indicating a more efficient shunting of propionate carbon to gluconeogenesis than either TE (64) or CE (56 1. The glucose:CO2 ratio supports this observation for [2- 14C]propionate utilization; the ratios for TE and CE (6.8 and 7.6) tended to be lower than the ratio for CW ( 10.4). Calculated from data reported by Faulkner and Pollock (19861, the glucose:CO2 ratio was 9.6 when isolated sheep hepatocytes were incubated with 5 mM [2-14C]propionate. However, the sex of the sheep was not reported, and the age ranged from 1 to 4 yr. The g1ucose:COz ratios for Na-[l-14Clpropionate were .22, .28, and .30 for TE, CE, and CW, respectively, compared with .37 calculated from Faulkner and Pollock (1986).

Our data are consistent with a more efficient partitioning of propionate to gluconeogenesis in CW, which may be partly responsible for the more efficient growth of lean tissue in male or castrate animals. The lack of a similar effect for TE suggests that the timing of androgen exposure was not appropriate to sexually differentiate hepatic propionate metabolism in TE.

P-Oxidutioa of Fnfty Acids. Rates of the initial cycle of @oxidation of Na-[l-14Clpalmitate by liver homogenates from lambs are presented in Table 7. Total oxidation represents the combined contributions of mitochondria and peroxisomes to 0-oxidation. Sex or prenatal androgenization did not affect the capacity for total or peroxisomal first cycle of 0-oxidation of palmitate. Short-term treatment of rats with the steroid precursor DHEA caused mitochondrial and peroxisomal proliferation (Bellei et al., 1992) and increased peroxisomal ,&oxidation (Mohan et al., 1990). Although ultrastructural changes in hepato- cytes from our lambs were not evaluated, the slightly larger livers of TE (Hansen et al., 1995) apparently did not correspond to enhanced rate or total liver capacity for the first cycle of &oxidation of palmitate by peroxisomes. The similarity of these measurements among groups suggests that &oxidation capacity is not a sexually differentiated response.

Little is known about the regulation and develop- ment of peroxisomal 0-oxidation activity in ruminant

PRENATALLY ANDROGENIZED LAMBS: METABOLISM 1709

Table 7. First cycle of @-oxidation of Na-[l-14C]palmitate by liver homogenates from lambs

Treatmenta

Measurementb TE CE CW SEM

nmol oxidized.2 h-l/100 mg of liver Total oxidation 1,660 1,700 1,500 180 Peroxisomal oxidation 800 740 700 80 Peroxisoma1:total .48 .44 .47 .02

mmol oxidized.h-l/liver Total hepatic Oxidation' 9.7 9.1 9.6 1.1 Total hepatic peroxisomal oxidation' 4.7 3.9 4.5 .5

aTE = treatment ewe lambs; CE = control ewe lambs; CW = control wether lambs. bAccumulation of Na-[l-14C]palmitate in C02 and acid soluble metabolites. 'Liver weights averaged 1.198, 1.046, and 1.177 kg for TE, CE, and CW (Hansen et al., 1995), respectively.

animals. Grum et al. (1994) reported that the peroxisomal capacity for first-cycle P-oxidation in liver from lactating dairy cows was approximately 50% of the total capacity for the initial cycle of @-oxidation of palmitate. The peroxisomal contribution to total first cycle of P-oxidation in sheep is consistent with the observations for dairy cows. Although total palmitate oxidation is greater in rats, peroxisomal oxidation represents only 15 to 30% of the total fatty acid oxidation (Kondrup and Lazarow, 1985; Veerkamp and van Moerkerk, 1986).

Mitochondrial Respiration. Measurements of mitochondrial respiration generally were not affected

by sex or prenatal androgenization (Table 8). State 3 respiration (characterized as maximal activity with unlimited substrate and ADP supply) and state 4 respiration (activity limited by ADP availability) were similar among treatment groups within each substrate. In contrast, Mohan and Cleary (1989) observed that state 3 mitochondrial respiration in rat liver was inhibited in vitro in the presence of succinate and either DHEA or testosterone. Mitochondria from livers of rats administered DHEA had increased state 3 and state 4 respiration rates in the presence of succinate or glutamate plus malate, both on a per gram of liver and per liver basis (Mohan et al., 1990).

Table 8. Respiratory activity in mitochondria (mito) of liver from lambs

Treatmenta

Measurementb TE CE cw SEM

State 3 respiration nmol Oz,min-'/rng of mito protein Pyruvate + malate 9.84 10.46 10.23 .73 Succinate 19.62 21.88 21.36 1.77 Glutamate + malate 13.32 14.35 13.04 1.07

State 4 respiration nmol OZ.min-'/mg of mito protein Pyruvate + malate Succinate Glutamate + malate

Respiratory control ratio Pyruvate + malate Succinate Glutamate + malate

ADP:O Pyruvate + malate Succinate Glutamate + malate

2.56 5.21 2.66

3.87 3.78C 5.03

2.63 5.33 2.90

3.83 4.08d 4.91

pmo1:patom 2.82 2.91 1.89 1.93 2.72 2.70

2.80 5.39 2.87

3.82

4.63 3.99Cd

.22

.42

.22

.26

. l 1

.l8

2.82 1.91 2.74

.07

.04

.05 mgig of liver

Liver mitochondrial protein 25.32 25.45 23.95 2.09

aTE = treatment ewe lambs; CE = control ewe lambs; CW = control wether lambs. bMitochondrial respiratory measurement in the presence of 5 m M pyruvate plus 5 mM malate, 10 mM succinate, or 5 mM glutamate plus

C,dMeans within a row lacking a common superscript letter differ ( P < . lo). 5 mM malate.

1710 HANSEN ET AL.

Table 9. Composition of liver from lambs (wet weight basis)

Treatmenta

TE CE cw SEM

Lipid Triglyceride Glycogen Protein

1.67 .32

2.74 20.1be

1.69 .28

2.82 21.2b

1.57 .24

2.86 18.8'

.06

.03

.28

.7

TE = treatment ewe lambs; CE = control ewe lambs; CW = control wether lambs. bvcMeans within a row lacking common superscript letter differ ( P < .05).

Bellei et al. (1992) noted that DHEA administration to rats caused mitochondrial proliferation and transi- tion from expanded to condensed configuration; these configurations are associated with low and high rates of respiration, respectively. Although mitochondrial proliferation was observed, long-term treatment with DHEA caused degeneration of mitochondria; therefore, the proliferative activity may have been a compensa- tory mechanism to sustain respiration at a certain rate (Bellei et al., 1992). Brown et al. (1988) reported that testosterone administration to female goats tended to increase and masculinize state 4 hepatic mitochondrial respiration in the presence of succinate.

The respiratory control ratios ([respiration rate in presence of substrate and ADPHrate after consump- tion of ADP]; i.e., [state 3 respiratory ratel/[state 4 respiratory ratel) and ADP:O ratios were not affected by prenatal testosterone exposure, with the exception of the tendency for a greater ( P < . l o ) RCR in the presence of succinate for mitochondria from CE than for those from TE. High RCR values are indicative of an abundant energy-conserving capacity, whereas lower values may indicate a loss of organelle integrity. The ratio of ADP:O consumption represents the coupling of respiration with oxidative phosphorylation. Bellei et al. (1992) and Brown et al. (1988) reported RCR and ADP:O values similar to ours, with no changes due to DHEA treatment of rats or testoster- one treatment of goats, respectively.

The content of mitochondrial protein per gram of liver was not affected by prenatal exposure t o testosterone (Table 8). In contrast, mitochondrial protein increased in hepatocytes from rats treated with DHEA (Mohan et al., 1990; Bellei et al., 1992). Although mitochondrial proliferation and altered mitochondrial respiration have been observed in rats administered DHEA (Mohan et al., 1990; Bellei et al., 1992) and goats treated with testosterone (Brown et al., 1988), these measurements were made while the steroids were being administered and no measure- ments were made after the withdrawal to evaluate the duration of the modifications. If fetal mitochondrial activity or ultrastructure were altered by prenatal androgenization, there may have been reversion to the normal state after removing the testosterone implants

from the pregnant ewes. The similarity of mitochon- drial measurements between CW and CE, however, suggests that little sexual differentiation occurs in mitochondrial function.

Composition. Contents of lipid, triglyceride, and glycogen in liver were not different among treatment groups (Table 9). However, the content of protein in livers of CW was greater ( P c .05) than that in livers of CE; TE were intermediate for this measurement. Because livers were enlarged by prenatal androgeniza- tion (DeHaan et al., 1987; Jenkins et al., 1988; Hansen et al., 19951, and mitochondrial protein was not affected by sex or prenatal androgenization (Table 81, there may have been testosterone-dependent hypertrophy of other hepatic organelles or increased cytosolic proteins that contributed to the increased total protein content observed in this experiment. In contrast, increased liver weights of DHEA-treated rats were attributed t o diffuse hypertrophy of hepatocytes, because contents of DM and protein were unaffected by treatment and glycogen content decreased with DHEA treatment (Bellei et al., 1992).

Conclusions

Prenatal androgenization improved ADG but did not affect feed efficiency or carcass characteristics (Hansen et al., 1995); these results were accompanied by minimal alterations of metabolic activities in AT and liver. Our data suggest that maximal masculini- zation was not achieved for TE in the present study, in contrast to data reported earlier (DeHaan et al., 1987). The lesser degree of masculinization may be caused by the larger range of gestational ages during which ewes were implanted with testosterone (40 to 70 d ) compared with an earlier study (40 to 60 d; DeHaan et al., 1987). Changes also may have been more pronounced if all lambs had been fed for a constant length of time or to a common backfat thickness.

Few differences in lipolytic or lipogenic capacities were observed among treatments for either AT site. Lipolytic and lipogenic rates might have been different earlier during the finishing phase or during the growth phase, and differences may have occurred if measured at constant body weight or age of lambs.

PRENATALLY ANDROGENIZED LAMBS: METABOLISM 1711

Ingle et al. (1972a) and Smith et al. (1987) reported that lipogenic rates and activities of lipogenic enzymes decreased with age in growing and finishing lambs.

The lack of plasma glucose response in TE or CW to pre- and post-feeding conditions (Hansen et al., 19951, combined with decreased incorporation of glucose into fatty acids and glycerol in subcutaneous AT of TE, suggest an insulin insensitivity attributable to prenatal androgen exposure. Whether the changes in capacity for glucose production from propionate in the liver, decreased utilization of glucose for synthesis of lipid precursors in subcutaneous AT, alterations in maximal lipolytic capacity, or altered regulation of lipolysis are factors that contribute to decreased carcass fat deposition in TE or CW need t o be studied in experimental situations where carcass lipid is decreased by prenatal androgen exposure. Studies including intact males may be useful to further delineate the effects of fetal exposure to testosterone on postnatal tissue metabolism. Furthermore, in vitro metabolic activities of subcutaneous and perirenal AT and hepatic tissue represent the metabolic potential in optimal conditions with excess substrate availability. The in vivo rates of glucose and fatty acid metabolism are tightly regulated by numerous homeostatic and homeorhetic mechanisms that merit further considera- tions in animals prenatally exposed to testosterone.

Implications

Minimal changes of metabolic activities were de- tected in adipose tissue and liver of prenatally androgenized ewe lambs, which was consistent with the lesser improvements in ADG and feed efficiency and the lack of decrease in carcass fat content compared with earlier studies. Alterations for glucose usage for triglyceride synthesis in subcutaneous adi- pose tissue, alterations in lipolysis, insulin insensitiv- ity or resistance, and changes in liver metabolism of propionate are factors that potentially may affect performance and carcass composition in prenatally androgenized lambs.

Literature Cited

Alexander, G. 1978. Quantitative development of adipose tissue in foetal sheep. Aust. J. Biol. Sci. 31:489.

Aprille, J. R., and G. K. Asimakis. 1980. Postnatal development of rat liver mitochondria: State 3 respiration, adenine nucleotide translocase activity, and the net accumulation of adenine nucleotides. Arch. Biochem. Biophys. 201:564.

Baldner, G. L., R. E. Flatt, R. N. Shaw, and D. C. Beitz. 1985. Fatty acid biosynthesis in liver and adipose tissue from dogs. Comp. Biochem. Physiol. 82B:153.

Ballard, F. J., 0. H. Filsell, and I. G. Jarrett. 1972. Effects of carbohydrate availability on lipogenesis in sheep. Biochem. J. 226:193.

Bellei, M., D. Battelli, C. Fornieri, G. Mori, U. Muscatello, H. Lardy, and V. Bobyleva. 1992. Changes in liver structure and function

after short-term and long-term treatment of rats with de- hydroepiandrosterone. J . Nutr. 122:967.

Bergman, E. N., D. J. Starr, and S. S. Reulein, Jr . 1968. Glycerol metabolism and gluconeogenesis in the normal and hypoglycemic ketotic sheep. Am. J . Physiol. 215:874.

Bligh, E. G., and W. J. Dyer. 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37:911.

Broad, T. E., and A. S. Davies. 1980. Pre- and postnatal study of the carcass growth of sheep. 1. Growth of dissectable fat and its components. Anim. Prod. 31:63.

Broad, T. E., A. S. Davies, and G. Y. Tan. 1980. Pre- and postnatal study of the carcass growth of sheep. 2. The cellular growth of adipose tissue. Anim. Prod. 31:73.

Brown, D. R., D. C. Beitz, P. G. Detilleux, and P. D. Ekema. 1988. Effects of testosterone on respiratory metabolism and ultras- tructure of caprine hepatic mitochondria. Biochem. Arch. 4:189.

Carpenter, Z. L., G. T. King, F. A. Orts, and N. L. Cunningham. 1964. Factors influencing retail carcass value of lambs. J. Anim. Sci. 23:741.

Cremin, J. D., Jr . , J . K. Drackley, D. E. Grum, L. R. Hansen, and G. C. Fahey, Jr. 1994. Effects of reduced phenolic acids on metabo- lism of propionate and palmitate in bovine liver tissue in vitro. J. Dairy Sci. 77:3608.

DeHaan, K. C., L. L. Berger, P. J. Bechtel, D. J. Kesler, F. K. McKeith, and D. L. Thomas. 1990. Effect of prenatal testoster- one treatment on nitrogen utilization and endocrine status of ewe lambs. J . Anim. Sci. 68:4100.

DeHaan, K. C., L. L. Berger, D. J. Kesler, F. K. McKeith, D. L. Thomas, and T. G. Nash. 1987. Effect of prenatal androgeniza- tion on lamb performance, carcass composition and reproduc- tive function. J. h i m . Sci. 65:1465.

Duquette, P. F., C. G. Scanes, and L. A. Muir. 1984. Effects of ovine growth hormone and other anterior pituitary hormones on lipolysis of rat and ovine adipose tissue in vitro. J. Anim. Sci. 58:1191.

Eggstein, M., and E. Kuhlmann. 1974. Triglycerides and glycerol determination after alkaline hydrolysis. In: H. U. Bergmeyer (Ed.) Methods of Enzymatic Analysis (2nd Ed.). Vol. 3. pp 1825-1835. Academic Press, New York.

Estabrook, R. W. 1967. Mitochondrial respiratory control and the polarographic measurement of ADP:O ratios. In: R. W. Es- tabrook and M. E. Pullman (Ed. ) Methods in Enzymology. Vol. 5. p 41. Academic Press, New York.

Etherton, T. D., D. E. Bauman, and J . R. Romans. 1977. Lipolysis in subcutaneous and perirenal adipose tissue from sheep and dairy steers. J. Anim. Sci. 44:llOO.

Etherton, T. D., C. M. Evock, and R. S. Kensinger. 1987. Native and recombinant bovine growth hormone antagonize insulin action in cultured bovine adipose tissue. Endocrinology 121:699.

Faulhaber, J . D., H. U. Klor, and H. Diyschuneit. 1972. New device for preparing thin slices of adipose tissue for metabolic studies in vitro. J. Lipid Res. 13:816.

Faulkner, A., and H. T. Pollock. 1986. Propionate metabolism and its regulation by fatty acids in ovine hepatocytes. Comp. Bio- chem. Physiol. 84B:559.

Folch, J., M. Lees, and G.H.S. Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:497.

Foster, L. B., and R. T. Dunn. 1973. Stable reagents for determina- tion of serum triglycerides by a colorimetric Hantzsch conden- sation method. Clin. Chem. 19:338.

Grum, D. E., L. R. Hansen, and J . K. Drackley. 1994. Peroxisomal (3- oxidation of fatty acids in bovine and rat liver. Comp. Biochem. Physiol. 109B:281.

Hansen, L. R., J . K. Drackley, L. L. Berger, and D. E. Grum. 1995. Prenatal androgenization of lambs: I. Alterations of growth, carcass characteristics, and metabolites in blood. J . h i m . Sci. 73:1694.

Hanson, R. W., and F. J. Ballard. 1967. The relative significance of acetate and glucose as precursors for lipid synthesis in liver and adipose tissue from ruminants. Biochem. J. 105:529.

1712 HANSEP

Haugebak, C. D., H. B. Hedrick, and J. M. Asplund. 1974. Adipose tissue accumulation and cellularity in growing and fattening lambs. J. Anim. Sci. 39:1016.

Herbein, J. H., R. W. Van Maanen, A. D. McGilliard, and J. W. Young. 1978. Rumen propionate and blood glucose kinetics in growing cattle fed isoenergetic diets. J. Nutr. 108:994.

Iliou, J. P,, and Y. Demame. 1987. Evolution of the sensitivity of isolated adipocytes of ewes to the lipolytic effects of different stimuli during pregnancy and lactation. Int. J. Biochem. 19: 253.

Ingle, D. L., D. E. Bauman, and U. S. Garrigus. 1972a. Lipogenesis in the ruminant: In vitro site of fatty acid synthesis in sheep. J. Nutr. 102:609.

Ingle, D. L., D. E. Bauman, and U. S. Garrigus. 1972b. Lipogenesis in the ruminant: In vivo site of fatty acid synthesis in sheep. J. Nutr. 102:617.

Jenkins, T. G., J . J. Ford, and J. Klindt. 1988. Postweaning growth, feed efficiency and chemical composition of sheep as affected by prenatal and postnatal testosterone. J. h i m . Sci. 66:1179.

Kelley, T. F. 1965. Improved method for microtitration of fatty acids. Anal. Chem. 37:1078.

Klindt, J., T. G. Jenkins, and J. J. Ford. 1987. Prenatal androgen exposure and growth and secretion of growth hormone and prolactin in ewes postweaning. Proc. Soc. Exp. Biol. Med. 185: 201.

Kondrup, J., and P. B. Lazarow. 1985. Flux of palmitate through the peroxisomal and mitochondrial &oxidation systems in isolated rat hepatocytes. Biochim. Biophys. Acta 835147.

Leng, R. A. 1970. Glucose synthesis in ruminants. Adv. Vet. Sci. Comp. Med. 14:209.

Lo, S., C. Russell, and A. W. Taylor. 1970. Determination of glycogen in small tissue samples. J. Appl. Physiol. 28:234.

Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265.

Lyle, R. R., K. D. Birkmeyer, and J. W. Young. 1984. In vitro hepatic gluconeogenesis and ketogenesis as affected by prolonged ketonemia-glucosuria and fasting in steers. J. Dairy Sci. 67: 2283.

McCann, J. P., and T. J . Reimers. 1985. Glucose response to ex- ogenous insulin and kinetics of insulin metabolism in obese and lean heifers. J . h i m . Sci. 61:612.

McNamara, J. P., and J. K. Hillers. 1986. Adaptations in lipid metabolism of bovine adipose tissue in lactogenesis and lacta- tion. J. Lipid Res. 27:150.

Mears, G. J., and V. E. Mendel. 1974. Correlation of food intake in lambs with adipocyte glucose metabolism and NEFA. J. Phys- iol. 240:625.

J ET AL.

Mersmann, H. J., J. Goodman, J. M. Houk, and S. Anderson. 1972. Studies on the biochemistry of mitochondria and cell morphol- ogy in the neonatal swine hepatocyte. J. Cell Biol. 53:335.

Mills, S. E., L. E. Armentano, R. W. Russell, and J. W. Young. 1981. Rapid and specific isolation of radioactive glucose from biologi- cal samples. J. Dairy Sci. 64:1719.

Mohan, P. F., and M. P. Cleary. 1989. Dehydroepiandrosterone and related steroids inhibit mitochondrial respiration in vitro. Int. J. Biochem. 21:1103.

Mohan, P. F., J. S. Ihnen, B. E. Levin, and M. P. Cleary. 1990. Effects of dehydroepiandrosterone treatment in rats with diet- induced obesity. J. Nutr. 120:1103.

Pocius, P. A., and J. H. Herbein. 1986. Effects of in vivo administra- tion of growth hormone on milk production and in vitro hepatic metabolism in dairy cattle. J. Dairy Sci. 69:713.

SAS. 1985. SAS User’s Guide: Statistics (Version 5 Ed.). SAS Inst. Inc., Cary, NC.

Shelton, M., and Z. L. Carpenter. 1972. Influence of sex, stilbestrol treatment and slaughter weight on performance and carcass traits of slaughter lambs. J. h i m . Sci. 34203.

Sidhu, K. S., R. S. Emery, A. F. Parr, and R. A. Merkel. 1973. Glyceride synthesis in ovine adipose homogenates. J. h i m . Sci. 37971 (Abstr.).

Sjostrom, L., P. Bjorntorp, and J. Vrana. 1971. Microscopic fat cell size measurements on frozen-cut adipose tissue in comparison with automatic determinations of osmium-fmed fat cells. J . Lipid Res. 12:521.

Smith, S. B., T. Jenkins, and R. L. Prior. 1987. Carcass composition and adipose tissue metabolism in growing sheep. J. Anim. Sci. 65:1525.

Terrettaz, J., and B. Jeanrenaud. 1990. Contribution of glycerol and alanine to basal hepatic glucose production in the genetically obese (falfa) rat. Biochem. J. 270:803.

Veerkamp, J. H., and H.T.B. van Moerkerk. 1986. Peroxisomal fatty acid oxidation in rat and human tissues. Effect of nutritional state, clofibrate treatment and postnatal development in the rat. Biochim. Biophys. Acta 875:301.

Vernon, R. G . 1982. Effects of growth hormone on fatty acid synthe- sis in sheep adipose tissue. Int. J. Biochem. 14:225.

Vernon, R. G., and E. Finley. 1988. Roles of insulin and growth hormone in the adaptation of fatty acid synthesis in white adipose tissue during the lactation cycle in sheep. Biochem. J. 256:873.

Watt, P. W., E. Finley, S. Cork, R. A. Clegg, and R. G. Vernon. 1991. Chronic control of the p- and a2-adrenergic systems of sheep adipose tissue by growth hormone and insulin. Biochem. J. 273: 39.

Xu, X., G. De Pergola, and P. Bjorntorp. 1990. The effects of andro- gens on the regulation of lipolysis in adipose precursor cells. Endocrinology 126:1229.