Regression of brown adipose tissue ... - Widener University

Regression of brown adipose tissue mitochondrial function and structure in neonatal goats ITZICK VATNICK, ROBERT S. TYZBIR, JAMES Department of Human Nutrition and Foods and Animal University of Vermont, Burlington, Vermont 05405

G. WELCH, AND ALICE P. HOOPER Science, College of Agriculture and Life Sciences,

VATNICK, ITZICK, ROBERT S. TYZBIR, JAMES G. WELCH, AND ALICE P. HOOPER. Regression of brown adipose tissue mitochondria function and structure in neonatal goats. Am. J.

ante as well as its cellular structure. These changes occur at various rates in different species (3, 4, 8, 20). The rapid morphological changes in the perirenal BAT of the Physiol. 252 (Endocrinol. Metab. 15): E391-E395,1987.-Rapneonatal goat (20) make the kid an ideal model for idly regressingperirenal brown adiposetissue (BAT) of neo- studying the correlation of mitochondrial function with natal goatswas studied to correlate changesin mitochondrial structure in a rapidly regressing tissue. metabolism and thermogenic capacity with changesin mitoIn this study we first characterized the morphological chondrial structure. The cu-glycerophosphate shuttle activity of perirenal BAT mitochondria declined 60% from birth to 6 days and biochemical properties of perirenal BAT mitochonof age. Oxygen consumption and thermogenic capacity mea- dria in neonatal kids and then examined the relationship biochemical activity, and thersuredby ion conductancepeaked at birth and declined to low among the morphology, levelsat 6 days. Sampleelectron micrographsof perirenal BAT mogenic capacity of this tissue in neonatal kids. showedintact electron-densemitochondria with many cristae and little matrix area at 2 days. However, by 6 days the mitochondria were very relaxed with large matrix area, few cristae, and observabledegradation. These resultsindicate that the morphological changes exhibited by rapidly regressing goat’sperirenal BAT in the 1st wk postpartum are accompanied by dramatic alterations in BAT mitochondrial function. thermogenesis;oxygen consumption; electron microscopy

(1) cited Gesner who described brown adipose tissue (BAT) more than four hundred years ago, but only in the last twenty-five years has the key role of brown adipose tissue in adaptive thermogenesis been recognized (for reviews see Refs. 6, 13, 16, 18). BAT is capable of thermogenesis due to a unique 32,000 mol wt protein in its mitochondrial membrane. This protein allows dissipation of the electron gradient in the mitochondria without comcomitant stoichiometric production of ATP. The uncoupling of oxidative phosphorylation from oxidation in BAT allows the energy to dissipate in the form of heat instead of being used to synthesize ATP as in conventional coupled mitochondria (11). All mammalian species examined possess BAT (17). In species in which BAT is most active during the neonatal period, this tissue may ensure the survival of neonates (18). All neonatal ruminants studied have easily identifiable BAT (2, 3, 5, 14, 20) and BAT biochemical properties have been characterized in prenatal and neonatal sheep (7, 10). Recent work with arctic ruminants (muskoxen and reindeer) showed that BAT thermogenesis plays a key role in the survival of the newborn of these species as well (2, 14). In ruminants, BAT undergoes morphological changes prenatally from late pregnancy through the first postnatal weeks. The tissue changes its macroscopic appearAFZELIUS

METHODS

Animals. Thirty French Alpine and Toggenburg kids aged from 6 h to 7 days and weighing from 2.53 to 5.00 kg were used. Kids born in early spring in an open barn were separated from their mothers within 2 days of birth and kept in a stall at ambient temperature (-25°C). Kids were fed 120 ml colostrum 3-4 times a day for the first 2 days and the same amount of whole goat or cow milk thereafter. Kids were killed by an intravenous injection of pentobarbital sodium within the first 24 h (birth), between 24 and 72 h (a-day group), between 72 and 120 h (4-day group), or between 120 and 168 h (6day group). Tissue sampling. The kidney and the fat pad adhering directly to the kidney were excised. The fat pad was removed from the kidney, washed in isotonic ice-cold buffered sucrose (0.25 M sucrose and 5 mM TES-HCl, pH 7.2), blotted, and weighed. A tissue sample weighing -2 g was homogenized in 10 ml ice-cold buffered sucrose by using a polytron (Brinkmann Instruments, Westbury, NY). Mitochondria isolution. Mitochondria were isolated from the homogenate by differential centrifugation (23). Briefly, a small tissue sample (-2 g) was homogenized in ice-cold buffered sucrose and spun slowly for 5 min (410 g) to precipitate the nuclear fraction. The supernatant then was spun 10 min at 8,880 g. The mitochondrial pellet obtained was then washed once with ice-cold buffered sucrose. The freshly isolated mitochondria were resuspended in a volume of cold sucrose buffer equal to the original weight of the tissue sample. An aliquot was taken and diluted to a final buffer volume of ten times the original sample weight for determination of the aglycerophosphate shuttle and ion conductance (21, 23). An aliquot without further dilution was used for the

0193-1849/87 $1.50 Copyright0 1987 the AmericanPhysiological Society

E391

E392

REGRESSION

OF BAT MITOCHONDRIA

usin .g a Yellow measure ment of oxygen consumption Springs biologica 1 oxyge n monitor m ode1 53 (Yellow Springs Instruments, Yellow Springs, OH). Mitochondrial protein was determined by the Bio-Rad method (Bio-Rad Laboratories, Richmond, CA) with bovine serum albumin used as the standard. Brown adipose tissue mitochondrial ion conductance.

BAT ion conductance was measured spectrophotometrically at 420 nm (21). The mitochondria were incubated in a medium of 100 mM KCl, 5 mM TES-HCl, 5 PM rotenone, and 0.5 PM valinomycin at pH 7.2. The decrease in absorbance within the 1st min after the addition of 0.1 ml mitochondria resuspension to 1.9 ml swelling solution was recorded (OD U/min). A second cuvette containing 0.1 ml mitochondria, 1.8 ml swelling solution, and 0.1 ml GDP (0.1 mM) was also incubated for 1 min, and the decrease in absorbance was measured. The differen .ce in op tical density units per minute between cu. vette 1 and 2 was calculated as the inhibition of swelling due to GDP. Specific activity of ion conductance was expressed as optical density units per minute per milligram mitochondrial protein. Brown adipose tissue cu-glycerophosphate shuttle. The glycerophosphate shuttle activity depends on both the cytosolic as well as the mitochondrial glycerol-3-phosphate dehydrogenases. Because our study was designed to measure mitochondrial changes, we reconstructed the shuttle in such a manner that the measurement of the glycerophosphate shuttle reflected the activity of the mitochondrial dehydrogenase. The shuttle was measured spectrophotometrically at 340 nm by following the conversion- of NADH to NAD (22). To reconstitute the (resuspension 1:lO wt/vol) shuttle 0.1 ml mitochondri was added to a final volume f 3 ml of mediu m containing 150 mM KCl, 10 mM tetraethylammonium-HCl, 5 mM MgC12-Hz06, 1 mM EDTA, 10 mM K2HP04, 1 pg rotenone/ml, 0.15 mM NADH, 0.1% bovine serum albumin, 10 mM DL-2-glycerophosphate, 1 mM ADP, and 5 U cyglycerophosphate dehydrogenase, pH 7.4. Specific shuttle activity measured as complete system minus shuttle components was expressed as nanomoles of NADH converted to NAD per minute per milligram mitochondrial protein. Brown adipose tissue mitochondrial oxygen consumption. Oxygen consumption was measured with a Yellow

Springs biological oxygen monitor model 53 (23). An aliquot of 0.1 ml of mitochondria resuspension (1:l wt/ vol) was incubated at 30°C in 2.9 ml of medium similar to the medium used for the shuttle with the exclusion of the NADH, rotenone, and shuttle components. Endogenous respiration was stimulated by the addition of 3.3 mM DL-carnitine and 0.9 mM ATP. When endogenous respiration reached a steady state, 3.3 mM succinate was added and substrate-stimulated respiration was recorded. Oxygen consumption was expressed as nanomoles of oxygen consumed per minute per milligram mitochondrial protein. Electron microscopy tissue sampling. Small tissue samples (-1 mm cube) from each perirenal BAT pad were taken immediately after the tissue was excised. The samples were rinsed in buffer and stored in 4.5% glutar-

IN NEONATAL

GOATS

aldehyde at pH 7.2 in phosphate buffer for not more than 8 wk at 4°C. Samples were postfixed by placing them in 1% osmium tetroxide for 2 h. After postfixing, the tissue was rinsed twice in phosphate buffer (15 min/ rinse). The samples were then dehydrated in ascending concentrations of ethanol (gold grade). Samples were placed in 25, 50, 70, 95, and 100% (twice) ethanol containing 1.3 mM CaC12 for 15 min in each concentration. Tissue fragments were embedded in a spurrite resin by placing them in ascending concentrations of resin. Samples were placed in a 3:1, l:l, and 1:3 ethanol:spurrite mixtures for l-2 h in each concentration. Finally samples were placed in 100% spurrite for 2 h, moved to new spurrite, and left to bake in embedding molds at 60°C overnight. The next day samples were sliced with an ultramicrotome (Reichert OM-U2) to silver sections and collected on 200-mesh copper grids. Selected sections were stained with uranyl acetate and lead citrate and examined at 60 kB under a transmission electron microscope (Philips 201). Statistical analysis. Statistical comparisons among groups were made by one-way analysis of variance adjusted for unequal group size and a nonparametric leastsignificance difference method for all possible comparisons (19). RESULTS

Somatometric parameters. Table 1 summarizes the somatometric measurements of the kids used. Body weights of kids was highly variable due to differences in litter size (i.e., singles vs. twins or triplets) and other factors such as maternal nutrition, and gestational length. It is not surprising to find no significant differences in body weight among the groups. There was a significant difference (P < 0.05) in liver weights between kids at birth and 2 days of age vs. older kids. This difference may indicate that liver weight increases during the 1st .wk of life faster than total body weight. Kidney and perirenal BAT weight were not significantly different during the first postnatal week in the goat kid (Table 1). a-Glycerophosphate shuttle. The a-glycerophosphate shuttle specific activity expressed as nmoles of NADH converted to NAD per minute per milligram mitochondrial protein was high at birth, declined (P c 0.05) to less than half of the original level by 2 days, and stayed at this low level until 6 days after birth (Table 2).

1. Somatometric parameters of neonatal goats during 1st wk postpartum TABLE

Parameter

Birth

Days Postpartum 2 days

4 days

6 days

Body wt, kg 3.4kO.3” 3.6t0.3” 4.lt0.2* 3.8tl.O” Liver wt, g 83.4*6.3* 90.2t7.0* 106.4+8.4t lll.lA5.37 Kidney wt, g 11.1t1.2” 11.4AIO.9” 13.3kO.3” 11.8t0.8* Perirenal BAT, g 3.2*0.2*? 2.9t0.2” 3.5*0.1t 3.2*0.3*t BAT, mg mitochon6.5k1.2” 3.6+0.3? 4.1+0.2*f 4.3+0.6*-t drial protein recovered/g Values are means t SE; n = 4-6 at birth, 4-8 at 2 days, 6-10 at 4 days, and 5 at 6 days postpartum. BAT, brown adipose tissue. Means across a row with different superscript differ significantly (P < 0.05).

REGRESSION

OF

BAT

MITOCHONDRIA

Days Birth

GP shuttle Specific activity, NADH oxidized min-l . mg mitochondrial protein-’ Mean total activity Ion conductance Specific activity, U. min-l . mg mitochondrial protein-’ Mean total activity recovered Oxygen consumption Specific activity, nmol Oz. min-l mg mitochondrial protein-’ Mean total activity recovered

47t2*

977.6 92t14*

1,913.6

4 days

19+4f

11+4t

198.4 89t9*

929.2

157.9 33+3t

473.6

GOATS

E393

165.1 16k5t

DISCUSSION

Postpartum

2 days

NEONATAL

ity was high. At 2 days the mitochondria were dense, cristae rich, and had a very small matrix space. A representative electron micrograph of BAT mitochondria of a 6-day-old kid, when biochemical and thermogenic capacity had decreased to lowest levels, is shown in the bottom panel of Fig. 1. At 6 days the mitochondria were relaxed, cristae poor, and had a large matrix. All samples were treated in the same manner and preparation for electron microscopy was done on the same day for each of the different age groups to minimize artifacts due to tissue preparation.

2. Brown adipose tissue mitochondrial function in neonatal goats during 1st wk postpartum TABLE

BAT

IN

6 days

12t5t

220.2

55*1*

107+9t

70+8$

28k4§

1,144

1,117

1,016

385

l

Values are mean t SE; n = 4-6 at birth, 4-8 at 2 days, 6-10 at 4 days, and 5 at 6 days postpartum. Mean total activity was calculated using the mean values for mitochondrial protein, tissue weight, and specific activity. BAT, brown adipose tissue. Means across a row with different superscript differ significantly (P c 0.05).

Glycerophosphate shuttle specific activity at 2, 4, and 6 days was not statistically different. The postnatal decline in mean total recovered shuttle activity, calculated as mM NADH oxidized per minute per total tissue, was similar to the decline in specific glycerophosphate shuttle activity (Table 2). Ion conductance. Specific activity of ion conductance measured in neonatal goat kids and expressed as units per minute per milligram of mitochondrial protein was very high at birth and 2 days but declined dramatically (P < 0.05) to low levels at 4 and 6 days after birth (Table 2). At 6 days after birth, ion conductance declined to half of the value measured at 4 days; however, this difference was not statistically significant. The calculated total recovered ion conductance expressed as units per minute per total tissue was also very high at birth but declined to half this value at 2 days. This activity continued to decline almost at the same rate to 6 days after birth (Table 2). Oxygen consumption. Oxygen consumption (Table 2) expressed as nmoles oxygen consumed per minute per milligram mitochondrial protein increased twofold from birth to 2 days (P < 0.05) then declined progressively to one-half the value at birth by 6 days postpartum, respectively. Oxygen consumption was different (P < 0.05) among all age groups (Table 2). Total recovered oxygen consumption calculated as nanomoles O2 per minute per total tissue was high at birth through 4 days and then declined to low levels at 6 days after birth (Table 2). Histological observations. Electron micrographs of perirenal BAT from representative kids from each group were examined. In the top panel of Fig. 1 is a representative electron micrograph of BAT mitochondria from a Z-day-old kid, when biochemical and thermogenic capac-

There are several methods used to measure thermogenie capacity: oxygen consumption, ion conductance, GDP binding, and more recently quantification of thermogenin by enzyme-linked immunosorbent assay or radioimmunoassay. Nedergaard and Cannon (12) compared ion conductance with these other methods and concluded that basal specific oxygen consumption, GDP binding, and the amount of thermogenin measured by an immunological test yielded estimates of thermogenic capacity of BAT that were in excellent agreement with each other and with the estimates obtained from measuring ion conductance (12). The progressive decline in both specific and total BAT ion conductance from birth to 6 days after birth indicates that kids rely heavily on BAT thermogenesis to keep warm immediately after birth, but the tissue progressively loses thermogenic capacity during the first postnatal week. The specific oxygen consumption peaked at 2 days and declined thereafter. In contrast, total mean recovered mitochondrial oxygen consumption stayed high during the first 4 days of life and declined to low levels by 6 days (Table 2). We measured succinate-stimulated oxygen consumption as an indicator of the capacity of the BAT mitochondria to oxidize exogenous substrate. Although succinate-stimulated oxygen consumption may not be closely correlated with BAT thermogenic capacity the decline in total and specific oxygen consumption from high levels at birth to low levels at 6 days postpartum indicates that the tissue not only loses its thermogenie capacity but also other mitochondrial activity during the 1st wk postpartum. Oxygen consumption in highly thermogenic mitochondria may be regulated by substrate availability, because they are loosely coupled and the phosphate potential cannot exert respiratory control (5). Our findings cannot be compared with those of Nedergaard and Cannon (l2), because they correlated basal specific oxygen consumption to ion conductance and we measured succinate-induced respiration. In BAT the a-glycerophosphate shuttle is the regulator of glycolysis and the balance between lipolysis and lipogenesis (9). The cY-glycerophosphate ensures continuous glycolytic activity by maintaining adequate concentration of NAD in the cytoplasm (22). Glycolysis is important in thermogenesis, because it provides part of the ATP required for fatty acid activation when uncoupled respiration results in low production of this important intermediate (22). Glycolysis also provides pyruvate,

E394

REGRESSION

OF

BAT

MITOCHONDRIA

IN

NEONATAL

GOATS

FIG. 1. Sample electron micrograph of brown adipose tissue (BAT) mitochondria of 2- and 6day-old kids at a magnification of ~30,000. BAT mitochondria of P-day-old kid shown on top are condensed and cristae rich, with sparse matrix indicating very active mitochondria. Mitochondria of B-day-old goat shown in bottom are relaxed and cristae poor, with a Iarge matrix space, and several were observed to be in various stages of degradation.

which feeds into the Krebs cycle in the mitochondria to form oxaloacetate, which couples to acetyl coenzyme A to form citrate. The high acetyl coenzyme A concentration may require high glycolytic activity to maintain the flux of glucose to oxaloacetate to maintain appropriate levels of Krebs cycle intermediates. Kids at birth had high specific and total glycerophosphate shuttle activity (Table 2) similar to lambs (10). This high-glycerophosphate shuttle activity is able to support high thermogenesis, which peaks during this period. The decline in glycerophosphate shuttle activity after birth (Table 2) may be another indicator of the decline in the thermogenie capacity and total decline in the tissue metabolism. No quantitative findings are reported for the cellular morphological changes in the BAT in kids in this study, because electron micrographs were done only on three to four randomly selected tissue samples of randomly selected three to four kids from each group. However, we

observed the same type of cellular morphological changes reported to occur in lambs (8) and kids (20), i.e., a transition from multilocular adipocytes to monolocular fat cells. Furthermore, we saw a change in mitochondria structure (Fig. 1). This observation confirms an earlier study (2O), which demonstrated a change in multilocular perirenal BAT cells to unilocular cells resembling white adipocytes by 7 days postpartum. Our study extends these results by showing that the regression in mitochondrial structure was synchronous with the decline in the thermogenic capacity and biochemical activity of this tissue. The goat kid is a good model to study this regression, because the morphological changes in BAT cells occur very rapidly, unlike other ruminant species such as lambs (3). In summary, BAT thermogenic capacity, metabolism, and biochemical activity in kids is very high during a short period after birth and, then regresses very rapidly

REGRESSION

OF

BAT

MITOCHONDRIA

to a quiescent state. These changes correlate with the morphological changes that occur at both the cellular and mitochondrial level during the first postnatal week.

IN

lo* 11.

Address for reprint Nutrition and Foods, 0148. Received

8 May

requests: University

1986; accepted

Dr. Robert S. Tyzbir, Dept. of Human of Vermont, Burlington, VT 05405in final

form

29 October

12.

1986.

REFERENCES AFZELIUS, B. A. Brown Adipose Tissue: its anatomy, histology and cytology. In: Brouln Adipose Tissue, edited by 0. Lindberg. New York: Elsevier, 1970, p. 1-31. ALEXANDER, G., AND A. W. BELL. Quantity and calculated oxygen consumption during summit metabolism of brown adipose tissue in new born lambs. Biol. Neonate 26: 214-220,1975. ALEXANDER, G., J. W. BENNETT, AND R. T. GEMMELL. Brown adipose tissue in a new born calf (bos taurus). J. Physiol. Land. 244: 223-239, 1974. ALEXANDER, G., AND D. WILLIAMS. Shivering and non-shivering thermogenesis during summit metabolism in young lambs. J. Physiol. Land. 198: 251-276, 1968. BLIX, A. S., H. J. GRAV, AND K. A. MARKUSSEN. Modes of thermal protection in newborn muskoxen (Ovibus moschatus). Actu Physiol. Stand. 122: 443-453, 1984. CANNON, B., AND J. NEDERGAARD. The biochemistry of an inefficient tissue: brown adipose tissue. Essays Biochem. 20: 111-164, 1985. CANNON, B., L. ROMERT, U. SUNDIN, AND T. BARNARD. Morphology and biochemical properties of perirenal tissue from lamb (ovis aries). A comparison with brown adipose tissue. Comp. Biothem. Physiol. 568: 87-99, 1977. GEMMELL, R. T., A. W. BELL, AND G. ALEXANDER. Morphology of adipose cells in lambs at birth and during subsequent transition of brown adipose to white adipose tissue in cold and warm conditions. Am. J. Anut. 133: 143-164, 1972. HOUSTI~K, J., B. CANNON, AND 0. LINDBERG. Glycerol-3-phosphate shuttle and its function in intermediary metabolism of

13. 14.

15.

16. 17.

18. 19.

20. 21.

22.

23.

NEONATAL

GOATS

E395

hamster brown adipose tissue. Eur. J. Biochem. 54: 11-18, 1975. KLEIN, A. H., A. REVICZKY, P. CHOU, J. PADBURY, AND D. A. FISHER. Development of brown adipose tissue thermogenesis in the ovine fetus and newborn. Endocrinology 112: 1662-1666, 1983. LINDBERG, O., J. NEDERGAARD, AND B. CANNON. Mitochondria and Microsomes, edited by C. P. Lee. Menlo Park, CA: AddisonWesley, 1981, p. 93-119. NEDERGAARD, J., AND CANNON, B. Regulation of thermogenin expression and activity in brown adipose tissue. In: Thermal Physiology, edited by J. R. S. Hales. New York: Raven, 1984, p. 169173. NICHOLLS, D. G., AND R. M. LOCKE. Thermogenic mechanisms in brown fat. Physiol. Rev. 64: l-64, 1984. MARKUSSE, K. A., A. ROGMO, AND A. S. BLIX. Some aspects of thermoregulation in newborn reindeer calves (Rangifer tarandus tarandus). Actu. Physiol. Scund. 123: 215-220, 1985. OHKAWA, K., M. T. VOGT, AND E. FARBER. Unusually high mitochondrial alpha-glycerophosphate dehydrogenase activity in rat brown adipose tissue. J. Cell. BioZ. 41: 441-449, 1969. ROTHWELL, N. J., AND M. J. STOCK. Diet-induced thermogenesis. Adv. Nutr. Res. 5: 201-220, 1983. ROWLATT, U., N. MROSOVSKY, AND A. ENGLISH. A comparative survey of brown fat in the neck and axilla of mammals at birth. Biol. Neonate 17: 53-84, 1971. SMITH, R. E., AND B. A. HORWITZ. Brown fat and thermogenesis. Physiol. Rev. 49: 330-425, 1969. SNEDECOR, G. W., AND W. C. COCHRAN. Factorial experiments. In: Statistical Methods. Ames, IA: Iowa State Univ. Press, 1980, chapt. 12, p. 339-377. THOMPSON, G. E., AND D. M. JENKINSON. Adipose tissue in new born goat. Res. Vet. Sci. 11: 102-103, 1970. TYZBIR, R. S. Altered brown adipose tissue function in neonates born to rats overfed foods of various protein content. J. Nutr. 114: 234-237,1984. TYZBIR, R. S., D. E. HEWETT, C. A. SWAN, AND L. B. CAREW. The effects of pre- and postdietary protein levels on mitochondrial metabolism in developing liver and interscapular brown adipose tissue (IBAT) in rats. J. Nutr. 112: 1536-1545, 1982. TYZBIR, R. S., A. S. KUNIN, E. A. SIMS, AND E. J. DANFORTH. Influence of diet composition, serum triiodothyronine (Ts) concentration, hepatic mitochondrial metabolism and shuttle system activities in rats. J. Nutr. 111: 252-259, 1981.