Camp. Biochem. Physiol.Vol. IOlA, No. 2, pp. 281-293,
0300-9629/92$5.00+ 0.00 0 1992Pergamon Press plc
1992
Printed in Great Britain
LOCALIZATION, CELLULAR MORPHOLOGY AND RESPIRATORY CAPACITY OF “BROWN” ADIPOSE TISSUE IN NEWBORN REINDEER PGVI SOPPELA,*~RAIJASORMUNEN,$ SEP~OSAARELA,$PIRKKOHUTTLJNEN~~ and MAURI NIEMINEN* *Finnish Game and Fisheries Research Institute, Reindeer Research, Koskikatu 33 A 17,961OO Rovaniemi, Finland. Telephone: 358-60-23040; Fax: 358-60-23040; $University of Oulu, Department of Pathology, Kajaanintie 52 D, 90220 Oulu, Finland; $University of Oulu, Department of Zoology, 90570 Oulu, Finland; [IDepartment of Forensic Medicine, Kajaanintie 52 D, 90220 Oulu, Finland (Received 24 May 1991)
Abstract-l. The localization and cellular morphology of adipose tissue was studied by light, fluorescence and electron microscopy in reindeer between 2 weeks pre partum and 4.5 months post partum during calving, and the subsequent growth period. The respiratory capacity of the adipose tissue was examined in terms of morphometric mitochondrial volume and cytochrome-c oxidase or succinate dehydrogenase activity. 2. Adipose tissue was located at specific anatomical sites in the newborn reindeer (from 0 to 2 days of age). The perirenal-abdominal depot was the largest location (32%) followed by the inter@re)scapular (18%) and sternal (12%) depots. Internal depots dominated over external or peripheral depots (66-34%). The locations of adipose tissue were largely similar in foetal, newborn and young reindeer. 3. The adipose tissue of the newborn reindeer had all the typical cell morphological characteristics of brown adipose tissue: abundant mitochondria, multilocular fat, high vascularization and a dense spot-like sympathetic innervation between the adipocytes. In the young reindeer, however, it resembled white adipose tissue, being almost totally unilocular with few mitochondria. 4. There was a significant correlation between morphometric mitochondrial volume and cytochrome-c oxidase activity (r = 0.848) in the adipose tissue. Mitochondrial volume, cytochrome-c oxidase and succinate dehydrogenase activity were highest after birth and decreased to almost an undetectable level during the first month. A parallel decrease occurred in the amount of brown adipose tissue from birth (l-2%) to the age of about one month (0.3%). 5. It is concluded that the distinct cell morphological features and high respiratory capacity of the adipose tissue indicate the presence of brown adipose tissue at specific anatomical locations in newborn reindeer. A marked progression towards the characteristics of white adipose tissue then takes place at the same locations during the first month. The results suggest the fundamental significance of brown adipose tissue for non-shivering thermogenesis in newborn reindeer.
INTRODUCTION The semi-domesticated Finnish reindeer (Rungifer tarundus rarundus L.) gives birth at the earliest occasion in the northern spring. Calving reaches its peak in May, when the pastures are usually still covered by snow and the ambient temperature frequently falls below 0°C. The “precocial” (cf. Blix and Steen, 1979) newborn reindeer is well-developed and is soon able to follow and suckle its mother. Previous research (Hissa et al., 1981; Markussen et al., 1985; Soppela et al., 1986a) has demonstrated an effective capacity for thermoregulation in reindeer calves, which possess the particular ability to produce heat by non-shivering thermogenesis in response to cold. The unreliable weather conditions and complete dependence on the milk from the mother may, however, expose the calves to the most severe thermal or energetic stresses of their lives at the moment of birth and during the critical postnatal development. Pre-
tTo whom all correspondence
should be addressed, 281
vious findings have suggested the presence of adipose tissue with the histological characteristics of brown adipose tissue (BAT) in newborn reindeer (Krog et al., 1977; Hissa et al., 1981; Soppela et al., 1986b), but our knowledge of the amount, anatomical distribution and nature of the adipose tissue in reindeer calves is still fragmentary. Mammalian white adipose tissue (WAT) and BAT are both able to store energy in the form of large amounts of fat, but they differ in their cellular morphology and ultimate function (reviewed by Nt?chad, 1986; Cannon and Nedergaard, 1985; Himms-Hagen, 1989). Typical brown adipocytes have abundant mitochondria and multilocular fat, in contrast to sparse mitochondria of unilocular white adipocytes. The primary function of BAT is to produce heat by non-shivering thermogenesis, which results from an effective uncoupling of oxidative phosphorylation (Nicholls and Locke, 1984) and is caused by sympathetic stimulation (Jansky, 1973). On the other hand, WAT provides primary long-term energy storage for other tissues. BAT is prominent in newborn and hibernating mammals and in small
P&VI SOPPELAet al.
282
cold-acclimated adult mammals (Smith and Horwitz, 1969). On the basis of developmental compa~son, BAT is assumed to account for most fat stores in precocial mammalian neonates, as in the newborn lamb (Gemmel et al., 1972) and bovine calf (Alexander ef al., 1975). Large arctic precocials have been studied only incidentally in this respect, however, and the functional basis for non-shivering thermogenesis is poorly established. The objective of the present work was to examine the nature of the adipose tissue in reindeer at birth and during the perinatal development by (1) comparing systematically the amount, precise anatomical location and distribution of adipose tissue in foetal, newborn and young reindeer, (2) by identifying and characte~~ng the cellular morphology of adipose tissue by a combination of light, fluorescence and electron microscopy, and (3) by assessing the respiratory capacity of adipose tissue in terms of morphometric mitochondrial volume (from electron micrographs) and cytochrome-c oxidase activity. Special attention was paid to the abundant mitochondria and dense sympathetic spot-like innervation between adipocytes as the most important morphological features of BAT as compared to WAT, but the appearance of multilocular fat was not considered a major criterion, since it is variable in both adipose tissues, as shown previously by several authors (e.g. Smith and Horwitz, 1969; Himms-Hagen, 1989). ~li~na~ findings from the part of this work have been presented previously in the form of an abstract (Soppela et al., 1986b). MATERIALS AND MKI’HODS The 42 reindeer examined (Rang@ tarandus tarandus L.) were either from the experimental herd of the Finnish Reindeer Herders’ Association, located in Kaamanen, near Inari (69”lO’N) in Northern Finland, or originated from various reindeer herding co-operatives in Finnish Lapland, mainly the Paistunturi and Muotkatunturi Reindeer Herding Districts near Inari. The characteristics of the adipose tissue were examined in reindeer calves of both sexes during the calving period from May to July 1985-1990. The animals ranged in age from 2 weeks pre partum to 4.5 months post partum. They included 29 animals which had died acoidentaliy in the field and 13 which were killed for the present purposes. The calves that were killed were examined immediately and the others within 48 hr of death. The reindeer had been exposed to normal seasonal ambient temperatures (mean daily r, from -6 to 24°C) and photoperiod (23 hr light: 1 hr dark in May, and 24 hr light in June-July). The calves had usually been accompanying their mothers, which were freely grazing on the pastures. A sunnlementarv feed containing 11So! crude protein (wt/wt, by‘hry matter) and a metabolizable energy density of 10.2 MJlka (Foron-Herkku, Raision Tehtaat OY, Finland) was me-affable daily for the hinds in the xaamanen reindeer herd. Water was freely available. The main food of the calves consisted of their mothers’ milk until 4-6 weeks of age, by which time they had adapted to the fermentation of green plants. Anatomical examination of adipose tissue The animals were weighed and the adipose tissue was identified anatomically in the autopsies. The colour and appearance of the tissue was recorded. All visible adipose tissue was dissected from each location and fragments of
muscle and connective tissue were carefully removed. The wet weight of each depot was dete~n~ to an accuracy of 0.01 g. Since a large number of the newborn calves that had died accidentally were small and undernourished, they were divided into two groups: underweight and normal weight calves (P < 0.001) to assess any differences in adipose tissue deposition. A boundary between the groups was set at 3.6 kg. The young calves aged 1 month, also differed in their body weight, which varied from 11.9 to 21.5 kg, and were again studied in two separate groups: normal-sized and large. Light microscopy The gross cellular appearance of the adipose tissue was characterized in light microscopy samples prepared from inter(pre)scapular locations in one foetal, three newborn and three young reindeer. Small pieces were removed from the middle of each depot, 5xed in Bouin fluid, dehydrated through an ethanol series and embedded in paraflln. The sections (5-7 pm) were stained with Mayers’ haematoxylin and eosine procedure. Histofiuorescence microscopy
Samples of inter(pre)scapular adipose tissue from the same tissues as used for light microscopy were quickly frozen with liquid nitrogen for the examination of adrenergic sympathetic innervation. The samples were stored for 2 weeks at - 70”C, after which lo-15 pm sections were cut in a cryostat. A sucrose-potassium phosphateglyoxylic acid (SPG) method (De La Torre, 1980) was used to demonstrate catecholaminergic nerve fibres in the tissue. The specimens were examined with a 5uorescence microscope (Polyvar, Rei~hert-Jung). Electron microscopy
The cellular morphology of the adipose tissue was identified and characterized by electron microscopy in each anatomical location in 11 newborn reindeer, and in the major locations (perirenal, inter@re)scapular, sternal) in one foetal and nine young reindeer. Small pieces of l-2 mm3 were removed from the middle of each tissue depot and 5xed in a cold (4°C) 1% glutaraldehyde +4% formaldehyde mixture in 0.1 M phosphate buffer. The fixed samples were washed in 0.1 M phosphate buffer overnight and postftxed in 1% osmium tetroxide in 0.1 M phosphate buffer for 2 hr. They were then stained in a 1% uranyl acetate-water solution, dehydrated in an acetone gradient and embedded in Epon LX 112. Thin sections (70 nm) were cut with a Reichert 4 E ultramicrotome and examined with a Philips 4 10 LS transmission electron microscope using an acceleration voltage of 60 kV. Morphometric mitochondrial volume
Altogether six photographs were taken from randomly chosen areas of the thin segments of perirenal, inter@re)scapular and sternal adipose tissue from each calf at final magnifications of 4800 x for morphometric calculations. The volume of mitochondria was calculated with a specific template provided with 100 statistical calculation points. Volume was calculated in percentages directly from the template as the mean of six measurements. Respiratory capacity
Samples from the major adipose tissues and ~sion~y from other available sites were assayed for sue&ate dehydrogenase (SDH) activity. Aliquots of l-3 g adipose tissue were weighed out for analysis, frozen rapidly on solid carbon dioxide and stored at -40°C until analysed. SDH activity was determined as micromoles of succinate oxidized per minute per milligram of protein from both total homogenate and mitochondria as described previously by Kinnula et al. (1983). SDH was determined using phenazinemethosulphate and 2,6-dichloroindophenol as the
Brown adipose tissue in newborn reindeer electron acceptor system (King, 1967). Maximal activity was obtained by preincubating the sample at 37°C in a potassium succinate mixture (Kimura et al., 1967). The mitochondria were separated, and mitochondrial proteins were determinated according to Lowry et ul. (1951). Cytochrome-c oxidase (COX) activity was analysed polarographically in the same adipose tissues as were used to determine morphometric mitochondrial volumes. COX activity was determined in tissue total homogenates at 25°C by a Clark type oxygen electrode (Bachofer) as described previously (Rafael ef al., 1970, modified by Saarela et al., 1989). The mitochondria were separated, and the mitochondrial protein content analysed by a method of Rafael et al. (1970). Statistical method
A one-way analysis of variance accompanied by posterior
LSD multiple range analysis (from Stargraphics Statistical Granhics Svstem, Grauhic Software Systems Inc.) was used for ihe comparisons between groups-The method of least squares was used for the simple regression analyses. Correlation coefficients were calculated using Pearson’s correlation test. RRSULTS
A~peur~ce ufadipose tissue The adipose tissue from the well-nourished newborn reindeer calves (from 0 to 2 days of age) sampled immediately after death was soft, lobular and light to reddish brown in colour. In some newborn calves which had been exposed to cold or disturbances in suckling the tissue was reddish to dark brown. It was sometimes totally depleted of fat and resembled a red jelly. The adipose tissue of the newborn reindeer at the dissection point was regarded macroscopically as “brown” adipose tissue. The f&eta1 adipose tissue (- 14 or 0 days pre partum) was soft and light brown
283
or yellowish in colour. The adipose tissue in the young calves (from 1 to 4.5 months of age) was usually light brown or white and more waxy than in the foetal or newborn reindeer. Location of “brown” adipose tissue The location of the “brown” adipose tissue was examined comprehensively in the newborn reindeer (from 0 to 2 days of age). The tissue was composed of well-defined and diffuse depots in specific locations in the body, i.e. in the major body cavities and near the vital organs and major blood vessels (Fig. 1). The perirenal-abdominal depot covered the kidneys and abdominal wall lymph nodes and extended dorsally to the pelvis. The inter@re)scapular depot was composed of bilateral strips which covered the prescapular lymph nodes, extended beneath the cervical muscles and ended in the region of the major cervical vessels. The sternal depot was found on both sides of the sternum, with the major depot beneath the pectoral muscle on the sternum, and a smaller one situated on the opposite side of the bone. Minor costal sheets were found beneath the muscle layers at the starting points of the ribs. Diffuse intralumbar adipose tissue was found covering small lymph nodes in the pelvic channel and lining the genitals. Bilateral, longitudinal strips were found in the vertebral region ventrolaterally from the last rib to the last lumbar vertebra and along the dorsal aorta, while a more diffuse mass was situated in the tracheal cavity along the trachea, oesophagus and the carotid vessel and was connected with the deeper lying ends of the bilateral inter@re)scapular tissue. Diffuse adipose tissue was found alo$g the omentum or mesenterium among the duodenum and colon. Bilateral depots were found analo~cally beneath the
Fig. 1. Anatomical locations of adipose tissue in r&born reindeer. All the locations except those marked with an asterisk are suggested as sites for “brown” adipose tissue. For the dist~bution of the tissues between these locations, see Fig. 2.
284
PiNI
%PPELA
muscles in the axillar and inquinal regions, and smaller depots existed in the pericardial sac and along the coronary groove around the heart. A certain amount of this tissue occurred regularly in the orbital area behind the eye. Light adipose tissue was occasionally found in the dorsal area of the neck; the occipital area, and in the aural area behind the ear. There was seldom any visible subcutaneous (caudal) adipose tissue in the newborn reindeer. Adipose tissue distribution
The adipose tissue was located in virtually the same manner in the foetal (14 or 0 days pre partum) and young reindeer (ca 1 month of age) as in the newborn reindeer (Fig. l), the perirenal-abdominal depot being the largest in all the age groups. It was nevertheless larger in the foetal and newborn reindeer than in the young reindeer (P < 0.01). Considerable amounts of adipose tissue also existed in the inter@re)scapular, sternal-costal, intralumbar, vertebral and tracheal locations in all groups (Fig. 2A,B), and
r
et a/.
in minor amounts in the inguinal or axillar locations, the pericardial and coronary areas, and the orbital, occipital and aural regions. The large l-month-old calves (body weight 20.3 kg) had a larger proportion of adipose tissue in the omental-mesenteral and subcutaneous locations than the normal-sized or newborn calves. The proportion of internal or visceral adipose tissue was usually larger than that of external or peripheral depots in the newborn reindeer (see Fig. 2A,B). Amount of adipose tissue
Total adipose tissue comprised 59.2 g, or 1.1 f 0.1% of body weight, in the normal weight newborn calves (mean 5.2 f 0.3 kg, N = 13) and 24.7 g, or 1.0 f O.l%, in the underweight ones (2.5 + 0.3 kg, N = 6). Since no significant differences were found between the newborns in the proportion of this tissue, their figures were combined. Adipose tissue comprised 1.l f 0.1% of the body weight (range 0.7-1.6%, N = 19) in the whole group of
A. Internal Foetal
=Perirenal-abdominal a =Intralumbar
locations
Newborn
Young
Young,
large
q=Vertebral
Q =Omental-mesenteral
B =Tracheal
q =Pericardial-coronary
B. External
locations
fl =lnter(prs)scapular
a =Axillar-inguinal
0 =Subcutaneous
0 =Sternal-costal
qaOrbital
m =Occipital
m PAural
Fig. 2. Proportions of adipose tissue found at the various anatomical locations in foe@ newborn and young reindeer at 1 month of age. Internal depots (A) refer to the adipose tissue inside the body cavities, and external depots (B) indicate peripheral locations. For further information Table 1.
on the reindeer, see
Brown adipose tissue in newborn reindeer
285
Table 1. Adipose tissue wet weight, proportion per unit of body weight, and distribution between internal and external locations (mean + SE) in reindeer during late gestation, at birth and at about 1 month of age. For further information on the distribution of adipose tissue, see Fig. 2A, B Foetal Age in days Body weight (kg) Adipose tissue wet weight (g) Proportion (%) Internal locations (%) External locations (%) No. of animals
-5.6 f 4.7 f 95.6 f 2.0 * 59.8 f 40.2 * 5
3.4 0.3 15.1 0.2 1.3 1.3
newborn calves (Table l), with no difference between the sexes. Adipose tissue wet weight and body weight showed a positive correlation in this group (r = 0.907). The amount of adipose tissue was highest pre partum, a mean of 2.0% of body weight, in the five foetuses at 14 or 0 days pre partum (cf Table 1). The normal-sized young reindeer aged 1 month had only 0.2-0.4% adipose tissue whereas the large ones had l&1.6% (Table 1). Cellular morphology
Light microscopy showed lobular adipocytes with multilocularly dispersed lipid vacuoles in the newborn reindeer (Fig. 3A), while the adipocytes were predominantly unilocular in the foetal reindeer 2 weeks pre partum and in the young reindeer from the first month onwards (Fig. 3B,C). The nucleus was usually central in the adipocytes from the newborns, but peripheral in the foetal and young reindeer. Fluorescence microscopy showed green fluorescent staining to be typically located around the adipocytes
Newborn
Young
Young, large
0.9 * 0.1 4.3 + 0.4 48.2 f 5.5 1.1 fO.l 65.6 + 1.7 34.4 * 1.7 19
24.0,35.0 11.9, 15.5 25.4, 56.6 0.2,0.4 38.0,60.1 62.0,39.9 2
28.0,32.0 19.0,21.5 266.4, 354.3 1.4, 1.6 47.4, 61.7 52.6, 38.3 2
in the newborn reindeer (Fig. 3A), and mainly around the arteries in the young reindeer. The foetal reindeer had sparse fluorescent arterial elements between the adipocytes (Fig. 3B,C). Adipose tissue of the newborn reindeer (Fig. 3A) showed a typical appearance of “brown” adipose tissue in electron microscopy. The polygonal adipocytes were tightly packed with large mitochondria, and varying amounts of multilocular lipid vacuoles were situated next to these. A dense network of capillaries was prominent. The adipocytes in the foetal reindeer at 2 weeks pre partum (Fig. 3B) resembled the perinatal adipose tissue, containing numerous mitochondria which varied in both shape and size and lined predominantly large lipid vacuoles. Abundant extracellular space and capillaries (some very large) were present. Preadipocytes and fibroblasts were prominent in the connective tissue strands between the adipocytes. The cellular morphology of the adipose tissue in the young reindeer was different, however (Fig. 3C), in that the adipocytes were large and usually unilocular, small mitochondria were
Fig. 3(A)
286
PANI So~pm.4 ei al.
Fig. 3(B)
Fig. 3(C) Fig. 3. Interscapular adipose tissue from (A) a 2-day-old male reindeer weighing 5.8 kg which had been exposed to 11°C after birth, (B) a foetal male reindeer at 2 weeks pre partum weighing 5.0 kg and (C) a 3.5month-old male reindeer weighing 20 kg exposed for ca 24 hr to 3°C. The cellular morphology of the tissue is illustrated by light (LM), fluorescence (FM) and electron micrographs (EM). C, capillary: L, lipid droplet and m, mitochondria. The bars represent 6Opm in the plates of LM and FM.
Brown adipose tissue in newborn reindeer randomly seen in the cytoplasm, and capillaries were few. Adipocytes from the perirenal-abdominal, inter(pre)scapular, sternal-costal, intralumbar, vertebral and tracheal locations in the newborn reindeer had the distinct characteristics of “brown” adipose tissue in electron microscopy (cf. Figs 1 and 3-5A). These sites were used to calculate the proportion of “brown” adipose tissue with age (Fig. 6). In addition, the adipose tissues in the orbital and omental-mesenteral locations are tentatively suggested to
287
adipose tissue, since the “brown” represent adipocytes contained relatively large multilocular or unilocular lipid vacuoles lined by a single or double row of mitochondria. However, light aural, occipital and subcutaneous (caudal) adipose tissue with irregular presence and predominantly unilocular adipocytes with few mitochondria are suggested to represent white adipose tissue in the newborn reindeer. Lipid vacuoles were totally absent in some coldexposed, poorly-fed newborn calves, and the
(A)
Fig. 4(A) and (B). Electron micrographs of perirenal “brown” adipose tissue from (A) a well-nourished, 2-day-old reindeer calf exposed to 13”C, and (B) a poorly-fed, 2-day-old calf exposed to 0°C after birth. Note the abundant multilocular lipid in the adipocytes of the well-nourished calf and the complete lack of lipid in the undernourished, cold-exposed calf. The mitochondria contain a dense inner membrane in both cases. Bars represent 1 pm in the inserts.
288
P6rvr S0ppE~~et al.
capillaries were expanded and filled with red corpuscles (cf. Fig. 4A). The cytoplasm was packed with mitochondria which had signs of mitochondriogenesis, i.e. intramitochondrial inclusions or dividing units. The adipocytes of the well-nourished, properly fed calves contained large amounts of multilocular lipid (Fig. 4B). Moreover, a progression in the cellular morphology of the brown adipose tissue towards characteristics of white adipose tissue occurred during the first postnatal month (Fig. SA-D). The lipid droplets grew larger with age and were unilocular in young reindeer, and the cytoplasmic mitochondria became sparse and eventually disappeared entirely. At the age of about 1.5 months the tissue was indistinguishable by electron microscopy from white adipose tissue (see Fig. SC,D).
Respiratory capacity
Succinate dehydrogenase (SDH) activity was prominent in all the four major “brown” adipose tissue locations studied in the newborn reindeer (cf. Table 2). Mitochondrial maximal SDH activities were ca twice those measured in total homogenates. No regular pattern of differences was found between the various locations, although the maximal SDH activity was higher in the total homogenates of the perirenal-abdominal adipose tissue than in those of the sternal tissue (P
0300-9629/92$5.00+ 0.00 0 1992Pergamon Press plc
1992
Printed in Great Britain
LOCALIZATION, CELLULAR MORPHOLOGY AND RESPIRATORY CAPACITY OF “BROWN” ADIPOSE TISSUE IN NEWBORN REINDEER PGVI SOPPELA,*~RAIJASORMUNEN,$ SEP~OSAARELA,$PIRKKOHUTTLJNEN~~ and MAURI NIEMINEN* *Finnish Game and Fisheries Research Institute, Reindeer Research, Koskikatu 33 A 17,961OO Rovaniemi, Finland. Telephone: 358-60-23040; Fax: 358-60-23040; $University of Oulu, Department of Pathology, Kajaanintie 52 D, 90220 Oulu, Finland; $University of Oulu, Department of Zoology, 90570 Oulu, Finland; [IDepartment of Forensic Medicine, Kajaanintie 52 D, 90220 Oulu, Finland (Received 24 May 1991)
Abstract-l. The localization and cellular morphology of adipose tissue was studied by light, fluorescence and electron microscopy in reindeer between 2 weeks pre partum and 4.5 months post partum during calving, and the subsequent growth period. The respiratory capacity of the adipose tissue was examined in terms of morphometric mitochondrial volume and cytochrome-c oxidase or succinate dehydrogenase activity. 2. Adipose tissue was located at specific anatomical sites in the newborn reindeer (from 0 to 2 days of age). The perirenal-abdominal depot was the largest location (32%) followed by the inter@re)scapular (18%) and sternal (12%) depots. Internal depots dominated over external or peripheral depots (66-34%). The locations of adipose tissue were largely similar in foetal, newborn and young reindeer. 3. The adipose tissue of the newborn reindeer had all the typical cell morphological characteristics of brown adipose tissue: abundant mitochondria, multilocular fat, high vascularization and a dense spot-like sympathetic innervation between the adipocytes. In the young reindeer, however, it resembled white adipose tissue, being almost totally unilocular with few mitochondria. 4. There was a significant correlation between morphometric mitochondrial volume and cytochrome-c oxidase activity (r = 0.848) in the adipose tissue. Mitochondrial volume, cytochrome-c oxidase and succinate dehydrogenase activity were highest after birth and decreased to almost an undetectable level during the first month. A parallel decrease occurred in the amount of brown adipose tissue from birth (l-2%) to the age of about one month (0.3%). 5. It is concluded that the distinct cell morphological features and high respiratory capacity of the adipose tissue indicate the presence of brown adipose tissue at specific anatomical locations in newborn reindeer. A marked progression towards the characteristics of white adipose tissue then takes place at the same locations during the first month. The results suggest the fundamental significance of brown adipose tissue for non-shivering thermogenesis in newborn reindeer.
INTRODUCTION The semi-domesticated Finnish reindeer (Rungifer tarundus rarundus L.) gives birth at the earliest occasion in the northern spring. Calving reaches its peak in May, when the pastures are usually still covered by snow and the ambient temperature frequently falls below 0°C. The “precocial” (cf. Blix and Steen, 1979) newborn reindeer is well-developed and is soon able to follow and suckle its mother. Previous research (Hissa et al., 1981; Markussen et al., 1985; Soppela et al., 1986a) has demonstrated an effective capacity for thermoregulation in reindeer calves, which possess the particular ability to produce heat by non-shivering thermogenesis in response to cold. The unreliable weather conditions and complete dependence on the milk from the mother may, however, expose the calves to the most severe thermal or energetic stresses of their lives at the moment of birth and during the critical postnatal development. Pre-
tTo whom all correspondence
should be addressed, 281
vious findings have suggested the presence of adipose tissue with the histological characteristics of brown adipose tissue (BAT) in newborn reindeer (Krog et al., 1977; Hissa et al., 1981; Soppela et al., 1986b), but our knowledge of the amount, anatomical distribution and nature of the adipose tissue in reindeer calves is still fragmentary. Mammalian white adipose tissue (WAT) and BAT are both able to store energy in the form of large amounts of fat, but they differ in their cellular morphology and ultimate function (reviewed by Nt?chad, 1986; Cannon and Nedergaard, 1985; Himms-Hagen, 1989). Typical brown adipocytes have abundant mitochondria and multilocular fat, in contrast to sparse mitochondria of unilocular white adipocytes. The primary function of BAT is to produce heat by non-shivering thermogenesis, which results from an effective uncoupling of oxidative phosphorylation (Nicholls and Locke, 1984) and is caused by sympathetic stimulation (Jansky, 1973). On the other hand, WAT provides primary long-term energy storage for other tissues. BAT is prominent in newborn and hibernating mammals and in small
P&VI SOPPELAet al.
282
cold-acclimated adult mammals (Smith and Horwitz, 1969). On the basis of developmental compa~son, BAT is assumed to account for most fat stores in precocial mammalian neonates, as in the newborn lamb (Gemmel et al., 1972) and bovine calf (Alexander ef al., 1975). Large arctic precocials have been studied only incidentally in this respect, however, and the functional basis for non-shivering thermogenesis is poorly established. The objective of the present work was to examine the nature of the adipose tissue in reindeer at birth and during the perinatal development by (1) comparing systematically the amount, precise anatomical location and distribution of adipose tissue in foetal, newborn and young reindeer, (2) by identifying and characte~~ng the cellular morphology of adipose tissue by a combination of light, fluorescence and electron microscopy, and (3) by assessing the respiratory capacity of adipose tissue in terms of morphometric mitochondrial volume (from electron micrographs) and cytochrome-c oxidase activity. Special attention was paid to the abundant mitochondria and dense sympathetic spot-like innervation between adipocytes as the most important morphological features of BAT as compared to WAT, but the appearance of multilocular fat was not considered a major criterion, since it is variable in both adipose tissues, as shown previously by several authors (e.g. Smith and Horwitz, 1969; Himms-Hagen, 1989). ~li~na~ findings from the part of this work have been presented previously in the form of an abstract (Soppela et al., 1986b). MATERIALS AND MKI’HODS The 42 reindeer examined (Rang@ tarandus tarandus L.) were either from the experimental herd of the Finnish Reindeer Herders’ Association, located in Kaamanen, near Inari (69”lO’N) in Northern Finland, or originated from various reindeer herding co-operatives in Finnish Lapland, mainly the Paistunturi and Muotkatunturi Reindeer Herding Districts near Inari. The characteristics of the adipose tissue were examined in reindeer calves of both sexes during the calving period from May to July 1985-1990. The animals ranged in age from 2 weeks pre partum to 4.5 months post partum. They included 29 animals which had died acoidentaliy in the field and 13 which were killed for the present purposes. The calves that were killed were examined immediately and the others within 48 hr of death. The reindeer had been exposed to normal seasonal ambient temperatures (mean daily r, from -6 to 24°C) and photoperiod (23 hr light: 1 hr dark in May, and 24 hr light in June-July). The calves had usually been accompanying their mothers, which were freely grazing on the pastures. A sunnlementarv feed containing 11So! crude protein (wt/wt, by‘hry matter) and a metabolizable energy density of 10.2 MJlka (Foron-Herkku, Raision Tehtaat OY, Finland) was me-affable daily for the hinds in the xaamanen reindeer herd. Water was freely available. The main food of the calves consisted of their mothers’ milk until 4-6 weeks of age, by which time they had adapted to the fermentation of green plants. Anatomical examination of adipose tissue The animals were weighed and the adipose tissue was identified anatomically in the autopsies. The colour and appearance of the tissue was recorded. All visible adipose tissue was dissected from each location and fragments of
muscle and connective tissue were carefully removed. The wet weight of each depot was dete~n~ to an accuracy of 0.01 g. Since a large number of the newborn calves that had died accidentally were small and undernourished, they were divided into two groups: underweight and normal weight calves (P < 0.001) to assess any differences in adipose tissue deposition. A boundary between the groups was set at 3.6 kg. The young calves aged 1 month, also differed in their body weight, which varied from 11.9 to 21.5 kg, and were again studied in two separate groups: normal-sized and large. Light microscopy The gross cellular appearance of the adipose tissue was characterized in light microscopy samples prepared from inter(pre)scapular locations in one foetal, three newborn and three young reindeer. Small pieces were removed from the middle of each depot, 5xed in Bouin fluid, dehydrated through an ethanol series and embedded in paraflln. The sections (5-7 pm) were stained with Mayers’ haematoxylin and eosine procedure. Histofiuorescence microscopy
Samples of inter(pre)scapular adipose tissue from the same tissues as used for light microscopy were quickly frozen with liquid nitrogen for the examination of adrenergic sympathetic innervation. The samples were stored for 2 weeks at - 70”C, after which lo-15 pm sections were cut in a cryostat. A sucrose-potassium phosphateglyoxylic acid (SPG) method (De La Torre, 1980) was used to demonstrate catecholaminergic nerve fibres in the tissue. The specimens were examined with a 5uorescence microscope (Polyvar, Rei~hert-Jung). Electron microscopy
The cellular morphology of the adipose tissue was identified and characterized by electron microscopy in each anatomical location in 11 newborn reindeer, and in the major locations (perirenal, inter@re)scapular, sternal) in one foetal and nine young reindeer. Small pieces of l-2 mm3 were removed from the middle of each tissue depot and 5xed in a cold (4°C) 1% glutaraldehyde +4% formaldehyde mixture in 0.1 M phosphate buffer. The fixed samples were washed in 0.1 M phosphate buffer overnight and postftxed in 1% osmium tetroxide in 0.1 M phosphate buffer for 2 hr. They were then stained in a 1% uranyl acetate-water solution, dehydrated in an acetone gradient and embedded in Epon LX 112. Thin sections (70 nm) were cut with a Reichert 4 E ultramicrotome and examined with a Philips 4 10 LS transmission electron microscope using an acceleration voltage of 60 kV. Morphometric mitochondrial volume
Altogether six photographs were taken from randomly chosen areas of the thin segments of perirenal, inter@re)scapular and sternal adipose tissue from each calf at final magnifications of 4800 x for morphometric calculations. The volume of mitochondria was calculated with a specific template provided with 100 statistical calculation points. Volume was calculated in percentages directly from the template as the mean of six measurements. Respiratory capacity
Samples from the major adipose tissues and ~sion~y from other available sites were assayed for sue&ate dehydrogenase (SDH) activity. Aliquots of l-3 g adipose tissue were weighed out for analysis, frozen rapidly on solid carbon dioxide and stored at -40°C until analysed. SDH activity was determined as micromoles of succinate oxidized per minute per milligram of protein from both total homogenate and mitochondria as described previously by Kinnula et al. (1983). SDH was determined using phenazinemethosulphate and 2,6-dichloroindophenol as the
Brown adipose tissue in newborn reindeer electron acceptor system (King, 1967). Maximal activity was obtained by preincubating the sample at 37°C in a potassium succinate mixture (Kimura et al., 1967). The mitochondria were separated, and mitochondrial proteins were determinated according to Lowry et ul. (1951). Cytochrome-c oxidase (COX) activity was analysed polarographically in the same adipose tissues as were used to determine morphometric mitochondrial volumes. COX activity was determined in tissue total homogenates at 25°C by a Clark type oxygen electrode (Bachofer) as described previously (Rafael ef al., 1970, modified by Saarela et al., 1989). The mitochondria were separated, and the mitochondrial protein content analysed by a method of Rafael et al. (1970). Statistical method
A one-way analysis of variance accompanied by posterior
LSD multiple range analysis (from Stargraphics Statistical Granhics Svstem, Grauhic Software Systems Inc.) was used for ihe comparisons between groups-The method of least squares was used for the simple regression analyses. Correlation coefficients were calculated using Pearson’s correlation test. RRSULTS
A~peur~ce ufadipose tissue The adipose tissue from the well-nourished newborn reindeer calves (from 0 to 2 days of age) sampled immediately after death was soft, lobular and light to reddish brown in colour. In some newborn calves which had been exposed to cold or disturbances in suckling the tissue was reddish to dark brown. It was sometimes totally depleted of fat and resembled a red jelly. The adipose tissue of the newborn reindeer at the dissection point was regarded macroscopically as “brown” adipose tissue. The f&eta1 adipose tissue (- 14 or 0 days pre partum) was soft and light brown
283
or yellowish in colour. The adipose tissue in the young calves (from 1 to 4.5 months of age) was usually light brown or white and more waxy than in the foetal or newborn reindeer. Location of “brown” adipose tissue The location of the “brown” adipose tissue was examined comprehensively in the newborn reindeer (from 0 to 2 days of age). The tissue was composed of well-defined and diffuse depots in specific locations in the body, i.e. in the major body cavities and near the vital organs and major blood vessels (Fig. 1). The perirenal-abdominal depot covered the kidneys and abdominal wall lymph nodes and extended dorsally to the pelvis. The inter@re)scapular depot was composed of bilateral strips which covered the prescapular lymph nodes, extended beneath the cervical muscles and ended in the region of the major cervical vessels. The sternal depot was found on both sides of the sternum, with the major depot beneath the pectoral muscle on the sternum, and a smaller one situated on the opposite side of the bone. Minor costal sheets were found beneath the muscle layers at the starting points of the ribs. Diffuse intralumbar adipose tissue was found covering small lymph nodes in the pelvic channel and lining the genitals. Bilateral, longitudinal strips were found in the vertebral region ventrolaterally from the last rib to the last lumbar vertebra and along the dorsal aorta, while a more diffuse mass was situated in the tracheal cavity along the trachea, oesophagus and the carotid vessel and was connected with the deeper lying ends of the bilateral inter@re)scapular tissue. Diffuse adipose tissue was found alo$g the omentum or mesenterium among the duodenum and colon. Bilateral depots were found analo~cally beneath the
Fig. 1. Anatomical locations of adipose tissue in r&born reindeer. All the locations except those marked with an asterisk are suggested as sites for “brown” adipose tissue. For the dist~bution of the tissues between these locations, see Fig. 2.
284
PiNI
%PPELA
muscles in the axillar and inquinal regions, and smaller depots existed in the pericardial sac and along the coronary groove around the heart. A certain amount of this tissue occurred regularly in the orbital area behind the eye. Light adipose tissue was occasionally found in the dorsal area of the neck; the occipital area, and in the aural area behind the ear. There was seldom any visible subcutaneous (caudal) adipose tissue in the newborn reindeer. Adipose tissue distribution
The adipose tissue was located in virtually the same manner in the foetal (14 or 0 days pre partum) and young reindeer (ca 1 month of age) as in the newborn reindeer (Fig. l), the perirenal-abdominal depot being the largest in all the age groups. It was nevertheless larger in the foetal and newborn reindeer than in the young reindeer (P < 0.01). Considerable amounts of adipose tissue also existed in the inter@re)scapular, sternal-costal, intralumbar, vertebral and tracheal locations in all groups (Fig. 2A,B), and
r
et a/.
in minor amounts in the inguinal or axillar locations, the pericardial and coronary areas, and the orbital, occipital and aural regions. The large l-month-old calves (body weight 20.3 kg) had a larger proportion of adipose tissue in the omental-mesenteral and subcutaneous locations than the normal-sized or newborn calves. The proportion of internal or visceral adipose tissue was usually larger than that of external or peripheral depots in the newborn reindeer (see Fig. 2A,B). Amount of adipose tissue
Total adipose tissue comprised 59.2 g, or 1.1 f 0.1% of body weight, in the normal weight newborn calves (mean 5.2 f 0.3 kg, N = 13) and 24.7 g, or 1.0 f O.l%, in the underweight ones (2.5 + 0.3 kg, N = 6). Since no significant differences were found between the newborns in the proportion of this tissue, their figures were combined. Adipose tissue comprised 1.l f 0.1% of the body weight (range 0.7-1.6%, N = 19) in the whole group of
A. Internal Foetal
=Perirenal-abdominal a =Intralumbar
locations
Newborn
Young
Young,
large
q=Vertebral
Q =Omental-mesenteral
B =Tracheal
q =Pericardial-coronary
B. External
locations
fl =lnter(prs)scapular
a =Axillar-inguinal
0 =Subcutaneous
0 =Sternal-costal
qaOrbital
m =Occipital
m PAural
Fig. 2. Proportions of adipose tissue found at the various anatomical locations in foe@ newborn and young reindeer at 1 month of age. Internal depots (A) refer to the adipose tissue inside the body cavities, and external depots (B) indicate peripheral locations. For further information Table 1.
on the reindeer, see
Brown adipose tissue in newborn reindeer
285
Table 1. Adipose tissue wet weight, proportion per unit of body weight, and distribution between internal and external locations (mean + SE) in reindeer during late gestation, at birth and at about 1 month of age. For further information on the distribution of adipose tissue, see Fig. 2A, B Foetal Age in days Body weight (kg) Adipose tissue wet weight (g) Proportion (%) Internal locations (%) External locations (%) No. of animals
-5.6 f 4.7 f 95.6 f 2.0 * 59.8 f 40.2 * 5
3.4 0.3 15.1 0.2 1.3 1.3
newborn calves (Table l), with no difference between the sexes. Adipose tissue wet weight and body weight showed a positive correlation in this group (r = 0.907). The amount of adipose tissue was highest pre partum, a mean of 2.0% of body weight, in the five foetuses at 14 or 0 days pre partum (cf Table 1). The normal-sized young reindeer aged 1 month had only 0.2-0.4% adipose tissue whereas the large ones had l&1.6% (Table 1). Cellular morphology
Light microscopy showed lobular adipocytes with multilocularly dispersed lipid vacuoles in the newborn reindeer (Fig. 3A), while the adipocytes were predominantly unilocular in the foetal reindeer 2 weeks pre partum and in the young reindeer from the first month onwards (Fig. 3B,C). The nucleus was usually central in the adipocytes from the newborns, but peripheral in the foetal and young reindeer. Fluorescence microscopy showed green fluorescent staining to be typically located around the adipocytes
Newborn
Young
Young, large
0.9 * 0.1 4.3 + 0.4 48.2 f 5.5 1.1 fO.l 65.6 + 1.7 34.4 * 1.7 19
24.0,35.0 11.9, 15.5 25.4, 56.6 0.2,0.4 38.0,60.1 62.0,39.9 2
28.0,32.0 19.0,21.5 266.4, 354.3 1.4, 1.6 47.4, 61.7 52.6, 38.3 2
in the newborn reindeer (Fig. 3A), and mainly around the arteries in the young reindeer. The foetal reindeer had sparse fluorescent arterial elements between the adipocytes (Fig. 3B,C). Adipose tissue of the newborn reindeer (Fig. 3A) showed a typical appearance of “brown” adipose tissue in electron microscopy. The polygonal adipocytes were tightly packed with large mitochondria, and varying amounts of multilocular lipid vacuoles were situated next to these. A dense network of capillaries was prominent. The adipocytes in the foetal reindeer at 2 weeks pre partum (Fig. 3B) resembled the perinatal adipose tissue, containing numerous mitochondria which varied in both shape and size and lined predominantly large lipid vacuoles. Abundant extracellular space and capillaries (some very large) were present. Preadipocytes and fibroblasts were prominent in the connective tissue strands between the adipocytes. The cellular morphology of the adipose tissue in the young reindeer was different, however (Fig. 3C), in that the adipocytes were large and usually unilocular, small mitochondria were
Fig. 3(A)
286
PANI So~pm.4 ei al.
Fig. 3(B)
Fig. 3(C) Fig. 3. Interscapular adipose tissue from (A) a 2-day-old male reindeer weighing 5.8 kg which had been exposed to 11°C after birth, (B) a foetal male reindeer at 2 weeks pre partum weighing 5.0 kg and (C) a 3.5month-old male reindeer weighing 20 kg exposed for ca 24 hr to 3°C. The cellular morphology of the tissue is illustrated by light (LM), fluorescence (FM) and electron micrographs (EM). C, capillary: L, lipid droplet and m, mitochondria. The bars represent 6Opm in the plates of LM and FM.
Brown adipose tissue in newborn reindeer randomly seen in the cytoplasm, and capillaries were few. Adipocytes from the perirenal-abdominal, inter(pre)scapular, sternal-costal, intralumbar, vertebral and tracheal locations in the newborn reindeer had the distinct characteristics of “brown” adipose tissue in electron microscopy (cf. Figs 1 and 3-5A). These sites were used to calculate the proportion of “brown” adipose tissue with age (Fig. 6). In addition, the adipose tissues in the orbital and omental-mesenteral locations are tentatively suggested to
287
adipose tissue, since the “brown” represent adipocytes contained relatively large multilocular or unilocular lipid vacuoles lined by a single or double row of mitochondria. However, light aural, occipital and subcutaneous (caudal) adipose tissue with irregular presence and predominantly unilocular adipocytes with few mitochondria are suggested to represent white adipose tissue in the newborn reindeer. Lipid vacuoles were totally absent in some coldexposed, poorly-fed newborn calves, and the
(A)
Fig. 4(A) and (B). Electron micrographs of perirenal “brown” adipose tissue from (A) a well-nourished, 2-day-old reindeer calf exposed to 13”C, and (B) a poorly-fed, 2-day-old calf exposed to 0°C after birth. Note the abundant multilocular lipid in the adipocytes of the well-nourished calf and the complete lack of lipid in the undernourished, cold-exposed calf. The mitochondria contain a dense inner membrane in both cases. Bars represent 1 pm in the inserts.
288
P6rvr S0ppE~~et al.
capillaries were expanded and filled with red corpuscles (cf. Fig. 4A). The cytoplasm was packed with mitochondria which had signs of mitochondriogenesis, i.e. intramitochondrial inclusions or dividing units. The adipocytes of the well-nourished, properly fed calves contained large amounts of multilocular lipid (Fig. 4B). Moreover, a progression in the cellular morphology of the brown adipose tissue towards characteristics of white adipose tissue occurred during the first postnatal month (Fig. SA-D). The lipid droplets grew larger with age and were unilocular in young reindeer, and the cytoplasmic mitochondria became sparse and eventually disappeared entirely. At the age of about 1.5 months the tissue was indistinguishable by electron microscopy from white adipose tissue (see Fig. SC,D).
Respiratory capacity
Succinate dehydrogenase (SDH) activity was prominent in all the four major “brown” adipose tissue locations studied in the newborn reindeer (cf. Table 2). Mitochondrial maximal SDH activities were ca twice those measured in total homogenates. No regular pattern of differences was found between the various locations, although the maximal SDH activity was higher in the total homogenates of the perirenal-abdominal adipose tissue than in those of the sternal tissue (P
