Cell Tissue Res (1991) 266:149-161
Cell and Tissue Research 9 Springer-Verlag 1991
Convertible adipose tissue in mice Dragutin Lon~ar* The Wenner-GrenInstitute, Universityof Stockholm, Stockholm,Sweden Accepted March 25, 1991
Summary. Ability to express uncoupling protein (UCP) and establish UCP-dependent thermogenesis was analyzed in anatomical areas of mice that are generally considered to be white adipose tissue: mesenterial, perimetral, epididymal, inguinal, and superficial layer of interscapular white adipose tissue. The mice were acclimatized for 1 week to 4~ C; the following week they were exposed to cold stress (1 h at - 2 0 ~ C, 2-3 times daily). In such conditions in inguinal adipose tissue, slot-blot analysis detected significant amount of UCP mRNA and lipoprotein lipase mRNA. Immuno-electron-microscopic localization of UCP showed that developed mitochondria of cold-stressed inguinal adipocytes contained UCP in the same amount as uncoupled (UC)-mitochondria of brown adipocytes. Morphological and morphometrical analysis showed that such inguinal adipose tissue appeared as brown adipose tissue. Since in control mice, inguinal adipose tissue was UCP-negative and tissue appeared as white adipose tissue, the duration of this white-to-brown adipose tissue conversion was analyzed. Mice, cold stressed for 1 week, were rewarmed at 28~ C and their inguinal adipose tissue was analyzed in comparison with interscapular brown adipose tissue and epididymal white adipose tissue for another 37 days. During that time inguinal adipocytes ceased expressing UCP mRNA; UC-mitochondria in inguinal adipocytes were destroyed and replaced with common, C-mitochondria; and UCP was undetectable immunohistochemically. Adipocytes accumulated lipids, and the tissue morphologically once again resembled white adipose tissue. Described changes showed that besides typical brown and white adipose tissue in mice, there existed a third type of adipose tissue described as convertible adipose tissue. Key words: Adipose tissue, brown - Adipose tissue, convertible - Adipose tissue, white - Adipocyte - Cold stress - Lipoprotein lipase - Mitochondria - Uncoupling protein Mouse (NMRI strain) Department of BiologyB-022, Universityof California at San Diego, La Jolla, CA 92093, USA * Present address and address f o r offprint requests:
Depending on its anatomical location and function, adipose tissue in mammals has been subdivided into two groups - white adipose tissue (WAT) and brown adipose tissue (BAT; Greenwood and Johnson 1983; Krsti6 1984; Slavin 1987). Besides the difference in structure (innervation, vascularization, size and shape of adipocytes, and the amount of lipids and mitochondria in adipocytes), the most prominent difference between these two types of adipose tissues, and their adipocytes, is in their function. While white adipocytes serve as a lipid store, brown adipocytes possess unique mitochondria that make brown adipocytes serve as a heater. Present in the mitochondrial cristae of brown adipocytes is a 32-kDa protein known as uncoupling protein (UCP; Ricquier and Bouillaud 1986). This type of mitochondria has been named UC-mitochondria, as distinct from common, coupled or C-mitochondria (Lon~ar and Afzelius 1989). During the oxidation process, protons are pumped from the mitochondrial matrix into intermembrane space. UCP in the mitochondrial cristae of brown adipocytes permits most of the protons to return to the mitochondrial matrix through UCP instead of through ADP/ATPase (Nicholls and Locke 1984; Katiyar and Shrago 1989). As a result of protons passing through the UCP-dimer, heat is released, i.e., UC-mitochondria produce mostly heat instead of ATP (Nicholls et al. 1986). Biochemical data suggest (Rafael and Heldt 1976; Ricquier and Kader 1976; Cannon et al. 1982; Lean and James 1983; Afong et al. 1985) and immuno-electronmicroscopic analysis has shown (King and Lean 1987; Lon~ar et al. 1988b; Lon~ar 1990) that of all organs and tissues examined (liver, kidney, heart, lacrimal gland, brain, skeletal muscle, smooth muscle, white adipose tissue, endothelial cells, and cells of connective tissue), UCP is expressed only in brown adipocytes. UCP presence has been accepted as the main criterion for distinguishing between brown and white adipocytes or between brown and white adipose tissue, respectively (see Cannon and Nedergaard 1985). Although the expression of UCP is influenced by several hormones (Peachey et al. 1988; Silva 1988; Geloen and Trayhurn 1990),
150 the simplest w a y to p r o v o k e expression o f U C P is to inject rodents with n o r a d r e n a l i n or to expose them to cold or to other conditions that induce the release o f noradrenalin, such as p h a e o c h r o m o c y t o m a (Lean et al. 1986a) or cigarette smoke (Yoshida et al. 1989). Previous studies with rats exposed to severe cold stress s h o w e d that a l t h o u g h epididymal adipocytes undergo d r a m a t i c changes, they c a n n o t express U C P , i.e., white adipocytes c a n n o t a d o p t U C P - d e p e n d e n t t h e r m o genic function ( L o n 6 a r et al. 1988a, b). O n the other hand, the biochemical assays as well as m y i m m u n o electron-microscopic analysis o f b r o w n adipocytes o f rats, mice, and hamsters showed that U C P is f o u n d uniformly in their b r o w n adipocyte m i t o c h o n d r i a ( L o n 6 a r 1990). Studies f r o m several laboratories have s h o w n that adipose tissue with adipocytes possessing b o t h U C P a n d U C - m i t o c h o n d r i a exist perinatally in certain a n a t o m i c a l areas o f the h u m a n (Lean et al. 1986b), calf (Casteilla et al. 1987), sheep ( T h o m p s o n et al. 1989), d o g (Ashwell et al. 1987), rabbit ( R o z o n et al. 1989), and guinea pig (Rafael and Heldt 1976). As they age, these brownlike adipocytes change structure, lose their U C - m i t o c h o n d r i a and their U C - d e p e n d e n t thermogenic capacity, and convert into whitelike adipocytes ( L o n 6 a r and Afzelius 1989; see also Hassi 1977; A s t r u p et al. 1984; A s t r u p 1986). Since only particular adipose tissues have this capability, we concluded that this type o f adipose tissue is distinct and p r o p o s e d the n a m e convertible adipose tissue ( C A T ) for this type (Lon6ar and Afzelius 1989). F o r the present study the ability to establish U C dependent thermogenesis, i.e., the ability to express U C P m R N A a n d U C P , was analyzed in the adipose tissue in six a n a t o m i c a l areas in mice exposed to cold stress. The results show that in mice, besides typical W A T (epididymal, perimetrial, mesenterial) and B A T (interscapular BAT), the inguinal areas also exhibited C A T . Adipocytes f r o m inguinal adipose tissue that in t h e r m o n e u tral conditions h a d the same appearance and function as white adipocytes, in cold stress conditions became m o r p h o l o g i c a l l y like b r o w n adipocytes, expressed U C P m R N A , a n d possessed U C - m i t o c h o n d r i a . After the cold-stressed mice were returned to a t h e r m o n e u t r a l environment, the expression o f U C P m R N A stopped, U C m i t o c h o n d r i a were destroyed a n d replaced with C-mitochondria, a n d U C P was undetectable i m m u n o h i s t o chemically. These changes, together with an accumulation o f lipids and a disappearance o f n u m e r o u s interadipocyte contacts, p r o m p t e d r e w a r m e d inguinal adipose tissue to re-appear as W A T . The described structuralfunctional changes s h o w e d that in mice, besides the typical white and b r o w n adipose tissue, convertible adipose tissue also existed.
Materials and methods
Animals Mice, 3~4 weeks old, were caged individually with free access to food and water. The control mice remained at a thermoneutral temperature (28~ C). The experimental group was kept at a con-
stant cold temperature (4~ C) for 1 week. The following week, the animals from the cold environment (4~ C) were exposed to cold stress (for 1 h at - 2 0 ~ C, two-three times per day). After 1 week of cold stress (2 weeks of cold temperature exposure), the animals were divided into five groups. The first group of cold-stressed mice were sacrificed with mice from the control group. The other four groups of cold-stressed animals were transferred at thermoneutral environment and sacrificed, one group per day, 4 days, 10 days, 17 days, and 37 days after cold stress.
Electron- and immuno-electron-microscopic preparation of adipose tissue Four animals from each group were anesthetized and transcardially perfused with fixative as described in detail earlier (Lon6ar et al. 1988 a). Male mice donated all but their perimetrial adipose tissue. During dissection of perimetral adipose tissue, contamination with perirenal adipose tissue was prevented by collecting the caudal part of perimetral adipose tissue. Interscapular brown adipose tissue and, separately, its superficial layer of white adipose tissue, mesenterial, perimetral inguinal, and epididymal adipose tissue were dissected out and prepared for electron-microscopical (Lon6ar et al. 1988a) and immuno-electron-microscopical analysis (Lon6ar 1990). Detailed morphometrical measurement, described in detail in Lon6ar et al. (1988 a), was done on interscapular brown, epididymal white, and inguinal adipose tissue.
Antibody and specifity of labeling Polyclonal rat anti-UCP antibody was prepared as described by Cannon et al. (1982). Specificity of UCP-labeling was tested as described earlier (Lon~ar 1990).
Determination of mRNA Four mice per group were sacrificed by cervical dislocation. The males' interscapular BAT (separated from a superficial layer of WAT), mesenterial, inguinal (dissected carefully from inguinal lymphatic nodes), and epididymal adipose tissue as well as the females' caudal part of the perimetral adipose tissue were excised and homogenized in guanidine extraction buffer. Total RNA was isolated according to Jacobsson et al. (1985). For the slot blots, an amount of the RNA preparation corresponding to 4 gg RNA was dissolved in 300 gl 10 x SCC/18% formaldehyde, with water to yield a total of 400 gl. After incubation for 15 rain at 65~ C, this solution was applied to a Zetaprobe filter in a Minifold slot-blot apparatus, washed with 400 gl 10 x SCC, and dried at room temperature. After prehybridization with salmon sperm DNA (Sigma) and poly A/ poly C mixture (Jacobsson et al. 1985), the filter paper was hybridized with cDNA probes, nick translated with a Bethesda Research Laboratories kit. The uncoupling protein probe utilized was the one characterized earlier by Jacobsson et al. (1985) and the lipoprotein lipase (LPL) probe was the one characterized by Kirchgessner et al. (1987). The filter papers were washed, dried, and then exposed to Kodak X-Omat AR film at -- 80~ C, and the amount of darkening evaluated with an LKB laser densitometer.
Results
Animah" and their adipose tissue after cold stress A few days after the mice were transferred to the cold e n v i r o n m e n t (4 ~ C), they built nests, where they shivered
151 Table 1. The effect of cold stress on the weight gain and amount of adipose tissue of mice. IBAT, Interscapular brown adipose tissue; ING, inguinal adipose tissue; EPID, epididymal adipose tissue
Control (28~ c) % of body weight Cold stress for 1 week (+ 4~ C and - 20~ C) % of body weight
Body weight (g)
Body gain (g)a
IBAT (g)
ING (g)
EPID (g)
30.8•
6.1•
0.14• 0.4
0.81• 2.6
0.29• 0.9
25.2•
3.9•
0.28• 1.1
0.38• 1.5
0.13• 0.5
" Body gain in period between 2 weeks of cold exposure (1 week of cold acclimatization and 1 week of cold stress)
Table 2. Summary of morphological analyses and uncoupling protein (UCP) mRNA determinations of adipose tissues in mice cold stressed for 1 week. UC, Mitochondria containing UCP; C, common mitochondria (without UCP); WAT, white adipose tissue Adipose tissue
IBAT
Superficial WAT in interscapular area
Mesenterial
Perimetrial
Inguinal
Epididymal
Morphological appearance Expression of UCP mRNA Mitochondria
Multilocular
Unilocular
Unilocular
Unilocular
Multilocular
Unilocular
+
_a
_
_
+
_
UC
C
C
C
UC
C
a Adipose tissue pooled from four animals
intensely. The mice left their nests only to satisfy metabolic requirements. After the mice had spent I h at - 2 0 ~ C, their rectal temperature dropped to an average of 34+0.6~ C compared with 37.1_+0.2~ C in control mice. Five hours after the mice were transferred from - 2 0 ~ C to + 4 ~ C, their rectal temperature increased to 36.6_+0.7 ~ C. The temperature was measured with the thermometer placed 12 m m deep in the rectum. One week of cold stress caused the mice to gain an average of 3.9_+0.6 g compared with 6.1+_0.6 g in the control mice. With a lower weight gain during the first week of cold exposure (4 ~ C) the experimental mice weighed an average of 5.6 g less than control mice (Table 1). After transcardial perfusion with Ringer solution (Lon6ar et al. 1988a), which removed blood, only inguinal adipose tissue of cold-stressed mice of all tissues analyzed became brownish, and in this way macroscopically resembled the interscapular BAT. Epididymal, mesenterial, and perimetral adipose tissue and the superficial layer of W A T in the interscapular area were white as in the control mice. Ultrastructural analysis, determination of U C P m R N A , and immuno-electron-microscopic analysis with anti-UCP antibody confirmed the macroscopic observation about the structural-functional changes in inguinal adipose tissue of cold-stressed mice (Table 2; compare Figs. 1 d and e; 2c and d). Since the results presented here and in an earlier study (Lon6ar et al. 1988b) showed that there was epididymal W A T that could not adopt UCP-dependent thermogenesis (Figs. 1 a, b; 2a, b), cold-stressed inguinal adipose tissue was analyzed further and compared with epididymal white and interscapular brown adipose tissue.
Inguinal adipose tissue of cold-stressed mice had the structure of B A T After cold stress the amount of inguinal adipose tissue dropped from 2.6% to 1.5% of total body mass. Epididymal adipose tissue dropped from 0.9% to 0.5% of body mass. Simultaneously the amount of BAT doubled and interscapular BAT increased to 1.1% total body mass (Table 1). These changes in the amount of adipose tissue correspond to changes in adipocyte size and the amount of lipids in adipocytes (Table 4). Brown adipocytes and epididymal white adipocytes showed no significant change in diameter and lipid amount, but the volume of cold-stressed inguinal adipocytes was reduced by almost half compared to the control mice (Table 4). The cellularity of adipose tissue was also changed. Semithin sections of Epon-embedded tissue showed that intralobular areas of inguinal adipose tissue of control mice contained about 4 0 % - 5 5 % of adipocyte nuclei, as did epididymal WAT. After cold stress the number of adipocyte nuclei in intralobular areas of inguinal adipose tissue became similar to that in interscapular BAT, i.e., adipocyte nuclei represented 2 0 % - 3 5 % of all nuclei only. The number of capillaries in contact with inguinal adipocytes increased, which, with decreased adipocyte surface, resulted in about 35% of adipocyte's surface being covered with endothelial cells (Fig. 1 e). Inguinal adipocytes had a multilocular appearance with a central nucleus (Fig. 1 e). Extracellular matrix was faintly present and numerous gap junctions developed between adipocytes (not shown). Cold stress increased the amount
152
153 Table 3. Effect of environmental temperature on expression of UCP mRNA and lipoprotein lipase mRNA (LPL mRNA) in adipose tissues of mice Temperature
28~ C (control)
Anatomical area
IBAT
ING
UCP mRNA LPL mRNA
1.0 1.0
0 0.3_+0.1
+4~
~ C for 1 week
EPID
IBAT
ING
EPID
IBAT
ING
EPID
0 1,2_+0.1
3.9_+0.4 1.6+_0.2
0.9+_0.2 0.9+_0.3
0 1.4_+0.1
0.9_+0.2 0.9_+0.2
0 0.4_+0.1
0 1.3-t-0.2
of mitochondria and their cristae. Inguinal adipocytes of cold-stressed mice had both the appearance and mitochondrial capacity (expressed as the surface area of mitochondrial cristae per unit volume of cytoplasm) of interscapular brown adipocytes (compare Fig. 1 e, h, Table 4).
Cold-stressed inguinal adipocytes expressed UCP mRNA Tables 2 and 3 show the effect of environmental temperature on expression of UCP m R N A . In control mice only interscapular BAT expressed UCP m R N A . Cold stress caused inguinal adipose tissue to express the U C P m R N A at a level equal to that in interscapular BAT of control mice. But, unlike in BAT, the expression of UCP m R N A in inguinal adipose tissue was only temporary. A return of cold-stressed mice to a thermoneutral environment for 37 days caused the amount of U C P m R N A to decline to the value control. At that time the expression of U C P m R N A in inguinal adipose tissue had completely ceased, and at thermoneutral conditions there was no detectable UCP m R N A in inguinal adipose tissue (Table 3). Cold stress also increased the expression of lipoprotein lipase (LPL) m R N A in all three analyzed adipose tissues (Table 3). With the return of coldstressed mice to a thermoneutral environment, the amount of L P L m R N A reached values similar to those in control mice (Table 3).
Cell and Tissue Research 9 Springer-Verlag 1991
Convertible adipose tissue in mice Dragutin Lon~ar* The Wenner-GrenInstitute, Universityof Stockholm, Stockholm,Sweden Accepted March 25, 1991
Summary. Ability to express uncoupling protein (UCP) and establish UCP-dependent thermogenesis was analyzed in anatomical areas of mice that are generally considered to be white adipose tissue: mesenterial, perimetral, epididymal, inguinal, and superficial layer of interscapular white adipose tissue. The mice were acclimatized for 1 week to 4~ C; the following week they were exposed to cold stress (1 h at - 2 0 ~ C, 2-3 times daily). In such conditions in inguinal adipose tissue, slot-blot analysis detected significant amount of UCP mRNA and lipoprotein lipase mRNA. Immuno-electron-microscopic localization of UCP showed that developed mitochondria of cold-stressed inguinal adipocytes contained UCP in the same amount as uncoupled (UC)-mitochondria of brown adipocytes. Morphological and morphometrical analysis showed that such inguinal adipose tissue appeared as brown adipose tissue. Since in control mice, inguinal adipose tissue was UCP-negative and tissue appeared as white adipose tissue, the duration of this white-to-brown adipose tissue conversion was analyzed. Mice, cold stressed for 1 week, were rewarmed at 28~ C and their inguinal adipose tissue was analyzed in comparison with interscapular brown adipose tissue and epididymal white adipose tissue for another 37 days. During that time inguinal adipocytes ceased expressing UCP mRNA; UC-mitochondria in inguinal adipocytes were destroyed and replaced with common, C-mitochondria; and UCP was undetectable immunohistochemically. Adipocytes accumulated lipids, and the tissue morphologically once again resembled white adipose tissue. Described changes showed that besides typical brown and white adipose tissue in mice, there existed a third type of adipose tissue described as convertible adipose tissue. Key words: Adipose tissue, brown - Adipose tissue, convertible - Adipose tissue, white - Adipocyte - Cold stress - Lipoprotein lipase - Mitochondria - Uncoupling protein Mouse (NMRI strain) Department of BiologyB-022, Universityof California at San Diego, La Jolla, CA 92093, USA * Present address and address f o r offprint requests:
Depending on its anatomical location and function, adipose tissue in mammals has been subdivided into two groups - white adipose tissue (WAT) and brown adipose tissue (BAT; Greenwood and Johnson 1983; Krsti6 1984; Slavin 1987). Besides the difference in structure (innervation, vascularization, size and shape of adipocytes, and the amount of lipids and mitochondria in adipocytes), the most prominent difference between these two types of adipose tissues, and their adipocytes, is in their function. While white adipocytes serve as a lipid store, brown adipocytes possess unique mitochondria that make brown adipocytes serve as a heater. Present in the mitochondrial cristae of brown adipocytes is a 32-kDa protein known as uncoupling protein (UCP; Ricquier and Bouillaud 1986). This type of mitochondria has been named UC-mitochondria, as distinct from common, coupled or C-mitochondria (Lon~ar and Afzelius 1989). During the oxidation process, protons are pumped from the mitochondrial matrix into intermembrane space. UCP in the mitochondrial cristae of brown adipocytes permits most of the protons to return to the mitochondrial matrix through UCP instead of through ADP/ATPase (Nicholls and Locke 1984; Katiyar and Shrago 1989). As a result of protons passing through the UCP-dimer, heat is released, i.e., UC-mitochondria produce mostly heat instead of ATP (Nicholls et al. 1986). Biochemical data suggest (Rafael and Heldt 1976; Ricquier and Kader 1976; Cannon et al. 1982; Lean and James 1983; Afong et al. 1985) and immuno-electronmicroscopic analysis has shown (King and Lean 1987; Lon~ar et al. 1988b; Lon~ar 1990) that of all organs and tissues examined (liver, kidney, heart, lacrimal gland, brain, skeletal muscle, smooth muscle, white adipose tissue, endothelial cells, and cells of connective tissue), UCP is expressed only in brown adipocytes. UCP presence has been accepted as the main criterion for distinguishing between brown and white adipocytes or between brown and white adipose tissue, respectively (see Cannon and Nedergaard 1985). Although the expression of UCP is influenced by several hormones (Peachey et al. 1988; Silva 1988; Geloen and Trayhurn 1990),
150 the simplest w a y to p r o v o k e expression o f U C P is to inject rodents with n o r a d r e n a l i n or to expose them to cold or to other conditions that induce the release o f noradrenalin, such as p h a e o c h r o m o c y t o m a (Lean et al. 1986a) or cigarette smoke (Yoshida et al. 1989). Previous studies with rats exposed to severe cold stress s h o w e d that a l t h o u g h epididymal adipocytes undergo d r a m a t i c changes, they c a n n o t express U C P , i.e., white adipocytes c a n n o t a d o p t U C P - d e p e n d e n t t h e r m o genic function ( L o n 6 a r et al. 1988a, b). O n the other hand, the biochemical assays as well as m y i m m u n o electron-microscopic analysis o f b r o w n adipocytes o f rats, mice, and hamsters showed that U C P is f o u n d uniformly in their b r o w n adipocyte m i t o c h o n d r i a ( L o n 6 a r 1990). Studies f r o m several laboratories have s h o w n that adipose tissue with adipocytes possessing b o t h U C P a n d U C - m i t o c h o n d r i a exist perinatally in certain a n a t o m i c a l areas o f the h u m a n (Lean et al. 1986b), calf (Casteilla et al. 1987), sheep ( T h o m p s o n et al. 1989), d o g (Ashwell et al. 1987), rabbit ( R o z o n et al. 1989), and guinea pig (Rafael and Heldt 1976). As they age, these brownlike adipocytes change structure, lose their U C - m i t o c h o n d r i a and their U C - d e p e n d e n t thermogenic capacity, and convert into whitelike adipocytes ( L o n 6 a r and Afzelius 1989; see also Hassi 1977; A s t r u p et al. 1984; A s t r u p 1986). Since only particular adipose tissues have this capability, we concluded that this type o f adipose tissue is distinct and p r o p o s e d the n a m e convertible adipose tissue ( C A T ) for this type (Lon6ar and Afzelius 1989). F o r the present study the ability to establish U C dependent thermogenesis, i.e., the ability to express U C P m R N A a n d U C P , was analyzed in the adipose tissue in six a n a t o m i c a l areas in mice exposed to cold stress. The results show that in mice, besides typical W A T (epididymal, perimetrial, mesenterial) and B A T (interscapular BAT), the inguinal areas also exhibited C A T . Adipocytes f r o m inguinal adipose tissue that in t h e r m o n e u tral conditions h a d the same appearance and function as white adipocytes, in cold stress conditions became m o r p h o l o g i c a l l y like b r o w n adipocytes, expressed U C P m R N A , a n d possessed U C - m i t o c h o n d r i a . After the cold-stressed mice were returned to a t h e r m o n e u t r a l environment, the expression o f U C P m R N A stopped, U C m i t o c h o n d r i a were destroyed a n d replaced with C-mitochondria, a n d U C P was undetectable i m m u n o h i s t o chemically. These changes, together with an accumulation o f lipids and a disappearance o f n u m e r o u s interadipocyte contacts, p r o m p t e d r e w a r m e d inguinal adipose tissue to re-appear as W A T . The described structuralfunctional changes s h o w e d that in mice, besides the typical white and b r o w n adipose tissue, convertible adipose tissue also existed.
Materials and methods
Animals Mice, 3~4 weeks old, were caged individually with free access to food and water. The control mice remained at a thermoneutral temperature (28~ C). The experimental group was kept at a con-
stant cold temperature (4~ C) for 1 week. The following week, the animals from the cold environment (4~ C) were exposed to cold stress (for 1 h at - 2 0 ~ C, two-three times per day). After 1 week of cold stress (2 weeks of cold temperature exposure), the animals were divided into five groups. The first group of cold-stressed mice were sacrificed with mice from the control group. The other four groups of cold-stressed animals were transferred at thermoneutral environment and sacrificed, one group per day, 4 days, 10 days, 17 days, and 37 days after cold stress.
Electron- and immuno-electron-microscopic preparation of adipose tissue Four animals from each group were anesthetized and transcardially perfused with fixative as described in detail earlier (Lon6ar et al. 1988 a). Male mice donated all but their perimetrial adipose tissue. During dissection of perimetral adipose tissue, contamination with perirenal adipose tissue was prevented by collecting the caudal part of perimetral adipose tissue. Interscapular brown adipose tissue and, separately, its superficial layer of white adipose tissue, mesenterial, perimetral inguinal, and epididymal adipose tissue were dissected out and prepared for electron-microscopical (Lon6ar et al. 1988a) and immuno-electron-microscopical analysis (Lon6ar 1990). Detailed morphometrical measurement, described in detail in Lon6ar et al. (1988 a), was done on interscapular brown, epididymal white, and inguinal adipose tissue.
Antibody and specifity of labeling Polyclonal rat anti-UCP antibody was prepared as described by Cannon et al. (1982). Specificity of UCP-labeling was tested as described earlier (Lon~ar 1990).
Determination of mRNA Four mice per group were sacrificed by cervical dislocation. The males' interscapular BAT (separated from a superficial layer of WAT), mesenterial, inguinal (dissected carefully from inguinal lymphatic nodes), and epididymal adipose tissue as well as the females' caudal part of the perimetral adipose tissue were excised and homogenized in guanidine extraction buffer. Total RNA was isolated according to Jacobsson et al. (1985). For the slot blots, an amount of the RNA preparation corresponding to 4 gg RNA was dissolved in 300 gl 10 x SCC/18% formaldehyde, with water to yield a total of 400 gl. After incubation for 15 rain at 65~ C, this solution was applied to a Zetaprobe filter in a Minifold slot-blot apparatus, washed with 400 gl 10 x SCC, and dried at room temperature. After prehybridization with salmon sperm DNA (Sigma) and poly A/ poly C mixture (Jacobsson et al. 1985), the filter paper was hybridized with cDNA probes, nick translated with a Bethesda Research Laboratories kit. The uncoupling protein probe utilized was the one characterized earlier by Jacobsson et al. (1985) and the lipoprotein lipase (LPL) probe was the one characterized by Kirchgessner et al. (1987). The filter papers were washed, dried, and then exposed to Kodak X-Omat AR film at -- 80~ C, and the amount of darkening evaluated with an LKB laser densitometer.
Results
Animah" and their adipose tissue after cold stress A few days after the mice were transferred to the cold e n v i r o n m e n t (4 ~ C), they built nests, where they shivered
151 Table 1. The effect of cold stress on the weight gain and amount of adipose tissue of mice. IBAT, Interscapular brown adipose tissue; ING, inguinal adipose tissue; EPID, epididymal adipose tissue
Control (28~ c) % of body weight Cold stress for 1 week (+ 4~ C and - 20~ C) % of body weight
Body weight (g)
Body gain (g)a
IBAT (g)
ING (g)
EPID (g)
30.8•
6.1•
0.14• 0.4
0.81• 2.6
0.29• 0.9
25.2•
3.9•
0.28• 1.1
0.38• 1.5
0.13• 0.5
" Body gain in period between 2 weeks of cold exposure (1 week of cold acclimatization and 1 week of cold stress)
Table 2. Summary of morphological analyses and uncoupling protein (UCP) mRNA determinations of adipose tissues in mice cold stressed for 1 week. UC, Mitochondria containing UCP; C, common mitochondria (without UCP); WAT, white adipose tissue Adipose tissue
IBAT
Superficial WAT in interscapular area
Mesenterial
Perimetrial
Inguinal
Epididymal
Morphological appearance Expression of UCP mRNA Mitochondria
Multilocular
Unilocular
Unilocular
Unilocular
Multilocular
Unilocular
+
_a
_
_
+
_
UC
C
C
C
UC
C
a Adipose tissue pooled from four animals
intensely. The mice left their nests only to satisfy metabolic requirements. After the mice had spent I h at - 2 0 ~ C, their rectal temperature dropped to an average of 34+0.6~ C compared with 37.1_+0.2~ C in control mice. Five hours after the mice were transferred from - 2 0 ~ C to + 4 ~ C, their rectal temperature increased to 36.6_+0.7 ~ C. The temperature was measured with the thermometer placed 12 m m deep in the rectum. One week of cold stress caused the mice to gain an average of 3.9_+0.6 g compared with 6.1+_0.6 g in the control mice. With a lower weight gain during the first week of cold exposure (4 ~ C) the experimental mice weighed an average of 5.6 g less than control mice (Table 1). After transcardial perfusion with Ringer solution (Lon6ar et al. 1988a), which removed blood, only inguinal adipose tissue of cold-stressed mice of all tissues analyzed became brownish, and in this way macroscopically resembled the interscapular BAT. Epididymal, mesenterial, and perimetral adipose tissue and the superficial layer of W A T in the interscapular area were white as in the control mice. Ultrastructural analysis, determination of U C P m R N A , and immuno-electron-microscopic analysis with anti-UCP antibody confirmed the macroscopic observation about the structural-functional changes in inguinal adipose tissue of cold-stressed mice (Table 2; compare Figs. 1 d and e; 2c and d). Since the results presented here and in an earlier study (Lon6ar et al. 1988b) showed that there was epididymal W A T that could not adopt UCP-dependent thermogenesis (Figs. 1 a, b; 2a, b), cold-stressed inguinal adipose tissue was analyzed further and compared with epididymal white and interscapular brown adipose tissue.
Inguinal adipose tissue of cold-stressed mice had the structure of B A T After cold stress the amount of inguinal adipose tissue dropped from 2.6% to 1.5% of total body mass. Epididymal adipose tissue dropped from 0.9% to 0.5% of body mass. Simultaneously the amount of BAT doubled and interscapular BAT increased to 1.1% total body mass (Table 1). These changes in the amount of adipose tissue correspond to changes in adipocyte size and the amount of lipids in adipocytes (Table 4). Brown adipocytes and epididymal white adipocytes showed no significant change in diameter and lipid amount, but the volume of cold-stressed inguinal adipocytes was reduced by almost half compared to the control mice (Table 4). The cellularity of adipose tissue was also changed. Semithin sections of Epon-embedded tissue showed that intralobular areas of inguinal adipose tissue of control mice contained about 4 0 % - 5 5 % of adipocyte nuclei, as did epididymal WAT. After cold stress the number of adipocyte nuclei in intralobular areas of inguinal adipose tissue became similar to that in interscapular BAT, i.e., adipocyte nuclei represented 2 0 % - 3 5 % of all nuclei only. The number of capillaries in contact with inguinal adipocytes increased, which, with decreased adipocyte surface, resulted in about 35% of adipocyte's surface being covered with endothelial cells (Fig. 1 e). Inguinal adipocytes had a multilocular appearance with a central nucleus (Fig. 1 e). Extracellular matrix was faintly present and numerous gap junctions developed between adipocytes (not shown). Cold stress increased the amount
152
153 Table 3. Effect of environmental temperature on expression of UCP mRNA and lipoprotein lipase mRNA (LPL mRNA) in adipose tissues of mice Temperature
28~ C (control)
Anatomical area
IBAT
ING
UCP mRNA LPL mRNA
1.0 1.0
0 0.3_+0.1
+4~
~ C for 1 week
EPID
IBAT
ING
EPID
IBAT
ING
EPID
0 1,2_+0.1
3.9_+0.4 1.6+_0.2
0.9+_0.2 0.9+_0.3
0 1.4_+0.1
0.9_+0.2 0.9_+0.2
0 0.4_+0.1
0 1.3-t-0.2
of mitochondria and their cristae. Inguinal adipocytes of cold-stressed mice had both the appearance and mitochondrial capacity (expressed as the surface area of mitochondrial cristae per unit volume of cytoplasm) of interscapular brown adipocytes (compare Fig. 1 e, h, Table 4).
Cold-stressed inguinal adipocytes expressed UCP mRNA Tables 2 and 3 show the effect of environmental temperature on expression of UCP m R N A . In control mice only interscapular BAT expressed UCP m R N A . Cold stress caused inguinal adipose tissue to express the U C P m R N A at a level equal to that in interscapular BAT of control mice. But, unlike in BAT, the expression of UCP m R N A in inguinal adipose tissue was only temporary. A return of cold-stressed mice to a thermoneutral environment for 37 days caused the amount of U C P m R N A to decline to the value control. At that time the expression of U C P m R N A in inguinal adipose tissue had completely ceased, and at thermoneutral conditions there was no detectable UCP m R N A in inguinal adipose tissue (Table 3). Cold stress also increased the expression of lipoprotein lipase (LPL) m R N A in all three analyzed adipose tissues (Table 3). With the return of coldstressed mice to a thermoneutral environment, the amount of L P L m R N A reached values similar to those in control mice (Table 3).
